Biomedical Ethics A Canadian Focus Third Edition 9780199022281 0199022283

OXFORD THIRD EDITION
BIOMEDICA
EllICS
A CANADIAN FOCUS
Roma
111
EDITED BY
FISHER RUSSELL BROWNE r BURKHOLDER
1.I
CCM
©40
THIRD EDITION
5\A
I L_
A CANADIAN FOCUS
EDITED BY
JOHNNA FISHER I J.S. RUSSELL

ALISTER BROWNE I LESLIE BURKHOLDER
OXFORD
UNIVERSITY PRESS
11 III I W I I ill 1111111
OXFORD
UNIVERSITY PRESS
Oxford University Press is a department of the University of Oxford.

It furthers the University's objective of excellence in research, scholarship,
and education by publishing worldwide. Oxford is a registered trade mark of
Oxford University Press in the UK and in certain other countries.
Published in Canada by
Oxford University Press
8 Sampson Mews, Suite 204,
Don Mills, Ontario M3C OHS Canada
wwwoupcanada.com
Copyright 0 Oxford University Press Canada 2018

The moral rights of the authors have been asserted
Database right Oxford University Press (maker)
First Edition published in 2009
Second Edition published in 2013
All rights reserved. No part of this publication may be reproduced, stored in

a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by licence, or under terms agreed with the appropriate reprographics
rights organization. Enquiries concerning reproduction outside the scope of the
above should be sent to the Permissions Department at the address above
or through the following url: www.oupcanada.com/permission/permission_request.php
Every effort has been made to determine and contact copyright holders.
In the case of any omissions, the publisher will be pleased to make
suitable acknowledgement in future editions.
Library and Archives Canada Cataloguing in Publication
Biomedical ethics : a Canadian focus / edited by Johnna Fisher,

J.S. Russell, Alister Browne and Leslie Burkholder. — Third edition.
Includes bibliographical references and index.

Issued in print and electronic formats.
ISBN 978-0-19-902228-1 (softcover).—ISBN 978-0-19-902229-8 (PDF)
1. Medical ethics Canada. 2. Bioethics—Canada. 3. Medical

ethics—Canada—C se studies. I. Fisher, Johnna, 1964-, editor
R724.8565 2018 174.20971 C2017-906106-2

C2017-906107-0
Cover image: Assembly/Iconica/Getty Images

Cover and interior design: Sherill Chapman
Oxford University Press is committed to our environment.

Wherever possible, our books are printed on paper which comes from
responsible sources.
Printed and bound in the United States of America
1 2 3 4-21 20 19 18
Contents
Preface xiii
1 Morality and Moral Decision-Making: A Brief Introduction 1

1.1 The Indispensability of Biomedical Ethics 1
1.2 The Case of Hassan Rasouli 2
1.3 Taking Morality Seriously 3
1.4 Normative Ethical Theories 7
1.5 Moral Reasoning in Biomedical Ethics 17
1.6 Professional Codes of Ethics and Law 19
1.7 Looking Ahead 20
1.8 Study Questions 21
1.9 Suggested Readings and Resources 22
2 Medical Decision-Making: Self-Determination and

Deciding for Others 23
2.1 Introduction 23
2.2 Determining Decision-Making Capacity 32
Standards of Competence 32
Allen E. Buchanan and Dan W. Brock
A Relational Approach to Autonomy in Health Care 35

Susan Sherwin
2.3 Informed Consent 45

The Nuts and Bolts of Obtaining Consent to Treatment 45
L.E. and F.A. Rozovsky
The Concept of Informed Consent 47

Ruth R. Faden and Tom L Beauchamp
Transparency: Informed Consent in Primary Care 51

Howard Brody
Culture, Power, and Informed Consent:The Impact of Aboriginal

Health Interpreters on Decision-Making 56
Joseph Kaufert and John O'Neil
■o i LI I I JII III Wil
® Contents
Informed Consent and Public Health 60

Onora O'Neill
2.4 Substituted Judgments 63

Enough:The Failure of the Living Will 63
Angela Fagerlin and Carl E. Schneider
Advance Directives for Resuscitation and Other Life-Saving or

Life-Sustaining Measures 76
Canadian Medical Association
2.5 Best-Interest Judgments 76

Involving Children in Medical Decisions 76
Christine Harrison, Nuala P Kenny, Mona Sidarous, and Mary Rowell
Position Statement: Treatment Decisions Regarding Infants, Children,

and Adolescents 80
Christine Harrison; Canadian Paediatric Society, Bioethics Committee
Withholding or Withdrawing Life-Sustaining Treatment in Children 81

Royal College of Paediatrics and Child Health
Deciding to Forego Life-Sustaining Treatment 82

Judicial Council, American Medical Association
Ethical Relativism in a Multicultural Society 83

Ruth Macklin
2.6 Cases 91
Case 1 Scott Starson: Refusing Treatment while Incompetent 91
Case 2 No Chemotherapy for Anael: Surrogate Refusal ofTreatment for a
Minor Child 92
Case 3 Do Everything for Mom: Advance Directives and a Surrogate's Right
to Demand Treatment 93
Case 4 Treatment for Jehovah's Witnesses: Adults and Children 94
Case 5 Markayla Sault: Caring for an Aboriginal Patient 95
2.8 Suggested Further Reading 96
3 Management of Medical Information 98

3.1 Introduction 98
3.2 Patient Access to Information 106
On the Supposed Right to Lie from Benevolent Motives 106
Immanuel Kant
Telling the Truth to Patients: A Clinical Ethics Exploration 107

David C. Thomasma
Telling Patients the Truth 110

Robert Scott Stewart
Medical Secrecy: Patients' Right of Access to Medical Records 113

Arthur Schafer
3.3 Confidentiality of Information 117

A Defense of Unqualified Medical Confidentiality 117
Kenneth Kipnis
Contents
Breaching Confidentiality 128

Dave Unger
How Should Health Data Be Used? Privacy, Secondary Use, and Big Data Sales 134
Bonnie Kaplan
3.4 Genetic Information 147

Genetic Exceptionalism and "Future Diaries": Is Genetic Information Different
from Other Medical Information? 147
Thomas H. Murray
3.5 Cases 153

Case 1 George VI's Cancer 153
Case 2 Arndt v Smith 154
Case 3 The Crash of Germanwings Flight 9525 155
Case 4 Personal Genome Mapping 156

4 Professional Ethics 159

4.1 Introduction 159
4.2 The Nature and Limits of Professional Autonomy and Professional
Responsibility 165
Patient and Physician Autonomy. Conflicting Rights and Obligations in the
Physician-Patient Relationship 165
Edmund D. Pellegrino
The Problem with Futility 176

Robert D. Truog, Allan S. Brett, and Joel Frader
Should Physicians Be Gatekeepers of Medical Resources? 181

Milton C. Weinstein
Caring in a Crisis:The Ethical Obligations of Physicians and Society during a

Pandemic 188
Why Medical Professionals Have No Moral Claim to Conscientious Objection

Accommodation in Liberal Democracies 191
Udo Schuklenk and Ricardo Smalling
The Functions and Limitations of Professional Codes of Ethics 201

Dale Beyerstein
4.3 Cases 205
Case 1 Hassan Rasouli and Medical Futility 205

Case 2 MRS in Toronto: A Duty to Care 206
Case 3 Is It Ever Right to Violate Rules of Rationing? 207
Case 4 BC Physicians "Cherry-Picking" Patients for Ease of Care, Refusing
Those in Need 207
Case 5 Helping Residents to Live at Risk 208
Case 6 When Is Treatment Futile? 209

11 I II ill 111111
Contents
5 Reproductive Ethics 211

5.2 Starting Reproduction 218
Reproductive Freedom, Autonomy, and Reproductive Rights 218
Christine Overall
Preconception Gender Selection 226

John A. Robertson
Preconception Arrangements 233

Royal Commission on New Reproductive Technologies
5.3 Stopping Reproduction 238
Why Abortion Is Immoral 238
Don Marquis
On the Moral and Legal Status of Abortion 248
Maly Anne Warren
A Defence of Abortion 253

Judith Ja►vis Thomson
A Third Way 262

L.W. Sumner
Abortion through a Feminist Ethics Lens 268
Susan Sherwin
5.4 Completing Reproduction 277

Judicial Intervention in Pregnancy and Birth 277
Royal Commission on New Reproductive Technologies
Dissent on Judicial Intervention in Pregnancy and Birth
(Royal Commission on New Reproductive Technologies) 281
Suzanne Rozell Scorsone
5.5 Cases 282
Case 1 Sex Selection in Canada 282
Case 2 Prenatal Diagnosis and Abortion or Infanticide through
Declining Treatment 282
Case 3 Abortion of Suspected Female Fetus 283
Case 4 Ms G and Refusal of Treatment while Pregnant 283
Case 5 Endangering Behaviour in a Pregnant Woman 284
Case 6 Abortion of a Fetus Due to Diagnosis of Down Syndrome 284

6 End-of-Life Decision-Making 287

6.2 Withholding or Withdrawing Life-Sustaining Treatment and
Physician-Assisted Death 295
Withholding and Withdrawal of Potentially Life-Sustaining Treatment 295
Health Law Institute, Dalhousie University
Voluntary Active Euthanasia 297

Dan W. Brock
Contents ®
Active and Passive Euthanasia 309

James Rachels
When Self-Determination Runs Amok 312

Daniel Callahan
Medical Ethics and Double Effect:The Case of Terminal Sedation 316

Joseph M. Boyle
6.3 Physician-Assisted Dying in Canada 321

Controlling the Risks of PAD 321
British Columbia Supreme Court
Legislative Background: Medical Assistance in Dying (Bill C-14) 323

Department ofJustice
Mature Minors, Mental Illness, Advance Directives, and

Conscientious Objection 328
Provincial-Territorial Expert Advisory Group on Physician-Assisted Dying and the Special Joint
Committee on Physician-Assisted Dying
6.4 The Concept of Death and Its Practical Implications 338

Defining Death 338
Alister Browne
6.5 Cases 341

Case 1 Nancy B.: Withdrawing Life-Sustaining Treatment 341
Case 2 Dr Nancy Morrison: Nonvoluntary Active Euthanasia
of an Adult 341
Case 3 Tracy Latimer: Nonvoluntary Active Euthanasia of a Minor 342
Case 4 Mr McCullough: Recommending Voluntary Passive Euthanasia 344
Case 5 Elizabeth and Eric MacDonald: Assisted Suicide 344
7 Delivery of Health Care and Resource Allocation 349

7.2 Access to Health Care in Canada 357
Sustaining Medicare:The Commission on the Future of Health Care in Canada 357
Roy Romanow
7.3 The Right to Health Care, Macroallocation, and Setting Priorities 367
The Right to a Decent Minimum of Health Care 367
Allen E Buchanan
Justice and the High Cost of Health 372

Ronald Dworkin
Why UndervaluingnStatisticar People Costs Lives 377

Tony Hope
QALYs vs DALYs vs LYs Gained: What Are the Differences, and What Difference
Do They Make for Health Care Priority Setting? 382
Bjarne Robberstad
O Contents
7.4 Microallocation: Decisions at the Bedside 391

The Allocation of Exotic Medical Life-Saving Therapy 391
Nicholas P. Rescher
Just Caring: In Defense of Limited Age-Based Healthcare Rationing 399

Leonard M. Fleck
7.5 Increasing Resources 407

Human Organs, Scarcities, and Sale: Morality Revisited 407
R.R. Kishore
7.6 Cases 412

Case 1 Optimal Care versus Cost Containment: What Is a Doctor to Do? 412
Case 2 Having an Estate Sale of One's Organs 413
Case 3 Rationing Services to an Elder Who Is Responsible for
His Medical Condition 413
Case 4 Buying a Kidney in India but Requesting Canadian After-Care 413
Case 5 Does Clifford Olson Deserve Medical Treatment? 414
8 Public Health 418

8.2 Collective Action Problems 427
The Measles and Free Riders: California's Mandatory Vaccination Law 427
Katharine Browne
Free Riding and Organ Donation 431

Walter Glannon
8.3 Screening for Disease 434

Direct-to-Consumer Genetics and Health Policy: A Worst-Case Scenario? 434
Timothy Caulfield
The Ethics of Screening: Is "Screeningitis" an Incurable Disease? 436

Darren Shickle and Ruth Chadwick
8.4 Harm Reduction Programs 443

Canada's Highest Court Unchains Injection Drug Users; Implications
for Harm Reduction as Standard of Healthcare 443
Dan Small
Virtue Ethics as an Alternative to Deontological and Consequential

Reasoning in the Harm Reduction Debate 451
Timothy Christie, Louis Groarke, and William Sweet
8.5 Inequality in Health 458

When Are Health Inequalities Unjust?—The Social Determinants of Health 458
Norman Daniels
8.6 Cases 471

Case 1 Health Care Workers and Flu Shots 471
Case 2 Students and Study-Enhancing Drugs 472
Contents G
Case 3 Breast Cancer Screening 473

Case 4 Water in First Nations Communities 474

9 Research with Humans 476

9.2 The Dark History of Human Research in Canada 485
Dr Ewen Cameron 485
Colin A. Ross
Biomedical Conflicts of Interest: A Defence of the Sequestration Thesis—

Learning from the Cases of Nancy Olivieri and David Healy 489
Arthur Schafer
9.3 Ethical Issues in the Design and Conduct of Clinical Research 495
An Ethical Framework for Biomedical Research 495
Ezekiel J. Emanuel, David Wendler, and Christine Grady
A Critique of Clinical Equipoise:Therapeutic Misconception in the

Ethics of ClinicalTrials 509
Franklin G. Miller and Howard Brody
Protecting Communities in Biomedical Research 518

Charles Weyer and Ezekiel J. Emanuel
Repairing Research Integrity 523

Sandra L Titus, James A. Wells, and Lawrence J. Rhoades
9.4 Use of Vulnerable Populations in Research 526

Vulnerability in Research and Health Care; Describing the Elephant in the
Room? 526
Sarnia A. Hurst
Best Practices for Health Research Involving Children and Adolescents:

Genetic, Pharmaceutical, and Longitudinal Studies 537
D. Avard, L Black, J. Samuel, G. Griener, and BM. Knoppers
9.5 Cases 547

Case 1 Dr Roger Poisson: Fraud in Breast Cancer Study 547
Case 2 Mr Halushka: Human Research and Harm to Participants 548
Case 3 Luka Magnotta and the Legal Recognition of Research
Confidentiality 549
Case 4 Jesse Gelsinger: Research Conflicts and Ethical Review 549

Glossary 553
Index 557
Pr ice
We are honoured to continue the work of our late colleague, Johnna Fisher, by editing the
third edition of her Biomedical Ethics: A Canadian Focus. Johnna Fisher was an outstand-
ing teacher and a respected bioethicist who, tragically, was just coming into her own as
a young academic when she died. Her editorship of the first two editions of this book
established a benchmark for biomedical ethics texts in Canada.
The current edition is substantially revised over previous editions, and this requires
some comment. We amended and extended Biomedical Ethics in response to helpful and
often copious advice from peer reviewers and also from Oxford University Press's Leah-Ann
Lymer. Each of the new editors has also brought his perspective to the book. The years
since the second edition was published have produced many important developments in
biomedical ethics in Canada that required careful attention. As a consequence, each chapter
introduction has been written anew, two chapters have been added, and 60 per cent of the
readings are new to this edition. But we have retained many of Johnna's outstanding case
studies (not surpassed anywhere, in our view) and the classic readings she had selected.
We thank all of the individuals who contributed to the revision of this book, and we hope
we have extended Johnna's work in fruitful ways.
J.S. Russell, Alister Browne, and Leslie Burkholder

May 2017
Reviewers
We wish to thank the reviewers, both named and anonymous, whose thoughtful comments
helped to shape this edition.
Ken Kirkwood, Western University
Chris Lowry, University of Waterloo
Jeff McLaughlin, Thompson Rivers University
Andrea Nicki, Simon Fraser University
Michael Poellet, University of Saskatchewan
Milly Ryan-Harshman, University of Ontario Institute of Technology
Andrew Sneddon, University of Ottawa
Curtis Sommerlatte, Concordia University
MoraIlty and MoraII
Dedision-Makring
A Brief Introduction
-1
1.1 The Indispensability of Biomedical Ethics : ,=-
Health care poses diverse, complex moral challenges that touch nearly everyone at some
time in their lives. These challenges tend also to be at the very forefront of public debate
and controversy. Euthanasia, abortion, organ donation, human experimentation, genetic
therapies, access to health care, the role of children in medical decision-making, sex selec-
tion of fetuses, and treatment of those responsible for their illness are just a few examples
of issues that are encountered in the lives of patients and health care providers and that
require the most thoughtful public engagement. They test our moral understanding at a
deep level. They also pose difficult conceptual, metaphysical, psychological, economic,
political, and sociological problems. (What is death? What is futility? What is the capacity
to consent? What can a society afford to spend on health care? Will sex selection lead to
social unrest?) There is no escape from grappling with these sorts of questions. Indeed, the
practical moral problems that we confront in health care frequently cannot be separated
from ideas, issues, and theories covering a range of disciplines. And then there are the
specifically medical questions. (What is the patient's diagnosis? Will the treatment work?
What are the risks?) All these questions place obvious, often onerous, demands on patients,
their families, health care providers, and public policy-makers.(lt explains why a discipline
of biomedical ethics has emerged and is indispensable to the proper conduct of health care.
This text attempts to provide a systematic introduction to biomedical ethics that will
educate students about the range of moral issues and problems that health care faces today.
Its focus is primarily Canadian. It takes a Canadian look at the field by including, whenever
possible, standards, cases, and articles that address biomedical moral issues and problems
in a Canadian context. As such, it will be of particular interest to anyone wanting to come
to grips with moral issues and problems that arise in health care in Canada. While the
articles and cases tend to focus primarily on health care provider-patient relations, they
also address conceptual issues in health care that will be of interest to anyone with a deeper
philosophical interest in biomedical ethics and the public policy questions it raises. -
The current chapter provides an introduction to moral theo6,) and 'oral reasorl.ng
that will help to ground an understanding of issues and debates-in biomedicarelMait is a
vital start to a proper appreciation of biomedical ethics and becoming an informed observer
and participant in related discussions and debates. Those who are new to studying morality
are sometimes perplexed and intimidated by the abstractness of some of these ideas and
un I II :,111 111111
1 Morality and Moral Decision-Making
by the diversity of apparently conflicting views. Some of this is understandable. There are
important disagreements among many of the moral theories that we will examine. These
disagreements are often reflected in the debates that are addressed in later chapters. That is
one good reason to have a sense of them, since they deepen understanding of those debates
and what is at stake in them. But another reason to begin with a treatment of moral theories
is to see the extent to which they overlap and can even be complementary. In this sense, the
diversity of theoretical views in morality is no different from that in other spheres of inquiry
where such disagreement reflects the difficulty of accurately describing and understanding
complex phenomena. Competing theories often purport to capture content that is not
covered or as well explained by other theories. In this sense, moral inquiry at more theor-
etical levels is no different from other subject areas, like economics, psychology, chemistry,
and physics. It also helps to explain why we should regard morality as a Kational sphere,
of inquiry:By this, we mean an area of discourse and investigation that is 'evidence-based,
one in which problems are identified and solutions attempted (at least provisionally) in
arguments that are based in the principles of logical reasoning and the best, most relevant
ideas, information, and theories that can currently be found. As with any sphere of rational
inquiry, we are looking for reasons, supported by the best evidence, for any claims that are
made. In stating that morality is evidence-based, we are not supposing that moral reason-
ing is strictly empirical in nature. That is a matter of dispute within philosophy. But as we
will see in the rest of this chapter, this dispute is consistent with recognizing that moral
decisions are grounded in reasons that are themselves drawn from the best evidence that
can be found to support those decisions.
We begin with a Canadian case involving end-of-life decision-making to draw out
ideas about morality and moral reasoning and to illustrate their practical relevance. Later in
the chapter, we give some further guidance about how to think of the relationship between
moral theories and biomedical ethical issues. We will also offer specific remarks about how
to approach moral decision-making in biomedical ethics and will discuss briefly the role
of professional codes of ethics and law in biomedical ethics. The chapter concludes with
some topics for discussion and a list of suggested readings and resources on moral theory
and biomedical ethics.
1.2 The Case of Hassan Rasouli

Hassan Rasouli and his family immigrated to Canada from Iran in 2010. Not long after their
arrival, Mr Rasouli was diagnosed as having a small non-cancerous tumour in his brain. The
tumour was successfully removed at a Toronto-area hospital. But complications arose following the
surgery, and Mr Rasouli had a heart attack and a serious stroke. He soon became comatose and
was put on life support. Early in 2011, he was diagnosed as being in a continuing unconscious state
(or "persistent vegetative state"), the initial step toward a diagnosis of permanent vegetative state.
The doctors believed that Mr Rasouli was unconscious and would not recover. His family disagreed.
The movements that he showed from time to time with his fingers and eyes convinced them that he
was conscious. They thought that he had shown signs of improvement and had even given them a
"thumbs up" sign, indicating his contentment with his situation. The family's Islamic religious views
held that human life is sacred, and they wanted treatment to continue.
Mr Rasouli's situation was re-examined later in 2012. It was thought that he might be "min-
imally conscious," one step above being in a persistent vegetative state. His doctors, however,
thought the results were equivocal, as was any claim that there had been an improvement in his
medical condition. They reported to Mr Rasouli's family a decision to remove his life support. The
family then applied to the courts to prevent the doctors from acting. The doctors argued in court
that life support was not conferring any medical benefit to Mr Rasouli and so was futile. As such,
neither life support nor its withdrawal constituted treatment, and so the doctors did not require the
consent of the family to remove it.
1.2 The Case of Hassan Rasouli 0
The case made its way to the Supreme Court of Canada in 2013. The Court rejected the argu-
ment that life support and its removal were not treatments.' It recognized that Mr Rasouli's consent
was required to withdraw treatment, or if his wishes were unknown as the court believed and the
doctors and the family disagreed over whether further treatment was in Mr Rasouli's best interests,
the doctors must approach Ontario's Capacity and Consent Board to resolve the dispute. To date,
Mr Rasouli's doctors have not taken this step, and he remains on life support?
1.3 Taking Morality Seriously
Is Morality Relative?
A response to difficult cases like Hassan Rasouli's by individuals who are new to studying
morality or biomedical ethics is to wonder whether their resolution ever reflects anything
more than mere opinion:, Moral judgments, it is suggested, are not backed by reasons or
evide,--ice in any genuine sense. They reflect only moral beliefs or preferences and nothing
more Philosophers describe this sort of position as(relativism about moral value. One way
of expressing this idea that is often attractive te students is that morality depends wholly
on what a culture or religion believes is right. Thus, whatever moral standards a culture
or religion believes in or endorses are morally correct, and that is all that can be said in
determining what is right. This position is commonly described as cultural relativism cultural relativism what is
(implicitly extending the notion of culture to include religious groups). Morality, in this morally right is determined by
whatever moral standards a
view, is purely conventional and arbitrary, and disagreements between cultural or religious
culture endorses.
outlooks are, ultimately, not amenable to rational evaluation and resolution.
It may be tempting to tliink-that the Rasouli case is a reflection of cultural relativism.
Rasouli's family had recently immigrated to Canada from Iran where they had been im-
mersed in Islamic culture, so perhaps the disagreement in this case simply reflects differ-
ent cultural and religious attitudes toward end-of-life decision-making. Philosophers and
ethicists generally agree that this is too superficial a way of thinking about morality. What
follows in the next few paragraphs is an explanation for this rejection of cultural relativism
and relativism about morality generally, along with the beginning of an extended argument
for taking morality seriously as a sphere of rational inquiry.
It is a striking fact about the Rasouli case that neither the doctors nor Mr Rasouli's
family claims authority for their position simply on the ground that "this is what my culture
believes." Each party states evidence for their view as part of a dialogue over what should
be accepted by all parties as relevant to resolving their dispute. Despite their disagreement
over withdrawing Mr Rasouli's life support, the parties agree on a great deal. For example,
no one wants to cause Mr Rasouli unnecessary suffering, so they share the values of not
doing harm and preventing suffering. The family believes Mr Rasouli is conscious and
content, and so keeping him alive preserves his well-being and dignity and recognizes the
sacredness of his life. It is doubtful that the doctors think that Mr Rasouli's well-being or
dignity is unimportant, so here are other values that they share. The doctors also agree with
the Rasouli family that human life is a profoundly important value. Indeed, what else would
explain the measures they took to preserve it? It is likely, too, that some of the doctors agree
that human life is sacred. But the doctors have doubts that Mr Rasouli is conscious. They
believe that even if he were conscious, it would be consciousness of a very low quality, most
likely to the point where he would have no awareness of self. If this is true, they would
perhaps understandably argue that talk about Mr Rasouli's well-being and dignity is largely
misplaced. Given their assumptions about Mr Rasouli's mental state and their belief that
continued life support is futile, the doctors also believe that treating him involves a misuse
of scarce resources that could be better used elsewhere to preserve the life, well-being, and
dignity of others. They would undoubtedly argue that even accepting that human life is
sacred unavoidably requires making difficult decisions when resources are scarce and not
every human life that requires those resources can be preserved. A difficult choice needs
11 11 I W II all 111141
O 1 Morality and Moral Decision-Making
to be made that best respects the value or sacredness of all human life. The family never
rejected this argument. But they are confident that Mr Rasouli has an awareness of self and
is the sort of person who would be quite content with even a low level of consciousness.
They also believe that the doctors are too pessimistic about his prognosis and so treatment
is not futile and resources are not being misspent or misused.
Perhaps the only thing that the parties clearly and fundamentally disagree about is
Mr Rasouli's mental status and whether it had progressed and will continue to progress.
These disagreements are not about moral values per se. They are disagreements about cer-
non-moral facts states of tain non-moral facts, principally certain psychological facts and facts about Mr Rasouli's
affairs that can be wholly prognosis. Thus, the parties to this case appear to agree on much, maybe all, of what
or adequately described
is important to making a decision. They just disagree about what the evidence is for
separately from moral values.
For example, biological,
some of the things they all agree are relevant to making a decision. Unfortunately, per-
psychological, social, haps their most important disagreement cannot be resolved in a definitive manner at
physical facts. the present time, since we do not yet have (and may never have) the tools for assessing
the psychological states of non-responsive, or mainly non-responsive, individuals. Unless
one side can make a reasonably convincing case that the evidence regarding some of
the non-moral facts favours its side, the evidence seems to support a moral tie of sorts.
In effect, the Supreme Court recognized the need for further adjudication of this evidence
by requiring the doctors to present their case for withdrawing treatment to Ontario's Con-
sent and Capacity Board.
Why Philosophers and Ethicists Are Generally Not

Convinced by Relativism
It is possible to see in this discussion why philosophers and ethicists are not impressed
by cultural relativism. Despite the evident multicultural character of the exchanges in
the Rasouli case, there is no evidence that any of the parties was merely pressing cultur-
ally or religiously specific values and opinions. In fact, there is agreement about several
familiar values, and this raises different, non-relativist questions about their status. Why
do people from different cultural backgrounds agree on them? If they are not treated
merely as personal or cultural preferences, do they have a status that is independent
of what people happen to believe about them? Are there values, then, that transcend
culture? Moreover, even a modestly careful analysis such as conducted here suggests that
much that passes as moral disagreement is not disagreement about moral value per se.
Indeed, the contested non-moral facts in the Rasouli case are not limited to whether he
is conscious. They include issues such as whether lie is experiencing suffering, what his
chances of recovery are, and conceptual questions such as what futility and treatment are.
Such non-moral disagreements can also extend to metaphysical questions about whether
a supreme being exists or not.
Recognizing the central role of non-moral facts presents a richer account of
moral reasoning, one that turns out to be particularly important in biomedical ethics.
Often, we can underestimate the extent of agreement about moral values by failing to
notice that much moral disagreement depends on disputes about non-moral facts. Indeed,
the emotional heat in moral debate is often generated by such disputes rather than by
disputes about values per se. Certainly this was true in Mr Rasouli's case. His family ada-
mantly disagreed with the doctors' assessment of his mental status and prognosis. By
contrast, people rarely debate the fundamental importance of prevention of suffering, the
promotion of well-being, the protection of human dignity, or the importance of human life.
Perhaps a cultural relativist will say this only reflects a fortunate agreement between cultures
about value. But this claim cannot just be assumed to be true. It stands in need of support,
in part because the striking nature of the shared values, apparently across cultures, invites
inquiry about whether something more can be said in defence of them that would explain
why they are widely shared.
13 Taking Morality Seriously 0
Before turning to some attempts to describe sources for these shared values in
section 1.4, it will be useful to explain other ways in which relativist ideas about morality
can be misleading and problematic. People who are attracted by relativistic ideas in ethics
(or want to see them discussed) often have the following evidence in mind:
1. The presence o f widespread disagreement in morality

2. The belief that relativism is a good theory for promoting tolerance and respect between
cultures and religions
3. Relativism is a good way to resolve the question about "who gets to decide" what is
right and what is wrong.
None of these claims represents good evidence for cultural relativism. Some of them con-
tradict the very idea of it. To begin with claim 1, the existence of disagreement, even
widespread and difficult to resolve disagreement, does not in itself offer any evidence for
relativism. Disagreement between individuals or cultures is consistent with there being
a fact of the matter that is independent of what people or cultures believe. There were
differing beliefs about the shape of the earth for millennia. There was nevertheless a fact
of the matter about this. The same may be true about difficult-to-resolve moral disagree-
ments. As well, disagreement is consistent with the evidence weighing more heavily, even
convincingly, in favour of one position over another. This is often the case in moral reason-
ing, as we shall see throughout this text. Moreover, as we have already seen, the Rasouli
case suggests ways in which moral disagreements can be overstated. It is an interesting
(and recommended) exercise to consider whether it is disagreement about value that lies
behind many apparently vexing moral controversies from past to current times. Bad biol-
ogy, bad sociology, and bad psychology probably do more to explain the denial of women's
rights and the justification of slavery over millennia than does disagreement about specific
moral values. Finally many moral disagreements are simply not worth taking seriously.
The moral, psychological, biological, and sociological evidence we have now overwhelm-
ingly favours recognizing women's rights, for example. In cultures where these rights are
not recognized, moral disagreements over women's rights cannot be taken seriously any
longer, though complex political and social problems obviously remain about what to do
in light of opposition to them.
Claims 2 and 3 contradict relativism and actually provide evidence against it. Cultural
relativism cannot be defended on the ground that it is a tolerant, respectful outlook, for tol-
erance and respect are moral values. That would mean relativism is committed to specific
moral values that apply to all cultures, contradicting the idea that it is the culture that
gets to decide what its moral values are going to be. A consistent reading of cultural rela-
tivism would have to remain neutral between tolerance and intolerance and respect and
disrespect. Indeed, far from being a tolerant outlook, relativism about moral values looks
extraordinarily bleak. It places the morality of Nazis and head-hunters on an equal status
beside the morality of peaceful multicultural democratic communities. Even the issue of
"who gets to decide" betrays a commitment to moral values, if it is motivated by concern,
as it typically is, that those who decide will improperly impose their views on others.
This also implies a general belief in the values of tolerance, open-mindedness, autonomy,
and individual choice and so contradicts relativism.
By contrast, if we take an evidence-based approach to morality, the answer to the "who
gets to decide" question is, roughly, that no one gets to decide. It is the best available evidence
that should be followed in answering questions about right and wrong. This is reflected
in the Supreme Court's insistence that the doctors in the Rasouli case take their plan to
the Ontario Consent and Capacity Board so that the evidence has the best opportunity to
be properly presented and then evaluated by experts. This is also why we need to tolerate
and encourage open debate about morality so that we can continually test moral ideas and
claims to make sure they are as well supported as possible, revising and improving them
I I In 111111
I0 1 Morality and Moral Decision-Making
as the evidence changes or is clarified. Morality, including biomedical ethics, like any other
sphere of rational inquiry, needs to be committed to principles of tolerance and fair- and
open-mindedness in order to encourage productive, ongoing discussion and evaluation of
evidence. In fact, it is ironic that cultural relativists would be concerned about the "who
gets to decide" issue, since cultural relativism itself gives an answer that seems so clearly at
odds with concern about intolerant, authoritarian imposition of moral values. What goes
in ethics, according to the cultural relativist, is just whatever a culture agrees to endorse,
including deeply oppressive moral standards.
Two Alternatives to Relativism: Subjectivism and Moral Realism

The lesson from this discussion is that the relativism often found among beginners to moral
philosophy is a superficial, almost certainly wrong account of the status of morality. Those
who believe they are relativists about morality are typically confused about what it means.
Usually, they are better described as liberals or multiculturalists rather than relativists, people
who believe in general values of tolerance, respect, and open-mindedness. The same objec-
tions apply to arguments that are given for an individually based relativism that claims
morality is just whatever moral standards an individual happens to endorse, since reasons
given for individual moral relativism tend to be the same as those given for cultural relativ-
ism. A position that seems close to an individual moral relativism is the view that moral
judgments are in some sense fundamentally expressions of human emotions. Philosophers
subjectivism morality is sometimes refer to this position as subjectivism.3 It is similar to moral relativism in claim-
fundamentally about the ing that morality is a matter of preference or opinion in some sense. But its formulation
expression of feeling or
is different from moral relativism. Moral relativism claims that morality is whatever an
emotion.
individual or culture believes is moral. Subjectivism thinks that morality is not about beliefs
per se. Rather, moral claims use language to express feelings, in particular "pro" and "con"
feelings toward certain states of affairs. "Abortion is wrong" thus expresses a negative feeling
toward abortion (something like "down with abortion!" or "bah abortion!") and encourages
others to adopt a similar attitude.
Subjectivism has had somewhat more credibility among moral philosophers, in part
because it does not commit the basic errors that are associated with the student's moral
relativism. But it is also controversial. It is far from clear that the moral decisions and values
in the Rasouli case are simply expressions of human emotion. To the extent that evidence
can be given for values reflected there (discussed further below), this does not seem a
promising position. Moreover, we seem to debate moral matters to try to get beyond our
emotions and come to agreement on values and non-moral facts that are supported by the
evidence. Thus, morality is apparently about identifying reasons for certain choices that
all should be able to accept independently of their feelings or beliefs. Subjectivists have
responses to these positions that are beyond the scope of this introduction. Suffice it to say
that philosophical subjectivists also generally agree that moral debate is an evidence-based
moral realism moral facts sphere of inquiry. It proceeds as if something like moral realism is true. Moral realism is
exist Independently of the the view that there are moral facts that exist independently of the evidence for them and
evidence for them and about
about which we have some, at least approximate, knowledge. Subjectivist views reject
which we can have at least
approximate knowledge.
moral realism, but they acknowledge that moral debate proceeds very much like evidence-
based discussions in other disciplines.4 Moral realism has evidence for it in the way that
moral debate is conducted, and it is fair to say that subjectivism struggles to accommodate
these features of moral reasoning.
It is not necessary, however, to choose between moral realism and subjectivism (and
various positions on a continuum between these outlooks) for the purposes of engaging in
the material in this text. All that is necessary is to acknowledge that moral discourse has
roughly the same sort of general shape as discourse in other areas of inquiry. This is simply
to say that, after all is said and done, we want an answer to a problem, reasons why that
answer was selected, and reasons why alternatives were not. Nothing short of this should
13 Taking Morality Seriously 10
satisfy anyone in any area of inquiry. We have already seen quite a bit of evidence for this in
the discussion in this section. We now turn to a consideration of some general theoretical
views about morality that will provide further support for this position and will suggest
further ways in which specific moral debates in biomedical ethics can be profitably framed.
1.4 Normative Ethical Theories

The discussion in the previous section focused partly on important questions about the
status of morality, in particular whether morality is merely a matter of preference or opin-
ion or feeling in some sense. These questions are part of what philosophers refer to as
metaethics. As important as metaethics is for understanding morality and moral reason- metaethics the inquiry
ing, normative ethics is the main focus of this book. Normative ethics inquires more into the nature and status of
morality.
directly and systematically into what the standards are that determine how to act morally
(or ethically) and lead a moral (or ethical) life. (The terms "morality" and "ethics" will be
used interchangeably throughout this text.) A rough definition of normative ethics, then, normative ethics the
inquiry into the standards
is that it is the inquiry into what makes right actions right or wrong. Normative ethics
that determine how to act
can be divided into general theoretical spheres, which try to identify broad theories that morally and lead a moral life.
explain the whole, or large parts, of what it is to act morally, and applied ethics, which
addresses particular moral issues and problems, sometimes by drawing on the resources of
applied ethics the
more general theories and often by drawing on moral principles found in everyday moral application of moral theories,
reasoning (such as those discussed in the Rasouli case) that the more general theories seek principles, and ideas to
to explain and justify. We will begin now by reviewing some of the major general theories specific moral problems.
of normative ethics. We shall see that these theories can help to explain and make better
sense of many of the moral ideas found in the Rasouli case, in health care generally, and in
various cultural and religious settings. Toward the end of the chapter, we discuss applied
ethics and its relationship to these more general theories.
A major division in normative ethics is between consequentialist and deontological consequentialism rightness
theories. This abstract, perhaps intimidating, jargon expresses fairly simple ideas. Con- is a function of promoting
sequentialist morality defines rightness as a function of promoting intrinsically good states good consequences.
of affairs, or good consequences, for short. Deontological morality, by contrast, claims

that rightness is not exclusively, and perhaps not at all, a matter of promoting good conse- deontology rightness is not
exclusively, and perhaps not
quences. The task for deontologists, then, is to explain what rightness is as something partly
at all, a function of promoting
or wholly independent of good consequences. Today, the most prominent consequentialist good consequences.
theory is utilitarianism, and the most prominent deontological theory is Immanuel Kant's
ethics, both of which are discussed below. Commitments to each of these positions are
suggested in the discussion of the Rasouli case. As we shall see, the parties' interest in pre-
venting suffering and preserving and promoting well-being most dearly reflects utilitarian-
type reasoning. The idea of respecting Mr Rasouli's dignity appears to be a Kantian ideal.
We will now present these theories and then consider some objections to them and their
overall relevance to biomedical ethics.
Utilitarianism
There are many formulations of utilitarianism, but the most common formulation is that utilitarianism right actions
right actions are ones that maximize overall happiness. Jeremy Bentham (1748-1832) are those that maximize
overall happiness or
was the first Western thinker to give a systematic account of utilitarianism. John Stuart
well-being.
Mill (1806-73) gives perhaps its classic formulation, or at least the most read and cited
formulation, in his essay Utilitarianism.5 Both authors defend versions of hedonistic utili-
tarianism, the view that happiness consists in pleasure and the absence of pain and what
is right is what maximizes net overall pleasure in the world. It is worth emphasizing that
this pleasure must be impartially considered. For both Bentham and Mill, in deliberating
about what we ought to do, we must take into account our own pleasure and that of others
In I I I UM 1111.0
® 1 Morality and Moral Decision-Making
and be as impartial between them as a disinterested spectator would be. Everyone counts
for one, no one for more than one, and it is only the amount of pleasure that is morally rel-
evant. Mill seems to have qualified Bentham hedonism, however, by arguing that there are
higher and lower pleasures—for example, intellectual and aesthetic pleasures as contrasted
with bodily pleasures—and that higher pleasures are to be preferred over lower pleasures.
Bentham's and Mills views are often described as "classical utilitarianism."
Utilitarianism has evolved significantly since Mill and Bentham. Philosophers have
debated whether pleasure represents a rich enough conception of happiness. Some have
argued that "enjoyment" or "satisfaction" or "preferences" are better descriptions. Others
have argued that happiness cannot be limited to psychological states and that it consists
in having certain character traits (e.g., virtues), developing distinctive human capacities
(e.g., intellectual, aesthetic, and athletic excellences), and having certain types of relations
with others (friendship and love).6 "Well-being" or "welfare" has often replaced happiness
as the end of utilitarian action. It is not clear that it is necessary to take sides in these
debates for the purposes of addressing questions in biomedical ethics. Utilitarians incor-
porate all these ideas of what contributes to happiness or well-being into their analysis of
right actions by according them intrinsic or instrumental value. These classes of value refer,
respectively, to what is valued for itself (i.e., as an end) or for the sake of something else
(i.e., as a means). From a biomedical ethical perspective, it makes sense to consider all such
values in trying to assess the well-being of patients and the consequences of actions. It is
often common in biomedical ethics to simplify the assessment of good consequences by
thinking in terms of preference satisfaction. This is imperfect, but it is a good rough guide in
ethical decision-making in health care.
Utilitarianism and Biomedical Ethics

It is easy to draw from utilitarianism a number of consequentialist principles that are rel-
non-maleficence do not evant to biomedical ethics, in particular principles of non-maleficence and beneficence.
cause harm. Non-maleficence is the idea, familiar in health care in the Hippocratic Oath, "to do no
harm." Beneficence is the idea that we should go beyond not harming and do positive
beneficence prevent harm, things to prevent harm, reduce harm, and do good. Harm prevention is the idea that it is
reduce harm, and do good. always a reason for acting if it prevents harm. Doing good is, of course, the idea that it is a
reason to act if the action brings about net positive good consequences overall.
Each of these consequentialist/utilitarian principles is reflected in the Rasouli case.
The family argued that withdrawing Mr Rasouli life support would cause him harm, thus
violating a consequentialist principle of non-maleficence. They argued that maintaining
his life .support would preserve his well-being and so would be doing good. By contrast,
the doctors expressed concern that if Mr Rasouli were in a "locked in" state (i.e., is con-
scious but cannot communicate), he may be suffering greatly, so withdrawing treatment
might prevent harm. We have already seen that settling the disagreement between these
consequentialist principles crucially requires that we make some judgments about Mr
Rasouli psychological state. This disagreement is thus not about the relevance of the
principles but whether the psychological evidence favours one over the other in assessing
Mr Rasoulfs well-being.
harm principle interference A further utilitarian principle defended by Mill (in On Liberty) is his famous harm
in the conduct of competent principle: "The only purpose for which power can be rightfully exercised over any member
persons is only justified to
of a civilized community, against his will, is to prevent harm to others."7 Mill's principle
prevent harm to others.
is widely accepted by utilitarians and others as a principle that protects individual liberty
and autonomy, since it limits the power of government and society over individual choice
to actions that cause harm to others. Actions that affect only the person performing them
should not be interfered with, even if they appear contrary to their well-being, assuming
the person is competent to make their own decisions. It was Mill's view that by balancing
individual liberty and societal interference, individuals will be secure and yet free to behave
1.4 Normative Ethical Theories ®
and lay out their existence as they see fit. Thus, happiness in society will be maximized.
We will return to a discussion of the harm principle below when we address the topic of paternalism interfering with
paternalism in biomedical ethics. someone for their own good.
Objections to Utilitarianism and Responses

It will perhaps be evident now why utilitarianism is a prominent normative ethical theory.
It reflects principles that are widely accepted and useful in thinking about moral problems.
More generally, utilitarians think that their theory aptly reflects ideas of impartiality that are
fundamental to ethical decision-making and proposes a useful, intuitively attractive way of
resolving moral disputes—that is, by weighing the consequences of actions and choosing
those that promote the most overall happiness or well-being of those affected.
However, many theorists do not find utilitarianism convincing. Perhaps the most
common and general objection to utilitarianism is that it does not adequately respect the
inherent dignity of individuals. If it would maximize happiness to enslave the talented to
work on our behalf, or have vast inequalities in wealth, or mistreat some minorities, or find
innocent persons guilty in a court of law, or harvest the organs of one unsuspecting patient
to save five, utilitarianism seems committed to saying it is our moral obligation to do so.
Another objection is that utilitarianism is too impartial and personally demanding, requir-
ing more than what morality demands. For example, does it really mean that people cannot
give any preference to promoting the well-being of intimates, family members and friends,
if happiness could be better promoted by helping others? Does utilitarianism demand that
I give up one of my kidneys if that would help others? Must I not donate the kidney to
my son if giving it to my neighbour would more effectively promote happiness? Another
objection is that morality is often not forward-looking in the way that utilitarianism im-
plies. Often, we have moral obligations because of agreements and relationships that we
entered into in the past. It might do more to improve happiness if I break my contract with
a business partner, but arguably I have a duty to keep the contract because of my earlier
agreement. Indeed, the point of a contract is to bind the parties to their previous agreement
in such circumstances.
These sorts of objections show something of the complexity of our attitudes toward
morality and the difficulty utilitarianism has in explaining that complexity. However, utili-
tarians have responses. The most common response is that a more careful analysis of each
of the objections shows it is not plausible that utilitarianism is committed to these sorts of
morally problematic positions. It is argued that a more careful analysis shows that enslaving
some, accepting great disparities in wealth, and so on would not maximize happiness.
More generally, utilitarians can say that these objections confuse the utilitarian prin-
ciple (what is right is what maximizes overall happiness) with method for making moral
decisions. Utilitarians argue that actions are right to the extent that they promote happi-
ness, but this standard does not require that we consciously deliberate over and aim to
maximize happiness with each of our choices. Rather, they argue that the most effective
method for promoting happiness, given human cognitive and emotional limitations, is
to follow a set of easy to understand and follow general rules and principles of morality
including commitments to basic rights. These should be put aside only in circumstances
when it is evident that they will not contribute to overall well-being. We should then try
to incorporate these precedents back into our understanding of a set of basic welfare-
supporting rules and principles.
A second response is for utilitarians to try to brazen it out. Rather than argue (as the
above does) that there is no unacceptable gap between our ordinary moral intuitions and
the dictates of utilitarianism, some utilitarians admit that there can be but then argue that
it is our intuitions that should give way. After all (so the argument goes), what people
ordinarily think about moral matters undeniably contains elements of prejudice, supersti-
tion, misinformation, stupidity, and bad logic. Thus, if we are sympathetic and benevolent
uii I LI II al uu
10 1 Morality and Moral Decision-Making
persons, we will adopt the Principle of Utility and (at least sometimes) revise our ordinary
moral intuitions in the light of that rather than reject that Principle because it conflicts with
those intuitions. For example, we ordinarily do think it is wrong to frame and execute an
innocent man to save more lives (by preventing a riot, say). But a utilitarian might respond
by saying that further reflection should make us reconsider. There is no morally relevant
difference in the bare distinction between doing something (framing) and letting something
happen (a disastrous riot). The only thing that counts is consequences, and in this regard,
since all the parties in question are innocent, the choice is between bad consequences for
one or for many. Is it thus so clear (the utilitarian might now ask) that utilitarianism gives
improper moral advice? And just as we might revise our first judgment here, a utilitarian
might argue that we should not be too quick to reject the theory because it conflicts with
common moral intuitions.
A third option is for utilitarians to retreat to what is called Negative Utilitarianism.
On this view, there is no obligation to maximize happiness. Our sole duty is not to harm
others or, if harm is unavoidable, not to do more harm to others than would be done by
some alternative that we could have chosen instead. Negative Utilitarianism immediately
removes the objections that utilitarianism places unacceptably heavy burdens on individ-
uals and forces them to do repugnant things to promote the general good. Some utilitar-
ians have embraced this view, and some have argued that it is Mill's view. Others, such as
Peter Singer—who is perhaps the most influential practical utilitarian of our time—have
held a view that is intermediate between maximizing and Negative Utilitarianism. Singer's
position is that if we can prevent something very bad from happening without thereby
sacrificing anything of comparable moral significance, we are wrong not to do it.8 Negative
Utilitarianism may seem not to impose strong enough obligations to help others. Maximiz-
ing utilitarianism may seem to impose excessively onerous obligations to do so. Singer
attempts to steer a middle course between these two views.
Whether utilitarianism can be defended in the above sorts of ways against the stock
objections to it is still a matter of controversy. But it must be admitted that these utilitarian
responses have plausibility, and the theory is very much alive today. Nevertheless, oppon-
ents of utilitarianism find the responses ultimately too indirect, question-begging, ad hoc,
and filled with problems of their own, including whether the responses and modifications
just considered end up collapsing into a modified deontological outlook. We will not settle
this dispute here. Suffice it to say that opponents of utilitarianism think that there are better
ways to account for the nature of morality. It is to these theories that we now turn.
Kantian Ethics
Immanuel Kant (1724-1804) was an eighteenth-century philosopher who made important
contributions to all spheres of philosophy and is generally regarded as the most import-
ant philosopher of the Enlightenment. Kant's views about morality have been particularly
influential and are arguably reflected all around us in modern systems of constitutional
democracy and human rights codes and laws. They also appear to play a prominent role
in health care and biomedical ethics. For Kant, morality is deontological because it is not
about promoting good consequences. Consequentialist moralities overlook the key aspect
of morality which Kant argues is a matter of acting with a proper moral motive or good will.
Acting with a moral motive means acting with respect toward others by recognizing their
inherent dignity. Thus, Kant's ethics are often described as an ethic of respect for the dignity
of persons. According to Kant,to act with a moral motive is to be motivated to act accord-
categorical imperative or ing to a principle that he calls the "categorical imperative."
the supreme principle of Kant calls moral duties "categorical imperatives" and contrasts them with "hypothet-
the categorical imperative ical imperatives." Understanding the latter helps us to understand the former. Hypothetical
the central concept in
Kant's ethics; given several
imperatives are things that we ought to do if we want or desire certain ends, as when one
formulations. says, "If you want to lose weight, eat less." You can get rid of the imperative "eat less" by
1.4 Normative Ethical Theories ®
simply changing your desire to lose weight. By contrast, the imperative force of categorical
imperatives cannot be removed in that way. They are directives grounded in reason refer-
ring to things that we must do whatever our desires may be. We must not exploit others
and must show gratitude to those who have helped us, for example, whether or not we
want or do not want to. This binding force is what characterizes a moral duty. They are
things that we must do on pain of acting wrongly, whatever our desires happen to be.
The categorical imperative occupies the same place in Kant's ethics as the Principle
of Utility does in Bentham's and Mill's. Kant gives several formulations of what he means
by this. Each formulation of the categorical imperative is supposed to be equivalent to
the others, though this is controversial. The basic ideas behind his notion of a categorical
imperative are reflected in two formulations:
1. The Formula of the Universal Law: Act only in accordance with that maxim through
which you can at the same time will that it become a universal law9
2. The Formula of Humanity. So act that you use humanity, whether in your own person
or that of another, always at the same time as an end, never merely as a means.10
Those new to Kant often find these formulas frustratingly difficult to understand,
but they reflect simple, reasonably compelling ideas that can be readily explained. The
first formula is designed to recognize that moral actions must be internally consistent and
follow principles that are consistently applied in relevantly similar contexts. Thus, morality
is not merely a matter of mutable emotional choices or whims or mere conventions but
is a matter of acting on reasons that guide actions in all relevantly similar circumstances.
If we make a moral choice, it must be one that is coherently expressed (for example, not
self-contradictory) and consistently observed. The first formula is thus formal, covering all
choices, and separates morality from mere emotion-driven preference or social convention.
However, the first formula gives limited specific guidance about what our moral maxims
should be. Here is where the second formula enters. It is the most distinctive and compel-
ling aspect of Kant's theory and explains why it has been attractive to many who reject
consequentialist moralities.
The second formula is generally recognized as articulating a theory of respect for per-
sons. We do this by respecting their ultimate value and thus their inherent dignity. This is
what Kant means when he says we must always treat ourselves and others as ends. Kant
argues this is reflected in the recognition that each of us has of our own ultimate value or
worth. We have a conception of ourselves as having worth "beyond price," as he puts it,
because of our capacity for reasoned choice. Of course, if we recognize this of ourselves,
it is a simple requirement of consistency that we must recognize it of others, who have a
similar conception of their worth and dignity. Thus, a universal morality emerges as an ex-
tension of an agent's recognition of her own absolute worth. But this still does not explain
how we protect and preserve the dignity of ourselves and others. Here is where the notion
of not treating persons as mere means plays a key role. According to Kant, humans must
unavoidably use each other as means to get what we want. The key, however, is to do so in a
way that respects them and preserves their dignity. Kant says we do this by not using them
as mere means—that is, as mere objects or tools. Humans are not mere objects or tools.
It is their capacity for choice or autonomy that so distinguishes them and, according to
Kant, gives them a sense of their dignity. We avoid treating persons as mere means and
respect them as ends by respecting their capacity for choice or autonomy.
Kantian Ethics and Biomedical Ethics

Kant's ethical theory may still seem fairly abstract, but the ideas are intuitive and prac-
tical nevertheless. The simple upshot of all this is that when we enter into relations with
others, we preserve their dignity by obtaining their consent before we ask them to do things.
We have seen this idea already in the priority the Supreme Court put on recognizing that
111 H i1 11111
Mr Rasouli had a right to consent to withdrawal of treatment. We can see it also in his
family's position. They believed that Mr Rasouli was consenting to continued treatment
with his "thumbs-up" gestures. They also believed that he is the sort of person who would
consent to remain in his current circumstances. The Supreme Court's recognition that even
if Mr Rasouli's preferences were unknown, its ruling that a decision should be based on
what was in his best interests reflects this idea. A Kantian would say that this should be
understood, so far as possible, as what a reasonable person in Mr Rasouli's circumstances
would accept as being in his best interests. Thus, the doctors' decision to remove treatment
might have been motivated partly by the idea that being conscious in a "locked in" state
would be a profoundly bleak circumstance that few, if any, would find tolerable, and so we
can be reasonably confident that Mr Rasouli would consent to withdrawal of treatment if
he were in such a circumstance.
Kantian ethics has been attractive to many. One key advantage is that it appears to
avoid the objections levelled at utilitarianism. (As a good exercise to see if you understand
Kantian ethics, take the description of it just given and see how it motivates many of the
above objections to utilitarianism.) It is grounded in the idea that humans are not mere
objects but are fundamentally self-determining beings. As such, the worst thing that can
happen to them—the loss of their dignity—means that they have become, or are being
used as, mere objects. It nicely explains why the loss of our mental faculties is perhaps our
most profound fear, something that patients and their health care providers will readily
appreciate. It explains why assaults, fraud, lying, emotional manipulation, exploitation,
and so on are immoral (they treat individuals as mere objects by ignoring their right to free
and informed consent) and even why we cannot simply interfere with others for their own
good without their consent as consequentialism can seem to condone.
Kantian ethics' relevance to biomedical ethics is clear, then, given the central role
that consent to treatment plays in health care decisions. Kantian ethics even seems to
explain and support the idea of democracy as government by consent of the people. More-
over, Kantian ethics does not seem as demanding as consequentialist ethics. Society is not
required to adopt policies to promote maximum happiness overall but to establish the
circumstances whereby individuals can exercise meaningful autonomy and so preserve
human dignity. A Kantian social safety net, then, would be demanding fiscally but arguably
not as demanding as a utilitarian one. And it would not be so demanding that a father
could not choose to give his kidney to his son, if that is what he wishes to do.
rights moral claims with Finally, Kantian ethics also gives substance to the idea of rights, which are claims with
special normative, apparently special normative, non-consequentialist force. For example, rights to autonomy and to
non-consequentialist, force.
consent to medical treatment, rights to freedom of expression, security of the person, and
not be discriminated against on the basis of race or sex (or other grounds) all protect us
from being used as mere means and would seem to be justified whether or not respecting
them would maximize overall happiness in every case or generally. In this sense, rights are
sometimes said to "trump" conflicting consequentialist claims based in promoting overall
well-being or happiness. Kantian ethics help to make sense of these ideas (though con-
sequentialists will respond with their own account of rights as, in effect, important goods
or benefits that are to be widely upheld).
Objections to Kantian Ethics and Responses

However, important objections to Kantian ethics have been raised. Often, Kant's theory is
criticized for its absolutism and lack of flexibility. Kant contributes to this impression in
some o f his remarks—for example, about lying and suicide. But it is fair to say that critics of
Kant often fail to capture the complexity of his views on these subjects. For example, Kant's
published views about lying are quite nuanced and do not clearly reflect the position that
is often attributed to him that one must never lie, including to "the murderer at the door"
who asks for the whereabouts of his intended yid im.il And it is important to distinguish
Kant's application of his own theory (and potential misapplication of it) from the theory
itself, which is often considerably more flexible than Kant and some of his opponents
recognize. For example, it seems to follow from Kant's theory that we may be able to lie to
the murderer at the door if it is the only reasonable option to prevent a murder. This could
be made into a universal law without contradiction under Kant's first formula. It directly
prevents another person from being treated as a mere means as per the second formula.
A Kantian might also say that the murderer cannot object to being lied to and being treated
as a mere means, since the plan to murder someone presupposes the belief and motive that
it is acceptable to use others as a mere means. More generally, it would make a sham of the
whole Kantian system of respect for the dignity of persons if people who plan to use others
as mere means were protected, and even facilitated, in carrying out their actions in ways
that undermine the very protections the theory seeks to provide.12
Another objection to Kantian ethics is the concern that its anti-consequentialism
goes too far. Surely consequences do matter in moral deliberations, but Kantian ethics
sometimes appears to reject their relevance completely. Reading Kant as thinking this
is also a misunderstanding. Kant objects only to the idea that the goal of morality is
consequentialist. He claims, rather, that it is acting with a correct motive. But for Kant,
far from being irrelevant, consequences are crucial in assessing whether a motive is mor-
ally correct or not. A Kantian must consider whether the consequences of an agent's
behaviour treat the individuals it affects as ends and not as mere means—for example,
whether consent is needed. So a Kantian can admit that a certain type and range of
consequences are relevant. What a Kantian denies is only that the ultimate goal of moral-
ity is to produce good consequences—happiness or the greatest good overall. Although
it is beyond this introduction to explain, Kant also argues that we have some duties to
promote well-being, since the promotion of well-being is necessary to establishing the
conditions for autonomous choice. A Kantian would say that to make the production of
good consequences into an absolute duty (as utilitarians do) is inappropriate for a variety
of reasons, including that it is too demanding and may be used to override the dignity
and autonomy of particular individuals.
However, there are other, arguably more serious, objections to Kantian ethics, in par-
ticular with respect to its attitudes toward non-rational beings and to suffering in general.
Kantian ethics seems to allow that we can use non-rationals--for example, incompetent
children, the mentally ill, the senile, and animals—as mere means, perhaps even using
them in cruel medical experiments if it suited us to do so. This is a frightening prospect.
However, Kantians can reasonably respond that we have responsibilities to children and
others to bring them, or return them, to autonomy, so the objection does not work for such
beings. Kantians can also say that if there are rational non-human animals, they are entitled
to respect too. But there are still the cases of the permanently mentally ill, the senile, and
non-rational animals. Their dignity and suffering do not seem to matter at all in Kantian
ethics. To be fair, Kant does say that we should treat these categories of beings humanely.
We should do this in order not to harden ourselves as a way of practising respectful behav-
iours that are necessary for our relations with autonomous persons. But to many, this is an
inadequate response. They argue that these beings also deserve respect as ends and that
causing suffering is always bad and a moral reason not to do it. These are reasonable con-
cerns about Kantian ethics. However, the objections do not undermine the Kantian insight
of the importance of respecting dignity and not using others as mere means. Rather than
undermining Kant's theory, the objections suggest a need to extend the circle of respect
further than Kant recognized. After all, these other classes are conscious beings and so are
not mere objects. Suffering is something bad that happens to them and that cannot happen
to mere objects. It might be appropriate, then, to extend an ethic of respect and recognize
that these beings also have legitimate claims based on a recognition of their dignity.
As with utilitarianism, there do not seem to be any clear, knock-down objections to
Kantian ethics. Moreover, its appeal seems to lie in attractive ideas that are not so clearly
or directly captured by utilitarian theory. We turn now to consider some other theoretical
on 1 I I :111 111111
approaches to normative ethics. These approaches are not as prominent as either utilitar-
ianism or Kantianism, but it will be useful to describe and examine them to see what they
might add to our understanding of morality.
Virtue Ethics
virtue ethics morality Is Virtue ethics reflects a long tradition in the West and elsewhere that emphasizes aspects of
principally a matter of realizing character that go into being a good person. An implication of virtue theory is that morality
in action character traits
is not so much, and perhaps not at all, a matter of observing or following the sorts of rules
(virtues) that express being a
good person.
and principles that are reflected in consequentialist and deontological moralities. When we
face a moral problem, virtue ethicists say that we should look to the example set by good
people and ask questions like "what would so-and-so do?" or "what would a good person
do?" Virtue ethics reflects another familiar idea, then—namely, that morality is concerned,
at least in part, with what makes a good person. Virtue ethicists often claim that the com-
plexity of moral questions cannot be resolved by recourse to moral rules. When we face
novel or difficult moral situations, we are often better off asking ourselves what certain
moral paragons or exemplars with whom we are familiar would do.
Virtue ethics was given its classic formulation in Western philosophy by Aristotle
(384-322 BCE), but virtue ethics theories are present in other cultures as well—for ex-
ample, in Confucian and Buddhist thought. Contemporary virtue theorists give formula-
tions that differ at least somewhat from Aristotle's. However, we focus here on Aristotle's
account, since it remains the point of departure for virtue ethics and contrasts effectively
with utilitarianism and Kantian ethics. According to Aristotle, virtues are psychological
states expressed in action that are necessary to functioning well as a human being and,
therefore, represent the dominant, necessary goods in a happy (or eudaimon)
There are two classes of virtues: intellectual and moral (or character) virtues. Both are
necessary to making correct choices. Today, we might say that intellectual virtues are those
required for applying the canons of evidence-based inquiry including virtues involved
in empirical observation and deductive and non-deductive reasoning. They would also
include virtues like fair-mindedness in the evaluation of evidence, willingness to consider
opposing ideas, and ability to be self-critical and change one's position in light of new
evidence or better understanding of existing evidence.
Moral virtues, by contrast, are expressed in actions that reflect the training and ha-
bituation of emotions in ways that conform with and support correct action. As such,
character virtues reflect a mean between excess and deficiency of feeling. The mean is
not arithmetical. The appropriate mean is always contextual, but there can be excess or
deficiency of feeling in any context. The mean reflects the idea that correct action is threat-
ened by having too much or too little of a relevant feeling. For example, courage cannot
reflect too much or too little fear of danger. Too little fear will lead to foolhardy actions,
and too much fear will lead to cowardice. Thus, character virtues exist as a mean between
vices reflecting excess and deficiency of feeling. Training of character virtues by practice
develops habits so that we are able to do the right thing at the right time with the correct
motive. Courage, temperance, friendliness, truthfulness, wit, generosity are a few examples
of character virtues discussed by Aristotle, but it is easy to think of many others (e.g., resili-
ence, patience, persistence, compassion, fidelity, tolerance).
It is pretty obvious that all of the character virtues just described, and others be-
sides, play a central role in making good doctors, nurses, and other health care providers
and should be cultivated as part of their training. It is less clear how useful they are as
decision-making tools in many circumstances, however. For example, it is difficult to see
how asking "what would a good doctor do?" could resolve the dispute in the Rasouli case,
for there is little reason to think that the doctors or any of the parties in this case lacked
virtue in any significant measure. Remember that the Supreme Court did not say that the
doctors' wish to withdraw treatment was wrong but that it needed to be justified before
a legal tribunal that had been set up to adjudicate disagreements between doctors and
patients. We saw, also, that the dispute might be resolved if we could answer some key
disputed questions about non-moral facts. It is thus not clear how asking what a good
person would do could resolve this question or determine what should be done in the
absence of an answer to what the relevant non-moral facts were. Moreover, asking what
a good doctor would do seems to evolve quickly into a consideration of principles of the
sort already discussed: good doctors act to prevent harm, promote well-being, and respect
autonomy and dignity. Finally, we often want to know in difficult situations what it is that
makes a right action right in order to understand our choices and to help guide us in
similar situations. Simply asking what a good person would do without relying on such
principles does not give us that clarity.
Many virtue theorists accept these sorts of criticisms. They argue that Aristotle is,
nevertheless, correct that there is practical value in training moral and intellectual virtues.
These help us to "tune" our psychological capacities in ways that can assist us in recogniz-
ing moral problems and relevant information and therefore in making better moral deci-
sions. And Aristotle may also be right that the virtuous person has the dominant necessary
components for leading a happy life, assuming (along with Aristotle) that certain other
goods are also present, including material means, friends, a good education, and a prosper-
ous, stable social environment.
A final note about virtue ethics and the Rasouli case: Aristotle thought that virtues do
not consist strictly in possessing certain psychological dispositions but are states expressed
in action. Thus, he says, we would not consider someone who slept his whole life virtuous
or happy, even if he had the requisite virtuous dispositions.'4 If the virtues are the domin-
ant necessary parts of a happy life, we might imagine Aristotle arguing that Mr Rasouli
could not be said to be a virtuous person or to have a meaningful, let alone happy, life if
his consciousness was compromised and he was unable to act. Mr Rasouli's family appears
to disagree with Aristotle, as would most utilitarians and Kantians. They would say that if
Mr Rasouli is content in his circumstances, or if he chooses to live this way without violat-
ing the dignity of others, his life has meaningful value.
Care Ethics
Care ethics is a recent development in normative ethics. It emerged initially from the care ethics morality
empirical work of Harvard psychologist Carol Gilligan in the 1980s." Gilligan's exam- is fundamentally about
caring, fostering positive
ination of adolescent boys' and girls' responses to moral dilemmas suggested that the
relationships, and related
boys' solutions tended to reflect traditional rule or principle-oriented approaches to concepts.
ethical theory, such as those found in utilitarian and, especially, Kantian ethics. The
girls, by contrast, seemed to approach the dilemmas "in a different voice." Gilligan found
that they took a more contextual approach that prized the preservation of relationships
and caring. Gilligan claimed that the boys' behaviour reflected traditional psychological
views of moral development, which give priority to principles and logical reasoning
and tend to systematically overlook the different, caring ethical attitudes reflected in
the girls' approaches to moral decision-making. It is fair to say that Gilligan's empirical
findings have been controversial. Nevertheless, they led to attempts to develop a norma-
tive ethic of care, either as a stand-alone normative ethical theory or as a supplement
to existing normative theories. The main features of the theory are evolving and so
difficult to summarize. We can do no better than give the succinct account offered by
Johnna Fisher, the late author of this chapter in previous editions of this text, herself a
care ethicist, who described care ethics as committed to two basic principles (given in
order of importance):
1. To minimize or avoid harm

2. To create, maintain, and protect positive relations
I I III 111111
G 1 Morality and Moral Decision-Making
Also reflected in these principles is care ethicists' scepticism about the commitments
to impartiality that are elements of utilitarian and Kantian ethics. They argue that moral
decision-making is more contextual than these theories allow and that they contain in-
adequate provision for the protection of the vulnerable and for the fostering of positive
relations. We have seen already that there appears to be some truth to these criticisms of
Kantian ethics, although a path forward for addressing them may be possible. It is difficult
to see how utilitarian ethics systematically overlooks either of the principles of care, in light
of the evident utility involved in minimizing and avoiding harm and fostering positive,
caring relations. And both theories are more contextually sensitive than their critics often
admit. But there is, nevertheless, reason to think that the importance of relations of caring
have not received their due in normative ethical theory generally, and care ethics has at
least helped us to develop a more nuanced understanding of caring in moral deliberations.
Care ethics is also sometimes seen as an extension of virtue ethics. Caring is pretty ob-
viously a virtue, but again it has not been given any notice in traditional virtue theory. Care
ethics is also often taken as a critique of patriarchal biases found in traditional normative
ethics. It emphasizes the protection of the vulnerable and repairing the status of those who
are exploited or discriminated against (e.g., women and some minorities). It is understand-
ably taken as a foundation for feminist ethics, although many care ethicists have resisted
the label of a feminist ethic. Care ethics has also been attractive to those working in health
care. Nurses in particular have taken an interest in the theory.
Critics of care ethics are generally concerned that its focus as a normative ethical
theory is too narrow. The protection of the vulnerable and fostering positive caring rela-
tions does not tell us how to deal with many ethical conflicts. For example, it is not clear
what assistance care ethics can offer in the Rasouli case. Everyone involved agrees that a
caring approach to Mr Rasouli is necessary, including preservation of the positive rela-
tions he has with his family and health care providers. But what underlies taking a caring
approach to Mr Rasouli appears to be concern about his well-being and respect for his
autonomy, his dignity, and his family's preferences. There is also an obvious issue about just
allocation of scarce resources. We have a duty to care for Mr Rasouli, but we also have a
duty to care for others who are vulnerable and who need the scarce resources Mr Rasouli is
using. Principles such as preserving caring relations, minimizing suffering, and protecting
the vulnerable are of little help in resolving these conflicts because they apply to all parties
who need the resources. If so, we may need to look outside care ethics to find solutions to
such problems.
Moreover, it is hardly clear that utilitarian and Kantian ethics are as bloodlessly impar-
tial as care ethicists sometimes suppose. A utilitarian can defend a mother who chooses to
rescue her own drowning child over rescuing another child. There is obviously substantial
utility in preserving caring maternal relationships. It is essential to promoting the well-being
of children and establishing important familial relations of trust that dependent children
need to rely on. As well, both families would be emotionally devastated by the loss of a
child, but a mother who does not choose to rescue her own child would have to live with
the added psychological disutility of choosing to let her own child die, a disutility she
would not experience by letting an unrelated child die. A Kantian could say that the mother
was free to act on her preference for saving her own child, since making this choice does
not use anyone as a mere means or violate anyone else's rights. Perhaps a Kantian would
also say that the choice to engage in acts of procreation brings with it duties others do not
have to nurture offspring to autonomous adult personhood. Thus, a Kantian could say that
the mother has a special prior, autonomously chosen obligation to care for her child, an
obligation that she does not have toward other nearby children who are also in danger.
Care ethicists have responded to criticisms by recognizing the importance of prin-
ciples of justice, equality, impartiality, and autonomy. This presses the issue as to whether
the theory is distinctive or only shows how other theories have overlooked important
elements of morality that should be incorporated within them. But even if care ethics must
1.4 Normative Ethical Theories e
rely on principles imported from outside, it emphasizes something standard normative

ethics does not—namely, the importance of trust and relationships. To be sure, Mill and
Kant envisioned ethics as a practical discipline designed to prescribe or prevent certain
actions, but they did not have the particular problems of clinical ethics at the centre of
their thought. Care ethicists do—the theory was invented to better handle the kinds of
conflicts that routinely occur in these contexts—and one of their enduring contributions
is to emphasize the importance of forming and fostering relationships of trust between the
parties in question. Once this is done, conversation and communication will be better, and
negotiation over irreconcilable goals can take place with greater prospects of success rather
than deteriorating into the ultimatums we find in the Rasouli case.
Caring can, then, be crucial to productive moral decision-making and inquiry.
Of course, the best efforts to produce the appropriate relationships may fail—this may
indeed have happened in the Rasouli case—but the importance of these relationships in
clinical ethics is emphasized in care ethics as it is not in any other theory. Again, we do
not have to settle disagreements about the status of care ethics to do work in the area of
biomedical ethics. As with the other normative theories we have considered, we may draw
on the insights offered by care ethics as they appear relevant to moral decision-making.
1.5 Moral Reasoning in Biomedical Ethics
The Principles of Biomedical Ethics

A now-standard approach in biomedical ethics is to begin moral deliberation by recogniz-
ing the role of the following four principles of biomedical ethics'° in identifying and principles of biomedical
deliberating about ethical problems in health care: ethics non-maleficence,
beneficence, autonomy, and
justice.
1. Non-maleficence (do not cause harm)
2. Beneficence (do good, prevent harm)
3. Autonomy (respect preferences)
4. Justice (treat like cases alike; be fair)
These principles are all suggested in the normative ethical theories we have reviewed.
Non-maleficence and beneficence are often said to be consequentialist or utilitarian prin-
ciples, whereas autonomy and justice are Kantian principles, but in fact all normative ethical
theories sign on to these ideas in some way or another. They are prima facie principles.
That is, they are not absolute requirements but represent reasons for action that hold in the
absence of countervailing reasons. Thus, for example, to say that we ought prima facie to
respect the autonomy of patients does not mean that we must always do what patients want
but rather that we must act in accordance with their wishes unless we have a good reason
not to (e.g., that doing so will put others at risk or treat them unfairly). Prima facie duties
thus shift the onus onto those who do not want to carry them out to justify overriding them.
This is consistent with a recognition that there is no agreed-upon theory of normative ethics
but that there are, nevertheless, principles that can be drawn from these theories, are re-
flected in everyday moral deliberations, and are generally agreed upon in biomedical ethics.
These so-called "four principles" may not be the only principles relevant to biomedical
ethics, but their generality and the fact that they are recognized in theory and practice
make them a useful starting point for thinking through moral problems. We have seen
that these principles must be supplemented by recognizing the fundamental role that non-
moral facts play in ethical decision-making. In health care, these facts include clinical
facts (diagnoses, prognoses, treatments, risks, and benefits); quality of life/death with and
without treatment; the preferences of patients, their families, and health care providers;
and any other relevant contextual facts (such as costs, what the law requires, and who else's
interests are affected)."
YY I YI
I I .111 our
This suggests a simple methodology for thinking about moral issues and problems in
health care:
1. There is no moral problem if the non-moral facts are clear and the principles point to
the same course of action.
2. There is a moral problem only i f a) there are conflicts between the principles or within
them, or b) there is uncertainty about the meaning or relevance of the principles, or c)
the non-moral facts are unclear.
The four-principles approach to moral problems sketched above provides a way of finding
the best evidence, and so the weightiest reasons, for resolving a dilemma or problem in
clinical or health care contexts generally. Undoubtedly, moral deliberation is more compli-
cated than this description suggests. Indeed, the principles of biomedical ethics are often
criticized as being abstract and incomplete. But their relevance and power to assist with
moral decision-making cannot be readily denied, and we have seen that their role in nor-
mative ethics is generally accepted or implied in every theory we have examined.
Ranking Principles
As a way of enhancing the use of the above methodology, a limited ranking of these princi-
ples is generally accepted in health care ethics and law. As we saw earlier, paternalism is the
concept of interfering with someone's liberty for their own good. It thus recognizes a pos-
weak paternalism it is not sible conflict between beneficence and autonomy. In ethics and law, weak paternalism is
permissible to interfere with a a recognized principle that gives important priority to autonomy over beneficence. It is the
competent person's informed
view that individuals who have the autonomous capacity to consent (i.e., can understand
decision to accept or refuse an
offered treatment.
the nature and consequences of their decisions, are fully informed, and consent voluntarily)
have an absolute right to accept or refuse an offered treatment. That is to say no treatment
can be imposed on them against their will. Autonomy trumps beneficence in these situa-
tions. We have seen this position defended in Kantian ethics and in utilitarian ethics in the
form of Mill's harm principle. Kantian ethics says to treat an autonomous person against
their will is to use them as a mere means, which we must never do. Mill's harm principle
offers a different justification. Mill's view is that autonomous people generally know their
own interests best. He also believed that people are by nature meddlesome and will inter-
fere gratuitously in the lives of others if given the opportunity. From these considerations,
strong paternalism it is Mill defends weak paternalism and (like Kant) opposed strong paternalism or the view
permissible to interfere with that we can interfere with either non-autonomous or autonomous persons against their will
a competent or incompetent
any time it is in their best interests.
person's decision about
treatment if it is in their best
Weak paternalism is usually held to have further implications for interfering with non-
interest. autonomous persons against their will. In fact, it is only with these implications that the
notion of paternalism, or interfering with someone for their own good, really applies to
weak paternalism. Specifically, interfering with non-autonomous or incompetent persons
is permissible only if:
1. The preferences of the individual while autonomous are not known.

2. The individual is at risk of harming themselves.
3. We can intervene without causing more harm than is prevented.
4. We choose the least intrusive measures available, including those involving the least
restrictions on the autonomy and liberty of the individual.
5. Our action is not discriminatory.
6. All reasonable efforts are taken to bring the person interfered with to see that the inter-
ference is justified. (While acknowledging that interference can take place without
such agreement, Mill makes the important point that it should not unless it is a case of
utmost necessity because to impose it always "partakes of the degradation of slavery.")
1.5 Moral Reasoning in Biomedical Ethics 19
The conditions weak paternalism establishes on interfering with non-autonomous in-

dividuals are designed to prevent meddlesome behaviour that is not in their interests and
to recognize that even non-autonomous persons are entitled to some respect for the choices
they may make. It is, however, often a hard principle to live by, for autonomous individuals
may refuse potentially life-saving interventions. But weak paternalism is generally accepted
in law and ethics today.
By contrast, strong paternalism is often tempting. It is difficult to stand by when prevent-
able tragedies are occurring as weak paternalism requires (for example, when a competent
Jehovah's Witness refuses a life-saving blood transfusion). Strong paternalist interventions
are sometimes justified on the grounds that patients will later thank the health care provid-
ers for their actions. It is important to see, however, that this defence of strong paternalism
makes it ambiguously distinct from weak paternalism, or a watered-down version of it.
It implies that patients do not truly appreciate their current choice and that they would
hypothetically consent to an intervention if they understood their situation better, since
they would later thank their health care providers. In effect, then, this version of strong
paternalism challenges the patienth capacity to make an autonomous decision. The problem
with such justification is that it is very difficult to know what people will accept later. Kant-
ians and utilitarians would say that it is a violation of dignity and simply presumptuous to
believe that others know an individual's best interests better than they do. It is not unreason-
able, then, to believe that thinking about respect for autonomy in this way—as hypothetical
consent to what others conscientiously believe is in a person's best interests—rationalizes
all manner of meddling in others' lives, which Kantian and utilitarian ethics guard against.
Arguably, it is for these reasons that weak paternalism has won out in mainstream biomed-
ical ethics—it is enshrined in every professional code of health care ethics in Canada—and
in Canadian law, though debates continue in philosophy journals and in hospital hallways.
1.6 Professional Codes of Ethics and Law
Students of biomedical ethics should be aware of and study the codes of ethics of the health
professions. These codes can further understanding of the ethical issues involved in health
care and can supplement the methodology just described. Health care professions have
also developed policy statements on specific health care issues that are ethically informative
and need to be consulted where applicable. For example, the Canadian Medical Associa-
tion has policies on end-of-life decision-making, abortion, physicians' obligations during
pandemics, protection of patient confidentiality, conflicts of interest with the pharma- conflict of interest
ceutical industry allocation of scarce resources, and research ethics, to name a few.'8 But it a situation in which a person
has an obligation to a person
is important to recognize the substantial limitations of ethics codes and even of carefully
or institution that is in
elaborated issue-specific policies for moral decision-making. opposition to another interest
To begin with, it is a misconception to suppose that ethics codes and policies can or obligation, which could
supply ready answers to all or even most genuine ethical problems.19 This is already im- corrupt the decision-making
plied in the discussion so far. For example, codes of ethics do not tell us how to resolve of that person.
conflicts among professional obligations. How does a doctor reconcile a patient like Mr Ra-
souli's right of access to health care with the competing claims of other patients who need
access to the resources he is using? What should be done when rights to confidentiality
conflict with duties to care for other patients—for example, when a genetic test discloses
information that may affect other family members? Also, codes of ethics do not tell us what
the relevant non-moral facts are in a case or how to resolve disputes about them. Nor do
codes of ethics answer conceptual questions that may be at the core of a disagreement—for
example, about what constitutes death or a futile treatment or confidentiality or harm.
Ethics codes and policies can help to guide decision-making to the extent that they
emphasize recognized standards of professional conduct that must be accounted for in
health care decision-making. But generally, the ethical heavy lifting happens outside of
I I I HI MIA
1 Morality and Moral Decision-Making
these codes when principles or policies conflict, conceptual problems need to be addressed,
or there is dispute or uncertainty about the contextual features of a situation. These issues
arise to some degree or other whenever a genuine ethical problem is identified that must
be worked through.
Like morality law is intended to guide human action. In any civilized community, law
is designed to follow morality. As a result, courts' reasons for judgments can often read
like the moral arguments they are, and they are often a rich source of ethical guidance. For
example, we will see in later chapters that the courts have offered helpful moral guidance
on issues regarding consent to treatment and end-of-life decision-making, in some cases
articulating novel moral arguments (e.g., the Supreme Court of Canada's recent decision
to legalize physician-assisted dying, discussed in Chapter 6). But law has its limitations
as an ethical tool. To begin with, if we are making moral decisions, we should gener-
ally consider the relevant moral features of a case first and what the law requires later.
This is because moral decision-making is about making moral decisions, not legal ones.
Sometimes law and morality conflict, and we have to be aware of this possibility and ask
what is to be done. Can the disagreement be managed in some way? Should a test case be
tried? Should efforts be made to change legislation? Should the law be ignored? Questions
like these often pose their own ethical problems and dilemmas. They may be obscured
if deliberation starts with a consideration of what is legally required. More generally, we
cannot suppose that what is legally correct is necessarily morally correct. Nor can we sup-
pose that what is morally correct requires a law to enforce it or that what is morally wrong
requires a law to prohibit it. These are all reasons to be careful not to confuse moral and
legal deliberations.
It is instructive to note that the law in modern democracies is restricted with respect
to imposing on individuals religious or other conceptions of what gives life meaning or
what life goals should be pursued. This is, once more, a reflection of respect for autonomy
and also the product of a hard-won recognition that social stability is best achieved when
people and governments are not permitted to impose on others their views about how
to live well. These considerations, which can be drawn from the ethical theories we have
reviewed, support and accommodate a multiculturalism that is often taken to be a hallmark
of Canadian democracy and was reflected in the Rasouli case. Arguably, the role of gov-
ernment in modern, pluralist, multicultural societies is to ensure that all-purpose means,
primary goods general sometimes called primary goods, like basic democratic rights, freedoms, and opportun-
all-purpose means, such ities, are fairly distributed to people so that they can make their own choices by their own
as liberty, opportunities,
lights about how to live, as long as those choices do not interfere with the rights of others
education, and health,
necessary for the pursuit of
to these general all-purpose means or primary goods. A similar point applies to decisions in
life goals. biomedical ethics. Health is one of those all-purpose means that modern democracies have
recognized a duty to protect and fairly promote. The goal of health care, then, should be
its provision. It should not be used as an occasion to impose on patients religious or other
conceptions about what choices or ends make a life worth living.
1.7 Looking Ahead
We hope that by now it is evident that biomedical ethics is an important, challenging, and,
indeed, exciting sphere of inquiry. It demands the thoughtful attention of anyone with a
serious interest in health care. Most of the ideas expressed in this introductory chapter will
be reflected on and developed in later chapters. In the chapter introductions dealing with
specific topics in biomedical ethics, we draw connections to the ideas and theories outlined
in this chapter to further clarify their relevance and to deepen understanding of the topics
covered. The text is designed in such a way that it is possible for a reader to examine the
chapter topics in no particular order. Each chapter contains a set of study questions, a
selection of actual cases that illustrate the ideas of the chapter in their natural habitat, and
1.7 Looking Ahead ®
a list of suggested readings. Although we have not attempted to cover every topical issue in
biomedical ethics, we hope that we have provided a solid general introduction to the field
that will encourage further interest, investigation, and study.
Notes
1. Cuthbertson v. Rasouli, 2013 5CC 53, https://scc-csc.lexum.com/scc-csc/scc-csden/item/13290/index.
do (accessed 10 May 2016).
2. The case study draws on material in Philip C. Hebert, Good Medicine: The Art of Ethical Care in
Canada (Toronto: Doubleday Canada, 2016), ch. vi.
3. James Rachels and Stuart Rachels, The Elements of Moral Philosophy, 8th edn (McGraw-Hill
Education, 2014).
4. Simon Blackburn, Essays in Quasi-Realism (Oxford University Press, 1993).
5. J.S. Mill, Utilitarianism, 1863, http://wwwutilitarianism.com/millhtm (accessed 10 May 2016).
6. David 0. Brink, Moral Realism and the Foundations of Ethics (Cambridge: Cambridge University Press,
1989), ch. 8.
7. J.S. Mill, On Liberty, 1859, http://www.utilitarianism.com/ol/one.html (accessed 10 May 2016).
8. Peter Singer, "Famine, Affluence, and Morality," in Steven M. Cahn and Peter Markie, eds, Ethics:
History, Theory, and Contemporary Issues (New York and Oxford: Oxford University Press, 1998), 800.
9. Immanuel Kant, "Groundwork of the Metaphysics of Morals," translated by Mary J. Gregor, in
The Cambridge Edition of the Works of Immanuel Kant: Practical Philosophy (Cambridge: Cambridge
University Press, 1999), 73.
10. Ibid., 80.
11. Allen W Wood, Kantian Ethics (Cambridge: Cambridge University Press, 2008), ch. 14.
12. Tamar Schapiro, "Compliance, Complicity, and the Nature of Nonideal Conditions," The Journal of
Philosophy 100, no. 7 (2003): 329-55.
13. Aristotle, Nicomachean Ethics, 2nd edn, translated by Terence Irwin (Indianapolis: Hackett, 1999).
14. Ibid., 1098b35-1099a7.
15. Carol Gilligan, In a Different Voice: Psychological Theory and Women's Development, revised edn
(Cambridge, MA: Harvard University Press, 1993).
16. Tom L. Beauchamp and James F Childress, Principles of Biomedical Ethics, 7th edn (Oxford University
Press, 2012).
17. Albert Jonsen, Mark Siegler, and William J. Winsdale, Clinical Ethics: A Practical Approach to Ethical
Decisions in Clinical Medicine, 8th edn (McGraw-Hill, 2010).
18. Canadian Medical Association, Issue-Specific Policies, https://www.cma.ca/En/Pages/issue-specific-
policies.aspx (accessed 9 May 2016).
19. Dale Beyerstein, "The Functions and Limitations of Professional Codes of Ethics," in Earl R. Winkler
and Jerrold R. Coombs, eds, Applied Ethics: A Reader (Cambridge, MA: Blackwells, 1992), 416-25.
1.8 Study Questions
1. How does the material in this chapter support the claim that morality is a rational
sphere of inquiry?
2. Explain why there is less fundamental disagreement about moral values than it would
appear from observing moral debates.
3. To what extent d o the normative ethical theories discussed overlap, and to what extent
do they diverge?
4. Take the methodology from section 1.5, and use it to conduct a formal analysis of
the Hassan Rasouli case in section 1.2. Give an argument (i.e., present the evidence)
for how you would resolve the disagreement between the doctors and Mr Rasouli's
family based on what you know about the case. What are the limits, if any, to this
methodology?
5. Look at an ethics code from one of the health care professions (readily available
online). Identify a list of concepts that require clarification in order to apply the code.
Try to clarify the meaning of those concepts.
11 111 i m II A 111111
e 1 Morality and Moral Decision-Making
6. Make a list of moral controversies in biomedical ethics and elsewhere, and identify
what the disputed non-moral facts are in each case. How many of these controversies
depend most crucially on the resolution of these disputes? Which ones do not?
7. How do the arguments given in section 1.3 show that respect for multiculturalism is
not a moral relativist position? How is such respect for multiculturalism evident in the
practice of health care in Canada?
1.9 Suggested Readings and Resources
General Introductions to Moral Theory

Pojman, Louis, and James Fieser, Ethics: Discovering Right and Wrong, 7th edn (Wadsworth
Publishing, 2012).
Rachels, James, and Stuart Rachels, The Elements of Moral Philosophy, 8th edn (McGraw-Hill
Education, 2014).
Schafer-Landau, Russ. Whatever Happened to Good and Evil? (Oxford University Press,
2004).
Classic Texts in Normative Ethics

Aristotle, Nicomachean Ethics, 2nd edn, translated by Terence Irwin (Indianapolis
Hackett, 1999).
Held, Virginia, The Ethics of Care: Personal, Political, Global (Oxford University Press, 2007).
Kant, Immanuel. "Groundwork of the Metaphysics of Morals," translated by Mary J. Gregor,
in The Cambridge Edition of the Works of Immanuel Kant: Practical Philosophy (Cambridge
University Press, 1999).
Mill, J. S . On Liberty, 1859, http://www.utilitarianism.com/ol/one.html.
Utilitarianism, 1863, http://www.utilitarianism.com/mill1.htm.
Singer, Peter, Practical Ethics, 3rd edn (Cambridge University Press, 2011).
Canadian Biomedical Ethics Books and Resources

Bioethics for Clinicians, Canadian Medical Association, http://www.cmaj.ca/site/misc/
bioethics_e.xhtml.
Canadian Bioethics Society, https://www.bioethics.ca.
Hebert, Philip C. Good Medicine: The Art of Ethical Care in Canada (Doubleday Canada, 2016).
University o f Toronto Joint Centre for Bioethics, http://jcb.utoronto.ca.
Yeo, Michael, Anne Moorehouse, Pamela Kahn, and Patricia Rodney, eds, Concepts and
Cases in Nursing Ethics, 3rd edn (Broadview Press, 2010).
Leading Biomedical Ethics Journals

Bioethics
Clinical Ethics
Cambridge Quarterly of Healthcare Ethics
Hastings Centre Report
IRB: Ethics and Human Research
Journal of Clinical Ethics
Journal of Medical Ethics
Nursing Ethics
Medical Decision-Making
Self-Determination and Deciding for Others
2.1 Introduction
For more than 2000 years, health care was dominated by the Hippocratic tradition. Ac-
cording to this tradition, health care providers had an obligation to act in the best interest
of patients. This meant that if what the patient wanted conflicted with what was good
for the patient, patient preferences were overridden. Health care providers knew best,
and health care was paternalistic. All this has now changed in Canada and the rest of the
Western world. The Hippocratic tradition has been replaced by an autonomy-based ethic
known as "patient/family-centred care" whereby, as much as possible, patients and families
are the decision-makers. This transformation reflects the triumph of weak paternalism
over strong paternalism for the reasons given in Chapter 1 (section 1.5, page 18), and the
triumph is complete. The philosophy of patient/family-centred care is accepted by the law
in every province and territory in Canada and in every Canadian professional health care
code of ethics
The acceptance of the philosophy of patient/family-centred care has momentous con-
sequences for medical decision-making. In this view, patients are entitled to make deci-
sions for themselves insofar as they have decision-making capacity and their decisions are
substitute decision-
voluntary and informed. (We will return shortly to examine these three key conditions.) maker a person who
They can accept or ref use any treatment that is offered. They can live at risk, with neglect, makes decisions for a patient
self-neglect, and abuse. Or they can choose not to live at all. Their lives and choices are when the patient lacks
their own, and all that others can do is explain and argue. decision-making capacity.
A substitute decision-maker
On the other hand, if the patient lacks one or more of these decision-making require-
is sometimes also called a
ments to qualify as the decision-maker and the lack or lacks cannot be remedied before surrogate decision-maker
the decision must be made, the decision will be made by a substitute decision-maker. or (when the person has been
This substitute decision-maker will typically be the person who is most knowledgeable legally designated as the
and concerned about the patient, but when such a person cannot be found it may be the decision-maker by the patient
physician or a court-appointed individual. when capable) a proxy
decision-maker.
The substitute decision-maker can make decisions on the basis of two possible kinds
of judgments.
The first is a substituted judgment. Substituted judgments are judgments that the substituted judgment the
judgment that a patient would
decision-maker thinks the patient would have made. In making decisions on the basis make for himself or herself.
of substituted judgments, the substitute decision-maker looks at the world with all the
patient's hopes, fears, beliefs, desires, and eccentricities and makes the decision he or she
best-interest judgment the
reckons the patient would have made. The second is a best-interest judgment. Best- judgment that the reasonable
interest judgments are judgments the reasonable person in the patient's position would person in the patient's
make. Here the substitute decision-maker puts himself or herself in the shoes of the position would make.
nu I II 111 111111
® 2 Medical Decision-Making
Patient
N
Decision-making capacity Lacks the characteristics necessary
to make decisions for him/herself
Voluntary decision
Full information
Patient decides Family + Physicians decide
N
Substituted judgment Best-interest judgment
Figure 2.1 Decision Tree
reasonable person and makes the decision on the basis of more objective societally de-
termined criteria, such as reduction of suffering and preservation of quantity and quality
of life. When substituted judgments and best-interest judgments conflict, the autonomy-
driven philosophy of patient/family-centred care tells us to give systematic precedence to
substituted judgments.
The philosophy of patient/family-centered care enables us to construct a decision tree
for medical decision-making. The best circumstance is when the patient has decision-
making capacity, can act voluntarily and is fully informed, and tells us what he or she
wants. The next best is when the patient fails to have one or more of these characteristics
but for whom there is a firm substituted judgment. The worst is when we do not know
what the patient would want and have to fall back on a best-interest judgment. This tree
can be set out as in Figure 2.1.
The decision tree in Figure 2.1 is widely accepted in both ethics and law But applying
it raises difficulties at every point. It is sometimes hard to determine whether the patient
has decision-making capacity, is acting voluntarily, or is fully informed. If the patient fails
to have one or more of these features, it can be hard to identify the appropriate substi-
tute decision-maker. Once the substitute decision-maker has been identified, there can be
daunting epistemological problems in arriving at what the patient would want or what the
reasonable person in the patient's position would want. The rest of the chapter is concerned
with exploring these matters. The section headings follow the order of topics in the deci-
sion tree reading from left to right.
Decision-Making by Patient
decision-making capacity
individuals can understand
the nature and consequences
Decision-Making Capacity
of their decisions and are not Three conditions must be satisfied for patients to have decision-making capacity. (These
subject to distorting mental
conditions must also be satisfied by a substitute decision-maker.)
illness or mental states.
Individuals who have these
abilities are also referred to as 1. The patient must be able to understand the nature and consequences of his or her decisions.
capable or competent. Understanding is a cognitive capacity, and it is a matter of degree. The clinical test
Decision-Making by Patient S
for determining the degree of understanding a patient has is to explain the situation
in ordinary language and ask the patient to repeat that back in his or her own words
and manner.
2. The patient must be free from distorting mental illness. Mental illness also comes in degrees
and can range from almost normal states of neuroses to significantly impairing states
such as manias, clinical depression, neurotic compulsions, and paranoia. Persons may
be able to understand everything but still have their decision-making capacity com-
promised by mental illness. A patient with clinical depression, for example, may have
no cognitive deficit but may simply not care about his or her well-being.
3. The patient must be free from other distorting mental states. One may be capable of
understanding everything and not be impaired by mental illness but still be subject
to distortion from mental states such as nervousness, passion, agitation, excitement,
pain, shock, and fatigue. These states can be induced by such factors as disease, injury,
alcohol, narcotics, or social circumstances and can be as impairing as an inability to
understand or think clearly by reason of mental illness.
Since decision-making capacity is analyzed in terms of the above conditions and they
are matters of degree, there are different degrees of decision-making capacity. The question
thus arises as to what the appropriate degree of capacity is. Allen E. Buchanan and Dan W.
Brock take up this question in "Standards of Competence" (see page 32). They argue that
the appropriate standard varies depending on the degree of risk involved in the decision.
If the risk of saying yes or no to offered treatment is great, the standard must be high. If
there is no significant risk, the standard may be low And if the risk is in between, the
standard must be proportionate to that risk. They thus arrive at what they call a "process"
standard of competence that tries to steer a middle course between respecting the choices
of the patient and protecting his or her well-being. Brock and Buchanan cite some medical
examples of such decision-making, and Cases 1 and 5 (see pages 91 and 95) provide a
couple more.
Voluntary Decision
Once patients are identified as having decision-making capacity, two further conditions
must be satisfied before they can make any decision. Their decisions must be voluntary and
informed, and both conditions call for further discussion.
To make a voluntary decision, the patient does not have to be free of all influences. voluntary decision
In medical contexts, the patient must take into account what the physician and health care a decision made when a
person is not subject to
providers say about diagnosis and prognosis, how the decision will affect the patient, what
undue influences.
the family thinks should be done, how the family and others will be affected, and so on.
These considerations will certainly influence the patient's decision, but the line between
voluntary and non-voluntary decisions is not drawn in terms of whether the decisions are
influenced or not but in terms of whether the influences are undue. Some influences are
compatible with voluntariness, but undue influences are not. Voluntary decisions can thus
be defined negatively. To say that a decision is voluntary is to say that it is not a result of
influences such as coercion, duress, manipulation, or adverse social or medical conditions,
and so on. Coercion and duress include things such as physical violence or threats of it,
as well as threats to interfere improperly with rights, such as rights to treatment or food.
Manipulation may take the form of withdrawing or threatening to withdraw one's help,
company, or affection. It is also often difficult to distinguish manipulation from forms
of persuasion that involve coaxing and imploring or pressure from physicians and other
health care providers. Adverse social conditions such as poverty, lack of social supports,
chronic pain, or dependency may likewise exert influences so that decisions are no longer
significantly free.
II All 111111
2 Medical Decision-Making
Evaluating whether any of these conditions is present, and if so to what degree, often
calls for subtle judgment on the part of physicians, nurses, and social workers. This is espe-
cially so because health care providers spend a good part of their time trying to talk patients
into accepting things patients initially do not want. This is an important part of the job and
may involve attempts to persuade, but it cannot involve coercing or manipulating. In the
end, the decision must still be the patient's, and distinguishing between caring concern and
coercion calls for delicate judgment.
Susan Sherwin, in "A Relational Approach to Autonomy in Health Care" (see page 35),
raises a more general problem about voluntariness from a feminist perspective. On her rela-
tional analysis of autonomy, all of an individual's choices are embedded in social structures,
and some of them can be (and in the case of women often are) oppressive. Thus, when
women make health care decisions, these decisions may not be truly autonomous even
though no undue influence such as coercion, manipulation, or duress can be identified.
Sherwin does not for this reason reject the current emphasis on autonomy in health care or
recommend that others make decisions for the oppressed. The ideal solution is to remove
the conditions of oppression. Until then, it may be necessary to provide the vulnerable with
greater support in decision-making.
Informed Decision
informed decision We now turn to the informed condition. For a patient to make an informed decision,
a decision made when a he or she must know everything that would make a difference to that decision. In "The
person knows everything he Nuts and Bolts of Obtaining Consent to Treatment" (see page 45), L.E. Rozovsky and F.A.
or she would want to know
before making the decision.
Rozovsky tell us that this level of disclosure requires (among other things that do not
concern us here) that physicians tell patients:
1. The diagnosis and probable prognosis;

2. The reasonable treatment options available, including that of non-treatment;
3. The nature of those treatments, i.e., whether they are surgery, medication, electro-
shock, and so on;
4. The potential risks and benefits of treatment and non-treatment.
Of special importance and interest is the disclosure of risk. In deciding what risks to
disdose, the physician must tell the patient enough so that the patient will not regret his
or her decision. The physician's aim is to avoid the circumstance in which patients can say
that if they had only known of such and such a risk, they would not have made the deci-
sion they did. On this the law and ethics agree, but the law and ethics can diverge on what
must be disclosed to avoid such regret.
According to Rozovsky and Rozovsky, the law does not require that the physician tell
the patient about those risks that can be taken to be common knowledge or are minor and
unlikely. But if the risk is serious (e.g., death, deafness, or paralysis), the law requires that
it be disclosed even if there is only a vanishingly small chance of it occurring, This require-
ment, however, can cause problems, and the four readings that follow are all concerned in
different ways with challenging that it is ethically required.
Ruth R. Faden and Tom L. Beauchamp, in "The Concept of Informed Consent" (see
page 47), are concerned that obsession with the level of disclosure required by law may
impair autonomous action. They begin by distinguishing two senses of informed consent.
Sense 1 amounts to "autonomous authorization." In this sense, the patient or subject has
given consent if the patient has a substantial understanding of an intervention, is substan-
tially free of the control of others, and intentionally authorizes a professional to perform
that intervention. By contrast, Sense 2 amounts to "effective authorization." In this sense,
an individual has given consent if the rules and regulations that define the institutional
practice of informed consent are satisfied. Thus, if the institution defines informed consent
Decision-Making by Patient ®
as the satisfaction of a set of rules Rl-Rn, as soon as those rules are satisfied informed
consent has been obtained.
They then argue that an informed consent in Sense 1 ("autonomous authorization")
can fail to be an informed consent in Sense 2 ("effective authorization") because it does not
conform to certain defining rules (e.g., witnessing or wait-period requirements). And con-
versely, "informed consent" in Sense 2 need not constitute "authorization." Giving decision-
makers a full list of alternatives, risks, and benefits is no guarantee that they understand
what they are consenting to, and indeed it may be more confusing than helpful.
Sense 1 and Sense 2 informed consent are not necessarily inconsistent. The ideal cir-
cumstance, in fact, is when the conditions that define Sense 2 informed consent also satisfy
Sense 1 informed consent. But these senses can diverge, and the worry is that the demand
for institutional efficiency and fear of legal liability, which encourage effective consent, will
overshadow the importance of obtaining informed consent in the sense of autonomous
authorization. And to the extent that it does this, the primary moral rationale for informed
consent will be lost.
Howard Brody, in "Transparency: Informed Consent in Primary Care" (see page 51),
likewise questions the requirement that informed consent demands the disclosure of all
serious risks, however improbable. Brody identifies four circumstances in which phys-
icians commonly make treatment recommendations:
1. The intervention is low-risk and routine, and the patient almost always chooses it.
2. The intervention i s not routine but seems t o offer a clear benefit with minimal risk.
3. The intervention offers substantial chances for benefit but also very substantial risks.
4. The intervention offers substantial risks and extremely questionable benefits, but all
alternatives do the same.
On the one hand, Brody contends that if the physician's decision to recommend an inter-
vention involves a careful consideration of risks and benefits (as it presumably would in
circumstances 3 and 4 above), those risks must be disclosed. On the other hand, Brody
claims that most interventions in primary care are not arrived at in such circumstances.
Far more commonly, treatment recommendations are made in circumstances like 1 and 2
above. Virtually all interventions have some remote and serious risks, but Brody argues that
if those risks do not figure in the physician's recommendation, they need not be disclosed.
The task of informing the patient is complete as soon as the physicians reasoning for the
recommendation has been made transparent to the patient, the patient is invited to ask
questions, and those questions are answered.
The article by Joseph Kaufert and John O'Neil, "Culture, Power, and Informed Con-
sent: The Impact of Aboriginal Health Interpreters on Decision-Making" (see page 56),
also challenges the legal standard b y identifying a circumstance i n which i t seems impos-
sible to meet and desirable to ignore. Their article details the difficulties of communicat-
ing with persons who have a radically different culture, language, and educational level
than the physician and hence of getting what we would recognize as informed consent.
In the case Kaufert and O'Neil report, the translator did much of the communicating,
and the patient was not told of the nature of the recommended procedures in any detail,
the risks and benefits, the available alternatives, or the possibility of non-treatment. The
patient was also largely silent, with the result that the physician proceeded more on the
basis of "assent," whereby the patient accepts treatment in the sense of not refusing it,
rather than of "consent," whereby the patient has a good understanding of what she is
getting into. One may lament this, and perhaps regard it as demeaning, but what is the
alternative?
Onora O'Neill, in "Informed Consent and Public Health" (see page 60), distinguishes
between two different contexts in which consent can be sought: therapeutic and public
health. After identifying a host of problems for therapeutic consent (which in her view
II WlI 1114W
are intractable), she contends that in public health contexts where consent is sought for
(say) immunization, the aim is radically different. In clinical contexts, the only interests
at stake are those of the patient. But in public health contexts, the public has a keen
interest in compliance with the proposed treatment because a public good (herd im-
munity in the case of vaccination) is achieved only if many participate and set back if
they do not. O'Neill then argues, relying on Mill's harm principle, which allows interfer-
ence with the liberty of individuals to prevent harm to others (see Chapter 1, page 18),
that when public health risks are sufficiently grave, informed consent is not necessary
for treatment. But even if withholding information (e.g., about the risks of vaccines)
will achieve a public health good, there is a question of whether that is appropriate in
an open democratic society as opposed to disclosing all and relying on public educa-
tion to secure compliance, supplemented as necessary by quarantines or penalties for
non-compliance.
Decision-Making by Substitute Decision-Makers
We now turn to the problem of making decisions for patients when they cannot make
them for themselves. Such decision-making will be done by substitute decision-makers
and involve making substituted judgments or best-interest judgments.
Substituted Judgments
Substituted judgments are the judgments that decision-makers think the patient would
have made and are the best basis for decision-making when patients lack decision-making
capability. To arrive at a substituted judgment, the decision-maker may rely on an informal
„
values inventory" of the person. By spending time with a person, discussing the news,
watching medical shows together, and so on, a substitute decision-maker can often get a
good idea of the patient's basic values and beliefs, which can be used to determine what the
patient would want in certain circumstances.
advance directive a legal Another way of arriving at what the patient would want comes from an advance
document made by persons directive made by the person when capable. Advance directives are documents that allow
with decision-making capacity
individuals to project their health care preferences into a time when they cannot make
to guide medical decisions
when they lack decision-
decisions for themselves. As such, they promise to make decision-making better for pa-
making capacity. tients, lessen the burden on health care providers and substitute decision-makers, and
simplify and (sometimes) by limiting treatment reduce costs to the system. For these rea-
sons, advance directives have become enormously popular. Many health authorities urge
individuals to make them, and they are endorsed by the Canadian Medical Association in
"Advance Directives for Resuscitation and Other Life-Saving or Life-Sustaining Measures"
(see page 76).
proxy directive an advance Advance directives come in two varieties. The first is a proxy directive, which allows
directive in which persons individuals to say who they would like to be their substitute decision-maker. In the absence
designate who is to make
of a proxy directive, the decision-maker will be selected from a legal line that begins with
decisions for them when they
lack decision-making capacity.
spouse, adult children, and parents and carries on until a single decision-maker is identi-
fied. Proxy directives are thus a good idea if one wants someone other than the first (or
any) person on the legal line to make decisions for him or her. The second is an instruc-
instructional advance tional advance directive, which allows individuals to say what they would like done in
directive an advance what circumstances. When the instructional advance directive is legally valid—i.e., made
directive in which persons in accordance with the laws of the jurisdiction and clearly identifies what the individual
designate what medical
wants done in the circumstances at hand—the directive has all the force of the patient's
decisions they would or would
not want made when they in-person instruction. As such, decisions made on its basis are not substituted judgments
lack decision-making capacity. but the patient's judgments. However, instructional advance directives are not always or
Decision-Making by Substitute Decision-Makers
even typically legally valid or perfectly clear, and in such cases substitute decision-makers
have to try to understand what the patient would have wanted and so make a genuine
substituted judgment.
Advance directives work best when patients unambiguously identify what care they
want and are unlikely to change their minds. They can thus be recommended if patients
have ideological views about certain procedures (e.g., no blood products because they are
a Jehovah's Witness) or have had experiences with specific interventions (e.g., ventilation)
that they do not want to repeat, or are visibly disabled and fear that they may not get
maximum care if they do not ask for it, and so on. But in the absence of such special views,
it is difficult for individuals to say what they would or would not want in what health care
conditions.
This difficulty may stem from not knowing what it is like to be in possible medical
conditions; what conditions they can adjust to; what their prognosis is with and without
treatment; and the nature, degree of invasiveness, and risks of available treatments. It is
thus hard for individuals to confidently say that they do not want life-saving treatment if
they become demented or quadriplegic or want electro-convulsive therapy if they suffer
from serious depression. There is also the difficulty that an individual's preferences may
change, so what they want now may not be what they will want or appreciate receiv-
ing in the future. Advance directives thus come with perils that individuals considering
them must weigh against their promised benefits. These issues are discussed in the reading
by Angela Fagerlin and Carl E. Schneider, titled "Enough: The Failure of the Living Will"
(see page 63). The reading is also notable for reviewing empirical evidence about the use-
fulness of living wills, much of which raises doubts about their value. While many who
have studied advance directives and seen medical decision-making up close enthusiastic-
ally support them, many others think that patients will do better by leaving the decision to
be made at the time by a loved one they trust.
Best-Interest Judgments
Best-interest judgments come into play when the patient does not have decision-making
capability and a substituted judgment is not available. Candidates for best-interest decision-
making include patients in all age groups from infants to the old and with varying degrees
of decision-making capability from the comatose to the almost-capable. The patient may be
a stranger or someone the substitute decision-maker knows very well but does not know
what he or she would want in the circumstances at hand.
The aim of best-interest judgments (as of substituted judgments) is to give patients the
health care they would have chosen for themselves if they were able to. But since what they
would have chosen is unknown, decision-makers have to gamble, and the standard used
to determine what treatment is appropriate is that of the "reasonable person," meaning
"most people in the patient's circumstances." The idea is that providing the care that most
people with the patient's medical condition, known religious or cultural beliefs, and social
situation would want has the best chance of giving the patient the care he or she would
have chosen.
Children and Infants

Some of the most poignant and difficult cases of best-interest judgments involve children,
and in this section we consider two sorts of hard cases. The first is presented by Christine
Harrison, Nuala P Kenny, Mona Sidarous, and Mary Rowell and addresses some of the
problems encountered in involving children in the decision-making. ("Involving Children
in Medical Decisions"; see page 76). They consider the case of an 11-year-old child who
has had experience with severe illness and the burdens of treatment and does not want
further treatment. The authors favour a family-centred approach to the decision-making
YY 1 dl II hill 1111.1.1
whereby the interests of everyone are taken into account. But any preferences that the
child, family, or treating team may have must be measured against the best interest of
the child. The view that we must always act in the best interest of the child is sometimes
challenged,' but it is the standard view, and the Bioethics Committee of the Canadian
Paediatric Society (CPS), in its position statement 'Treatment Decisions Regarding Infants,
Children, and Adolescents" (see page 80), is clear and firm that the interest of the child
must be paramount.
Our second sort of hard case concerns two of the most common and worrisome
problems facing physicians in the treatment of children. The first is when the family re-
quests more life-sustaining treatment than the physician thinks appropriate. The second
is when the family requests less than the physician wants to provide and asks the phys-
ician to withdraw or withhold life-sustaining treatment (wLST). When can physicians
say "no" to such requests? We will take the cases in turn.
The CPS, in "Treatment Decisions Regarding Infants, Children, and Adolescents" cited
above, addresses the first of these situations. It contends that physicians can WLST when
there is consensus that there is a high degree of probability that:
1. There is irreversible progression to imminent death;

2. Treatment is clearly ineffective or harmful;
3. Life will be severely shortened regardless of treatment, and the limitation or with-
drawal of interventions will allow greater palliative and comfort care;
4. Lives will be filled with intolerable distress and suffering that cannot be prevented or
alleviated.
In saying this, the CPS sets a limit on how much treatment families can request and
receive. But this view conflicts with that of the Judicial Council of the American Medical
Association (AMA) when it wrote: "In desperate situations involving newborns, the advice
and judgment of the physician should be readily available, but the decision whether to
exert maximal efforts to sustain life should be the choice of the parents" (see page 82).
Given that decision-makers must always act in the child's best interest, which of these
should be followed depends on which will best promote that interest. But it is hard to say
where that interest lies. The AMA can reasonably say that life is so valuable that it is in the
best interest of the child to take great risks to preserve it. On the other hand, the CPS can
say that continued life-sustaining treatment will likely do the child no good and cause it
great suffering and that, thus, providing maximal treatment is not a gamble that is in the
child's best interest. Most pediatricians accept some version of the CPS's view, but the AMA's
position is unrefuted, and the issue continues unresolved.
We now turn to the opposite problem of when the family wants the physician to WLST
and the physician disagrees. Since physicians must always act in the child's best inter-
est, they can clearly say "no" to family requests to WLST when they think life-sustaining
treatment will benefit the child. No one can disagree that beneficial treatment must be
provided, but disagreement can occur over what treatment is beneficial, and examples
of situations in which it is appropriate to WLST would be helpful. The Royal College of
Paediatrics and Child Health identifies five such situations. These are when (1) the child
is brain dead, (2) the child is in a permanent vegetative state, (3) treatment will only delay
death and not alleviate suffering, (4) the child will survive with unacceptable impairments,
and (5) "the child and/or family feel that in the face of progressive and irreversible illness
further treatment is more than can be borne," and they may request to have "a particular
treatment withdrawn or to refuse further treatment irrespective of the medical opinion that
it may be of some benefit" (see page 81).
Situations 1-4 are uncontroversial and generally accepted. Situation 5, however, intro-
duces a new and novel possibility and calls for further discussion. Situation 5 as stated
Decision-Making by Substitute Decision-Makers 31
above is taken from the "Summary" that the Royal College provides of its guideline (as
is the selection in the text). But in the body of the guideline, the phrase "in the face of
progressive and irreversible illness" is omitted. This is a substantial difference. If the phrase
is included, treatment can be removed because of intolerable suffering only when the child
is already in a seriously compromised condition and a healthy child would not emerge
whatever treatment is provided. On the other hand, if the phrase is omitted, treatment
can be removed if suffering is intolerable even if a healthy child could emerge. On this
view, just as substitute decision-makers for incompetent adults can refuse very burden-
some treatments (e.g., radical chemotherapy) even if a good result is possible, so could
families do the same for their children. The college never removes this ambiguity, and
readers should carefully consider which view is better. If intolerable suffering despite the
possibility of a good outcome is sometimes allowed as a reason for the family to say "no" to
life-sustaining treatment, the sphere of family autonomy is expanded and that of physician
autonomy is shrunk in the way described in Chapter 4 (see "Professional Autonomy and
Professional Ethics").
Religion and Culture

Other cases calling for a best-interest determination are those in which religious or cul-
tural views conflict with the prevailing views of Western society. One common such case
concerns Jehovah's Witnesses. It is now standard to allow competent adults to make a
voluntary and informed choice to forego potentially life-saving blood transfusions but not
to allow them to make such decisions for their children. Jehovah's Witness parents, how-
ever, often disagree with this stance, arguing that it deprives them of the most important
decisions they could make for their child—namely about their spiritual welfare and the
possibility of everlasting life.
Every court that has addressed the issue has rejected this parental claim, but the re-
jection is typically not accompanied with a philosophically satisfying explanation. If the
Jehovah's Witness belief is that to receive a blood transfusion is to lose eternal life, one may
ask why it is in the best interest of the child to receive the transfusion. It does not seem a
good gamble to risk infinite loss to make a finite gain, whatever the probabilities of each
may be. As long as it is possible that giving blood to a child will deprive that child of eternal
life, it is not clear why withholding blood should not be the best strategic choice for the
child. Thus, unless one takes the plunge and says that there is no possibility whatsoever
that the Jehovah's Witness belief is true and provides reasons for thinking that, the argu-
ment for forcing blood is not complete.
Cultural practices are also often in conflict with the norms of Western society. In "Eth-
ical Relativism in a Multicultural Society" (see page 83), Ruth Macklin discusses three
particularly difficult cases of this sort. One concerns what to do when the physician has
bad news about a father and the family says, "Don't tell dad," explaining that in their
culture such information is not shared. A second is when individuals in certain cultures
(Macklin cites the Navaho) do not want to hear about risks and possible bad outcomes and
the physician wants to secure consent for a serious surgery. Still another involves rituals
(such as female circumcision) or healing practices (such as inflicting burns) that are socially
mandatory or common in a family culture but child abuse in ours. These sorts of matters
are often dealt with briskly by appealing to hospital rules or the law, but the philosophic
and ethical rationales for whatever decisions are made are not always perfectly clear or
satisfying for a society that celebrates multiculturalism.
Note
1. By, for example, John Hardwig in "What about the Family?" The Hastings Center Report 20, no. 2
(March/April 1990): 5-10.
ii di] wu
C
I 2 Medical Decision-Making
2.2 Determining Decision-Making Capacity
Standards of Competence conception of well-being—but fails adequately to re-

spect patient self-determination.
Allen E. Buchanan and Dan W. Brock At bottom, a person's interest in self-determination
is his or her interest in defining, revising over time, and
I. Different Standards of Competence
pursuing his or her own particular conception of the
good life. [With so-called ideal or objective] theories of
A number of different standards of competence have the good for persons, there are serious practical or fal-
been identified and supported in the literature, al- libilist risks associated with any purportedly objective
though statutory and case law provide little help in standard for the correct decision—the standard may
articulating precise standards.' It is neither feasible ignore the patient's own distinctive conception of the
nor necessary to discuss here all the alternatives that good and may constitute enforcement of unjustified
have been proposed. Instead, the range of alternatives ideals or unjustifiably substitute another's conception
will be delineated, and the difficulties of the main stan- of what is best for the patient. Moreover, even such
dards will be examined in order to clarify and defend a standard's theoretical claim to protect maximally a
[our decision-relative analysis]. More or less stringent patient's well-being is only as strong as the objective
standards of competence in effect strike different bal- account of a person's well-being on which the standard
ances between the values of patient well-being and rests. Many proponents of ideal theories only assert the
self-determination. ideals and fail even to recognize the need for justifying
them, much less proceed to do so.
A. A Minimal Standard of Competence Although ascertaining the correct or best theory
An example of a minimal standard of competence of individual well-being or the good for persons is a
is that the patient merely be able to express a prefer- complex and controversial task, . . . any standard of
ence. This standard respects every expressed choice of individual well-being that does not ultimately rest on
a patient, and so is not in fact a criterion of competent an individual's own underlying and enduring aims and
choice at all.' It entirely disregards whether defects or values is both problematic in theory and subject to in-
mistakes are present in the reasoning process leading to tolerable abuse in practice. There may be room in some
the choice, whether the choice is in accord with the pa- broad policy decisions or overall theories of justice for
tientb own conception of his or her good, and whether more "objective" and interpersonal measures of well-
the choice would be harmful to the patient. It thus fails being that fail fully to reflect differences in individuals'
to provide any protection for patient well-being, and it own views of their well-being,3 but we believe there is
is insensitive to the way the value of self-determination much less room for such purportedly objective measures
itself varies both with the nature of the decision to be in the kind of judgments of concern here—judgments
made and with differences in people's capacities to about appropriate treatment for an individual patient.
choose in accordance with their conceptions of their Thus, a standard that judges competence by compar-
own good. ing the content of a patient's decision to some objective
standard for the correct decision may fail even to pro-
B. An Outcome Standard of Competence tect appropriately a patient's well-being.
At the other extreme are standards that look solely to
the content or outcome of the decision—for example, C. A Process Standard of Decision-Making
the standard that the choice be a reasonable one, or Competence
be what other reasonable or rational persons would An adequate standard of competence will focus primar-
choose. On this view, failure of the patient's choice ily not on the content of the patient's decision but on
to match some such allegedly objective outcome the process of the reasoning that leads up to that deci-
standard of choice entails that it is an incompetent sion. There are two central questions for any process
choice. Such a standard maximally protects patient standard of competence. First, a process standard must
well-being—although only according to the standard's set a level of reasoning required for the patient to be
Source: From Deciding for Others: The Ethics of Surrogate Decision Malting (Cambridge University Press: Cambridge, 1990), 48-57.
Copyright © 1990 Cambridge University Press. Reprinted with the permission of Cambridge University Press.
Buchanan/Brock: Standards of Competence 33
competent. In other words, how well must the patient alternatives will usually be determined by the phys-
understand and reason to be competent? How much ician. This assessment should focus on the expected
can understanding be limited or reasoning be defective effects of a particular treatment option in forwarding
and still be compatible with competence? The second the patient's underlying and enduring aims and values,
question often passes without explicit notice by those to the extent that these are known. When the patient's
evaluating competence. How certain must those per- aims and values are not known, the risk/benefit assess-
sons evaluating competence be about how well the ment will balance the expected effects of a particular
patient has understood and reasoned in coming to a treatment option in achieving the general goals of
decision? This second question is important because it health care in prolonging life, preventing injury and
is common in cases of marginal or questionable compe- disability, and relieving suffering as against its risks of
tence for there to be a significant degree of uncertainty harm. The table indicates that the relevant comparison
about the patient's reasoning and decision-making pro- is with other available alternatives, and the degree to
cess that can never be eliminated. which the net benefit/risk balance of the alternative
chosen is better or worse than that for optimal alterna-
II. Relation of the Process Standard of tive treatment options. It should be noted that a choice
Competence to Expected Harms and Benefits might properly require only low/minimal competence,
even though its expected risks exceed its expected
Because the competence evaluation requires striking a benefits or it is more generally a high-risk treatment,
balance between the two values of respecting patients' because all other available alternatives have substan-
rights to decide for themselves and protecting them tially worse risk/benefit ratios.
from the harmful consequences of their own choices, it Table 1 also indicates, for each level of compe-
should be clear that no single standard of competence— tence, the relative importance of different grounds for
no single answer to the questions above—can be ad- believing that a patient's own choice best promotes his
equate for all decisions. This is true because (1) the or her well-being. This brings out an important point.
degree of expected harm from choices made at a given For all patient choices, other people responsible for
level of understanding and reasoning can vary from deciding whether those choices should be respected
none to the most serious, including major disability or should have grounds for believing that the choice, if
death, and because (2) the importance or value to the it is to be honoured, is reasonably in accord with the
patient of self-determination can vary depending on patient's well-being (although the choice need not,
the choice being made. of course, maximally promote the patient's interests).
There is an important implication of this view that When the patient's level of decision-making compe-
the standard of competence ought to vary in part with tence need be only at the low/minimal level, as in the
the expected harms or benefits to the patient of acting in agreement to a lumbar puncture for presumed men-
accordance with the patient's choice—namely, that just ingitis, these grounds derive only minimally from the
because a patient is competent to consent to a treatment, fact that the patient has chosen the option in question;
it does not follow that the patient is competent to refuse they principally stem from others' positive assess-
it, and vice versa. For example, consent to a low-risk ment of the choice's expected effects on the patient's
lifesaving procedure by an otherwise healthy individual well-being.
should require only a minimal level of competence, but At the other extreme, when the expected effects
refusal of that same procedure by such an individual of the patient's choice for his or her well-being appear
should require the highest level of competence. to be substantially worse than available alternatives, as
Because the appropriate level of competence prop- in the refusal of a simple appendectomy, the require-
erly required for a particular decision must be adjusted ment of a high/maximal level of competence provides
to the consequences of acting on that decision, no grounds for relying on the patient's decision as itself
single standard of decision-making competence is ad- establishing that the choice best fits the patient's good
equate. Instead, the level of competence appropriately (his or her own underlying and enduring aims and
required for decision-making varies along a full range values). The highest level of competence should assure
from low/minimal to high/maximal. Table 1 illustrates that no significant mistakes in the patient's reasoning
this variation, with the treatment choices listed used and decision-making are present, and is required to
only as examples of any treatment choice with that rela- rebut the presumption that the choice is not in fact
tive risk benefit assessment. reasonably related to the patient's interests.
The net balance of expected benefits and risks When the expected effects for the patient's well-
of the patient's choice in comparison with other being of his or her choice are approximately comparable
SW 1 91 II di 1111:u
s 2 Medical Decision-Making
Table 1 Decision-Making Competence and Patient Well-Being

Others' risk/benefit
assessment of that
The patient's choice in comparison Level of decision-making Grounds for believing patient's choice best
treatment choice with other alternatives competence required promotes/protects own well-being
Patient consents to Net balance substantially Low/minimal Principally the benefit/risk assessment made
lumbar puncture for better than for possible by others
presumed meningitis alternatives
Patient chooses Net balance roughly Moderate/median Roughly equally from the benefit/risk
lumpectomy for breast comparable to that of assessment made by others and from the
cancer other alternatives patient's decision that the chosen alternative
best fits own conception of own good
Patient refuses Net balance substantially High/maximal Principally from patient's decision that the
surgery for simple worse than for another chosen alternative best fits own conception
appendectomy alternative or alternatives of own good
to those of alternatives, as in the choice of a lumpec- the decision in favour of a standard that focuses on
tomy for treatment of breast cancer, a moderate/median the process of the patient's reasoning. This may appear
level of competence is sufficient to provide reasonable inconsistent with our insistence here that the appro-
grounds that the choice promotes the patient's good and priate level of decision-making capacity required for
that her well-being is adequately protected. It is also competence should depend in significant part on the
reasonable to assume that as the level of competence re- effects for the patient's well-being of accepting his or
quired increases (from minimal to maximal), the instru- her choice, since what those effects are clearly depends
mental value or importance of respecting the patient's on the content or outcome of the patient's choice. How-
self-determination increases as well, specifically the part ever, there is no inconsistency. The competence evalua-
of the value of self-determination that rests on the as- tion addresses the process of the patient's reasoning,
sumption that persons will secure their good when they whereas the degree of defectiveness and limitation of,
choose for themselves. As competence increases, other and uncertainty about, that process that is compatible
things being equal, the likelihood of this happening with competence depends in significant part on the
increases. likely harm to the patient's well-being of accepting his
Thus, according to the concept of competence or her choice. To the extent that they are known, the ef-
endorsed here, a particular individual's decision- fects on the patient's well-being should be evaluated in
making capacity at a given time may be sufficient for terms of his or her own underlying and enduring aims
making a decision to refuse a diagnostic procedure and values, or, where these are not known, in terms
when forgoing the procedure does not carry a signifi- of the effects on life and health. Thus in our approach
cant risk, although it would not necessarily be suf- there is no use of an "objective" standard for the best
ficient for refusing a surgical procedure that would or correct decision that is known to be in conflict with
correct a life-threatening condition. The greater the the patient's own underlying and enduring aims and
risk relative to other alternatives—where risk is a values, which was the objectionable feature of a con-
function of the severity of the expected harm and the tent or outcome standard of competence.
probability of its occurrence—the greater the level of The evaluation of the patient's decision-making will
communication, understanding, and reasoning skills seek to assess how well the patient has understood the
required for competence to make that decision. It is nature of the proposed treatment and any significant
not always true, however, that if a person is competent alternatives, the expected benefits and risks and the
to make one decision, then he or she is competent to likelihood of each, the reason for the recommendation,
make another decision so long as it involves equal and then whether the patient has made a choice that
risk. Even if the risk is the same, one decision may reasonably conforms to his or her underlying and en-
be more complex, and hence require a higher level during aims and values. Two broad kinds of defect are
of capacity for understanding options and reasoning then possible: first, "factual" misunderstanding about
about consequences. the nature and likelihood of an outcome, for example
In the previous section, we rejected a standard of from limitations in cognitive understanding resulting
competence that looks to the content or outcome of from stroke or from impairment of short-term memory
Buchanan/Brock: Standards of Competence C
resulting from dementia; second, failure of the patient's enduring aims and values," or at least with what these
choice to be based on his or her underlying and endur- would be except for unfortunate distortions. If there
ing aims and values, for example because depression is no objective way to determine a person's under-
has temporarily distorted them so that the patient "no lying and enduring aims and values then the worry
longer cares" about restoration of the function he or she is that our view will lead to excessive paternalism.
had valued before becoming depressed.4 We acknowledge that this determination will often
A crude but perhaps helpful way of characterizing be difficult and uncertain, for example in cases like
the proper aim of the evaluator of the competence of a severe chronic depression, leading to genuine and
seemingly harmful or "bad" patient choice is to think of justified uncertainty about the patient's "true" aims
him or her addressing the patient in this fashion: "Help and values. But any claims that the aims and values
me try to understand and make sense of your choice. actually expressed by the patient are not his or her
Help me to see whether your choice is reasonable, not underlying and enduring aims and values should be
in the sense that it is what I or most people would based on evidence of the distortion of the actual aims
choose, but that it is reasonable for you in light of your and values independent of their mere difference with
underlying and enduring aims and values." This is the some other, "better" aims and values. Just as the pro-
proper focus of a process standard of competence. cess standard of competence focuses on the process of
Some may object that misguided paternalists will the patient's reasoning, so also it requires evidence of
always be ready to assert that their interference with a process of distortion of the patient's aims and values
the patient's choice is "deep down" in accord with to justify evaluating choices by a standard other than
what we have called the patient's "underlying and the patient's actually expressed aims and values. . . .
Notes
1. See especially Roth, L.H., Meisel, A., and Lidz, C.W., be used in evaluating persons' well-being within a theory
(1977), "Tests of Competency to Consent to Treatment," of justice; cf. Rawls, J. (1971), A Theory of Justice (Harvard
in American Journal of Psychiatry 134: 279-84; what they University Press: Cambridge, MA).
call "tests" are what we call "standards." An excellent 4. This second kind of decision-making defect illustrates the
discussion of competence generally, and of Roth et inadequacy of the tests that Roth, Meisel, and Lidz call
tests for competence in particular, is Freedman, B., (1981), "the ability to understand" and "actual understanding"
"Competence, Marginal and Otherwise," in International tests (cf. Roth et al., op. cit., 281-2). The clinically
Journal of Law and Psychiatry 4: 53-72. depressed patient may evidence no failure to understand
2. Cf. Freedman, op. cit. the harmful consequences of his choice, but instead
3. For example, John Rawls makes such claims for an evidence indifference to those consequences as result of
objective and interpersonal account of "primary goods" to his depression.
A Relational Approach to Autonomy . . . some difficulties I find with the usual interpreta-
tions of the concept, focusing especially on difficulties
in Health Care that arise from a specifically feminist perspective. In
Susan Sherwin response to these problems, I propose an alternative
conception of autonomy that I label "relational" though
the terms socially situated or contextualized would de-
Respect for patient autonomy (or self-direction) is scribe it equally well. To avoid confusion, I explicitly
broadly understood as recognition that patients have distinguish my use of the term relational from that of
the authority to make decisions about their own health some other feminist authors, such as Carol Gilligan
care. . . . (1982), who reserve it to refer only to the narrower
I propose a feminist analysis of autonomy, making set of interpersonal relations. I apply the term to the
vivid both our attraction to and distrust of the domin- full range of influential human relations, personal and
ant interpretation of this concept. I begin by reviewing public. Oppression permeates both personal and public
Source: From The Politics of Women's Health: Exploring Agency and Autonomy, by Susan Sherwin et al. Used by permission of Temple
University Press. Copyright © 1998 by Temple University. All rights reserved.
II SI ww
relationships; hence, I prefer to politicize the under- This problem is compounded within our increasingly
standing of the term relational as a way of emphasizing diverse urban communities, where differences in lan-
the political dimensions of the multiple relationships guage and culture between health care providers and
that structure an individual's selfhood, rather than to the patients they serve may create enormous practical
reserve the term to protect a sphere of purely private re- barriers to informed choice.
lationships that may appear to be free of political influ- There are yet deeper problems with the ideal of
ence.' I explain why I think the relational alternative is autonomy invoked in most bioethical discussions. The
more successful than the familiar individualistic inter- paradigm offered for informed consent is built on a
pretation at addressing the concerns identified. Finally, model of articulate, intelligent patients who are accus-
I briefly indicate some of the implications of adopting tomed to making decisions about the course of their
a relational interpretation of autonomy with respect to lives and who possess the resources necessary to allow
some of the issues discussed elsewhere in this book them a range of options to choose among. Decisions
[The Politics of Women's Health: Exploring Agency and Au- are constructed as a product of objective calculation
tonomy, ed. Susan Sherwin (Temple University Press: on the basis of near perfect information. Clearly, not
1988)], and I identify some of the changes that this all patients meet these ideal conditions (perhaps none
notion of relational autonomy suggests for the delivery does), yet there are no satisfactory guidelines available
of health services. . . . about how to proceed when dealing with patients who
do not fit the paradigm.
Problems with the Autonomy Ideal Feminist analysis reveals several problems inher-
ent in the very construction of the concept of auton-
. . . Despite this broad consensus about the value of omy that is at the heart of most bioethics discussions.3
a principle of respect for patient autonomy in health One problem is that autonomy provisions are some-
care, there are many problems with the principle as it times interpreted as functioning independently of and
is usually interpreted and applied in health care ethics. outweighing all other moral values. More specifically,
As many health critics have observed, we need to ques- autonomy is often understood to exist in conflict with
tion how much control individual patients really have the demands of justice because the requirements of the
over the determination of their treatment within the latter may have to be imposed on unwilling citizens.
stressful world of health care services. Even a casual en- Autonomy is frequently interpreted to mean freedom
counter with most modern hospitals reveals that wide from interference; this analysis can be invoked (as it fre-
agreement about the moral importance of respect for quently is) to oppose taxation as coercive and, hence, a
patient autonomy does not always translate into a set of violation of personal autonomy. But coercive measures
practices that actually respect and foster patient auton- like taxation are essential if a society wants to reduce
omy in any meaningful sense. Ensuring that patients inequity and provide the disadvantaged with access to
meet some measure of informed choice—or, more the means (e.g., basic necessities, social respect, educa-
commonly, informed consent2—bef ore receiving or tion, and health care) that are necessary for meaningful
declining treatment has become accepted as the most exercise of their autonomy. In contrast to traditional
promising mechanism for ensuring patient autonomy accounts of autonomy that accept and indeed presume
in health care settings, but, in practice, the effective- some sort of tension between autonomy and justice,
ness of the actual procedures used to obtain informed feminism encourages us to see the connections be-
consent usually falls short of fully protecting patient tween these two central moral ideals.
autonomy. This gap is easy to understand: attention to In fact, autonomy language is often used to hide
patient autonomy can be a time-consuming business the workings of privilege and to mask the barriers of
and the demands of identifying patient values and oppression. For example, within North America it
preferences are often sacrificed in the face of heavy seems that people who were raised in an atmosphere
patient loads and staff shortages. In addition, health of privilege and respect come rather easily to think of
care providers are often constrained from promoting themselves as independent and self-governing; it feels
and responding to patients' autonomy in health care natural to them to conceive of themselves as autono-
because of pressures they experience to contain health mous. Having been taught that they need only to apply
care costs and to avoid making themselves liable to themselves in order to take advantage of the oppor-
lawsuits. Moreover, most health care providers are tunities available to them, most learn to think of their
generally not well trained in the communication skills successes as self-created and deserved. Such thinking
necessary to ensure that patients have the requisite encourages them to be oblivious to the barriers that
understanding to provide genuine informed consent. oppression and disadvantage create, and it allows
Sherwin: A Relational Approach to Autonomy in Health Care 37
them to see the failures of others as evidence of the with respect to stereotypical assumptions about mem-
latters' unwillingness to exercise their own presumed bers of racial minorities, indigenous peoples, persons
autonomy responsibly. This individualistic approach with disabilities, welfare recipients, people from devel-
to autonomy makes it very easy for people of privil- oping countries, those who are non-literate, and so on.
ege to remain ignorant of the social arrangements that Minimally, then, health care providers must become
support their own sense of independence, such as the sensitive to the ways in which oppressive stereotypes
institutions that provide them with an exceptionally can undermine their ability to recognize some sorts of
good education and a relatively high degree of personal patients as being rational or competent.
safety. Encouraged to focus on their own sense of in- Consider, also, the second condition, which has
dividual accomplishment, they are inclined to blame to do with making a (reasonable) choice from the set
less well-situated people for their lack of comparable of available options. Here, the difficulty is that the set
success rather than to appreciate the costs of oppres- of available options is constructed in ways that may
sion. This familiar sort of thinking tends to interfere already seriously limit the patient's autonomy by pre-
with people's ability to see the importance of support- maturely excluding options the patient might have
ive social conditions for fostering autonomous action. preferred. There is a whole series of complex decisions
By focusing instead on the injustice that is associated that together shape the set of options that health care
with oppression, feminism helps us to recognize that providers are able to offer their patients: these can in-
autonomy is best achieved where the social conditions volve such factors as the forces that structure research
that support it are in place. Hence, it provides us with programs, the types of results that journals are willing
an alternative perspective for understanding a socially to publish, curriculum priorities in medical and other
grounded notion of autonomy. professional schools, and funding policies within the
Further, the standard conception of autonomy, es- health care system.' While all patients will face limited
pecially as it is invoked in bioethics, tends to place the choices by virtue of these sorts of institutional policy
focus of concern quite narrowly on particular decisions decisions, the consequences are especially significant
of individuals; that is, it is common to speak of specific for members of oppressed groups because they tend
health care decisions as autonomous, or, at least, of the to be underrepresented on the bodies that make these
patient as autonomous with respect to the decision at earlier decisions, and therefore their interests are less
hand. Such analyses discourage attention to the context likely to be reflected in each of the background deci-
in which decisions are actually made. Patient decisions sions that are made. In general, the sorts of institutional
are considered to be autonomous if the patient is (1) decisions in question tend to reflect the biases of dis-
deemed to be sufficiently competent (rational) to make criminatory values and practices. Hence, the outcomes
the decision at issue, (2) makes a (reasonable) choice of these multiple earlier decisions can have a significant
from a set of available options, (3) has adequate infor- impact on an oppressed patient's ultimate autonomy by
mation and understanding about the available choices, disproportionately and unfairly restricting the choices
and (4) is free from explicit coercion toward (or away available to her. Nevertheless, such background condi-
from) one of those options. It is assumed that these tions are seldom visible within discussions of patient
criteria can be evaluated in any particular case, simply autonomy in bioethics.
by looking at the state of the patient and her delibera- The third condition is also problematic in that
tions in isolation from the social conditions that struc- the information made available to patients is, inevit-
ture her options. Yet, each of these conditions is more ably, the information that has been deemed worthy of
problematic than is generally recognized. study and that is considered relevant by the health care
The competency criterion threatens to exclude providers involved. Again, research, publication, and
people who are oppressed from the scope of autonomy education policies largely determine what sorts of data
provisions altogether. This is because competency is are collected and, significantly, what questions are neg-
often equated with being rational,4 yet the rationality lected; systemic bias unquestionably influences these
of women and members of other oppressed groups is policies. Further, the very large gap in life experience
frequently denied. In fact, as Genevieve Lloyd (1984) between physicians, who are, by virtue of their profes-
has shown, the very concept of rationality has been sional status, relatively privileged members of society,
constructed in opposition to the traits that are stereo- and some of their seriously disadvantaged patients
typically assigned to women (e.g., by requiring that makes the likelihood of the former anticipating the
agents demonstrate objectivity and emotional dis- specific information needs of the latter questionable.
tance),5 with the result that women are often seen as While an open consent process will help reduce this
simply incapable of rationality.6 Similar problems arise gap by providing patients with the opportunity to raise
II w I11111
questions, patients often feel too intimidated to ask or of autonomy associated with her immediate decision
even formulate questions, especially when they feel so- about a particular treatment offered. We need a way
cially and intellectually inferior to their physicians and of acknowledging how oppressive circumstances can
when the physicians project an image of being busy interfere with autonomy, but this is not easily captured
with more important demands. Often, one needs some in traditional accounts.
information in order to know what further questions Finally, there are good reasons to be wary of the
to ask, and large gaps in perspective between patients ways in which the appearance of choice is used to mask
and their health care providers may result in a break- the normalizing powers of medicine and other health-
down in communication because of false assumptions related institutions. As Michel Foucault (1979, 1980)
by either participant. suggests, in modern societies the illusion of choice can
The fourth condition, the one that demands free- be part of the mechanism for controlling behaviour.
dom from coercion in exercising choice, is extremely Indeed, it is possible that bioethical efforts to guar-
difficult to evaluate when the individual in question antee the exercise of individual informed choice may
is oppressed. The task becomes even trickier if the actually make the exercise of medical authority even
choice is in a sphere that is tied to her oppression. The more powerful and effective than it would be under
condition of being oppressed can be so fundamentally more traditionally paternalistic models. In practice, the
restrictive that it is distorting to describe as autono- ideal of informed choice amounts to assuring patients
mous some specific choices made under such condi- of the opportunity to consent to one of a limited list
tions. For example, many women believe they have of relatively similar, medically encouraged procedures.
no real choice but to seek expensive, risky cosmetic Thus, informed consent procedures aimed simply at
surgery because they accurately perceive that their op- protecting autonomy in the narrow sense of specific
portunities for success in work or love depend on their choice among pre-selected options may ultimately
more closely approximating some externally defined serve to secure the compliance of docile patients who
standard of beauty. Similar sorts of questions arise with operate under the illusion of autonomy by virtue of
respect to some women's choice of dangerous, un- being invited to consent to procedures they are socially
proven experiments in new reproductive technologies encouraged to choose. Unless we find a way of identify-
because continued childlessness can be expected to ing a deeper sense of autonomy than that associated
have devastating consequences for their lives. In other with the expression of individual preference in select-
cases, women sometimes choose to have abortions being among a limited set of similar options, we run the
cause they fear that giving birth will involve them in risk of struggling to protect not patient autonomy but
unwanted and lifelong relationships with abusive part- the very mechanisms that ensure compliant medical
ners. Some women have little access to contraceptives consumers, preoccupied with the task of selecting
and find themselves choosing sterilization as the most among a narrow range of treatments.
effective way of resisting immediate demands of their
partners even if they might want more children in the Focus on the Individual
future. Or, some women seek out prenatal diagnosis
and selective abortion of cherished fetuses because they A striking feature o f most bioethical discussions about
realize that they cannot afford to raise a child born with patient autonomy is their exclusive focus on individual
a serious disability, though they would value such a patients; this pattern mirrors medicine's consistent ten-
child themselves. Many middle-class Western women dency to approach illness as primarily a problem of par-
choose hormone replacement therapy at menopause ticular patients.8 Similar problems are associated with
because they recognize that their social and economic each discipline. Within the medical tradition, suffering
lives may be threatened if they appear to be aging too is located and addressed in the individuals who experi-
quickly. When a woman's sense of herself and her range ence it rather than in the social arrangements that may
of opportunities have been oppressively constructed be responsible for causing the problem. Instead of ex-
in ways that (seem to) leave her little choice but to ploring the cultural context that tolerates and even sup-
pursue all available options in the pursuit of beauty or ports practices such as war, pollution, sexual violence,
childbearing or when she is raised in a culture that ties and systemic unemployment—practices that contribute
her own sense of herself to external norms of physical to much of the illness that occupies modern medicine—
appearance or fulfillment associated with childbear- physicians generally respond to the symptoms troub-
ing or, conversely, when having a(nother) child will ling particular patients in isolation from the context
impose unjust and intolerable costs on her, it does not that produces these conditions. Apart from population-
seem sufficient to restrict our analysis to the degree based epidemiological studies (which, typically, restrict
their focus to a narrow range of patterns of illness and personalized health care must continue to be made
often exclude or distort important social dimensions), available to those who become ill. Further, I want to be
medicine is primarily oriented toward dealing with in- clear that my critique does not imply that physicians
dividuals who have become ill (or pregnant, [in] fertile, or other direct care providers are necessarily the ones
or menopausal). This orientation directs the vast major- who ought to be assuming the task of identifying the
ity of research money and expertise toward the things social and environmental causes of disease. Health
that can be done to change the individual, but it often care training, and especially the training of physicians,
ignores key elements at the source of the problems. is directed at developing the requisite skills for the
For example, physicians tend to respond to in- extremely important work of caring for individuals
fertility either by trivializing the problem and telling who become ill. The responsibility for investigating
women to go home and "relax," or by prescribing the social causes of illness and for changing hazard-
hormonal and surgical treatment of particular women, ous conditions is a social one that is probably best met
rather than by demanding that research and public by those who undertake different sorts of training and
health efforts be aimed at preventing pelvic inflamma- study. The problem is that medicine, despite the limits
tory disease, which causes many cases of infertility, or of its expertise and focus, is the primary agent of health
by encouraging wide public debate (or private reflec- care activity in our society, and physicians are granted
tions) on the powerful social pressures to reproduce significant social authority to be the arbiters of health
that are directed at women. In similar fashion, the policy. Hence, when medicine makes the treatment of
mainstream scientific and medical communities re- individuals its primary focus, we must understand that
spond to the growth of breast cancer rates by promot- important gaps are created in our society's ability to
ing individual responsibility for self-examination and understand and promote good health.
by searching for the gene(s) that makes some women In parallel fashion, autonomy-focused bioethics
particularly susceptible to the disease; when it is found concentrates its practitioners' attention on the prefer-
in a patient, the principal medical therapy available ences of particular patients, and it is, thereby, complicit
is to perform "prophylactic" double mastectomies. in the individualistic orientation of medicine. It asks
Few physicians demand examination of the potential health care providers to ensure that individual patients
contributory role played by the use of pesticides or have the information they need to make rational deci-
chlorine, or the practice of feeding artificial hormones sions about their health care, yet it does not ask the
to agricultural animals. Or they deal with dramatically necessary questions about the circumstances in which
increased skin cancer rates by promoting the personal such decisions are made. The emphasis most bioethi-
use of sunscreens while resigning themselves to the cists place on traditional, individualistic understand-
continued depletion of the ozone layer. In another ings of autonomy reinforces the tendency of health
area, health care professionals generally deal with care providers and ethicists to neglect exploration of
the devastating effects of domestic violence by patch- the deep social causes and conditions that contribute
ing up its victims, providing them with medications to health and illness. Moreover, it encourages patients
to relieve depression and advice to move out of their to see their own health care decisions in isolation from
homes, and devising pathological names for victims those of anyone else, thereby increasing their sense of
who stay in violent relationships ("battered woman vulnerability and dependence on medical authority.
syndrome" and "self-defeating personality disorder"), The narrow individual focus that characterizes the
but few actively challenge the sexism that accepts male central traditions within both medicine and bioethics
violence as a "natural" response to frustration and fears obscures our need to consider questions of power,
of abandonment. dominance, and privilege in our interpretations and
Some qualifications are in order. Clearly these are responses to illness and other health-related matters as
crude and imprecise generalizations. They describe a well as in our interpretations of the ideal of autonomy.
general orientation of current health practices, but they These ways of structuring thought and practice make it
certainly do not capture the work of all those involved difficult to see the political dimensions of illness, and,
in medical research and practice. Fortunately, there are in a parallel way, they obscure the political dimensions
practitioners and researchers engaged in the very sorts of the conventional criteria for autonomous delibera-
of investigation I call for, but they are exceptional, not tion. As a result, they interfere with our ability to iden-
typical. Moreover, I do not want to imply that med- tify and pursue more effective health practices while
icine should simply abandon its concern with treating helping to foster a social environment that ignores and
disease in individuals. I understand that prevention tolerates oppression. In both cases, a broader political
strategies will not eliminate all illness and I believe that perspective is necessary if we are to avoid the problems
II aIII 111111
40 2 Medical Decision-Making
created by restricting our focus to individuals apart unjust barriers to care or may be acting in response to
from their location. oppressive circumstances; traditional conceptions are
Feminism offers just such a broader perspective. In inadequate to the extent that they make invisible the
contrast to the standard approaches in bioethics, fem- oppression that structures such decisions. By focus-
inism raises questions about the social basis for deci- ing only on the moment of medical decision-making,
sions about health and health care at all levels. Here, as traditional views fail to examine how specific decisions
elsewhere, feminists are inclined to ask whose interests are embedded within a complex set of relations and
are served and whose are harmed by the traditional policies that constrain (or, ideally, promote) an individ-
ways of structuring thought and practice. By asking ual's ability to exercise autonomy with respect to any
these questions, we are able to see how assumptions of particular choice.
individual-based medicine help to preserve the social To understand this puzzle it is necessary to dis-
and political status quo. For example, the current tinguish between agency and autonomy. To exercise
taxonomy in Canada designates certain sorts of condi- agency, one need only exercise reasonable choice." The
tions (e.g., infertility, cancer, heart disease, anxiety) as women who choose some of the controversial practices
appropriate for medical intervention, and it provides discussed (e.g., abortion to avoid contact with an abu-
grounds for ensuring that such needs are met. At the sive partner, cosmetic surgery to conform to artificial
same time, it views other sorts of conditions (e.g., norms of beauty, use of dangerous forms of reproduct-
malnutrition, fear of assault, low self-esteem) as falling ive technology) are exercising agency, clearly they are
beyond the purview of the health care system and, making choices, and, often, those choices are rational
therefore, as ineligible to draw on the considerable re- under the circumstances. They also meet the demands
sources allocated to the delivery of health services.9 In of conventional notions of autonomy that ask only that
this way, individualistic assumptions support a system anyone contemplating such procedures be competent,
that provides expert care for many of the health com- or capable of choosing (wisely), have available infor-
plaints of those with greatest financial privilege while mation current practice deems relevant, and be free of
dismissing as outside the scope of health care many of direct coercion. But insofar as their behaviour accepts
the sources of illness that primarily affect the disadvan- and adapts to oppression, describing it as autonomous
taged. A more social vision of health would require us seems inadequate. Together, the habits of equating
to investigate ways in which non-medical strategies, agency (the making of a choice) with autonomy (self-
such as improving social and material conditions for governance) and accepting as given the prevailing social
disadvantaged groups, can affect the health status of arrangements have the effect of helping to perpetuate
different segments of the community.1° oppression: when we limit our analysis to the quality
None of the concerns I have identified argues of an individual's choice under existing conditions (or
against maintaining a strong commitment to autonomy when we fail to inquire why some people do not even
in bioethical deliberations. In fact, I have no wish to seek health services), we ignore the significance of op-
abandon this ideal (just as I have no desire to abandon pressive conditions. Minimally, autonomous persons
patient-centred medical care). I still believe that a prin- should be able to resist oppression—not just act in
ciple of respect for patient autonomy is an important compliance with it—and be able to refuse the choices
element of good patient care. Moreover, I believe that oppression seems to make nearly irresistible. Ideally,
appeal to a principle of respect for autonomy can be an they should be able to escape from the structures of
important instrument in challenging oppression, and it oppression altogether and create new options that are
can actually serve as the basis for many of the feminist not defined by these structures either positively or
criticisms I present with respect to our current health negatively.
care system. In order to ensure that we recognize and address
What these criticisms do suggest, however, is that the restrictions that oppression places on people's
we must pursue a more careful and politically sensitive health choices, then, we need a wider notion of auton-
interpretation of the range of possible restrictions on omy that will allow us to distinguish genuinely autono-
autonomy than is found in most of the non-feminist mous behaviour from acts of merely rational agency.
bioethics literature. We need to be able to look at This conception must provide room to challenge the
specific decisions as well as the context that influ- quality of an agent's specific decision-making ability
ences and sometimes limits such decisions. Many of and the social norms that encourage agents to partici-
the troublesome examples I review above are entirely pate in practices that may be partially constitutive of
compatible with traditional conceptions of autonomy their oppression.'2 A richer, more politically sensitive
even though the patients in question may be facing standard of autonomy should make visible the impact
of oppression on a person's choices as well as on her distorting because, in fact, no one is fully independent.
very ability to exercise autonomy fully. Such a concep- As well, they observe that this model is exclusionary
tion has the advantage of allowing us to avoid the trap because those who are most obviously dependent on
of focusing on the supposed flaws of the individual others (e.g., because of disability or financial need)
who is choosing under oppressive circumstances (e.g., seem to be disqualified from consideration in ways
by dismissing her choices as "false consciousness"), for that others are not. Many feminists object that the
it is able to recognize that such choices can be reason- view of individuals as isolated social units is not only
able for the agent. Instead, it directs our attention to the false but impoverished: much of who we are and what
conditions that shape the agent's choice and it makes we value is rooted in our relationships and affinities
those conditions the basis of critical analysis. with others. Also, many feminists take issue with the
The problems that I identify with the conventional common assumption that agents are single-mindedly
interpretation of patient autonomy reveal a need to self-interested, when so much of our experience is de-
expand our understanding of the types of forces that voted to building or maintaining personal relationships
interfere with a patient's autonomy. On non-feminist and communities.14
accounts, these are irrationality, failure to recognize If we are to effectively address these concerns, we
that a choice is called for, lack of necessary information, need to move away from the familiar Western under-
and coercion (including psychological compulsion). standing of autonomy as self-defining, self-interested,
Since each of these conditions must be reinterpreted to and self-protecting, as if the self were simply some
allow for the ways in which oppression may be operat- special kind of property to be preserved." Under
ing, we must add to this list recognition of the costs most popular interpretations, the structure of the
and effects of oppression and of the particular ways in autonomy—heteronomy framework (governance by self
which oppression is manifested. But we must do more or by others) is predicated on a certain view of persons
than simply modify our interpretation of the four cri- and society in which the individual is thought to be
teria reviewed above. We also need an understanding somehow separate from and to exist independently
of the ways in which a person can be encouraged to of the larger society, each person's major concern is
develop (or discouraged from developing) the ability to to be protected from the demands and encroachment
exercise autonomy. For this task, we need to consider of others. This sort of conception fails to account for
the presence or absence of meaningful opportunities the complexity of the relations that exist between per-
to build the skills required to be able to exercise au- sons and their culture. It idealizes decisions that are
tonomy well (Meyers 1989), including the existence of free from outside influence without acknowledging
appropriate material and social conditions. In addition, that all persons are, to a significant degree, socially
our account should reflect the fact that many decision- constructed, that their identities, values, concepts, and
makers, especially women, place the interests of others perceptions are, in large measure, products of their
at the centre of their deliberations. Such an analysis social environment.
will allow us to ensure that autonomy standards reflect Since notions of the self are at the heart of au-
not only the quality of reasoning displayed by a patient tonomy discussions, alternative interpretations of
at the moment of medical decision-making but also the autonomy must begin with an alternative conception
circumstances that surround this decision-making. of the self. Curiously, despite its focus on individuals,
standard interpretations of autonomy have tended to
A Relational Alternative think of selves as generic rather than distinctive beings.
In the traditional view, individuals tend to be treated
A major reason for many of the problems identified as interchangeable in that no attention is paid to the
with the autonomy ideal is that the term is commonly details of personal experience. Hence, there is no space
understood to represent freedom of action for agents within standard conceptions to accommodate import-
who are paradigmatically regarded as independent, ant differences among agents, especially the effects that
self-interested, and self-sufficient. As such, it is part of a oppression (or social privilege) has on a person's ability
larger North American cultural ideal of competitive in- to exercise autonomy. In order to capture these kinds of
dividualism in which every citizen is to be left "free" to social concerns, some feminists have proposed turning
negotiate "his" way through the complex interactions to a relational conception of personhood that recog-
of social, economic, and political life." The feminist lit- nizes the importance of social forces in shaping each
erature is filled with criticism of such models of agency person's identity, development, and aspirations.m Fol-
and autonomy. for example, many feminists object lowing this suggestion, I now explore a relational inter-
that this ideal appeals to a model of personhood that is pretation of autonomy that is built around a relational
II 111 11111I
e 2 Medical Decision-Making
conception of the self that is explicitly feminist in its oppression. Feminists have long understood that one of
conception. the most insidious features of oppression is its tendency
Under relational theory, selfhood is seen as an to become internalized in the minds of its victims. This
ongoing process, rather than as something static or is because internalized oppression diminishes the cap-
fixed. Relational selves are inherently social beings that acity of its victims to develop self-respect, and, as sev-
are significantly shaped and modified within a web eral feminists have argued, reduced (or compromised)
of interconnected (and sometimes conflicting) rela- self-respect undermines autonomy by undermining
tionships. Individuals engage in the activities that are the individual's sense of herself as capable of making
constitutive of identity and autonomy (e.g., defining, independent judgments (Meyers 1989; Dillon 1992;
questioning, revising, and pursuing projects) within a Benson 1991, 1994). Moreover, as Susan Babbitt (1993,
configuration of relationships, both interpersonal and 1996) has argued, these oppression-induced barriers
political. By including attention to political relation- to autonomy cannot necessarily be rectified simply by
ships of power and powerlessness, this interpretation providing those affected with more information or by
of relational theory provides room to recognize how the removing explicit coercive forces (as the traditional
forces of oppression can interfere with an individual's view assumes). When the messages of reduced self-
ability to exercise autonomy by undermining her sense worth are internalized, agents tend to lose the ability
of herself as an autonomous agent and by depriving her even to know their own objective interests. According
of opportunities to exercise autonomy. Thus, it is able to Babbitt, in such cases transformative experiences
to provide us with insight into why it is that oppressed can be far more important to autonomy than access
people often seem less autonomous than others even to alternative information. Feminist theory suggests,
when offered a comparable range of choices. Under a then, that women and members of other oppressed
relational view, autonomy is best understood to be a groups can be helped to increase their autonomy skills
capacity or skill that is developed (and constrained) by being offered more opportunities to exercise those
by social circumstances. It is exercised within relation- skills and a supportive climate for practising them
ships and social structures that jointly help to shape the (Meyers 1989), by being provided with the opportun-
individual while also affecting others' responses to her ity to develop stronger senses of self-esteem (Benson
efforts at autonomy. 1994; Dillon 1992; Meyers 1989), by having the op-
Diana Meyers (1989) has developed one such portunity for transformative experiences that make vis-
theory of personal autonomy. She argues that auton- ible the forces of oppression (Babbitt 1993, 1996), and
omy involves a particular competency that requires the by having experiences of making choices that are not
development of specific skills. As such, it can be either influenced by the wishes of those who dominate them
enhanced or diminished by the sort of socialization (Babbitt 1993, 1996).
the agent experiences. Meyers shows how the specific Autonomy requires more than the effective exercise
gender socialization most (Western) women undergo of personal resources and skills, however, generally, it
trains them in social docility and rewards them for also demands that appropriate structural conditions be
defining their interests in terms of others, thereby rob- met. Relational theory reminds us that material restric-
bing them of the opportunity to develop the essential tions, including very restricted economic resources,
capacity of self-direction Such training relegates most ongoing fear of assault, and lack of educational op-
women to a category she labels "minimally autono- portunity (i.e., the sorts of circumstances that are often
mous" (as distinct from her more desirable categories part of the condition of being oppressed), constitute
of medially autonomous and fully autonomous). Rela- real limitations on the options available to the agent.
tional theory allows us to appreciate how each relation- Moreover, it helps us to see how socially constructed
ship a person participates in plays a role in fostering stereotypes can reduce both society's and the agent's
or inhibiting that individual's capacity for autonomous sense of that person's ability to act autonomously.
action by encouraging or restricting her opportunities Relational theory allows us to recognize how such
to understand herself as an autonomous agent and to diminished expectations readily become translated into
practice exercising the requisite skills. Such a concep- diminished capacities.
tion makes clear the importance of discovering the The relational interpretation I favour is feminist
ways in which oppression often reduces a person's abil- in that it takes into account the impact of social and
ity to develop and exercise the skills that are necessary political structures, especially sexism and other forms
for achieving a reasonable degree of autonomy. of oppression, on the lives and opportunities of indi-
For instance, relational theory allows us to see viduals. It acknowledges that the presence or absence
the damaging effects on autonomy of internalized of a degree of autonomy is not just a matter of being
Sherwin: A Relational Approach to Autonomy in Health Care
offered a choice. It also requires that the person have the requisite autonomy skills. The best way of course
had the opportunity to develop the skills necessary for to help oppressed people to develop autonomy skills
making the type of choice in question, the experience is to remove the conditions of their oppression. Short
of being respected in her decisions, and encourage- of that, long-term social projects can help to provide
ment to reflect on her own values. The society, not just educational opportunities to counter the psycho-
the agent, is subject to critical scrutiny under the rubric logical burdens of oppression. In the short term, it
of relational autonomy. may be necessary to spend more time than usual in
It is important, however, to avoid an account supporting patients in the deliberative process of
that denies any scope for autonomy on the part of decision-making and providing them with access to
those who are oppressed. Such a conclusion would relevant political as well as medical information when
be dangerous, since the widespread perception of they contemplate controversial procedures (e.g., in-
limited autonomy can easily become a self-fulfilling formation about the social dimensions of hormone
prophecy. Moreover, such a conclusion would be replacement therapy).
false. Many members of oppressed groups do manage Relational autonomy is not only about changing
to develop autonomy skills and, thus, are able to act the individual, however. It also demands attention to
autonomously in a wide variety of situations, though ways in which the range of choices before those who
the particular demands of acting autonomously under belong to oppressed groups can be modified to include
oppression are easily overlooked (Benson 1991). more non-oppressive options, that is, options that will
Some feminists, such as bell hooks (1990) and Sarah not further entrench their existing oppression (as often
Hoagland (1992), have observed that the marginal- happens, for example, when women choose cosmetic
ity associated with being oppressed can sometimes surgery or the use of many reproductive technologies).
provide people with better opportunities than are Whereas in traditional autonomy theory only the mode
available to more well-situated citizens for question- and quality of specific decisions are evaluated, femin-
ing social norms and devising their own patterns of ist relational autonomy regards the range and nature
resistance to social convention. Because those who are of available and acceptable options as being at least as
especially marginalized (e.g., those who are multiply important as the quality of specific decision-making.
oppressed or who are "deviant" with respect to im- Only when we understand the ways in which oppres-
portant social norms) may have no significant social sion can infect the background or baseline conditions
privilege to lose, they are, sometimes, freer than others under which choices are to be made will we be able to
to demand changes in the status quo. They may be modify those conditions and work toward the possibil-
far more likely to engage in resistance to the norms of ity of greater autonomy by promoting non-oppressive
oppression than are those who derive some personal alternatives.
benefits from oppressive structures (e.g., middle-class, As in health matters, it is important in relational
able-bodied, married women). discussions not to lose sight of the need to continue
Still, we must not make the mistake of romanti- to maintain some focus on the individual. Relational
cizing the opportunities available to the oppressed. autonomy redefines autonomy as the social project it
An adequate conception of autonomy should afford is, but it does not deny that autonomy ultimately res-
individuals more than the opportunity to resist oppres- ides in individuals. Our attention to social and political
sion; it should also ensure that they have opportunities contexts helps deepen and enrich the narrow and im-
to actively shape their world. A relational conception poverished view of autonomy available under individ-
of autonomy seems better suited than the traditional ualistic conceptions, but it does not support wholesale
models to handle the complexities of such paradoxes neglect of the needs and interests of individuals in
because it encourages us to attend to the complex ways favour of broader social and political interests. Rather,
in which the detailed circumstances of an individual it can be seen as democratizing access to autonomy by
social and political circumstances can affect her ability helping to identify and remove the effects of barriers to
to act in different kinds of contexts. autonomy that are created by oppression. A relational
When relational autonomy reveals the disadvan- approach can help to move autonomy from the largely
tage associated with oppression in terms of autonomy, exclusive preserve of the socially privileged and see
the response should not be that others are thereby that it is combined with a commitment to social justice
licensed to make decisions for those who are opin order to ensure that oppression is not allowed to
pressed; this response would only increase their continue simply because its victims have been deprived
powerlessness. Rather, it demands attention to ways of the resources necessary to exercise the autonomy re-
in which oppressed people can be helped to develop quired to challenge it. . . .
II III uuJJ
O 2 Medical Decision-Making
Notes
1. Some Network members prefer the terms "contextual" 8. I focus primarily on medicine since it is the dominant health
or "situated" as a way of avoiding all confusion with profession and is responsible for organization of most health
those feminists who reserve the term "relational" to refer services in developed countries. Most health professions
exclusively to interpersonal relations. I feel that this involve a similar bias toward treatment of individuals, though
usage perpetuates the misleading sense that interpersonal some (e.g., social work) pride themselves on attending to
relations are themselves "apolitical." I have, therefore, social structures as well as individual need, and most health
chosen to insist on a thoroughly political reading of the professions, including medicine, include subspecialties
term "relational" that applies to both interpersonal and concerned with matters of public health.
more public sorts of relations. 9. Because health care is a provincial responsibility, there are
2. Informed choice suggests a wider scope for patient differences in the precise services offered from province to
autonomy than informed consent in that it includes the province and from one administration to the next within
possibility of patients' initiating treatment suggestions, provinces The examples here are broad generalizations.
where informed consent implies that the role of the patient 10. Such considerations do play a role in health care planning
is merely to consent to the treatment proposed by the at a governmental level where the focus shifts from
physician; further, informed choice makes more explicit medical interventions to the idea of health determinants,
that patients ought also to be free to refuse recommended but here, too, there is excessive attention paid to what
treatments as well as to accept them. the individual can and should be doing ("healthism") and
3. Many of these concerns are not exclusive to feminists; insufficient concern about promoting egalitarian social
several have also been raised by other sorts of critics. I call conditions.
them feminist because I came to these concerns through 11. The language of agency and autonomy is quite varied
a feminist analysis that attends to the role in society of within feminist (and other) discourse For example, the
systems of dominance and oppression, especially those term agency is used throughout the collection Provoking
connected with gender. Agents: Gender and Agency in Theory and Practice (Gardiner
4. This reduction may be a result of a tendency to collapse 1995) in ways that sometimes appear to overlap with my
the ideal of personal autonomy central to bioethics usage of relational autonomy. Susan Babbitt (1996), on the
discussions with the concept of moral autonomy other hand, seems to use the two terms in ways analogous
developed by Immanuel Kant. to the use here.
5. It is often taken as a truism in our culture that emotional 12. In addition, we need the conceptual space to be able
involvement constitutes irrationality, that emotions are to acknowledge that restrictive definitions of health
direct threats to rationality. It is hard to see, however, how sometimes preempt autonomy analysis by limiting the
decisions about important life decisions are improved if opportunity of some people even to enter the relatively
they are made without any emotional attachment to the well-funded health care system for assistance with
outcomes. problems (e.g., poverty) that affect their health.
6. Susan Babbitt (1996) argues that the traditional 13. The agent imagined in such cases is always stereotypically
conception of rationality is defined in terms of masculine
propositional understanding in ways that obscure the 14. Feminist discussion of these and other critiques can be
experiences and needs of opprecced people. found in Gilligan 1982; Baier 1985; Code 1991; and Held
7. For example, research priorities have led to the situation 1993.
where birth control pills are available only for women, and 15. See Nedelsky 1989 for discussion of this view and its
this increases the pressure on women seeking temporary limitations.
protection against pregnancy to take the pill even when it 16. For example, Baier 1985; Code 1991; and Held 1993.
endangers their health.
References
Babbitt, Susan. 1993. "Feminism and Objective Interests," 1994. "Free Agency and Self-Worth," in Journal of
in Feminist Epistemologies, ed. Linda Alcoff and Elizabeth Philosophy 91 (12): 650-68.
Potter. Routledge: New York. Code, Lorraine. 1991. What Can She Know? Feminist Theory
---. 1996. Impossible Dreams: Rationality, Integrity, and and the Construction of Knowledge. Cornell University
Moral Imagination. Westview Press: Boulder, CO. Press: Ithaca, NY.
Baier, Annette. 1985. "What Do Women Want in a Moral Dillon, Robin. 1992. "Toward a Feminist Conception of Self-
Theory?," in Nous 19 (1): 53-63. Respect," in Hypatia 7 (1): 52-69.
Benson, Paul. 1991. "Autonomy and Oppressive Socializa- Foucault, Michel. 1979. Discipline and Punish. Vintage: New
tion," in Social Theory and Practice 17 (3): 385-408. York.
—. 1980. Power/Knowledge, ed. Colin Gordon. Har- Practice, ed. Susan Browning Cole and Susan Coultrap-
vester: Brighton, England. McQuin. Indiana University Press: Bloomington.
Gardiner, Judith Regan. 1995. Provoking Agents: Gender and Agency hooks, bell. 1990. Yearning: Race, Gender; and Cultural Politics.
in Theory and Practice. University of Illinois Press: Chicago. Between the Lines: Toronto.
Gilligan, Carol. 1982. In a Different Voice: Psychological Theory Lloyd, Genevieve. 1984. The Man of Reason: "Male" and
and Women's Moral Development. Harvard University "Female" in Western Philosophy. University of Minnesota
Press: Cambridge, MA. Press: Minneapolis.
Held, Virginia 1993. Feminist Morality: Transforming Culture, Meyers, Diana T. 1989. Self, Society, and Personal Choice. Col-
Society, and Politics. University of Chicago Press: Chicago. umbia University Press: New York.
Hoagland, Sarah Lucia 1992. "Lesbian Ethics and Female Nedelsky, Jennifer. 1989. "Reconceiving Autonomy," in Yale
Agency," in Explorations in Feminist Ethics: Theory and Journal of Law and Feminism 1 (1): 7-36.
2.3 Informed Consent
The Nuts and Bolts of Obtaining even in those provinces in which legislation or regula-
tions focus on so-called signed, written consents.
Consent to Treatment Consent is actually a process that culminates in
LE. and F.A. Rozovsky the parties agreeing to the permissible scope of a diag-
nostic, medical, or surgical procedure. As noted in the
accompanying illustration, there are several steps to
Introduction
the consent process. Some of these are "mechanical" in
Consent to treatment is one o f the most hotly discussed nature while others involve the legal requirements for
medicolegal issues in medicine. Physicians complain obtaining consent.
that courts and legislative assemblies have imposed The consent process can be broken down into
undue burdens upon them in dealing with their pa- seven different categories of legal concern:
tients. Many physicians are fearful of litigation for fail-
ing to meet appropriate legal standards in obtaining 1. Voluntary Atmosphere for Consent;
consent to treatment. 2. Legal Capacity to Give Consent;
The law of consent need not be looked upon as 3. Mental Competency to Give Consent;
a burdensome task nor as an unwarranted invasion of 4. Adequate Disclosure o f Information;
the physician—patient relationship. Moreover, phys- 5. Candid Replies to Patient's Questions;
icians need not be unduly alarmed about the prospect 6. Consent Must Be Specific to Procedure; and
of consent lawsuits. Although there have been many 7. Documentation of Consent.
court cases involving consent issues, it is becoming dear
that plaintiffs must overcome significant hurdles to be Getting to know the patient requires the physician
successful. to obtain a thorough personal and family history regard-
When consent is obtained properly physicians need ing such matters as prior illnesses, known or suspected
not concern themselves with such litigation. Obtaining allergies to foods or medications, current medications
a proper authorization for treatment not only safeguards taken by the patient, etc. This enables the physician to
the rights of the patient, it forms the basis for a solid tailor the consent process to the specific patient. The
defence against consent lawsuits. The key is for doctors patient should not be viewed in a vacuum; rather, the
to know the necessary steps in obtaining consent. totality of the patient's surroundings must be taken into
account including his lifestyle, economic and financial
Consent Is a Process Not a Form status, the probability of success in the procedure, and
whether it is an elective or an imperative procedure.
When a patient signs a consent form many assume that What a reasonable person would want to know is
the document constitutes consent. This is not correct, measured in terms of what is considered significant or
Source From Econ Medical Professional Advisory Program, Bulletin No. 3, 1985, pp. 1-3. Reprinted with permission.
® 2 Medical Decision-Making
AUTHORIZE
ESTABLISH DETERMINE DISCLOSE DEFINE
& DOCUMENT
Voluntary Legal capacity Risks, benefits, With the patient Express, implied;
atmosphere Assess mental reasonable permissible verbally or
Establish competency to alternatives scope of the in writing
relationship determine if Adequate procedure to be Consistent
performed VALID
by getting to patient can give disclosure of -70 style of CONSENT
know patient consent information by documentation
answering
patient's
questions
Figure 1 The Consent Process
material information in his decision-making process. Candid Responses to Patient Inquiries

This would include:
Doctors should respond in a candid manner to patients'
1. The nature and purpose o f the proposed procedure; questions. Inquiries made by patients may be indicative
2. Probable risks and benefits associated with the of a potential lapse in the consent process. It may mean
procedure, including the probability of serious that a patient has not completely understood some
injury or death; aspect of the information, or it may signify a new con-
3. Reasonable alternative medical or surgical proced- cern. Whatever the reason, questions posed by patients
ures and risks associated with them; should be addressed in a straightforward manner.
4. The risks of foregoing all diagnostic procedures or
treatment; Authorization for the Procedure
5. The length of time, if any, the patient can postpone to Be Forfeited
the procedure and the probable risks involved in
doing so; The consent of the patient must be specific to the pro-
6. The probable length of time that the patient will cedure to be performed. An authorization for treatment
be hospitalized and unable to resume his employ- is not an open-ended invitation to doctors to help them-
ment and regular activities; selves. The authorization obtained at the conclusion of
7. The probable need for follow-up care and additional the consent process should be highly specific, outlining
surgery, radiation treatment, or chemotherapy; and what the doctor and staff are and are not permitted to do.
8. The right of the patient to withdraw consent. Some flexibility, however, can be built into the
authorization in terms of the permissible source of the
It is clear that in the treatment context the courts procedure. For example, in an exploratory operation
do not expect clinicians to disclose ALL known risks of the abdomen, it may be difficult to define the extent
and benefits. The prospect of remote risks need not be of the procedure. The surgeon can reach an agreement
revealed. The focus is on what the average, reasonable, with the patient regarding the taking of tissue speci-
and prudent person would want to know in the cir- mens for immediate pathology assessment, the removal
cumstances. To be sure, the circumstances can and do of organs, the resection of bowel, the performance of a
change the scope of the disclosure. If the procedure is colostomy or an ileostomy.
novel or experimental there should be a greater degree Under no circumstances should the physician extend
of information imparted to the patient. the scope of the procedure beyond its permissible limits.
Furthermore, clinicians need not disclose infor- The only exception to this general rule is if an unantici-
mation about facts that are readily known to the lay pated problem occurs requiring an immediate response.
public. Taking aspirin to relieve pain, a high temper- Even at this stage, however, the measures undertaken
ature is indicative of infection, or the like is the type must be reasonable in the context of the problem.
of information the clinician may generally assume the If, for example, in the course of unrelated surgery
public understands. There are exceptions, however, as the surgeon discovered a bleeding ulcer that requires
in the case of patients from less developed nations or prompt surgical attention, he could lawfully extend the
different cultures who cannot be expected to appreciate surgery to correct the newly discovered problem. How-
what is "common knowledge." ever, if the problem that is detected does not require an
Rozovsky: The Nuts and Bolts of Obtaining Consent to Treatment 47
immediate response, the physician cannot forge ahead jeopardizes the value of the documentation in legal de-
and treat it. He must first obtain either an authorization fence. It should never be deferred until sometime after
from the patient or his legal representative to extend the the procedure has been performed. The best practice
procedure which was consented to originally. is to record consent in a consistent fashion, keeping in
mind provincial and institutional standards for the type
of documentation to be employed.
Documentation of Consent
Unless provincial law specifically requires it, the patient's Conclusion
consent to treatment need not be given in writing. Con-
sent may be expressed or implied, verbally or in writ- The process of consent has several purposes, the most
ing. However, some form of documentation is necessary important of which is to educate the patient about im-
for purposes of legal defence. The documentation used pending diagnostic, medical, or surgical interventions
may be a consent form filed in the patient's record or a and to provide a firm basis for a legal defence to al-
detailed note written in the patient's chart outlining the legations of inadequate authorization for care. The best
dialogue between the doctor and the patient. protection from consent litigation is adherence to the
Consent should be documented when the patient rules for consent, including adequate documentation
gives his authorization for treatment. To do otherwise of an authorization for treatment.
The Concept of Informed Consent acquiesce in, yield to, or comply with an arrangement
or a proposal. He or she actively authorizes the proposal
Ruth R. Faden and Tom L Beauchamp in the act of consent. John may assent to a treatment plan
without authorizing it. The assent may be a mere sub-
What is an informed consent? Answering this question mission to the doctor's authoritative order, in which case
is complicated because there are two common, en- John does not call on his own authority in order to give
trenched, and starkly different meanings of "informed permission, and thus does not authorize the plan. In-
consent." That is, the term is analyzable in two pro- stead, he acts like a child who submits, yields, or assents
foundly different ways—not because of mere subtle to the school principal's spanking and in no way gives
differences of connotation that appear in different con- permission for or authorizes the spanking. Just as the
texts, but because two different conceptions of informed child merely submits to an authority in a system where
consent have emerged from its history and are still at the lines of authority are quite clear, so often do patients
work, however unnoticed, in literature on the subject. Accordingly, an informed consent in senses
In one sense, which we label senses, "informed should be defined as follows: an informed consent is
consent" is analyzable as a particular kind of action an autonomous action by a subject or a patient that
by individual patients and subjects: an autonomous authorizes a professional either to involve the subject in
authorization. In the second sense, sense2, informed research or to initiate a medical plan for the patient (or
consent is analyzable in terms of the web of cultural both). We can whittle down this definition by saying
and policy rules and requirements of consent that col- that an informed consent in sense, is given if a patient
lectively form the social practice of informed consent or subject with (1) substantial understanding and (2) in
in institutional contexts where groups of patients and substantial absence of control by others (3) intention-
subjects must be treated in accordance with rules, poli- ally (4) authorizes a professional (to do intervention I).
cies, and standard practices. Here, informed consents All substantially autonomous acts satisfy condi-
are not always autonomous acts, nor are they always in tions 1-3; but it does not follow from that analysis
any meaningful respect authorizations. alone that all such acts satisfy 4. The fourth condition
is what distinguishes informed consent as one hind
Senses: Informed Consent as Autonomous of autonomous action. (Note also that the definition
Authorization restricts the kinds of authorization to medical and
research contexts.) A person whose act satisfies con-
The idea of an informed consent suggests that a patient ditions 1-3 but who refuses an intervention gives an
or subject does more than express agreement with, informed refusal.
Source: From A History and Theory of Informed Consent (Oxford: Oxford University Press: 1986), 145-9. © 1986 by Oxford
University Press. By permission of Oxford University Press, USA.
II rill 111111
The Problem of Shared Decision-Making the exclusive province of the physician, and thus some
This analysis of informed consent in sense, is delib- decisions are likely always to remain subject exclu-
erately silent on the question of how the authorizer sively to the physician's authorization. Moreover, in the
and agent(s) being authorized arrive at an agreement uncommon situation, a patient could autonomously
about the performance of "I." Recent commentators authorize the physician to make all decisions about
on informed consent in clinical medicine, notably Jay medical treatment, thus giving his or her informed
Katz and the Presidenth Commission, have tended to consent to an arrangement that scarcely resembles the
equate the idea of informed consent with a model of sharing of decision-making between doctor and patient.
"shared decision-making" between doctor and patient.
The Presidents Commission titles the first chapter of its Authorization
report on informed consent in the patient—practitioner In authorizing, one both assumes responsibility for
relationship "Informed Consent as Active, Shared what one has authorized and transfers to another
Decision Making," while in Katz's work "the idea of one's authority to implement it. There is no informed
informed consent" and "mutual decision-making" are consent unless one understands these features of the
treated as virtually synonymous terms.' act and intends to perform that act. That is, one must
There is of course an historical relationship in clin- understand that one is assuming responsibility and
ical medicine between medical decision-making and warranting another to proceed.
informed consent. The emergence of the legal doctrine To say that one assumes responsibility does not
of informed consent was instrumental in drawing atten- quite locate the essence of the matter, however, because
tion to issues of decision-making as well as authority a transfer of responsibility as well as of authority also
in the doctor—patient relationship. Nevertheless, it occurs. The crucial element in an authorization is that
is a confusion to treat informed consent and shared the person who authorizes uses whatever right, power,
decision-making as anything like synonymous. For one or control he or she possesses in the situation to endow
thing, informed consent is not restricted to clinical another with the right to act. In so doing, the author-
medicine. It is a term that applies equally to biomed- izer assumes some responsibility for the actions taken
ical and behavioural research contexts where a model by the other person. Here one could either authorize
of shared decision-making is frequently inappropriate. broadly so that a person can act in accordance with
Even in clinical contexts, the social and psychological general guidelines, or narrowly so as to authorize only a
dynamics involved in selecting medical interventions particular, carefully circumscribed procedure.
should be distinguished from the patient's authorization.
We endorse Katzs view that effective communica- Sense2: Informed Consent as
tion between professional and patient or subject is often Effective Consent
instrumental in obtaining informed consents (sense),
but we resist his conviction that the idea o f informed By contrast to sensep sense2, or effective consent, is
consent entails that the patient and physician "share a policy-oriented sense whose conditions are not
decision-making," or "reason together," or reach a con- derivable solely from analyses of autonomy and au-
sensus about what is in the patient's best interest. This thorization, or even from broad notions of respect for
is a manipulation of the concept from a too singular autonomy. "Informed consent" in this second sense
and defined moral perspective on the practice of medi- does not refer to autonomous authorization, but to a
cine that is in effect a moral program for changing the legally or institutionally effective (sometimes mislead-
practice. Although the patient and physician may reach ingly called valid) authorization from a patient or a
a decision together, they need not. It is the essence of subject. Such an authorization is "effective" because it
informed consent in sense, only that the patient or sub- has been obtained through procedures that satisfy the
ject authorizes autonomously; it is a matter of indifference rules and requirements defining a specific institutional
where or how the proposal being authorized originates. practice in health care or in research.
For example, one might advocate a model of shared The social and legal practice of requiring profes-
decision-making for the doctor—patient relationship sionals to obtain informed consent emerged in insti-
without simultaneously advocating that every medical tutional contexts, where conformity to operative rules
procedure requires the consent of patients. Even re- was and still is the sole necessary and sufficient condi-
lationships characterized by an ample slice of shared tion of informed consent. Any consent is an informed
decision-making, mutual trust, and respect would and consent in sense, if it satisfies whatever operative rules
should permit many decisions about routine and low- apply to the practice of informed consent. Sense, re-
risk aspects of the patient's medical treatment to remain quirements for informed consent typically do not focus
Faden/Beauchamp: The Concept of Informed Consent 49
on the autonomy of the act of giving consent (as sense, rules by appeal to something like sense,. For example,
does), but rather on regulating the behaviour of the disclosure conditions for informed consent are central
consent-seeker and on establishing procedures and rules to the history of "informed consent" in sense2, because
for the context of consent. Such requirements of pro- disclosure has traditionally been a necessary condition
fessional behaviour and procedure are obviously more of effective informed consent (and sometimes a sufficient
readily monitored and enforced by institutions. condition!). The legal doctrine of informed consent is
However, because formal institutional rules such primarily a law of disclosure; satisfaction of disclosure
as federal regulations and hospital policies govern rules virtually consumes "informed consent" in law This
whether an act of authorizing is effective, a patient or should come as no surprise, because the legal system
subject can autonomously authorize an intervention, needs a generally applicable informed consent mechan-
and so give an informed consent in sense,, and yet not ism by which injury and responsibility can be readily and
effectively authorize that intervention in sense2. fairly assessed in court. These disclosure requirements in
Consider the following example. Carol and Martie the legal and regulatory contexts are not conditions of "in-
are 19-year-old, identical twins attending the same formed consent" in sensei; indeed disclosure may be en-
university. Martie was born with multiple birth de- tirely irrelevant to giving an informed consent in sense,.
fects, and has only one kidney. When both sisters are If a person has an adequate understanding of relevant
involved in an automobile accident, Carol is not badly information without benefit of a disclosure, then it makes
hurt, but her sister is seriously injured. It is quickly no difference whether someone discloses that information.
determined that Martie desperately needs a kidney Other sense2 rules besides those of disclosure have
transplant. After detailed discussions with the trans- been enforced. These include rules requiring evidence of
plant team and with friends, Carol consents to be the adequate comprehension of information and the afore-
donor. There is no question that Carol's authorization mentioned rules requiring the presence of auditor wit-
of the transplant surgery is substantially autonomous. nesses and mandatory waiting periods. Sense2 informed
She is well informed and has long anticipated being in consent requirements generally take the form of rules fo-
just such a circumstance. She has had ample opportun- cusing on disclosure, comprehension, the minimization
ity over the years to consider what she would do were of potentially controlling influences, and competence.
she faced with such a decision Unfortunately, Carol's These requirements express the present-day mainstream
parents, who were in Nepal at the time of the accident, conception in the federal government of the United States.
do not approve of her decision. Furious that they were They are also typical of international documents and state
not consulted, they decide to sue the transplant team regulations, which all reflect a sense2 orientation
and the hospital for having performed an unauthorized
surgery on their minor daughter. (In this state the legal
The Relationship Between Sense, and Sense2
age to consent to surgical procedures is 21.)
According to our analysis, Carol gave her informed A sense, "informed consent" can fail to be an informed
consent in sense, to the surgery, but she did not give consent in sense2 by a lack of conformity to applicable
her informed consent in sense2. That is, she autono- rules and requirements. Similarly, an informed consent
mously authorized the transplant and thereby gave an in sense2 may not be an informed consent in sense,.
informed consent in sense, but did not give a consent The rules and requirements that determine sense2 con-
that was effective under the operative legal and institu- sents need not result in autonomous authorizations at
tional policy, which in this case required that the person all in order to qualify as informed consents.
consenting be a legally authorized agent. Examples of Such peculiarities in informed consent law have
other policies that can define sense2 informed consent led Jay Katz to argue that the legal doctrine of "in-
(but not sense) include rules that consent be witnessed formed consent" bears a "name" that "promises much
by an auditor or that there be a one-day waiting period more than its construction in case law has delivered."
between solicitation of consent and implementation of He has argued insightfully that the courts have, in
the intervention in order for the person's authorization effect, imposed a mere duty to warn on physicians, an
to be effective. Such rules can and do vary, both within obligation confined to risk disclosures and statements
the United States by jurisdiction and institution, and of proposed interventions. He maintains that "This ju-
across the countries of the world. dicially imposed obligation must be distinguished from
Medical and research codes, as well as case law and the idea of informed consent, namely, that patients have
federal regulations, have developed models of informed a decisive role to play in the medical decision-making
consent that are delineated entirely in a sense2 format, process. The idea of informed consent, though alluded
although they have sometimes attempted to justify the to also in case law, cannot be implemented, as courts
■ II I 1.11
I I Ell 111111
have attempted, by only expanding the disclosure re- institutional requirements for informed consent in sense2
quirements." By their actions and declarations, Katz be- should be intended to maximize the likelihood that the
lieves, the courts have made informed consent a "cruel conditions of informed consent in sense, will be satisfied.
hoax" and have allowed "the idea of informed consent A major problem at the policy level, where rules
. . . to wither on the vine."' and requirements must be developed and applied in the
The most plausible interpretation of Katz's con- aggregate, is the following: the obligations imposed to
tentions is through the sense,/sense2 distinction. If a enable patients and subjects to make authorization deci-
physician obtains a consent under the courts' criteria, sions must be evaluated not only in terms of the demands
then an informed consent (sense2) has been obtained. of a set of abstract conditions of "true" or sense, informed
But it does not follow that the courts are using the right consent, but also in terms of the impact of imposing such
standards, or sufficiently rigorous standards in light of a obligations or requirements on various institutions with
stricter autonomy-based model—or Idea" as Katz puts their concrete concerns and priorities. One must take ac-
it—of informed consent (sense).3 If Katz is correct that count of what is fair and reasonable to require of health
the courts have made a mockery of informed consent care professionals and researchers, the effect of alterna-
and of its moral justification in respect for autonomy tive consent requirements on efficiency and effectiveness
then of course his criticisms are thoroughly justified. At in the delivery of health care and the advancement of
the same time, it should be recognized that people can science, and—particularly in medical care—the effect of
proffer legally or institutionally effective authorizations requirements on the welfare of patients. Also relevant are
under prevailing rules even if they fall far short of the considerations peculiar to the particular social context,
standards implicit in sense,. such as proof, precedent, or liability theory in case law, or
Despite the differences between sense, and sense2, regulatory authority and due process in the development
a definition of informed consent need not fall into one of federal regulations and IRB consent policies.
or the other class of definitions. It may conform to Moreover, at the sense2 level, one must resolve
both. Many definitions of informed consent in policy not only which requirements will define effective
contexts reflect at least a strong and definite reliance on consent; one must also settle on the rules stipulating
informed consent in sense,. Although the conditions the conditions under which effective consent must be
of sense, are not logically necessary conditions for obtained. In some cases, hard decisions must be made
sense2, we take it as morally axiomatic that they ought about whether requirements of informed consent (in
to serve—and in fact have served—as the benchmark sense2) should be imposed at all, even though informed
or model against which the moral adequacy of a defin- consent (in sense2) could realistically and meaningfully
ition framed for sense2 purposes is to be evaluated. This be obtained in the circumstances and could serve as a
position is, roughly speaking, Katz's position. model for institutional rules. For example, should there
A defence of the moral viewpoint that policies be any consent requirements in the cases of minimal-
governing informed consent in sense2 should be formu- risk medical procedures and research activities?
lated to conform to the standards of informed consent This need to balance is not a problem for informed
in sense, is not hard to express. The goal of informed consent in sensel, which is not policy-oriented. Thus,
consent in medical care and in research—that is, the it is possible to have a morally acceptable set of require-
purpose behind the obligation to obtain informed ments for informed consent in sense2 that deviates con-
consent—is to enable potential subjects and patients to siderably from the conditions of informed consent in
make autonomous decisions about whether to grant or sense,. However, the burden of moral proof rests with
refuse authorization for medical and research interven- those who defend such deviations since the primary
tions. Accordingly, embedded in the reason for having moral justification of the obligation to obtain informed
the social institution of informed consent is the idea that consent is respect for autonomous action.
Notes
1. President's Commission. Making Health Care Decisions, 2. Katz, Jay. 1980. "Disclosure and Consent," in Genetics and
vol. 1, 15; Katz, Jay. 1984. The Silent World of Doctor and the Law iI, ed. A. Milunsky and G. Annas. Plenum Press:
Patient. The Free Press: New York, 87; and Katz, Jay. 1973. New York, 122, 128.
"The Regulation of Human Research—Reflections and 3. We have already noted that Katz's "idea" of informed
Proposals," in Clinical Research 21: 758-91. Katz does not consent as the active involvement of patients in the
provide a sustained analysis of joint or shared decision- medical decision-making process is different from our
making, and it is unclear precisely how he would relate senses.
this notion to informed consent.
Brody: Transparency: Informed Consent in Primary Care 8
Transparency: Informed Consent patient autonomy, one that interrupts the process of
medical care.
in Primary Care However, available empirical evidence strongly
Howard Brody suggests that this is precisely how physicians currently
view informed consent practices. Informed consent is
While the patient's right to give informed consent to still seen as bureaucratic legalism rather than as part
medical treatment is now well established both in US of patient care. Physicians often deny the existence of
law and in biomedical ethics, evidence continues to realistic treatment alternatives, thereby attenuating the
suggest that the concept has been poorly integrated perceived need to inform the patient of meaningful
into American medical practice, and that in many options. While patients may be informed, efforts are
instances the needs and desires of patients are not seldom made to assess accurately the patient's actual
being well met by current policies.' It appears that the need or desire for information, or what the patient
theory and the practice of informed consent are out then proceeds to do with the information provided.
of joint in some crucial ways. This is particularly true Physicians typically underestimate patients' desire to be
for primary care settings, a context typically ignored informed and overestimate their desire to be involved
by medical ethics literature, but where the majority of in decision-making. Physicians may also view informed
doctor—patient encounters occur. Indeed, some have consent as an empty charade, since they are confident
suggested that the concept of informed consent is vir- in their abilities to manipulate consent by how they dis-
tually foreign to primary care medicine where benign cuss or divulge in formation.3
paternalism appropriately reigns and where respect for A third assumption is that there are important
patient autonomy is almost completely absent.2 differences between the practice of primary care medi-
It is worth asking whether current legal standards cine and the tertiary care settings that have been most
for informed consent tend to resolve the problem or frequently discussed in the literature on informed
to exacerbate it. I will maintain that accepted legal consent. The models of informed consent discussed
standards, at least in the form commonly employed below typically take as the paradigm case something
by courts, send physicians the wrong message about like surgery for breast cancer or the performance of
what is expected of them. An alternative standard an invasive and risky radiologic procedure. It is as-
that would send physicians the correct message, a sumed that the risks to the patient are significant, and
conversation standard, is probably unworkable legally. the values placed on alternative forms of treatment are
As an alternative, I will propose a transparency stan- quite weighty. Moreover, it is assumed that the special-
dard as a compromise that gives physicians a doable ist physician performing the procedure probably does a
task and allows courts to review appropriately. I must fairly limited number of procedures and thus could be
begin, however, by briefly identifying some assump- expected to know exhaustively the precise risks, bene-
tions crucial to the development of this position even fits, and alternatives for each.
though space precludes complete argumentation and Primary care medicine, however, fails to fit this
documentation. model. The primary care physician, instead of per-
forming five or six complicated and risky procedures
Crucial Assumptions frequently, may engage in several hundred treatment
modalities during an average week of practice. In many
Informed consent is a meaningful ethical concept only cases, risks to the patient are negligible and conflicts
to the extent that it can be realized and promoted over patient values and the goals of treatment or non-
within the ongoing practice of good medicine. This treatment are of little consequence. Moreover, in con-
need not imply diminished respect for patient auton- trast to the tertiary care patient, the typical ambulatory
omy, for there are excellent reasons to regard respect for patient is much better able to exercise freedom of choice
patient autonomy as a central feature of good medical and somewhat less likely to be intimidated by either the
care. Informed consent, properly understood, must severity of the disease or the expertise of the physician;
be considered an essential ingredient of good patient the opportunities for changing one's mind once treat-
care, and a physician who lacks the skills to inform ment has begun are also much greater. Indeed, in pri-
patients appropriately and obtain proper consent as a mary care, it is much more likely for the full process of
non-medical, legalistic exercise designed to promote informed consent to treatment (such as the beginning
Source: From The Hastings Center Report 19, no. 5 (1989), 5-9. Reprinted by permission of The Hastings Center and the author.
II 41I1 1111;k1
C
and the dose adjustment of an antihypertensive medica- guidance to physicians. One feature would be that
tion) to occur over several office visits rather than at one physicians esteem the law as a positive force in guid-
single point in time. ing their practice, rather than as a threat to their
It might be argued that for all these reasons, the well-being that must be handled defensively. Another
stakes are so low in primary care that it is fully appro- element would be a prospective consideration by the
priate for informed consent to be interpreted only with law of what the physician could reasonably have been
regard to the specialized or tertiary care setting. I be- expected to do in practice, rather than a retrospective
lieve that this is quite incorrect for three reasons. First, review armed with the foreknowledge that some sig-
good primary care medicine ought to embrace respect nificant patient harm has already occurred.
for patient autonomy, and if patient autonomy is oper- Unfortunately, given the present legal climate, the
ationalized in informed consent, properly understood, physician is much more likely to get a mixed or an
then it ought to be part and parcel of good primary undesirable message from the reasonable patient stan-
care. Second, the claim that the primary care physician dard. The message the physician hears from the reason-
cannot be expected to obtain the patient's informed able patient standard is that one must exhaustively lay
consent seems to undermine the idea that informed out all possible risks as well as benefits and alterna-
consent could or ought to be part of the daily prac- tives of the proposed procedure. If one remembers to
tice of medicine. Third, primary care encounters are discuss fifty possible risks, and the patient in a par-
statistically more common than the highly specialized ticular case suffers the fifty-first, the physician might
encounters previously used as models for the concept subsequently be found liable for incomplete disclosure.
of informed consent.' Since lawsuits are triggered when patients suffer harm,
disclosure of risk becomes relatively more important
Accepted Legal Standards than disclosure of benefits. Moreover, disclosure of in-
formation becomes much more critical than effective
Most of the literature on legal approaches to informed patient participation in decision-making. Physicians
consent addresses the tension between the community consider it more important to document what they said
practice standard and the reasonable patient standard, to the patient than to document how the patient used
with the latter seen as the more satisfactory, emerging or thought about that information subsequently.
legal standard.' However, neither standard sends the In specialty practice, many of these concerns can
proper message to the physician about what is expected be nicely met by detailed written or videotaped con-
of her to promote patient autonomy effectively and sent documents, which can provide the depth of in-
to serve the informational needs of patients in daily formation required while still putting the benefits and
practice. alternatives in proper context. This is workable when
The community practice standard sends the wrong one engages in a limited number of procedures and
message because it leaves the door open too wide for can have a complete document or videotape for each.°
physician paternalism. The physician is instructed to However, this approach is not feasible for primary care,
behave as other physicians in that specialty behave, when the number of procedures may be much more
regardless of how well or how poorly that behaviour numerous and the time available with each patient may
serves patients' needs. Certainly, behaving the way be considerably less. Moreover, it is simply not realistic
other physicians behave is a task we might expect to expect even the best educated of primary care phys-
physicians to readily accomplish; unfortunately the icians to rattle off at a moment's notice a detailed list of
standard fails to inform them of the end toward which significant risks attached to any of the many drugs and
the task is aimed. therapeutic modalities they recommend.
The reasonable patient standard does a much better This sets informed consent apart from all other
job of indicating the centrality of respect for patient au- aspects of medical practice in a way that I believe is
tonomy and the desired outcome of the informed con- widely perceived by non-paternalistic primary care
sent process, which is revealing the information that a physicians, but which is almost never commented
reasonable person would need to make an informed upon in the medical ethics literature. To the physician
and rational decision. This standard is particularly obtaining informed consent, you never know when you
valuable when modified to include the specific infor- are finished. When a primary care physician is told to
mational and decisional needs of a particular patient. treat a patient for strep throat or to counsel a person
If certain things were true about the relationship suffering a normal grief reaction from the recent death
between medicine and law in today's society, the rea- of a relative, the physician has a good sense of what it
sonable patient standard would provide acceptable means to complete the task at hand. When a physician
Brody: Transparency: Informed Consent in Primary Care e
is told to obtain the patient's informed consent for a issues in informed consent, and could also be a power-
medical intervention, the impression is quite different. ful educational tool for highlighting the skills and at-
A list of as many possible risks as can be thought of may titudes that a physician needs to successfully integrate
still omit some significant ones. A list of all the risks this process within patient care.
that actually have occurred may still not have dealt with If primary care physicians understand informed
the patient's need to know risks in relation to benefits consent as this sort of conversation process, the idea
and alternatives. A description of all benefits, risks, that exact rules cannot be given for its successful
and alternatives may not establish whether the patient management could cease to be a mystery. Physicians
has understood the information. If the patient says he would instead be guided to rely on their own intui-
understands, the physician has to wonder whether he tions and communication skills, with careful attention
really understands or whether he is simply saying this to information received from the patient, to determine
to be accommodating. As the law currently appears to when an adequate job had been done in the informed
operate (in the perception of the defensively minded consent process.
physician), there never comes a point at which you can Moreover, physicians would be encouraged to see
be certain that you have adequately completed your informed consent as a genuinely mutual and participa-
legal as well as your ethical task. tory process, instead of being reduced to the one-way
The point is not simply that physicians are para- disclosure of information. In effect, informed consent
noid about the law; more fundamentally, physicians could be demystified, and located within the context
are getting a message that informed consent is very of the everyday relationships between physician and
different from any other task they are asked to perform patient, albeit with a renewed emphasis on patient
in medicine. If physicians conclude that informed con- participation.8
sent is therefore not properly part of medicine at all, Unfortunately, the conversation metaphor does not
but is rather a legalistic and bureaucratic hurdle they lend itself to ready translation into a legal standard for
must overcome at their own peril, blame cannot be at- determining whether or not the physician has satisfied
tributed to paternalistic attitudes or lack of respect for her basic responsibilities to the patient. There seems
patient autonomy. to be an inherently subjective element to conversation
that makes it ill-suited as a legal standard for review of
The Conversation Model controversial cases. A conversation in which one par-
ticipates is by its nature a very different thing from the
A metaphor employed by Jay Katz, informed consent same conversation described to an outsider. It is hard
as conversation, provides an approach to respect for to imagine how a jury could be instructed to determine
patient autonomy that can be readily integrated within in retrospect whether or not a particular conversation
primary care practice.' Just as the specific needs of an was adequate for its purposes. However, without the
individual patient for information, or the meaning that possibility for legal review, the message that patient
patient will attach to the information as it is presented, autonomy is an important value and that patients have
cannot be known in advance, one cannot always tell in important rights within primary care would seem to
advance how a conversation is going to turn out. One be severely undermined. The question then is whether
must follow the process along and take one's cues from some of the important strengths of the conversation
the unfolding conversation itself. Despite the absence model can be retained in another model that does allow
of any formal rules for carrying out or completing a better guidance.
conversation on a specific subject, most people have a
good intuitive grasp of what it means for a conversation The Transparency Standard
to be finished, what it means to change the subject in
the middle of a conversation, and what it means to later I propose the transparency standard as a means to oper-
reopen a conversation one had thought was completed ationalize the best features of the conversation model in
when something new has just arisen. Thus, the meta- medical practice. According to this standard, adequate in-
phor suggests that informed consent consists not in a formed consent is obtained when a reasonably informed
formal process carried out strictly by protocol but in patient is allowed to participate in the medical decision
a conversation designed to encourage patient partici- to the extent that patient wishes. In turn, "reasonably
pation in all medical decisions to the extent that the informed" consists of two features: (1) the physician
patient wishes to be included. The idea of informed discloses the basis on which the proposed treatment, or
consent as physician—patient conversation could, when alternative possible treatments, has been chosen; and (2)
properly developed, be a useful analytic tool for ethical the patient is allowed to ask questions suggested by the
I I Li .111111
disdosure of the physician's reasoning, and those ques- with the patient; secondly, she has to encourage and
tions are answered to the patient's satisfaction. answer questions; and third, she has to discover how
According to the transparency model, the key participatory he wishes to be and facilitate that level of
to reasonable disclosure is not adherence to existing participation. This seems a much more reasonable task
standards of other practitioners, nor is it adherence within primary care than an exhaustive listing of often
to a list of risks that a hypothetical reasonable patient irrelevant risk factors.
would want to know. Instead, disclosure is adequate There are also considerable advantages for the pa-
when the physician's basic thinking has been rendered tient in this approach. The patient retains the right to
transparent to the patient. If the physician arrives at ask for an exhaustive recital of risks and alternatives.
a recommended therapeutic or diagnostic intervention However, the vast majority of patients, in a primary
only after carefully examining a list of risks and bene- care setting particularly, would wish to supplement a
fits, then rendering the physician's thinking transparent standardized recital of risks and benefits of treatment
requires that those risks and benefits be detailed for with some questions like, "Yes, doctor, but what does
the patient. If the physician's thinking has not followed this really mean for me? What meaning am I supposed
that route but has reached its conclusion by other to attach to the information that you've just given?" For
considerations, then what needs to be disclosed to the example, in scenarios 1 and 2, the precise and specific
patient is accordingly different. Essentially, the trans- risk probabilities and possibilities are very small con-
parency standard requires the physician to engage in siderations in the thinking of the physician, and recit-
the typical patient management thought process, only ing an exhaustive list of risks would seriously misstate
to do it out loud in language understandable to the patient.9 just what the physician was thinking. If the physician
To see how this might work in practice, consider did detail a laundry list of risk factors, the patient
the following as possible general decision-making might very well ask, "Well, doctor, just what should
strategies that might be used by a primary physician: I think about what you have just told me?" and the
thoughtful and concerned physician might well reply,
1. The intervention, in addition to being presumably "There's certainly a small possibility that one of these
low-risk, is also routine and automatic. The phys- bad things will happen to you; but I think the chance is
ician, faced with a case like that presented by the extremely remote and in my own practice I have never
patient, almost always chooses this treatment. seen anything like that occur." The patient is very likely
2. The decision is not routine but seems to offer clear to give much more weight to that statement, putting
benefit with minimal risk the risks in perspective, than he is to the listing of risks.
3. The proposed procedure offers substantial chances And that emphasis corresponds with an understanding
for benefit, but also very substantial risks. of how the physician herself has reached the decision.
4. The proposed intervention offers substantial risks The transparency standard should further facilitate
and extremely questionable benefits. Unfortu- and encourage useful questions from patients. If a patient
nately, possible alternative courses of action also is given a routine list of risks and benefits and then is
have high risk and uncertain benefit. asked "Do you have any questions?" the response may
well be perfunctory and automatic. If the patient is told
The exact risks entailed by treatment loom much precisely the grounds on which the physician has made
larger in the physician's own thinking in cases 3 and 4 her recommendation, and then asked the same question,
than in cases 1 and 2. The transparency standard would the response is much more likely to be individualized
require that physicians at least mention the various and meaningful.
risks to patients in scenarios 3 and 4, but would not There certainly would be problems in applying the
necessarily require physicians exhaustively to describe transparency standard in the courtroom, but these do
risks, unless the patient asked, in scenarios 1 and 2. not appear to be materially more difficult than those
The transparency standard seems to offer some encountered in applying other standards; moreover,
considerable advantages for informing physicians what this standard could call attention to more important
can legitimately be expected of them in the promotion features in the ethical relationship between physician
of patient autonomy while carrying out the activities and patient. Consider the fairly typical case, in which
of primary care medicine. We would hope that the a patient suffers harm from the occurrence of a rare
well-trained primary care physician generally thinks but predictable complication of a procedure, and
before acting. On that assumption, the physician can then claims that he would not have consented had he
be told exactly when she is finished obtaining in- known about that risk Under the present "enlightened"
formed consent—first, she has to share her thinking court standards, the jury would examine whether a
Brody: Transparency: Informed Consent in Primary Care S
reasonable patient would have needed to know about think about therapeutic choice will prove to be too
that risk factor prior to making a decision on the pro- idiosyncratic and variable to serve as any sort of stan-
posed intervention. Under the transparency standard, dard. Perhaps disclosing basic thinking processes will
the question would instead be whether the physician impede rather than promote optimal patient participa-
thought about that risk factor as a relevant consideration in decisions.
tion prior to recommending the course of action to the But the transparency standard must be judged,
patient. If the physician did seriously consider that risk not only against ideal medical practice, but also
factor, but failed to reveal that to the patient, he was in against the present-day standard and the message it
effect making up the patient's mind in advance about sends to practitioners. I have argued that that message
what risks were worth accepting. In that situation, the is, "You can protect yourself legally only by guessing
physician could easily be held liable. If, on the other all bad outcomes that might occur and warning each
hand, that risk was considered too insignificant to play patient explicitly that he might suffer any of them."
a role in determining which intervention ought to be The transparency standard is an attempt to send the
performed, the physician may still have rendered his message, "You can protect yourself legally by convers-
thinking completely transparent to the patient even ing with your patients in a way that promotes their
though that specific risk factor was not mentioned. In participation in medical decisions, and more specific-
this circumstance, the physician would be held to have ally by making sure that they see the basic reasoning
done an adequate job of disclosing information.w A you used to arrive at the recommended treatment." It
question would still exist as to whether a competent seems at least plausible to me that the attempt is worth
physician ought to have known about that risk factor making.
and ought to have considered it more carefully prior to The reasonable person standard may still be the
doing the procedure. But that question raises the issue best way to view informed consent in highly spe-
of negligence, which is where such considerations cialized settings where a relatively small number
properly belong, and removes the problem from the of discrete and potentially risky procedures are the
context of informed consent. Obviously, the standard daily order of business. In primary care settings, the
of informed consent is misapplied if it is intended by best ethical advice we can give physicians is to view
itself to prevent the practice of negligent medicine. informed consent as an ongoing process of conversa-
tion designed to maximize patient participation after
Transparency in Medical Practice adequately revealing the key facts. Because the conver-
sation metaphor does not by itself suggest measures
Will adopting a legal standard like transparency change for later judicial review, a transparency standard, or
medical practice for the better? Ultimately only em- something like it, may be a reasonable way to oper-
pirical research will answer this question. We know ationalize that concept in primary care practice. Some
almost nothing about the sorts of conversations pri- positive side-effects of this might be more focus on
mary care physicians now have with their patients, or good diagnostic and therapeutic decision-making on
what would happen if these physicians routinely tried the physician's part, since it will be understood that
harder to share their basic thinking about therapeutic the patient will be made aware of what the physician's
choices. In this setting it is possible to argue that the reasoning process has been like, and better documen-
transparency standard will have deleterious effects. tation of management decisions in the patient record.
Perhaps the physician's basic thinking will fail to in- If these occur, then it will be clearer that the standard
clude risk issues that patients, from their perspective, of informed consent has promoted rather than im-
would regard as substantial. Perhaps how physicians peded high quality patient care.
Notes
1. Charles W Lidz et al., (1983), "Barriers to Informed Theory of Informed Consent (Oxford University Press: New
Consent," in Annals of Internal Medicine 99 (4): 539-43. York), 98-9 and associated footnotes.
2. Tom L. Beauchamp and Laurence McCullough, (1984), 4. For efforts to address ethical aspects of primary care
Medical Ethics: The Moral Responsibilities of Physicians practice, see Ronald J. Christie and Barry Hoffmaster,
(Prentice-Hall: Englewood Cliffs, NJ). (1986), Ethical Issues in Family Medicine (Oxford University
3. For a concise overview of empirical data about Press: New York); and Harmon L. Smith and Larry R.
contemporary informed consent practices see Ruth R. Churchill, (1986), Professional Ethics and Primary Care
Faden and Tom L. Beauchamp, (1986), A History and Medicine (Duke University Press Durham, NC).
II III I 411
IIan 1111.1.1
G 2 Medical Decision-Making
5. Faden and Beauchamp, A History and Theory of Informed 8 Howard Brody, (1987), Stories of Sickness (Yale University
Consent, 23-49 and 114-50. I have also greatly benefited Press: New Haven), 171-81.
from an unpublished paper by Margaret Wallace. 9. For an interesting study of physicians' practices on this point,
6. For a specialty opinion to the contrary, see W.H. Coles, see William C. Wu and Robert A. Pearlman, (1988), "Consent
et al., (1987), "Teaching Informed Consent," in Further in Medical Decision-making: The Role of Communication,"
Developments in Assessing Clinical Competence, ed. Ian R Hart inJournal of General Internal Medicine 3 (1): 9-14.
and Ronald M. Harden (Can-Heal Publications: Montreal), 10. A court case that might point the way toward this line of
241-70. This paper is interesting in applying to specialty care reasoning is Precourt v. Frederick, 395 Mass. 689 (1985).
a model very much like the one I propose for primary care. See William J. Curran, (1986), "Informed Consent in
7. Jay Katz, (1984), The Silent World of Doctor and Patient Malpractice Cases: A Turn Toward Reality," in New England
(Free Press: New York). Journal of Medicine 314 (7): 429-31
Culture, Power, and Informed Mediating Client and Physician Explanations of

Invasive Diagnostic Procedures: A Case Example
Consent: The Impact of
The case documents the communication with an Ab-
Aboriginal Health Interpreters on original patient who was asked to consent to gastro-
Decision-Making scopic and colonoscopic examinations. The 46-year-old
Cree-speaking woman was referred from a northern
Joseph Kaufert and John O'Neil nursing station for further investigation of anemia by a
gastroenterologist in Winnipeg, A series of encounters
Introduction
were videotaped at each stage of the diagnostic workup.
Several encounters involved the signing of formal consent
The signing of a consent agreement prior to surgery or agreements. In each encounter the physician worked with
invasive diagnostic or treatment procedures is a pivotal a Cree-speaking medical interpreter to explain diagnostic
event in the negotiation of trust in the doctor—patient re- and treatment options and to negotiate patient consent
lationship. Most analysts have focused on the legal, ethical for examinations of the stomach, small and large intestine.
or procedural aspects of consent. However, there is grow- In the initial encounter, the physician attempted to
ing recognition of the importance of considering political evaluate the patient's understanding of her own prob-
and cultural factors which lie outside the immediate con- lem and to explain his diagnostic model of the prob-
text of the medical encounter and beyond the control of able cause of anemia. Specifically he attempted to move
either physician or patient. This [reading] will examine from discussing the client's understanding of anemia
the processes through which consent is negotiated when (conceptualized by the patient in terms of weakness) to
the patient is a Native from one of the remote areas of a more complex model linking the loss of blood to the
northern Canada. It will explore the application of ethno- presence of lesions caused by anti-inflammatory medi-
medical approaches emphasizing explanatory models and cation. Following a cursory explanation of the general
an interactionist framework to understanding the impact diagnosis, the physician moved to a series of diagnostic
of intermediaries in cross-cultural negotiation of consent. questions about presence of blood in the patient's stool.
In our research on cross-cultural communication
in urban hospitals, it was apparent that negotiations DOCTOR: She's anemic and pale, which means she
around the signing of a consent form provided the clear- must be losing blood.
est illustration of the unequal knowledge and power of
the clinician and the patient. The clinician's approach INTERPRETER (Cree): This is what he says about
to obtaining consent was based primarily on a biomed- you. You are pale, you have no blood. (Cree term
ical understanding of a particular disease and associated for anemia connotes bloodless state.)
treatment procedures; the approach of the Native pa-
tient to giving consent was based on experiential and DOCTOR: Has she had any bleeding from the bowel
cultural knowledge of past and present illnesses, inter- when she's had a bowel movement?
pretations of the social meaning of hospital regulations
and health professional behaviour, and general attitudes INTERPRETER (Cree): When you have a bowel
defining intergroup relations in the wider society . . . movement, do you notice any blood?
Source: From Health and Canadian Society: Sociological Perspectives, 3rd edn., D. Coburn and C. D'Arcy (Toronto: University of
Toronto Press, 1998), 131-46. Reprinted by permission of the publisher.
Kaufert/O'Neil: Culture, Power, and Informed Consent C)
PATIENT (Cree): I'm not sure. from where you swallow. Nothing wrong that can
be seen.
INTERPRETER (Cree): Is your stool ever black, or
very light? What does it look like? PATIENT: (Nods, but makes no verbal response.)
PATIENT (Cree): Sometimes dark. DOCTOR: We're going to put a small tube in from
the colon. It's only this big. To have a look, to see if
At this point the patient told the interpreter that she there's any abnormality. It won't take too long and
did not understand how her "weakness" (anemia) was it will be very quick and you shouldn't be uncom-
related to questions about gastrointestinal symptomatol- fortable with it at all. Okay? . . . So we'll go ahead
ogy in the physician's reference to dark stools. Without and do that now while we can.
asking for additional explanation from the physician,
the interpreter attempted to link the patient's under- INTERPRETER (Cree): He wants to see you over here
standing of her anemia with the concept of blood loss. from where your bowel movements come from.
Something will be put there, like the first one (the
INTERPRETER (Cree): We want to know, he says, why tube you swallowed), but smaller. So you can be
it is that you are lacking blood. Thath why he asked examined "down there." Maybe somewhere "down
you what your stool looks like. Sometimes you lose there," it'll be seen that you are losing blood from
your blood from there, when your stool is black. there. The reason why you are lacking blood. That's
what h8 looking for. The cause for your blood loss.
In discussing the probable etiology of the woman's
anemia, the physician introduced a complex explana- During the exchange the patient's willingness to
tory model which explained gastric or intestinal extend the initial consent agreement to cover the inves-
bleeding in terms of the possible side effects of anti- tigation of the lower bowel is inferred from a nod and
inflammatory medication for rheumatism. The patient no real alternative was suggested by the physician. The
again indicated that she did not understand why the patient was asked to initial the addition to the consent
questions about her experience with medication for agreement, without formal translation of the English text.
rheumatism were relevant to the current diagnosis of Colonoscopic and radiological examination of
problems with weakness and blood loss. The inter- the intestine revealed a benign polyp. The physician
preter provided an unprompted explanation linking recommended that the polyp should be cauterized
the line of questioning about the side effects of anti- through a second colonoscopy and asked the patient
inflammatory drugs with the concept of blood loss. to sign a consent form for the additional procedure. Al-
though risks and benefits were not formally discussed
INTERPRETER (Cree): He says that those pills you the interpreter elaborated on the basic diagnostic infor-
are taking for rheumatism, sometimes they cause mation provided by the physician. The interpreter also
you to bleed inside, or you will spit up blood. Not introduced a more formal decision point at which the
everyone has these effects. This is why he wants to patient was asked to give her formal consent.
know about your medication.
DOCTOR: We x-rayed the bowel.
Following gastroscopy, the physician attempted to
explain the results of the gastric studies and at the same INTERPRETER (Cree): And this is what they did this
time to extend the initial consent agreement to permit morning—when you were x-rayed. The pictures
colonoscopic examination of the lower bowel. of the area you have bowel movements.
DOCTOR: Everything looked good. There was no PATIENT (Cree): Yes.

ulcer and no nasty disease in the stomach or the
esophagus. No bleeding. You're still anemic so we DOCTOR: And that shows a polyp, a small benign
still want to find out if there's any bleeding from tumour. And I have to take that out.
the lower end.
INTERPRETER (Cree): This picture they took this
INTERPRETER (Cree): He says this about you: morning. He saw it already. There's something
there's nothing visible in your stomach. Nothing, growing there. About this size. And it has to be
no sores, lumps, what they call "ulcers." Nothing removed, because you might bleed from there.
II II I 111 II 111111
PATIENT (Cree): Yes. INTERPRETER (Cree): Do you want to have this pro-
cedure done? Will you consent to have this growth
DOCTOR: Now I can take it out without an oper- removed—burned? Do you consent to have it
ation, by putting a tube inside the bowel, and put- done?
ting a wire around it and burning the polyp off.
That stops the bleeding, no need for an operation. PATIENT (Cree): I don't know.
INTERPRETER (Cree): He says they can put in a tube INTERPRETER: You know, if it's not removed it may
like before and burn off the growth. bleed. It may cause problems.
PATIENT: (Nods but makes no verbal response.) INTERPRETER (English directed at physician): Dr
, isn't it true that if it's not removed,
DOCTOR: If she wants to make the arrangements it can bleed and she can become anemic?
for the hospital admission she can come down and
sign the consent form. DOCTOR: That's correct, we feel that your anemia
may result from the bleeding of the polyp.
INTERPRETER (English): Will there be complications?
INTERPRETER (Cree): If it's not removed, you may
DOCTOR: There are a few complications but I think end up with cancer. You know? And you will not
it would be difficult to explain them all. have an operation. Its harder when a person has
an operation. You know? And [this procedure] that
INTERPRETER (Cree): After this procedure has been he's going to do will get it on time. Before it begins
done you won't be staying here at the hospital. You'll to bleed or starts to grow. You're lucky it's caught
be able to go home on Saturday. It will be done on on time. And it will bother you when you have a
Friday, then youll already be able to go home on bowel movement. This way there's no danger that
Saturday. It won't be long. But its entirely up to you. this growth will bleed.
In this exchange, the interpreter is providing more PATIENT (Cree): I still don't know.
than a simple elaboration of the risks, benefits, and
rationale for the procedure. The physician assumes that INTERPRETER (Cree): Well if you want to come in
his explanation of the reasons for doing the proced- for the procedure while you are here? It's all up to
ure will be sufficient to obtain the patient's consent. In you to think about.
addition the physician closes the exchange by asserting
that it would be too difficult to explain all the pos- Again, the interpreter has assumed responsibility
sible complications. However, in both instances, the for providing a rationale for the procedure and ex-
interpreter does not provide a literal translation of the plaining the potential benefits. Her explanations are
physician's side comments, but attempts to assure the also clearly based on her own medical knowledge, and
patient and justify the procedure through explanations her understanding of the patient's explanatory model.
addressing the patient's concern with the length of her The fear of cancer in the Native community is linked to
stay and desire to return to her home community. The general understandings about the history of infectious
interpreter assures the patient, "You'll be able to go disease epidemics that nearly destroyed Aboriginal so-
home on Saturday," emphasizing the expectation that ciety in North America. Cancer is increasingly viewed
the operation will be minor and she won't be separated as the new "epidemic." The interpreter is using her
from family and home for long. The interpreter's state- knowledge of these fears to negotiate the patient's con-
ment linking approval of the consent to the patient's sent, but she is also using Cree models of negotiation
early return to her community occurs in Cree and emphasizing individual autonomy. Her final statement
therefore is not accessible to the physician. emphasizes her client's ultimate personal responsibil-
As the end of the encounter the physician in- ity: "It's all up to you to think about."
cluded the consent agreement with the hospital At this point, the patient accompanied the in-
admission protocol and assumed it would be signed terpreter and physician to the appointment desk and
with the other paper work. The interpreter provided scheduled the colonoscopy for the following day. After
a more direct opportunity for the patient to give or a brief summary of the text (which was printed in
withhold consent. English) was provided by the interpreter, the patient
Kaufert/O'Neil: Culture, Power, and Informed Consent
signed the consent form. The formal act of signing the of the health care system and elaborating these treat-
form was immediately subordinated to a discussion of ment options.
specific arrangements for the client's discharge from However, in consent negotiations involving
hospital and travel arrangements for returning to the both Indian and Inuit patients, the presence of a
reserve community. language interpreter or patient advocate as an inter-
In this case, consent was negotiated by drawing mediary raises a number of ethical and sociopolitical
on expressions of trust by the patient about her rela- issues. For example, the involvement of interpreter-
tionship with the interpreter. The physician initially advocates in the doctor—patient relationship may shift
assumed that little explanation was required for con- responsibility and initiative for disclosure from the
sent, and indicated his unwillingness to negotiate. professional to the intermediary. The translator may
Information-sharing occurred gradually over the course exercise control through selective interpretation of
of the encounter. The interpreter assumed the negoti- information provided by the client. The interpreter
ator's role, based on shared cultural understanding of may also priorize and filter information about treat-
both biomedicine and Native culture. Throughout the ment options and associated risks or benefits. The
sequence of interaction the interpreter worked to elicit interpreters' function of mediating and priorizing
and clarify both the client's and clinician's interpreta- information occurs within a linguistic and cultural
tion of the condition and the program of treatment. "black box." Within this box the interpreter may
However, the interpreter's intervention introduced actively intervene on the patient's behalf, or use his
a third party into the clinician—client relationship. or her cultural knowledge and personal rapport with
She directly influenced the course of the decision by the patient to reinforce the clinician's definition of
independently introducing new information about ill- appropriate treatment choices. Consideration of this
ness and treatment options. She also imposed decision variable by clinicians in obtaining informed consent
points where the patient could actually exercise her is important.
option to consent. In summary, sociocultural analysis of real inter-
The patient's willingness to allow the interpreter action sequences in the negotiation of consent be-
to negotiate consent is also evident in her reluctance tween the clinician and client differs fundamentally
to ask the physician direct questions. In response to from legal or ethical analysis of the marker decisions.
the interpreter's questions about her understanding, For Native clients, agreements may reflect the emer-
she repeats at several junctures, "I don't know" This gence of trust relationships achieved through an ex-
provides a cue for the interpreter to introduce further tended, incremental process of exchange rather than a
information or elaborate relative risks. The final con- formal, final contract. Interactionist and ethnomedical
sent is passive, in the sense that the patient signs the approaches more clearly reveal the communication
forms without further resistance. processes and power relations which are part of the
process of translating and prioritizing information.
Power and Control in Consent Decisions Our analysis of the role of Native medical interpret-
among Native Canadians
ers in both case studies clearly indicates that dyadic
clinician—client interaction is strongly influenced
The case study demonstrates the role of interpreter- by intermediaries. Both confirm that translators,
advocates in redressing cultural and structural con- cultural brokers, and personal advocates negotiate
straints for Native people in consent negotiations in shared meanings and influence the balance of power
urban hospitals (Kaufert and Koolage 1984). In some in cross-cultural, clinical communication. As well as
urban hospitals interpreters have expanded their demarcating formal legal and ethical decision points,
role beyond narrow language translation functions cross-cultural consent agreements also function as in-
to assume advocacy roles which empower the client tegrative rituals through which participants reconcile
through providing information about the structure power imbalance and negotiate clinical trust.
References
Kaufert, J., and W. Koolage. 1984. "Role Conflict among This manuscript contains sections of a previously published
Culture Brokers: The Experience of Native Canadian manuscript, "Biomedical Rituals and Informed Consent,"
Medical Interpreters," in Social Science and Medicine 18 originally printed in Social Science Perspectives on Medical
(3): 283-6. Ethics, ed. G. Weisz, University of Pennsylvania Press, 1991.
I I UN 1111111
Informed Consent and Public Health undermine the fundamental concerns that are taken to
justify informed consent requirements or even show
Onora O'Neill disrespect for individuals or their autonomy? Should
proxy consent perhaps be set aside as mere pretence
1. Informed Consent and Public Health and replaced with greater reliance on professional
judgement of each patient's best interests? Or would
Medical ethics has been transformed during the past doing so revert to unacceptable medical paternalism
30 years. One conspicuous change has been a steadily and so fail to respect patients and their (faltering) au-
increasing focus on informed consent, which is now tonomy? Responses to these questions often propose
usually taken to be essential for any ethically acceptable ways of revising or refining the procedures used for
medical practice. The literature on informed consent requesting and recording consent to make consent-
in medical ethics is vast and repetitive. However, it has ing easier for less competent patients. But even the
significant limitations. Some of the difficulties are well most energetic and time-consuming presentation of
known and recalcitrant, although this seldom dents the user-friendly information, even the most elaborate and
enthusiasm of those who think informed consent es- detailed consent forms and procedures, will not make
sential to (or even sufficient for) ethically acceptable informed consent possible for numerous patients with
medical practice. In this paper I shall mention the various types of incapacity.
commonly discussed difficulties but shall concentrate However, these are not the deepest difficulties.
on some less discussed but philosophically deeper dif- Informed consent procedures are problematic not
ficulties that limit the use to which informed consent merely because their philosophical rationale is dis-
procedures can be put in public health provision. puted, and not merely because some individuals lack
Some of the most frequent disagreements about competence to consent, or lack it at some times. The
informed consent are about the basic reasons for think- most basic philosophical difficulties with informed con-
ing that it is ethically important. Is informed consent sent arise because consent is a propositional attitude.
required to respect persons or to respect the autonomy Consenting—like other cognitive states or acts such as
of persons? If the latter, which conception of autonomy knowing, believing, understanding, hoping, wonder-
is relevant? If some persons are more autonomous than ing, thinking, desiring, or fearing—takes a proposition
others, will informed consent procedures be more im- as its object. Hence consent is never directed at a med-
portant for them? Or will they, on the contrary, be more ical intervention as such but rather at some proposition
important for those with limited autonomy? Alterna- that describes an intended intervention. However, any
tively are informed consent procedures required be- intervention can be described in many different ways.
cause they provide a degree of assurance that patients Even in the easy cases, where competent patients con-
are not deceived or coerced in the course of clinical sent to a procedure or treatment described in a certain
practice? (Faden & Beauchamp 1986; Wolpe 1998; way, they may not be aware of and may not consent
O'Neill 2002a). to other true descriptions of the same intervention or
A second, even more frequently discussed, range of its more obvious effects. This can be the case even
of problems arises when patients cannot grasp the in- when the second description is entailed by the first or
formation that is essential to giving informed consent. when it refers to an obvious consequence of the state of
If they cannot understand the proposition to which affairs described by the first.
their consent is sought, they cannot give or refuse So propositional attitudes are opaque. A person
informed consent. The hard cases are numerous and may know or understand or hope that x but not know
intractable. Many patients cannot consent to medical or understand or hope that y even where x entails y.
intervention or treatment because they are too young, A patient may consent to an intervention, and the
too ill, too disabled, or too demented to understand the intervention as described may entail or bring about
information that they would have to grasp to make an certain conditions; however, the patient may not con-
informed choice. As they can hardly be denied medical sent to those conditions because he or she may not
treatment because of these difficulties, it must be given grasp the entailment relation or the causal connection.
without their consent. Should it then be given on the For example, a parent may consent to the removal of
basis of others' consent (e.g., that of parents, guardians, tissues from their dead child, and the undifferentiated
or relatives)? Or does the very idea of proxy consent reference to tissues will cover entire organs; however,
Source: From Philosophical Transactions of the Royal Society B 359 (2004), 1133-6. Reprinted with permission.
O'Neill: Informed Consent and Public Health 61
the parent may not know that this is so and be upset to implied. No programme for replacing implied with
discover that entire organs were removed on the basis explicit consent can be complete.
of general consent to the removal of tissues. The distinction between specific and generic con-
Because propositional attitudes are always opaque, sent applies to the propositions to which consent is
the basic difficulties of informed consent procedures given, rather than to processes of consenting. For ex-
cannot be reliably eliminated by making information ample, consent may be given to the removal of tissues,
more available or consent procedures easier to follow. For or alternatively to the removal of a specific sort of tissue,
unless the information is actually understood by those or even to the removal of tissue for a specific use, such
whose informed consent is requested, genuine consent as diagnosis, or as part of a cancer treatment or post-
will not stretch to the relevant proposition. This diffi- mortem to determine the cause of death. The descrip-
culty is ubiquitous within the central debates of medical tions to which consent is given are always incomplete.
ethics, although barely discussed (O'Neill 2002b). We can always add more detail. So those who believe
that informed consent should be highly specific need to
2. Explicit or Implied? Specific or Generic?
explain how specific it has to be to constitute ethically
adequate informed consent. Answering this question
These difficulties have been heightened rather than may be no easier than answering the pseudo-question
resolved by recent attempts to improve informed con- "how long is a piece of string?"
sent procedures. Two supposed improvements are often These problems are not "merely theoretical." They
advocated and raise particular difficulties. The first de- surface and create practical problems wherever data pro-
mands that all consent should be explicit rather than im- tection issues arise, in secondary data analyses, indeed
plied; the second that it be specific rather than generic. in the use of tissues for comparative study. If personal
The distinction between explicit and implied information is to be "processed" (obtained, recorded,
consent contrasts ways of consenting. Explicit consent processed, sorted, used) only in accordance with the
typically relies on documents, signatures, and formal consent of those to whom it pertains (the "data sub-
statements; it may require witnesses who confirm that jects"), then it will constantly turn out that if the consent
proper procedures for consenting have been followed. obtained is sufficiently specific to permit a certain use, it
The formal procedures are typically designed to create will also be sufficiently specific to preclude other uses.
enduring records, thereby reducing later uncertainty Similarly, if tissues and information from past
about the consent given and perhaps forestalling dis- patients are to be studied for purposes that could not
satisfaction, complaint, or litigation. Patients who con- have been anticipated, they must be studied without
sent explicitly to proposed interventions thereby accept specific consent, because such consent could not in
that they cannot later claim that they were injured or principle have been requested or given at the relevant
wronged and accept that they will not have grounds for time. The problem is not merely that in the past con-
complaint or litigation. sent procedures were too lax and that the relevant con-
By contrast, implied consent is inferred from a pa- sents were not obtained or not recorded with sufficient
tient's action. For example, agreement to blood being clarity (although that was often the case). The problem
taken or to having an injection is standardly signified is deeper, indeed irresolvable, because many valuable
by extending one's arm for the doctor to take the blood purposes could not have been anticipated at the time
or give the injection. No documentation of the consent that tissues were removed and stored. For example,
is required. It would be possible—but laborious—to nobody could tell in advance when information and
replace the implied consent that is currently seen as suf- tissues obtained in the course of treating past patients
ficient in these and similar cases with explicit consent will turn out to be useful for unanticipated research.
procedures. It would be possible—but strenuous—to Both clinical information and tissue samples pertaining
introduce explicit consent procedures for the most to deceased patients may later turn out to be vital for
minor and routine of medical interventions. However reaching a better understanding of new diseases. When
much we introduce additional explicit procedures and the first patients with VCJD died, the only way in which
consent forms for interventions that are now performed pathologists could determine whether this was a new
on the basis of implied consent, explicit consent always disease was by comparing their brain tissue with sam-
relies on background understandings that remain im- ples taken from patients who had died of Creutzfeldt—
plicit. The longest and most complex consent form Jakob disease across many decades in many countries.
cannot include a complete description of everything Tissue and information from those who died in the
that will be done. Much is taken as understood, and 1918 influenza epidemic may yet prove valuable in
consent based on those understandings can only be studying emerging diseases (Gamblin et al. 2004). It is
I I al II1La
difficult to see how secondary data analysis of this sort standards and formats that are also largely fixed and
can proceed if access to tissue samples from and data uniform and so cannot be treated as a matter for in-
pertaining to past patients, their treatment and their formed consent. The scaffolding of professional train-
clinical outcomes, cannot be consulted without specific ing, of institutional structures, of public funding,
prior consent to such studies. Although the incidence of physical facilities are all public goods. The public
of disease could be monitored on the basis of collecting provision of health care can reflect democratic process
deidentified data, linked data are needed for all more and thereby certain forms of collective choice; but its
elaborate forms of retrospective study and public health basic structures cannot be geared to individual choice.
research. Specific consent requirements undermine Unavoidably there are large areas of medical ethics in
secondary data analysis in medical and other areas of which informed consent can play no part, or at most a
inquiry; they would close down epidemiology. minor part. What then are the appropriate ethical and
However, a claim that specific consent is ethically re- other normative issues in these areas of medical ethics?
quired for retrospective study of linked patient informa-
tion and tissues is neither intuitive nor plausible, provided
4. Health and Justice
that standard safeguards such as the approval of ethics
committees and anonymization are in place. What would The ethical reasoning most commonly used in med-
we think of a patient who asks his doctor how he knows ical ethics focuses on transactions between individual
that a medicine will prove helpful, is told that it helped professionals and individual patients or between indi-
nearly all patients with the same condition, accepts the vidual professionals and individual research subjects.
treatment but then reflises consent to the further study of Individualistic approaches are not likely to prove to be
information—even anonymized information—about the useful for analysing ethical questions about the provi-
clinical outcome in his own case? sion of public goods, such as public health provision.
However, it may seem that theories of justice will also
provide an inappropriate account of the normative rea-
3. Public Health and Public Goods
soning most relevant to the provision of public goods or
There are further and more general reasons for rejecting to public health ethics. Most uses of theories of justice
the current tendency to suppose that informed consent in health care ethics have addressed distributive issues,
procedures are the touchstone of ethically acceptable such as the just distribution of clinical care. Discussions
medical provision. One of the most significant is that of health care allocation decisions—of rationing—are
informed consent procedures are inapplicable when- discussions of the just distribution of a good that can
ever the goods or benefits to be provided are public be made contingent on individual choice. Theories of
goods. Certain types of goods—consumer goods, clin- distributive justice also fail to address the distinctive
ical care—can be provided for individuals, and their ethical questions that arise in providing public goods
provision can in principle be made to be contingent on and so are not helpful for public health ethics.
individual consent. The difficulties that informed con- But there is more to justice than distributive jus-
sent requirements raise may prove irresolvable in some tice. All theories of justice also address the justification
cases and resolvable in others. By contrast, if public of compulsion and in particular the justification of the
goods are provided for any, they have to be provided forms of compulsion on which any legal order depends.
for many. Some types of public good must be provided There are many differing theories of justice, but by way
(or not provided) for whole populations; others may be of illustration I shall instance one theory, because it
provided (or not provided) for more restricted groups. gives a particularly large—supposedly maximal—role
For present purposes I leave these differences aside, to individual liberty and so to individual consent pro-
to make the simple point that the provision of public cedures. This theory is John Stuart Mill's form of lib-
goods cannot be made contingent on individual con- eralism, especially as developed in On Liberty. I choose
sent. For example, road safety, food safety water safety Mill's account of justice not because I assume or argue
safe medicines, and measures that protect against infec- that it is more plausible than other accounts but be-
tion cannot be tailored to individual choice. Because cause he explicitly opposes compulsion except in very
there are no obligations to do the impossible ("ought limited circumstances. Mill famously claimed that
implies can"), informed consent cannot be ethically
required for the provision of public goods. . . . the sole end for which mankind are warranted,
The implications of these thoughts about public individually or collectively, in interfering with the
goods are wider than may at first seem to be the case. liberty of any of their number is self-protection.
For example, clinical care itself has to be provided to (Mill 1989, ch. 1, p. 13)
O'Neill: Informed Consent and Public Health
Compulsion, on this view, is permitted only where these compulsory would hardly have seemed contro-
needed so as to protect others: it is unjust unless needed versial a century ago. It has come to seem controversial
to prevent harm to others. Public health provision is an on the basis of an illusory assumption that all medical
obvious area where Mill's arguments are relevant, be- provision, and with it public health provision, can be
cause compulsion may be needed to prevent individual organized on the basis of informed consent of individ-
action that might harm others' health. uals. It cannot.
Prevention of transmission of disease is a central Of course, there are always difficulties in judging
case for Mill's justification of compulsion. Given the how great a risk is and in deciding which forms of
death rate from severe acute respiratory syndrome compulsion are most effective and most readily justi-
and the seriousness of the illness, Mill would view it fied in a particular case. Such issues can be resolved
as legitimate to make certain forms of action or treat- only on a case-by-case basis, using the right expertise
ment compulsory because the risk of transmission can and the right information. Often there are deficits in
be moderately high and the risk of death for those information and in expertise, as with the many un-
who succumb high for some age groups. Depending certainties about the mode of transmission and likely
on the gravity of the risk, it might be permissible to spread of vCJD and other new transmissible diseases.
institute mandatory monitoring of those who may have But where information and expertise point to the
been exposed, mandatory vaccination (if a vaccine is likelihood of harm to others, there are even on a very
developed), restrictions on free movement, or quaran- strong liberal account of justice no good ethical argu-
tine. Similarly, where vaccination is safe and effective, ments to forbid all compulsion. To the contrary, ap-
Mill's argument would suggest that it could legitimately peals to individual consent do not offer a coherent, let
be made compulsory to produce the herd immunity alone an acceptable, way of approaching public health
that protects vulnerable individuals who cannot (yet) provision. Salus populi suprema lex is not an obsolete
be vaccinated. Making public health measures such as thought (Cicero 1928).
References
Cicero, M. T. 1928 De Legibus, III, iii, 8. In De Re Publica, De O'Neill, 0. 2002 Autonomy and trust in bioethics. Cambridge
Legibus (tr. Clinton Walker Keyes; ed. M. T. Cicero). Loeb University Press.
Classical Library. O'Neill, 0. 2002 Some limits of informed consent. J. Med.
Faden, R & Beauchamp, T. 1986 A history and theory of in- Ethics 28, 1-3. See ftp://ftpbmj.bma.org.uk/SJFauthor/
formed consent. Oxford University Press. JME/ feb03/1Fme021800.pdf.
Gamblin, S. J. (and 11 others) 2004 The structure and Wolpe, P. R. 1998 The triumph of autonomy in American
receptor-binding properties of the 1918 influenza hemag- bioethics: a sociological view. In Bioethics and society:
glutinin. Science 303, 1838-1842. (DOI 10.1126/science constructing the ethical (ed. R. DeVries & J. Subedi),
.1093155.) pp. 38-59. Englewood Cliffs, NJ: Prentice-Hall.
Mill, J. S. 1989 On liberty (1859). In On liberty and other writ-
ings (ed. S. Collini), p. 13. Cambridge University Press.
2.4 Substituted Judgments
Enough: The Failure of the Living Will We should have known it would fail: A notable
but neglected psychological literature always provided
Angela Fagerlin and Carl E. Schneider arresting reasons to expect the policy of living wills to
misfire. Given their alluring potential, perhaps they
were worth trying. But a crescendoing empirical litera-
By their fruits ye shall know them.
ture and persistent clinical disappointments reveal that
Enough. The living will has failed, and it is time to say so. the rewards of the campaign to promote living wills do
Source: From The Hastings Center Report 34,2 (2004), 30-42. Reprinted with permission of The Hastings Center and the author.
II i311 111111
not justify its costs. Nor can any degree of tinkering Courts and administrative agencies too have
ever make the living will an effective instrument of become advocates of living wills. The Veterans Ad-
social policy. ministration has proposed a rule to encourage the use
As the evidence of failure has mounted, living of advance directives, including living wills.5 Where
wills have lost some of their friends. We offer system- legislatures have not granted living wills legal status,
atic support for their change of heart. But living wills some courts have done so as a matter of common law,
are still widely and confidently urged on patients, and and where legislatures have granted them legal status,
they retain the allegiance of many bioethicists, doctors, courts have cooperated with eager enthusiasm.° Living
nurses, social workers, and patients. For these loyal ad- wills have assumed special importance in states that
vocates, we offer systematic proof that such persistence prohibit terminating treatment in the absence of strong
in error is but the triumph of dogma over inquiry and evidence of the patient's wishes.' One supreme court
hope over experience. summarized a common theme: "[A] written directive
A note about the scope of our contentions: First, would provide the most concrete evidence of the pa-
we reject only living wills, not durable powers of at- tient's decisions, and we strongly urge all persons to
torney: Second, there are excellent reasons to be skep- create such a directive."
tical of living wills on principle. For example, perhaps The grandees of law and medicine also give their
former selves should not be able to bind latter selves in benediction to the living will. The AMA's Council on
the ways living wills contemplate.' And many people Ethical and Judicial Affairs proclaims: "Physicians
do and perhaps should reject the view of patients, their should encourage their patients to document their treat-
families, and their communities that informs living ment preferences or to appoint a health care proxy with
wills.2 But we accept for the sake of argument that whom they can discuss their values regarding health
living wills desirably serve a strong version of patients' care and treatment."9 The elite National Conference of
autonomy: We contend, nevertheless, that living wills Commissioners on Uniform State Laws continues to
do not and cannot achieve that goal. promulgate the Uniform Health-Care Decisions Act, a
And a stipulation: We do not propose the elimina- prestigious model statute that has been put into law
tion of living wills. We can imagine recommending them in a still-growing number of states. Medical journals
to patients whose medical situation is plain, whose crisis regularly admonish doctors and nurses to see that pa-
is imminent, whose preferences are specific, strong, and tients have advance directives, including living wills.10
delineable, and who have special reasons to prescribe Bar journals regularly admonish lawyers that their
their care. We argue on the level of public policy: In clients—all their clients—need advance directives,
an attempt to extend patients' exercise of autonomy including living wills." Researchers demonstrate their
beyond their span of competence, resources have been conviction that living wills are important by the persis-
lavished to make living wills routine and even universal. tence of their studies of patients' attitudes toward living
This policy has not produced results that recompense wills and ways of inveigling patients to sign them.
its costs, and it should therefore be renounced. Not only do legislatures, courts, administrative
Living wills are a bioethical idea that has passed agencies, and professional associations promote the
from controversy to conventional wisdom, from the living will, but other groups unite with them. The Web
counsel of academic journals to the commands of abounds in sites advocating the living will to patients."
law books, from professors' proposal to professional The web site for our university's hospital plugs advance
practice. Advance directives generally are embodied in directives and suggests that it "is probably better to
federal policy by the Patient Self-Determination Act, have written instructions because then everyone can
which requires medical institutions to give patients read them and understand your wishes.""
information about their state's advance directives. In Our own experience in presenting this paper is
turn, the law of every state provides for advance direc- that its thesis provokes some bioethicists to disbelief
tives, almost all states provide for living wills, and most and indignation. It is as though they simply cannot
states "have at least two statutes, one establishing a bear to believe that living wills might not work. How
living will type directive, the other establishing a proxy can anything so intuitively right be proved so infuri-
or durable power of attorney for health care."3 Not only atingly wrong? And indeed, bioethicists continue to
are all these statutes very much in effect, but new legis- investigate ways the living will might be extended (to
lative activity is constant. Senators Rockefeller, Collins, deal with problems of the mentally ill and of minors,
and Specter have introduced bills to "strengthen" the for example) and developed for other countries.
PSDA and living wills,4 and state legislatures continue to Although some sophisticated observers have long
amend living will statutes and to enact new ones. doubted the wisdom of living wills," proponents have
Fag erli n/Schn eider: Enough:The Failure of the Living Will
tended to respond in one of three ways, all of which Do People Have Living Wills?
preserve an important role for living wills. First, pro-
ponents have supposed that the principal problem with At the level of principle, living wills have triumphed
living wills is that people just won't sign them. These among the public as among the princes of medicine.
proponents have persevered in the struggle to find People widely say they want a living will, and living
ways of getting more people to sign up.13 wills have so much become conventional medical
Second, proponents have reasserted the useful- wisdom "that involvement in the process is being por-
ness of the living wills. For example, Norman Cantor, trayed as a duty to physicians and others.' Despite
distinguished advocate of living wills, acknowledges this, and despite decades of urging, most Americans
that " [s] ome commentators doubt the utility or efficacy lack them." While most of us who need one have a
of advance directives" (by which he means the living property will, roughly 18 percent have living wills."
will), but he concludes that "these objections don't The chronically or terminally ill are likelier to prepare
obviate the importance of advance directives.' Other living wills than the healthy, but even they do so fit-
proponents are daunted by the criticisms of living wills fully." In one study of dialysis patients, for instance,
but offer new justifications for them. Linda Emanuel, only 35 percent had a living will, even though all of
another eminent exponent of living wills, writes that them thought living wills a "good idea."23
"living wills can help doctors and patients talk about Why do people flout the conventional wisdom?
dying" and can thereby "open the door to a positive, The flouters advance many explanations." They don't
caring approach to death."17 know enough about living wills,25 they think living
Third, some proponents concede the weaknesses wills are hard to execute,26 they procrastinate," they
of the living will and the advantages of the durable hesitate to broach the topic to their doctors (as their
power of attorney and then propose a durable power doctors likewise hesitate).28 Some patients doubt they
of attorney that incorporates a living will. That is, the need a living will. Some think living wills are for the
forms they propose for establishing a durable power elderly or infirm and count themselves in neither
of attorney invite their authors to provide the kinds of group." Others suspect that living wills do not change
instructions formerly confined to living wills.'8 the treatment people receive; 91 percent of the veterans
None of these responses fully grapples with the in one study shared that suspicion.3° Many patients are
whole range of difficulties that confound the policy content or even anxious to delegate decisions to their
promoting living wills. In fairness, this is partly because families,' often because they care less what decisions
the case against that policy has been made piecemeal are made than that they are made by people they trust.
and not in a full-fledged and full-throated analysis of Some patients find living wills incompatible with their
the empirical literature on living wills. cultural traditions.32 Thus in the large SUPPORT and
In sum, the law has embraced the principle of HELP studies, most patients preferred to leave final
living wills and cheerfully continues to this moment to resuscitation decisions to their family and physician
expound and expand that principle. Doctors, nurses, instead of having their own preferences expressly fol-
hospitals, and lawyers are daily urged to convince their lowed (70.8% in HELP and 78.0% in SUPPORT). "This
patients and clients to adopt living wills, and patients result is so striking that it is worth restating: not even a
hear their virtues from many other sources besides. third of the HELP patients and hardly more than a fifth
Some advocates of living wills have shifted the grounds of the SUPPORT patients "would want their own prefer-
for their support of living wills, but they persist in ences followed."33
believing that they are useful. The time has come to If people lacked living wills only because of ignor-
investigate those policies and those hopes systematic- ance, living wills might proliferate with education. But
ally. That is what this article attempts. studies seem not to "support the speculations found in
We ask an obvious but unasked question: What the literature that the low level of advance directives
would it take for a regime of living wills to function use is due primarily to a lack of information and en-
as their advocates hope? First, people must have living couragement from health care professionals and family
wills. Second, they must decide what treatment they members."34 Rather, there is considerable evidence
would want if incompetent. Third, they must accur- "that the elderly's action of delaying execution of ad-
ately and lucidly state that preference. Fourth, their vance directives and deferring to others is a deliberate,
living wills must be available to people making deci- if not an explicit, refusal to participate in the advance
sions for a patient. Fifth, those people must grasp and directives process."35
heed the living will instructions. These conditions are The federal government has sought to propagate
unmet and largely unmeetable. living wills through the Patient Self-Determination
• NI II II illl 111Li
Act,36 which essentially requires medical institutions from many, people grossly overestimate the effective-
to inform patients about advance directives. However, ness of CPR and in fact hardly know what it is.44 For
"empirical studies demonstrate that: the PSDA has gen- such information, people must rely on doctors. But
erally failed to foster a significant increase in advance doctors convey that information wretchedly even to
directives use; it is being implemented by medical incompetent patients making contemporaneous deci-
stitutions and their personnel in a passive manner, and sions. Living wills can be executed without even con-
the involvement of physicians in its implementation is sulting a doctor," and when doctors are consulted,
lacking."37 One commentator even thinks "the PSDA's the conversations are ordinarily short, vague, and
legal requirements have become a ceiling instead of a tendentious. In the Tulsky study, for example, doctors
floor."38 only described either "dire scenarios . . . in which few
In short, people have reasons, often substantial people, terminally ill or otherwise, would want treat-
and estimable reasons, for eschewing living wills, rea- ment" or "situations in which patients could recover
sons unlikely to be overcome by persuasion. Indeed, with proper treatment."'
persuasion seems quickly to find its limits. Numerous Let us put the point differently The conventional—
studies indicate that without considerable intervention, legal and ethical wisdom—insists that candidates for
approximately 20 percent of us complete living wills, even a flu shot give "informed consent." And that wisdom
but programs to propagate wills have mixed results.39 has increasingly raised the standards for disclosure." If
Some have achieved significant if still limited increases we applied those standards to the information patients
in the completion of living wills," while others have have before making the astonishing catalog of momen-
quite failed to do so.' tous choices living wills can embody, the conventional
Thus we must ask: If after so much propaganda so wisdom would be left shivering with indignation.
few of us have living wills, do we really want them, or Not only do people regularly know too little when
are we just saying what we think we ought to think and they sign a living will, but often (again, we're human)
what investigators want to hear? they analyze their choices only superficially before pla-
cing them in the time capsule. An ocean of evidence af-
firms that answers are shaped by the way questions are
Do People Know What They Will Want?
asked. Preferences about treatments are influenced by
Suppose, counterfactually, that people executed living factors like whether success or failure rates are used,"
wills. For those documents to work, people would have the level of detail employed," and whether long- or
to predict their preferences accurately. This is an ambi- short-term consequences are explained first.5° Thus in
tious demand. Even patients making contemporary one study, "201 elderly subjects opted for the interven-
decisions about contemporary illnesses are regularly tion 12 % of the time when it was presented negatively,
daunted by the decisions' difficulty. They are human. 18% of the time when it was phrased as in an advance
We humans falter in gathering information, misunder- directive already in use, and 30% of the time when it
stand and ignore what we gather, lack well-considered was phrased positively. Seventy-seven percent of the
preferences to guide decisions, and rush headlong to subjects changed their minds at least once when given
choice." How much harder, then, is it to conjure up the same case scenario but a different description of the
preferences for an unspecifiable future confronted with intervention. "51
unidentifiable maladies with unpredictable treatments? If patients have trouble with contemporaneous
For example, people often misapprehend crucial decisions, how much more trouble must they have
background facts about their medical choices. Oregon with prospective ones. For such decisions to be "true,"
has made medical policy in fresh and controversial patients' preferences must be reasonably stable. Sur-
ways, has recently had two referenda on assisted sui- prisingly often, they are not. A famous study of eight-
cide, and alone has legalized it. Presumably, then, its een women in a "natural childbirth" class found that
citizens are especially knowledgeable. But only 46 per- preferences about anesthesia and avoiding pain were
cent of them knew that patients may legally withdraw relatively stable before childbirth, but at "the beginning
life-sustaining treatment. Even experience is a poor of active labor (4-5 cm dilation) there was a shift in the
teacher "Personal experience with illness . . . and au- preference toward avoiding labor pains. . . . During the
thoring an advance directive . . . were not significantly transition phase of labor (8-10 cm) the values remained
associated with better knowledge about options."43 relatively stable, but then . . . the mothers' preferences
Nor do people reliably know enough about ill- shifted again at postpartum toward avoiding the use
nesses and treatments to make prospective life-or- of anesthesia during the delivery of her next child."32
death decisions about them. To take one example And not only are preferences surprisingly labile, but
Fagerlin/Schneider: Enough:The Failure of the Living Will ()
people have trouble recognizing that their views have missteps in predicting their tastes generally, we should
changed.53 This makes it less likely they will amend expect misapprehensions about end-of-life preferences.
their living wills as their opinions develop and more Indeed, those preferences should be especially volatile,
likely that their living wills will treasonously misrepre- since people lack experience deciding to die.
sent their wishes.
Instability matters. The healthy may incautiously
Can People Articulate What They Want?
prefer death to disability. Once stricken, competent
patients can test and reject that preference. They Suppose, arguendo, that patients regularly made sound
often do.54 Thus Wilfrid Sheed "quickly learned [that] choices about future treatments and write living wills.
cancer, even more than polio, has a disarming way Can they articulate their choices accurately? This
of bargaining downward, beginning with your whole question is crucially unrealistic, of course, because
estate and then letting you keep the game warden's cot- the assumption is false. People have trouble reaching
tage or badminton court; and by the time it has tried to well-considered decisions, and you cannot state clearly
frighten you to death and threatened to take away your on paper what is muddled in your mind. And indeed
very existence, you'd be amazed at how little you're people do, for instance, issue mutually inconsistent
willing to settle for."55 instructions in living wills.70
At least sixteen studies have investigated the sta- But assume this difficulty away and the problem
bility of people's preferences for life-sustaining treat- of articulation persists. In one sense, the best way to
ment.56 A meta-analysis of eleven of these studies found divine patients' preferences is to have them write their
that the stability of patients' preferences was 71 percent own living wills to give surrogates the patient's glori-
(the range was 57 percent to 89 percent).57 Although ously unmediated voice. This is not a practical policy.
stability depended on numerous factors (including the Too many people are functionally illiterate,71 and most
illness, the treatment, and demographic variables), the of the literate cannot express themselves clearly in
bottom line is that, over periods as short as two years, writing. It's hard, even for the expert writer. Further-
almost one-third of preferences for life-sustaining more, most people know too little about their choices
medical treatment changed. More particularly, illness to cover all the relevant subjects. Hence living wills
and hospitalization change people's preferences for are generally forms that demand little writing. But the
life-sustaining treatments.58 In a prospective study, the forms have failed. For example, "several studies sug-
desire for life-sustaining treatment declined signifi- gest that even those patients who have completed AD
cantly after hospitalization but returned almost to its forms . . . may not fully understand the function of
original level three to six months later.59 Another study the form or its language."72 Living wills routinely baffle
concluded that the "will to live is highly unstable among patients with their
terminally ill cancer patients."6° The authors thought
their findings "perhaps not surprising, given that only "syntactic complexity, concept density, abstract-
10-14% of individuals who survive a suicide attempt ness, organization, coherence, sequence of ideas,
commit suicide during the next 10 years, which sug- page format, length of line of print, length of para-
gests that a desire to die is inherently changeable." graph, punctuation, illustrations, color, and reader
The consistent finding that interest in life- interest." Unfortunately, most advance directive
sustaining treatment shifts over time and across conforms . . . often have neither a reasonable scope
texts coincides tellingly with research charting people's nor depth. They do not ask all the right questions
struggles to predict their own tastes, behavior, and emo- and they do not ask those questions in a manner
tions even over short periods and under familiar cir- that elicits clear responses.73
cumstances.61 People mispredict what poster they will
like,62 how much they will buy at the grocery store,' Doctors and lawyers who believe their clients are all
how sublimely they will enjoy an ice cream,64 and above average should ask them what their living will
how they will adjust to tenure decisions.65 And people says. One of us (CES) has tried the experiment. The
"miswant" for numerous reasons.66 They imagine a dif- modal answer is, in its entirety: "It says I don't want to
ferent event from the one that actually occurs, nurture be a vegetable."
inaccurate theories about what gives them pleasure,67 No doubt the forms could b e improved, but not
forget they might outwit misery, concentrate on salient enough to matter. The world abounds in dreadfully
negative events and ignore offsetting happier ones,68 drafted forms because writing complex instructions for
and misgauge the effect of physiological sensations the future is crushingly difficult. Statutes read horribly
like pain.69 Given this rich stew of research on people's because their authors are struggling to (1) work out
.11 II Jill III Vil
exactly what rule they want, (2) imagine all the circum- another. The failure to devise workable forms is not a
stances in which it might apply, and (3) find language failure of effort or intelligence. It is a consequence of
to specify all those but only those circumstances. Each attempting the impossible.
task is ultimately impossible, which is why statutes ex-
plicitly or implicitly confide their enforcers with some
Where Is the Living Will?
discretion and why courts must interpret—rewrite?—
statutes. However, these skills and resources are not Suppose that, mirabile dictu, people executed living
available to physicians or surrogates. wills, knew what they will want, and could say it. That
One might retort that property wills work and will not matter unless the living will reaches the people
that living wills are not that far removed from property responsible for the incompetent patient. Often, it does
wills. But wills work as well as they do to distribute not. This should be no surprise, for long can be the
property because their scope is—compared to living road from the drafter's chair to the icu bed.
wills—narrow and routinized. Most people have First, the living will may be signed years before
little property to distribute and few plausible heirs. it is used, and its existence and location may vanish
As property accumulates and ambitions swell, prob- in the mists of time." Roughly half of all living wills
lems proliferate. Many of them are resolvable because are drawn up by lawyers and must somehow reach the
experts—lawyers--exclusively draft and interpret hospital, and 62 percent of patients do not give their
wills. Lawyers have been experimenting for centuries living will to their physician." On admission to the
with testamentary language in a process which has pro- hospital, patients can be too assailed and anxious to
duced standard formulas with predictable meanings recall and mention their advance directives.78 Admis-
and standard ways of distributing property into which sion clerks can be harried, neglectful, and loath to ask
testators are channeled. Finally, if testators didn't say patients awkward questions.
it clearly enough in the right words and following the Thus when a team of researchers reviewed the
right procedures, courts coolly ignore their wishes and charts of 182 patients who had completed a living will
substitute default rules. before being hospitalized, they found that only 26 per-
The lamentable history of the living will demon- cent of the charts accurately recorded information about
strates just how recalcitrant these problems are. There those directives79 and only 16 percent of the charts con-
have been, essentially, three generations of living wills. tained the form. And in another study only 35 percent
At first, they stated fatuously general desires in ab- of the nursing home patients who were transferred to
surdly general terms. As the vacuity of over-generality the hospital had their living wills with them.8°
became clear, advocates of living wills did the obvious:
Were living wills too general? Make them specific. Were Will Proxies Read It Accurately?
they "one size fits all"? Make them elaborate question-
naires. Were they uncritically signed? "Require" prob- Suppose, per impossibile, that patients wrote living wills,
ing discussions between doctor and patient. However, correctly anticipated their preferences, articulated their
the demand for specificity forced patients to address desires lucidly, and conveyed their document to its
more questions than they could comprehend. So, gen- interpreters. How acutely will the interpreters analyze
eralities were insufficiently specific and insufficiently their instructions? Living wills are not self-executing:
considered. Specifics were insufficiently general and someone must decide whether the patient is incompe-
perhaps still insufficiently considered. What was a tent, whether a medical situation described in the
doctor—or lawyer—to do? Behold the "values history," living will has arisen, and what the living will then
a disquisition on the patient's supposed overarching commands.
beliefs from which to infer answers to specific ques- Usually, the patient's intimates will be central
tions.74 That patients can be induced to trek through among a living will's interpreters. We might hope that
these interminable and imponderable documents is intimates already know the patient's mind so that only
unproved and unlikely. That useful conclusions can be modest demands need be made on their interpreting
drawn from the platitudes they evoke is false. As Justice skills. But many studies have asked such surrogates
Holmes knew, "General propositions do not decide to predict what treatment the patient would choose.81
concrete cases."75 Across these studies, approximately 70 percent of the
The lessons of this story are that drafting instruc- predictions were correct—not inspiring success for life
tions is harder than proponents of living wills seem to and death decisions.
believe and that when you move toward one blessing Do living wills help? We know of only one study
in structuring these documents, you walk away from that addresses that question. In a randomized trial,
Fagerlin/Schneider: Enough:The Failure of the Living Will ()
researchers asked elderly patients to complete a disease- misunderstood by astonishing numbers of doctors and
and treatment-based or a value-based living will.82 A con- nurses. And so on.89
trol group of elderly patients completed no living will. But why don't living wills affect care?9° Joan Teno
The surrogates were generally spouses or children who and colleagues saw no evidence "that a physician uni-
had known the patient for decades. Surrogates who were laterally decided to ignore or disregard an AD." Rather,
not able to consult their loved one's living will predicted there was "a complex interaction of . . . three themes."
patients' preferences about 70 percent of the time. Strik- First (as we have emphasized), "the contents of ADs
ingly, surrogates who consulted the living will did no were vague and difficult to apply to current clinical
better than surrogates denied it. Nor were surrogates situations." The imprecision of living wills not only sty-
more successful when they discussed living wills with mies interpreters, it exacerbates their natural tendency
patients just before their prediction. to read documents in light of their own preferences.
What is more, a similar study found that primary Thus "(e)ven with the therapy-specific AD accom-
care physicians' predictions were similarly unimproved panied by designation of a proxy and prior patient—
by providing them with patients' advance directives.83 physician discussion, the proportion of physicians who
On the other hand, emergency room doctors (complete were willing to withhold therapies was quite variable:
strangers) given a living will more accurately predicted cardiopulmonary resuscitation, 100%; administration
patients' preferences than ER doctors without one.84 of artificial nutrition and hydration, 82%; administra-
tion of antibiotics, 80%; simple tests, 70%; and admin-
istration of pain medication, 13%."91
Do Living Wills Alter Patient Care?
Second, the Teno team found that "patients were
Our survey of the mounting empirical evidence shows not seen as 'absolutely, hopelessly ill,' and thus, it was
that none of the five requisites to making living wills never considered the time to invoke the AD." Living
successful social policy is met now or is likely to be. wills typically operate when patients become ter-
The program has failed and indeed is impossible. minally ill, but neither doctors nor families lightly con-
That impossibility is confirmed by studies of how clude patients are dying, especially when that means
living wills are implemented, which show that living ending treatment. And understandably. For instance,
wills seem not to affect patients' treatments. For in- "on the day before death, the median prognosis for
stance, one study concluded that living wills "do not patients with heart failure is still a 50% chance to live
influence the level of medical care overall. This finding 6 more months because patients with heart failure typ-
was manifested in the quantitatively equal use of diag- ically die quickly from an unpredictable complication
nostic testing, operations, and invasive hemodynamic like arrhythmia or infection."92 So by the time doctors
monitoring among patients with and without advance and families finally conclude the patient is dying, the
directives. Hospital and icu lengths of stay, as well as patient's condition is already so dire that treatment
health care costs, were also similar for patients with looks pointless quite apart from any living will. "In all
and without advance directive statements.' Another cases in which life-sustaining treatment was withheld
study found that in thirty of thirty-nine cases in which or withdrawn, this decision was made after a trial of
a patient was incompetent and the living will was in the life-sustaining treatment and at a time when the pa-
patient's medical record, the surrogate decision-maker tient was seen as 'absolutely, hopelessly ill' or 'actively
was not the person the patient had appointed.86 In yet dying' Until patients crossed this threshold, ADs were
a third study, a quarter of the patients received care that not seen as applicable." Thus "it is not surprising that
was inconsistent with their living will.87 our previous research has shown that those with ADs
But all this is normal. Harry Truman rightly pre- did not differ in timing of DNR orders or patterns of
dicted that his successor would "sit here, and he'll say, resource utilization from those without ADs."93
`Do this! Do that!' And nothing will happen. Poor Ike— Third, "family members or the surrogate desig-
it won't be a bit like the army. He'll find it very frustrat- nated in a [durable power of attorney] were not avail-
ing." (Of course, the army isn't like the army either, as able, were ineffectual, or were overwhelmed with their
Captain Truman surely knew) Indeed, the whole law own concerns and did not effectively advocate for the
of bioethics often seems a whited sepulchre for slaugh- patient." Family members are crucial surrogates be-
tered hopes, for its policies have repeatedly fallen woe- cause they should be: patients commonly want them
fully short of their purposes. Informed consent is a to be; they commonly want to be; they specially cher-
"fairytale."88 Programs to increase organ donation have ish the patient's interests. Doctors ordinarily assume
persistently disappointed. Laws regulating DNR orders families know the patient's situation and preferences
are hardly better. Legal definitions of brain death are and may not relish responsibility for life-and-death
II 1111 111111
decisions, and doctors intent on avoiding litigation hopeless and the most hopeful cases. Conversations
may realize that the only plausible plaintiffs are fam- tended to ignore "the more common, less clear-cut
ilies. The family, however, may not direct attention to predicaments surrounding end-of-life care." True, the
the advance directive and may not insist on its enforce- patients all thought "their physicians 'did a good job
ment. In fact, surrogates may be guided by either their talking about the issues,— but this only suggests that
own treatment preferences or an urgent desire to keep patients did not understand how little they were told.
their beloved alive.94 The second candidate for beneficial side effect
In sum, not only are we awash in evidence that the arises from evidence that living wills may comfort
prerequisites for a successful living wills policy are un- patients and surrogates. People with a living will ap-
achievable, but there is direct evidence that living wills parently gain confidence that their surrogates will
regularly fail to have their intended effect. That failure understand their preferences and will implement them
is confirmed by the numerous convincing explanations comfortably, and the surrogates concur.98 Improved
for it. And if living wills do not affect treatment, they satisfaction with decisions was also a rare positive
do not work. effect of the SUPPORT study (which devoted enormous
resources to improving end-of-life decisions and care
but made dismayingly little difference).99 In another
Do Living Wills Have Beneficial Side Effects?
study, living wills reduced the stress and unhappiness
Even if living wills do not effectively promote patients' of family members who had recently withdrawn life
autonomy, they might have other benefits that justify support from a relative. w° But even if living wills make
their costs. There are three promising candidates. patients and surrogates more confident and comfort-
First, living wills might stimulate conversation able, those qualities are apparently unrelated to the
between doctor and patient about terminal treatment. accuracy of surrogates' decisions. Thus we are left with
However, at least one study finds little association be- the irony that one of the best arguments for a tool for
tween patients' reports of executing an advance direc- enhancing peopleb autonomy is that it deceives them
tive and their reports of such conversations.95 Nor do into confidence.
these conversations, when they occur, appear satisfac- Third, because living wills generally constrain
tory." James Tulsky and colleagues asked experienced treatment, they might reduce the onerous costs of ter-
clinicians who had relationships with patients who minal illness. Although several studies associated living
were over sixty-five or seriously ill to "discuss advance wills with small decreases in those costs,' several
directives in whatever way you think is appropriate" studies have reached the opposite conclusion.102 The
with them. Although the doctors knew they were being old Scotch verdict, "not proven," seems apt.
taped, the conversations were impressively short and
one-sided: The median discussion "lasted 5.6 minutes
The Costs
(range, 0.9 to 15.0 minutes). Physicians spoke for a
median of 3.9 minutes (range, 0.6 to 10.9 minutes), There is no free living will, and the better (or at least
and patients spoke for the remaining 1.7 minutes more thorough and careful) the living will, the more it
(range, 0.3 to 9.6 minutes). . . . Usually, the conversa- costs. Living wills consume patients' time and energy.
tion ended without any specific follow-up plan." The When doctors or lawyers help, costs soar. On a broader
"[p]atients' personal values, goals for care, and reasons view, Jeremy Sugarman and colleagues estimated
for treatment preferences were discussed in 71% of that the Patient Self-Determination Act imposed on
cases and were explicitly elicited by 34% of physicians." all hospitals a start-up cost of $101,569,922 and im-
But doctors commonly "did not explore the reasons for posed on one hospital (Johns Hopkins) initial costs of
patientb preferences and merely determined whether $114,528,103. These figures omit the expenses, paid
they wanted specific interventions."97 even as we write and you read, of administering the
Nor were the conversations conspicuously in- program. And this money has bought only pro forma
formative: "Physicians used vague language to describe compliance.
scenarios, asking what patients would want if they These are real costs incurred when over 40 mil-
became 'very very sick' or 'had something that was very lion people lack health insurance and when we are
serious' . . ." Further, "[v]arious qualitative terms were spending more of our gross domestic product on
used loosely to describe outcome probabilities." In health care than comparable countries without buying
addition, these brief conversations considered almost commensurately better health. If programs to promote
exclusively the two ends of the continuum—the most and provide living wills showed signs of achieving the
Fagerlin/Schneider: Enough: The Failure of the Living Will 71
goals cherished for them, we would have to decide without adequate reflection, lack necessary informa-
whether their valuable but incalculable rewards ex- tion, and have fluctuating preferences anyway then
ceeded their diffuse but daunting costs. However, living wills will not lead surrogates to make the choices
since those programs have failed, their costs plainly patients would have wanted. Thus, as Pope suggests,
outweigh their benefits. the "PSDA, rather than promoting autonomy has 'done
a disservice to most real patients and their families
and caregivers.' It has promoted the execution of un-
What Is to Be Done?
informed and under-informed advance directives, and
Living wills attempt what undertakers like to call has undermined, not protected, self-determination.',104
"pre-need planning," and on inspection they are as If living wills have failed, we must say so. We must
otiose as the mortuary version. Critically, empiricists say so to patients. If we believe our declamations about
cannot show that advance directives affect care. This truth-telling, we should frankly warn patients how faint
is damning, but were it our only evidence, perhaps we is the chance that living wills can have their intended
might not be weary in well doing: for in due season we effect. More broadly, we should abjure programs in-
might reap, if we faint not. However, our survey of the tended to cajole everyone into signing living wills. We
evidence suggests that living wills fail not for want of should also repeal the PSDA, which was passed with
effort, or education, or intelligence, or good will but arrant and arrogant indifference to its effectiveness and
because of stubborn traits of human psychology and its costs and which today imposes accumulating paper-
persistent features of social organization. work and administrative expense for paltry rewards.'°5
Thus, when we reviewed the five conditions for Of course we recognize the problems presented by
a successful program of living wills, we encountered the decisions that must be made for incompetent pa-
evidence that not one condition has been achieved or, tients, and our counsel is not wholly negative. Patients
we think, can be. First, despite the millions of dollars anxious to control future medical decisions should be
lavished on propaganda, most people do not have told about durable powers of attorney. These surely do
living wills. And they often have considered and con- not guarantee patients that their wishes will blossom
siderable reasons for their choice. Second, people who into fact, but nothing does. What matters is that powers
sign living wills have generally not thought through its of attorney have advantages over living wills. First, the
instructions in a way we should want for life-and-death choices that powers of attorney demand of patients are
decisions. Nor can we expect people to make thought- relatively few, familiar, and simple. Second, a regime
ful and stable decisions about so complex a question of powers of attorney requires little change from cur-
so far in the future. Third, drafters of living wills have rent practice, in which family members ordinarily act
failed to offer people the means to articulate their informally for incompetent patients. Third, powers of
preferences accurately. And the fault lies primarily not attorney probably improve decisions for patients, since
with the drafters; it lies with the inherent impossibility surrogates know more at the time of the decision than
of living wills' task. Fourth, living wills too often do patients can know in advance. Fourth, powers of at-
not reach the people actually making decisions for in- torney are cheap; they require only a simple form easily
competent patients. This is the most remediable of the filled out with little advice. Fifth, powers of attorney
five problems, but it is remediable only with unsustain- can be supplemented by legislation (already in force
able effort and unjustifiable expense. Fifth, living wills in some states) akin to statutes of intestacy. These stat-
seem not to increase the accuracy with which surro- utes specify who is to act for incompetent patients who
gates identify patients' preferences. And the reasons have not specified a surrogate. In short, durable powers
we surveyed when we explained why living wills do of attorney are—as these things go—simple, direct,
not affect patients' care suggest that these problems are modest, straightforward, and thrifty.
insurmountable. In social policy as in medicine, plausible notions
The cost-benefit analysis here is simple: If living can turn out to be bad ideas. Bad ideas should be
wills lack detectable benefits, they cannot justify any renounced. Bloodletting once seemed plausible, but
cost, much less the considerable costs they now exact. when it demonstrably failed, the course of wisdom
Any attempt to increase their incidence and their was to abandon it, not to insist on its virtues and to
availability to surrogates must be expensive. And the scrounge for alternative justifications for it. Living
evidence suggests that broader use of living wills can wills were praised and peddled before they were fully
actually disserve rather than promote patients' auton- developed, much less studied. They have now failed
omy: If, as we have argued, patients sign living wills repeated tests of practice. It is time to say, "enough."
II IN 111111
Notes
1. R. Drescer, "Missing Persons: Legal Perceptions of 17. L. Emanuel, "Living Wills Can Help Doctors and
Incompetent Patients," Rutgers Law Review 46 (1994): Patients Talk about Dying," Western Journal of Medicine 173
609-695. (2000): 368.
2. J. Lynn, 'Why I Don't Have a Living Will," Law, Medicine & 18. For example, the form provided by a consortium of
Health Care 19, nos. 1-2 (1991): 101-104. the American Bar Association, the American Medical
3. C.P. Sabatino, "End-of-Life Legal Trends," ABA Commission on Association, and the American Association of Retired
Legal Problems of the Elderly 2 (2000). Persons "combines and expands the traditional Living
4. Health Care Assurance of 2001. S. 26. 107th Congress ed; Will and Health Care Power of Attorney into a single,
2001; The Advance Planning and Compassionate Care Act of comprehensive document" (http://www.ama-assn.org/
1999. S. 628. 106th Congress ed; 1999. public/booklets/livgwill.htm).
5. 38 CFR Part 17 RIN. 2900-AJ28. November 2,1998. 19. D.M. High, "Why Are Elderly People Not Using Advance
6. Knight v. Beverly Health Care. 820 Std 92; 2001. Directives?" Journal of Aging and Health 5, no. 4 (1993):
7. See Conservatorship of Wendland, where the California 497-515.
Supreme Court construed the state's Health Care Decisions 20. L.L. Emanuel, "Advance Directives for Medical Care;
Law as "requiring clear and convincing evidence of a Reply," NEJM 325 (1991): 1256; N.L. Cantor, "Making
conscious conservatee's wish to justify withholding life- Advance Directives Meaningful," Psychology, Public
sustaining treatment" but held that decision did not affect Policy, and Law 4, no. 3 (1998): 629-652; D.M. Cox
patients who had left "formal directions for health care." 28 and G.A. Sachs, "Advance Directives and the Patient
P.3d 151; 2001. Self-Determination Act," Clinics in Geriatric Medicine
8. In re Martin. 538 NW2d 399; Mich 1995. 10 (1994): 431-443; G.A.D. Havens, "Differences in
9. Council on Ethical and Judicial Affairs of the American the Execution/Nonexecution of Advance Directives
Medical Association, Surrogate Decision Making E8.081. http:// by Community Dwelling Adults," Research in Nursing
www.ama-assn.org. and Health 23 (2000): 319-333; D.M. High, "Advance
10. P.J. Aitken, "IncorporatingAdvance Care Planning into Family Directives and the Elderly: A Study of Intervention
Practice," American Family Physician," 59 (1999): 605-620; Strategies to Increase Use," Gerontologist 33, no. 3 (1993):
A.O. Calvin and A.P. Clark, "How Are You Facilitating 342-349; S.H. Miles, R. Koepp, and E.P. Weber, "Advance
Advance Directives in Your Clinical Nurse Specialist Practice?' End-of-Life Treatment Planning: A Research Review,"
Clinical Nur se Specialist 16, no. 6 (2002): 292-294. Archives of Internal Medicine 156, no. 10 (1996): 1062-
11. A document that "1,0 ives person responsible for making 1068; S.R. Steiber, "Right to Die: Public Balks at Deciding
medical decisions greater information, specificity and for Others," Hospitals 61 (1987): 572; J. Teno et al., "Do
insight about your specific health-care related decisions, Advance Directives Provide Instructions that Direct Care?
wishes, and objectives" is "A MUST FOR NEARLY SUPPORT Investigators. Study to Understand Prognoses
EVERYONE" (P.A. Meints, "A Trust and Estate Planning and Preferences for Outcomes and Risks of Treatment,"
Questionnaire for Families with Minor Children," The Journal of the American Geriatrics Society 45, no. 4 (1997):
Practical Tax Lawyer 16, no. 1, 12001]: 33). Providing 508-512.
living wills has also become a pro bono activity "Wills 21. Emanuel, "Advance Directives for Medical Care; Reply"
on Wheels was established by a committee of paralegals 22. Miles, Koepp, and Weber, "Advance End-of-Life
and consulting attorneys determined to provide . . . Treatment Planning"; J.L. Holley et al., "Factors
low-income adults with simple wills and living wills" Influencing Dialysis Patients' Completion of Advance
G.M. Price, "pro Bono and Paralegals: Helping to Make a Directives," American Journal of Kidney Diseases 30, no.
Difference," Colorado Lawyer (September 30, 2000), 55-56. 3 (1997): 356-360; L.C. Hanson and E. Rodgman,
12. See www.aarp.org/confacts/programs/endoflife.html. "The Use of Living Wills at the End of Life: A National
13. The form's critical paragraph reads: "My desires concerning Study," Archives of Internal Medicine 156, no. 9 (1996):
medical treatment are-." Then it leaves fourteen blank 1018-1022; Teno et al., "Do Advance Directives Provide
lines the patient may fill in. Available at www.med.umich Instructions that Direct Care?"
.edu/llibr/aha/umlega104.htm. 23. Holley et al., "Factors Influencing Dialysis Patients'
14. R. Dresser, "Precommitment: A Misguided Strategy for Completion of Advance Directives."
Securing Death with Dignity," Texas Law Review 81 (2003): 24. Cox and Sachs, "Advance Directives and the Patient Self-
1823-1847. Determination Act"; Miles, Koepp, and Weber, "Advance
15. A.R. Eiser and M.D. Weiss, "The Underachieving Advance End-of-Life Treatment Planning"; D.M. High, "All in
Directive: Recommendations for Increasing Advance the Family: Extended Autonomy and Expectations in
Directive Completion," American Journal of Bioethics 1 Surrogate Health Care Decision-Making," Gerontologist 28
(2001): 1-5. (suppl) (1988): 46-51.
16. N.L. Cantor, "Twenty-Five Years after Quinlan: A Review 25. L.L. Emanuel and E. J. Emanuel, "The Medical Directive: A
of the Jurisprudence of Death and Dying," Journal of Law, New Comprehensive Advance Care Document,"JAMA 261
Medicine & Ethics 29 (2001): 182-196. (1989): 3288-3293.
Fagerlin/Schneider: Enough:The Failure of the Living Will 8
26. High, "Advance Directives and the Elderly"; J.M. Roe et 35. Ibid.
al., "Durable Power of Attorney for Health Care: A Survey 36. Patient Self-Determination Act of 1990, of the Omnibus
of Senior Center Participants," Archives of Internal Medicine Reconciliation Act of 1990.
152 (1992): 292-296. 37. J.L. Yates and H.R. Glick, "The Failed Patient Self-
27. High, "Why Are Elderly People Not Using Advance Determination Act and Policy Alternatives for the Right to
Directives?"; Roe et al., "Durable Power of Attorney for Die,"Journal of Aging and Social Policy 29 (1997): 29,31.
Health Care." 38. M.T. Pope, "The Maladaptation of Miranda to Advance
28. High, "Why Are Elderly People Not Using Advance Directives: A Critique of the Implementation of the Patient
Directives?"; Roe et al., "Durable Power of Attorney Self-Determination Act," Health Matrix 9 (1999): 139.
for Health Care"; E.J. Emanuel, L.L. Emanuel, and D. 39. Cox and Sachs, "Advance Directives and the Patient Self-
Orentlicher, "Advance Directives," JAMA 266 (1991): Determination Act."
2563-63; G.A. Sachs, C.B. Stocking, and S.H. Miles, 40. J. Hare and C. Nelson, "Will Outpatients Complete Living
"Empowerment of the Older Patient? A Randomized, Wills? A Comparison of Two Interventions," Journal of
Controlled Trial to Increase Discussion and Use of Advance General Internal Medicine 6 (1991): 41-46.
Directives," Journal of the American Geriatrics Society 40, 41. Yamada et al., "A Multimedia Intervention on
no. 3 (1992): 269-273; L.L. Brunetti, S.D. Carperos, and Cardiopulmonary Resuscitation and Advance Directives";
R.E. Westlund, "Physicians' Attitudes towards Living Wills G.A. Sachs, S.H. Miles, and R.A. Levin, "Limiting
and Cardiopulmonary Resuscitation," Journal of General Resuscitation: Emerging Policy in the Emergency Medical
Internal Medicine 6 (1991): 323-329; T.E. Finucane et System," Annals of Internal Medicine 114 (1991): 151-154.
al., "Planning with Elderly Outpatients for Contingencies 42. C.E. Schneider, The Practice of Autonomy: Patients, Doctors,
of Severe Illness: A Survey and Clinical Trial," Journal and Medical Decisions (New York Oxford University Press,
of General Internal Medicine 3, no. 4 (1988): 322-325; 1998).
B. Lo, G.A. McLeod, and G. Saika, "Patient Attitudes to 43. M J. Silveira et al., "Patient's Knowledge of Options at the
Discussing Life-Sustaining Treatment," Archives of Internal End of Life Ignorance in the Face of Death," JAMA 284
Medicine 146, no. 8 (1986): 1613-1615; R. Yamada et (2000): 2483,2486-2487.
al., "A Multimedia Intervention on Cardiopulmonary 44. Yamada et al., "A Multimedia Intervention on
Resuscitation and Advance Directives," Journal of General Cardiopulmonary Resuscitation and Advance Directives";
Internal Medicine 14 (1999): 559-563. S.H. Miles, "Advanced Directives to Limit Treatment: The
29. Cox and Sachs, "Advance Directives and the Patient Need for Portability," Journal of the American Geriatrics
Self-Determination Act"; L.L. Emanuel and E Emanuel, Society 35, no. 1 (1987): 74-76; K.M. Coppola et al.,
"Advance Directives," Annals of Internal Medicine 116 "Perceived Benefits and Burdens of Life-Sustaining
(1992): 348-349; B.B. Ott, "Advance Directives: The Treatments: Differences among Elderly Adults, Physicians,
Emerging Body of Research," AmericanJournal of Critical and Young Adults," Journal of Ethics, Law, and Aging 4, no.
Care 8 (1999): 514-519. 1 (1998): 3-13.
30. J. Sugarman, M. Weinberger, and G. Samsa, "Factors 45. Roe et al., "Durable Power of Attorney for Health Care."
Associated with Veterans' Decisions about Living Wills," 46. J.A. Tulsky et al., "Opening the Black Box: How Do
Archives of Internal Medicine 152 (1992): 343-347. Physicians Communicate about Advance Directives?"
31. Cox and Sachs, "Advance Directives and the Patient Self- Annals of Internal Medicine 129 (1998): 441,444.
Determination Act"; Holley et al., "Factors Influencing 47. C.E. Schneider and M. Farrell, Information, Decisions,
Dialysis Patients' Completion of Advance Directives," High, and the Limits of Informed Consent (New York: Oxford
"All in the Family"; Roe et al., "Durable Power of Attorney University Press, 2000).
for Health Care"; Ott, "Advance Directives"; N.A. Hawkins 48. B.J. McNeil et al., "On the Elicitation of Preferences
et al., "Do Patients Want to Micro-manage Their Own for Alternative Therapies," NE JM 306 (1982):
Deaths? Process Preferences, Values and Goals in End-of- 1259-1262.
Life Medical Decision Making," Unpublished manuscript; 49. T.R. Malloy et al., "The Influence of Treatment
P.B. Terry et al., "End-of-Life Decision Making: When Descriptions on Advance Medical Directive Decisions,"
Patients and Surrogates Disagree," Journal of Clinical Ethics Journal of the American Geriatrics Society 40, no. 12
10, no. 4 (1999): 286-293. (1992): 1255-1260; D.J. Mazur and D.H. Hickman,
32. J. Carrese and L. Rhodes, "Western Bioethics on the "Patient Preferences: Survival versus Quality-of-Life
Navajo Reservation: Benefit or Harm?" JAMA 274 (1995): Considerations,"Journal of General Internal Medicine 8, no.
826-829; L. J. Blackhall et al., "Ethnicity and Attitudes 7 (1993): 374-377; D.J. Mazur and J.F. Merz, "How Age,
toward Patient Autonomy," JAMA 274 (1995): 820-825. Outcome Severity, and Scale Influence General Medicine
33. C.M. Puchalski et al., Patients Who Want Their Family Clinic Patients' Interpretations of Verbal Probability Terms"
and Physician to Make Resuscitation Decisions for Them: (See comments), Journal of General Internal Medicine 9
Observations from SUPPORT and HELP, JAGS 48 (1994): 268-271.
(2000): S84. 50. Miles, Koepp, and Weber, "Advance End-of-Life Treatment
34. High, "Why Are Elderly People Not Using Advance Planning."
Directives?" 51. Ott, "Advance Directives," pp. 514,517.
uu I Al
52. J.J. Christensen-Szalanski, "Discount Functions and the 59. Ditto et al., "A Prospective Study of the Effects of
Measurement of Patients' Values: Women's Decisions Hospitalization."
during Childbirth," Medical Decision Making 4, no. 1 60. H.M. Chochinov et al., "Will to Live in the Terminally Ill,"
(1984): 47-58. Lancet 354 (1999): 816,818.
53. R.M. Gready et al., "Actual and Perceived Stability of 61. D.T. Gilbert and T.D. Wilson, "Miswanting: Some
Preferences for Life-Sustaining Treatment,"Journal of Problems in the Forecasting of Future Affective States," in
Clinical Ethics 11, no. 4 (2000): 334-346. Feeling and Thinking: The Role of Affect in Social Cognition,
54. A. Upadya et al., "Patient, Physician, and Family Member ed. J.P. Forgas (New York: Cambridge University Press,
Understanding of Living Wills," American Journal of 2000): 178-197; C.H. Griffith 3rd et al., "Knowledge
Respiratory and Critical Care Medicine 166 (2002): 1433. and Experience with Alzheimer's Disease: Relationship to
55. W. Sheed, In Love with Daylight: A Memoir of Recovery (New Resuscitation Preference," Archives of Family Medicine 4,
York: Simon and Schuster, 1995): 14. no. 9 (1995): 780-784; T.M. Osberg and J.S. Shrauger,
56. Gready et al., "Actual and Perceived Stability of "Self-Prediction: Exploring the Parameters of Accuracy,"
Preferences for Life-Sustaining Treatment"; J.T. Berger Journal of Personality and Social Psychology 51, no. 5 (1986):
and D. Majerovitz, "Stability of Preferences for Treatment 1044-1057.
among Nursing Home Residents," Gerontologist 28, no. 62. Griffith 3rd et al., "Knowledge and Experience with
2 (1998): 217-223; S. Carmel and E Mutran, "Stability Alzheimer's Disease."
of Elderly Persons' Expressed Preferences Regarding the 63. Gilbert and Wilson, "Miswanting."
Use of Life-Sustaining Treatments," Social Science and 64. D. Kahneman and J. Snell, "Predicting a Changing
Medicine 49, no. 3 (1999): 303-311; M. Danis et al., Taste: Do People Know What They Will Like?"Journal of
"Stability of Choices about Life-Sustaining Treatments," Behavioral Decision Making 5, no. 3 (1992): 187-200.
Annals of Internal Medicine 120, no. 7 (1994): 567-573; 65. Gilbert and Wilson, "Miswanting."
P.H. Ditto et al., "A Prospective Study of the Effects of 66. Ibid.
Hospitalization on Life-Sustaining Treatment Preferences: 67. G. Loewenstein and D. Schkade, "Wouldn't It Be Nice?
Context Changes Choices," Unpublished manuscript; P.H. Predicting Future Feelings," in Hedonic Psychology: Scientific
Ditto et al., "The Stability of Older Adults' Preferences Approaches to Enjoyment, Suffering and Wellbeing, ed. N.
for Life-Sustaining Medical Treatment," Unpublished Schwartz and D. Kahneman (New York: Russell Sage
manuscript; E.J. Emanuel, "Commentary on Discussions Foundation, 1997).
about Life-Sustaining Treatments,"Journal of Clinical Ethics 68. D. Schkade, "Does Living in California Make People
5, no. 3 (1994): 250-251; L.L. Emanuel et al., "Advance Happy? A Focusing Illusion in Judgements of Life
Directives: Stability of Patients' Treatment Choices," Satisfaction," Psychological Science 9 (1998): 340-346.
Archives of Internal Medicine 154 (1994): 209-217; 69. Loewenstein and Schkade, "Wouldn't It Be Nice?"
M.A. Everhart and R.A. Pearlman, "Stability of Patient 70. A.S. Brett, "Limitations of Listing Specific Medical
Preferences Regarding Life-Sustaining Treatments," Chest Interventions in Advance Directives," JAMA 266 (1991):
97 (1990): 159-164; L. Ganzini et al., "The Effect of 825-828.
Depression Treatment on Elderly Patients' Preferences 71. I.S. Kirsch et al., Adult Literacy in America: A First Look
for Life-Sustaining Medical Therapy," American Journal at the Results of the National Adult Literacy Survey, US
of Psychiatry 151, no. 11 (1994): 1631-1636; N. Kohut Department of Education, August 1993, NCES 93275.
et al., "Stability of Treatment Preferences: Although Most 72. Cox and Sachs, "Advance Directives and the Patient Self-
Preferences Do Not Change, Most People Change Some Determination Act"; Miles, Koepp, and Weber, "Advance
of Their Preferences," Journal of Clinical Ethics 8, no. 2 End-of-Life Treatment Planning"; Silveira et al., "Patient
(1997): 124-135; M.D. Silverstein et al., "Amyotrophic Knowledge of Options at the End of Life"; Coppola et
Lateral Sclerosis and Life-Sustaining Therapy: Patients' al., "Perceived Benefits and Burdens of Life-Sustaining
Desires for Information, Participation in Decision Making, Treatments."
and Life-Sustaining Therapy," Mayo Clinic Proceedings 73. Pope, "The Maladaptation of Miranda to Advance
66 (1991): 906-913; J.S. Weissman et al., "The Stability Directives," pp. 139,165-166.
of Preferences for Life-Sustaining Care among Persons 74. D.J. Doukas and L.B. McCullough, "The Values History:
with AIDS in the Boston Health Study," Medical Decision The Evaluation of the Patients Values and Advance
Making 19 (1999): 16-26; K.M. Coppola et al., "Are Life- Directives," Journal of Family Practice 32, no. 2 (1991):
Sustaining Treatment Preferences Stable over Time? An 145-153.
Analysis of the Literature," Unpublished manuscript. 75. Lochner v. New York N. 198 U.S. 45: Supreme Court of
57. Coppola et al., "Are Life-Sustaining Treatment Preferences the United States; 1905.
Stable over Time?" 76. HI Silverman et al., "Implementation of the Patient
58. Danis et al., "Stability of Choices about Life-Sustaining Self-Determination Act in a Hospital Setting: An Initial
Treatments"; Ditto et al., "A Prospective Study of the Evaluation," Archives of Internal Medicine 155, no. 5
Effects of Hospitalization"; Weissman et al., "The Stability (1995): 502-510.
of Preferences for Life-Sustaining Care." 77. Roe et al., "Durable Power of Attorney for Health Care."
Fagerlin/Schneider: Enough:The Failure ofthe Living Will ®
78. R.S. Morrison et al., "The Inaccessibility of Advance C.H. Braddock 3rd et al., "Informed Decision Making in
Directives on Transfer from Ambulatory to Acute Care Outpatient Practice: Time to Get Back to Basics," JAMA
Settings," JAMA 274 (1995): 478-482. 282, no. 24 (1999): 2313-2320.
79. Ibid. 89. C.E. Schneider, 'The Best-Laid Plans," Hastings Center Report
80. M. Danis et al., "A Prospective Study of the Impact of 30, no. 4 (2000): 24-25; C.E. Schneider, "Gang Aft Agley,"
Patient Preferences on Life-Sustaining Treatment and Hastings Center Report 31, no. 1 (2001): 27-28.
Hospital Cost," Critical Care Medicine 24,11 (1996): 90. Teno et al., "Do Advance Directives Provide Instructions
1811-1817. That Direct Care?"
81. J.A. Druley et al., "Physicians' Predictions of Elderly 91. W.R. Mower and L.J. Baraff,"Advance Directives:
Outpatients' Preferences for Life-Sustaining Treatment," Effect of Type of Directive on Physicians' Therapeutic
Journal of Family Practice 37 (1993): 469-475; J. Hare, Decisions," Archives of Internal Medicine 153 (1993):
C. Pratt, and C. Nelson, "Agreement between Patients 375,378.
and Their Self-Selected Surrogates on Difficult Medical 92. J. Lynn, "Learning to Care for People with Chronic Illness
Decisions," Archives of Internal Medicine 152, no. 5 (1992): Facing the End of Life," JAMA 284 (2000): 2508-2509.
1049-1054; P.M. Layde et al., "Surrogates' Predictions 93. J. Teno et al., "The Illusion of End-of-Life Resource Savings
of Seriously Ill Patients' Resuscitation Preferences," with Advance Directives SUPPORT Investigators. Study to
Archives of Family Medicine 4, no. 6 (1995): 518-523; J.G. Understand Prognoses and Preferences for Outcomes and
Ouslander, A. J. Tymchuk, and B. Rahbar, "Health Care Risks of Treatment," Journal of the American Geriatrics Society
Decisions among Elderly Long-Term Care Residents and 45, no. 4 (1997): 513-518.
Their Potential Proxies," Archives of Internal Medicine 149, 94. A. Fagerlin et al., "Projection in Surrogate Decisions about
no. 6 (1989): 1367-1372; A.B. Seckler et al., "Substituted Life-Sustaining Medical Treatments," Health Psychology 20,
Judgment: How Accurate Are Proxy Predictions?" Annals no. 3 (2001): 166-175.
of Internal Medicine 115 (1991): 92-98; D.P. Sulmasy et 95. J. Virmani, U. Schneiderman, and R.M. Kaplan,
al., "The Accuracy of Substituted Judgments in Patients "Relationship of Advance Directives to Physician-Patient
with Terminal Diagnoses," Annals of Internal Medicine 128, Communication," Archives of Internal Medicine 154 (1994):
no. 8 (1998): 621-629; R.F. Uhlmann, R.A. Pearlman, 909-913.
and K.C. Cain, "Physicians' and Spouses' Predictions of 96. J.A. Tulsky, M.A. Chesney, B. Lo, "How Do Medical
Elderly Patients' Resuscitation Preferences," Journal of Residents Discuss Resuscitation with Patients?"Journal of
Gerontology 43, no. 5 (1988): M115-M121; R.F. Uhlmann, General Internal Medicine 10, no. 8 (1995): 436-442.
RA. Pearlman, and K.C. Cain, "Understanding of Elderly 97. Tulsky et al., "Opening the Black Box," pp. 441,445.
Patients' Resuscitation Preferences by Physicians and 98. P.H. Ditto et al., "Advance Directives as Acts of
Nurses," Western Journal of Medicine 150 (1989): 705-707; Communication: A Randomized Controlled Trial, Archives of
N.R. Zweibel and C.K. Cassell, 'Treatment Choices at the Internal Medicine 161, no. 3 (2001): 421-430.
End of Life: A Comparison of Decisions by Older Patients 99. R. Baker et al., "Family Satisfaction with End-of-Life
and Their Physician-Selected Proxies," Gerontologist 29, no. Care in Seriously Ill Hospitalized Adults," Journal of
5 (1989): 615-621. the American Geriatrics Society 48, no 5 (suppl) (2000):
82. L. Emanuel, "The Health Care Directive: Learning How to S61-S69.
Draft Advance Care Documents," Journal of the American 100. V.P. Tilden et al., "Family Decisionmaking to Withdraw
Geriatrics Society 39, no. 12 (1991): 1221-1228; P.H. Ditto Life-Sustaining Treatments from Hospitalized Patients,"
et al., "Fates Worse Than Death: The Role of Valued Life Nursing Research 50, no. 2 (2001): 105-115.
Activities in Health-State Evaluations," Health Psychology 101. Miles, Koepp, and Weber, "Advance End-of-Life
15, no. 5 (1996): 332-343. Treatment Planning."
83. K.M. Coppola et al., "Accuracy of Primary Care and 102. Teno et al., "The Illusion of End-of-Life Resource
Hospital-Based Physicians Predictions of Elderly Savings with Advance Directives"; E.J. Emanuel
Outpatients' Treatment Preferences with and without and L.L. Emanuel, "The Economics of Dying: The
Advance Directives," Archives of Internal Medicine 161, no. Illusion of Cost Savings at the End of Life," NEJM 330
3 (2001): 431-440. (1994): 540-544; L.J. Schneiderman et al., "Effects of
84. Ibid. Offering Advance Directives on Medical Treatments and
85. M.D. Goodman, M. Tamoff, and G.J. Slotman, "Effect Costs," Annals of Internal Medicine 117, no. 7 (1992):
of Advance Directives on the Management of Elderly 599-606.
Critically Ill Patients," Critical Care Medicine 26, no. 4 103. J. Sugarman et al., "The Cost of Ethics Legislation: A Look
(1998): 701-704. at the Patient Self-Determination Act," Kennedy Institute of
86. Morrison et al., "The Inaccessibility of Advance Directives." Ethics Journal 3, no. 4 (1993): 387-399.
87. M. Danis and J.M. Garrett, "Advance Directives for Medical 104. Pope, "The Maladaptation of Miranda to Advance
Care: Reply," NEJM 325 (1991). Directives," pp. 139,167.
88. J. Katz, "Informed Consent-A Fairy Tale? Laws Vision," 105. Yates and Glick, "The Failed Patient Self-Determination
University of Pittsburgh Law Review 39, no. 2 (1977): 137-174; Act"; Sugarman et al., "The Cost of Ethics Legislation."
II 1111 1111ild
Advance Directives for Resuscitation or ones essentially similar to them do not exist. On
the other hand, if an advance directive is so general
and Other Life-Saving or
that it applies to all possible events that could arise it
Life-Sustaining Measures is usually too vague to give any usable direction to the
Canadian Medical Association physician. In either case physicians will have to rely on
their professional judgment to reach a decision.
Some people want to specify in advance the types of Implementation

medical procedures they would or would not want to
undergo in the event that they became incompetent. A physician should assist a patient in a consultative
They can fulfill this desire through a written advance dir- capacity in the preparation of an advance directive
ective, or by appointing a proxy decision-maker, or both. concerning life-saving or -sustaining measures if the
Physicians should assist their patients in these endeav- patient requests such assistance. In the course of this
ours. They should honour a patient's advance directives consultative process, the physician should try to ensure
unless there are reasonable grounds for not doing so. that the patient understands the limits of such docu-
In recent years, patients' concerns over decision- ments. Also, the physician should impress upon the
making in the medical setting have increasingly focused patient the need to make advance directives reasonable
on advance directives for cardiopulmonary resuscita- and accessible. Any such directive should be in writing.
tion, resuscitation in general, and other life-saving or A patient's duly executed advance directive shall be
-sustaining measures. The CMA holds that the right to honoured by the attending physician unless there are rea-
accept or reject any treatment or procedure ultimately sonable grounds to suppose that it no longer represents
resides with the patient or appropriate proxy. This in- the wishes of the patient or that the patient's understand-
cludes the right to accept or ref use resuscitative as well ing was incomplete at the time the directive was prepared.
as other life-saving or -sustaining measures should they Some patients may not wish to execute an advance
become medically indicated. Furthermore, under cer- directive but are concerned about who will make health
tain circumstances it may be appropriate for a patient to care decisions for them when they are no longer able
indicate to the physician and other relevant people, by to do so. Physicians should explore with these patients
means of an advance directive, whether he or she wants the possibility of identifying a specific person who will
such resuscitative measures taken should the need arise. have the legal power to make health care decisions on
Patients frequently believe that an advance direc- their behalf in such an eventuality.
tive to ref use life-saving or -sustaining measures will be Physicians whose patients do wish to draw up
honoured under all circumstances. The reality of med- advance directives should explore with them the pos-
ical practice makes this impossible. If an advance direc- sibility of identifying a specific person who will have
tive is specific to a particular set of circumstances the the legal power to act as their proxy decision-maker
directive will have no force when these circumstances should the need arise for clarification of the directive.
Source: From wwwcma.ca/index.cfm/ci_id/33230/la_id/l.htm (1992). Copyright © 1992. This work is protected by copyright and
the making of this copy was with the permission of Access Copyright. Any alteration of its content or further copying in any form
whatsoever is strictly prohibited unless otherwise permitted by law.
2.5 Best-Interest Judgments
Involving Children in Medical Decisions Eleven-year-old Samantha is a bright, loving child who
was treated for osteosarcoma in her left arm. The arm
Christine Harrison, Nuala R. Kenny, Mona Sidarous, had to be amputated, and Samantha was given a course of
and Mary Rowell chemotherapy. She has been cancer-free for 18 months
Source: From Canadian Medical Association Journal 156, 6 (1997), 825-8. Copyright © 1997. This work is protected by copyright
and the making of this copy was with the permission of Access Copyright. Any alteration of its content or further copying in any
form whatsoever is strictly prohibited unless otherwise permitted by law.
Harrison/Kenny/Sidarous/Rowell: Involving Children in Medical Decisions 77
and is doing well in school. She is self-conscious about family members, their responsibilities toward one
her prosthesis and sad because she had to give away another, and the burdens and benefits of a decision
her cat, Snowy, to decrease her risk of infection. Recent for each member, while acknowledging the special
tests indicate that the cancer has recurred and metasta- vulnerability of the child patient.
sized to her lungs. Her family is devastated by this news A family-centred approach presents special chal-
but do not want to give up hope. However, even with lenges for the health care team, particularly when
aggressive treatment Samantha's chances for recovery there is disagreement between parent and child. Such
are less than 20 per cent. a situation raises profound questions about the nature
Samantha adamantly refuses further treatment. On of the physician—patient relationship in pediatric
earlier occasions she had acquiesced to treatment only practice. Integrity in this relationship is fundamental
to struggle violently when it was administered. She dis- to the achievement of the goal of medicine,6 which
trusts her health care providers and is angry with them has been defined as "right and good healing action
and her parents. She protests, "You already made me taken in the interest of a particular patient."' In the
give up Snowy and my arm. What more do you want?" care of adults, the physician's primary relationship
Her parents insist that treatment must continue. At the is with the particular capable patient. The patient
request of her physician, a psychologist and psych- family may be involved in decision-making, but it is
iatrist conduct a capacity assessment. usually the patient who defines the bounds of such
They agree that Samantha is probably incapable involvement.
of making treatment decisions; her understanding of The care of children, on the other hand, has been
death is immature and her anxiety level very high. described in terms of a "triadic" relationship in which
Nursing staff are reluctant to impose treatment; in the the child, his or her parents, and the physician all have
past, Samantha's struggling and the need to restrain her a necessary involvement (Dr Abbyann Lynch, Director,
upset them a great deal. Ethics in Health Care Associates, Toronto: personal
communication, 1992). When there is disagreement
Why Is It Important to Include Children in between parent and child, the physician may experi-
Medical Decision-Making? ence some moral discomfort in having to deal separ-
ately with the child and parent.
Ethics The assumption that parents best understand
Traditionally, parents and physicians have made all what is in the interest of their child is usually sound.
medical decisions on behalf of children. However, However, situations can arise in which the parents'
just as the concept of informed consent has developed distress prevents them from attending carefully to the
over the last 30 years with respect to competent adult child's concerns and wishes. Simply complying with
patients, so new ways of thinking about the role of the parents' wishes in such cases is inadequate. It is
children in medical decision-making have evolved. more helpful to and respectful of the child to affirm
Ethical principles that provide guidance in the the parents' responsibility for the care of their child
care of adults are insufficient in the context of caring while allowing the child to exercise choice in a meas-
for childrent-3 Issues related to the voluntariness of ure appropriate to his or her level of development and
consent, the disclosure of information, capacity as- experience of illness and treatment. This approach
sessment, treatment decisions, and bereavement are does not discount the parents' concerns and wishes,
more complex, as is the physician's relationship with but recognizes the child as the particular patient to
the patient and the patient's family.3'4 Adult models whom the physician has a primary duty of care. This
presume that the patient is autonomous and has a approach seeks to harmonize the values of everyone
stable sense of self, established values, and mature involved in making the decision.6
cognitive skills; these characteristics are undeveloped
or underdeveloped in children. Law
Although it is important to understand and The legal right to refuse medical treatment is related
respect the developing autonomy of a child, and to, but not identical with, the right to consent to treat-
although the duty of beneficence provides a start- ment. The patient's right to refuse even life-saving med-
ing point for determining what is in the child's best ical treatment is recognized in Canadian law8,9 and is
interest, a family-centred ethic is the best model for premised on the patient's right to exercise control over
understanding the interdependent relationships that his or her own body. Providing treatment despite a pa-
bear upon the child's situation.5 A family-centred tienth valid refusal can constitute battery and, in some
approach considers the effects of a decision on all circumstances, negligence.
■ ui I II 111 111111
To be legally valid, the refusal of medical treatment and parents to make decisions for very young patients
must be given by a person deemed capable of making in their best interest and states that " [p] arents and phys-
health care choices, that is, capable of understanding icians should not exclude children and adolescents from
the nature and consequences of the recommended decision-making without persuasive reasons.'
treatment, alternative treatments, and non-treatment.
In common law the notion of the "mature minor" rec- Empirical Studies
ognizes that some children are capable of making their As they grow, children develop decision-making skills,
own health care choices despite their age.10 In common the ability to reason using complex concepts, an under-
law and under the statutory law of some provinces standing of death,14 and the ability to imagine a future
patients are presumed capable regardless of age unless for themselves.15 Children with a chronic or terminal ill-
shown otherwise; in other provinces an age at which ness may have experiences that endow them with insight
patients are presumed capable is specified." When a and maturity beyond their years. Families often encour-
child's capacity is in doubt an assessment is required. age children to participate in decision-making. Allowing
In the case of children who are incapable of making even young children to make decisions about simple
their own health care decisions, parents or legal guard- matters facilitates the development of skills that they will
ians generally have the legal authority to act as surro- need to make more complex decisions later on.' '8
gate decision-makers. The surrogate decision-maker is Because tools developed to assess the capacity of
obliged to make treatment decisions in the best interest adults have not been tested with children, health care
of the child. Health care providers who believe that a professionals working with children should be sensi-
surrogate's decisions are not in the child's best inter- tive to the particular capacity of each child. Children
est can appeal to provincial child welfare authorities. are constantly developing their physical, intellectual,
The courts have the authority to assume a parens pat- emotional, and personal maturity. Although develop-
riae role in treatment decisions if the child is deemed mental milestones give us a general sense of capacities,
to be in need of protection. This issue has arisen most two children of the same age will not necessarily have
commonly with respect to Jehovah's Witnesses who the same ability to make choices. Even when they are
refuse blood transfusions for their children on religious deemed capable of making health care choices, children
grounds, and courts have authorized treatment in rec- need support for their decisions from family members
ognition of the state's interest in protecting the health and the health care team.
and well-being of children." Every province has child
welfare legislation that sets out the general parameters How Should I Determine the Appropriate
of the "best interest" standard. Courts are reluctant to Role of a Child in Medical Decision-Making?
authorize the withholding or withdrawal of medical
treatment, especially in the face of parental support for Most children fall into one o f three groups with respect to
such treatment. their appropriate involvement in decision-making.19,20
A special point to consider involves the use of
patient restraints. The wrongful or excessive use of re- Infants and Young Children
straints could prompt an action of false imprisonment or Preschool children have no significant decision-making
battery. Restraint can involve the use of force, mechan- capacity and cannot provide their own consent. As surro-
ical means, or chemicals. The use of restraint comprom- gate decision-makers, parents should authorize (or refuse
ises the dignity and liberty of the patient, including the authorization) on their child's behalf, basing their deci-
child patient. Restraints should never be used solely to sions on what they believe to be in the child's best interest.
facilitate care but, rather, only when the patient is likely
to cause serious bodily harm to himself or herself or to Primary-School Children
another. If restraint is required, the health care provider Children of primary-school age may participate in
should use the least restrictive means possible, and the medical decisions but do not have full decision-making
need for the restraint (as well as its effect on the patient) capacity. They may indicate their assent or dissent
should be assessed on an ongoing basis. without fully understanding its implications. None-
theless they should be provided with information ap-
Policy propriate to their level of comprehension. Although
The Canadian Paediatric Society has no policy regarding the child's parents should authorize or refuse to au-
the role of the child patient in medical decision-making thorize treatment, the child's assent should be sought
The American Academy of Pediatrics statement on this and any strong and sustained dissent should be taken
question articulates the joint responsibility of physicians seriously.21
Harrison/Kenny/Sidarous/Rowell: Involving Children in Medical Decisions
Adolescents The Case

Many adolescents have the decision-making capacity of
an adult.22.23 This capacity will need to be determined For Samantha, resuming aggressive treatment will
for each patient in light of his or her have a serious negative effect on her quality of life.
The chances of remission are small, yet a decision to
• ability to understand and communicate relevant discontinue treatment will likely result in her death.
information; Because death is an irreversible harm, and decisions
• ability to think and choose with some degree of with serious consequences require a high level of com-
independence; petence in decision-making,27 the capacity required
• ability to assess the potential for benefits, risks, would be very high. It has been determined that Sam-
or harms as well as to consider consequences and antha does not have this capacity.
multiple options; and Nevertheless, Samantha is included in discussions
• achievement of a fairly stable set of values.24 about her treatment options, and her reasons for re-
fusing treatment are explored. 28 Members of the team
Many children and adolescents, particularly those who work hard to re-establish trust. They and Samantha's
have been seriously ill, will need assistance in develop- parents come to agree that refusing treatment is not ne-
ing an understanding of the issues and in demonstrat- cessarily unreasonable; a decision by an adult patient in
ing their decision-making capacity. Age-appropriate similar circumstances to discontinue treatment would
discussions, perhaps with the assistance of teachers, certainly be honoured. Discussions address Samantha's
chaplains, play therapists, nurses, psychologists, or and her parents' hopes and fears, their understanding
others skilled in communicating with children, are of the possibility of cure, the meaning for them of the
helpful. The child's participation may be facilitated by statistics provided by the physicians, Samantha's role in
the use of art activities, stories, poems, role-playing, decision-making, and her access to information. They
and other techniques.25,28 are assisted by nurses, a child psychologist, a psych-
Physicians should ensure that good decisions iatrist, a member of the clergy, a bioethicist, a social
are made on behalf of their child patients. Although worker, and a palliative care specialist.
the interests of other family members are import- Discussions focus on reaching a common under-
ant and will influence decision-making, the child's standing about the goals of treatment for Samantha. Her
interests are most important and are unlikely to be physician helps her to express her feelings and concerns
expressed or defended by the child himself or her- about the likely effects of continued treatment. Con-
self. Anxious, stressed, or grieving family members sideration is given to the effects on her physical well-
may need assistance in focusing on what is best for being, quality of life, self-esteem, and dignity of imposing
the child. This may be especially difficult when a treatment against her wishes. Spiritual and psychological
cure is no longer possible; in such cases a decision support for Samantha and her family is acknowledged
to stop treatment may seem like a decision to cause to be an essential component of the treatment plan.
the child's death. Opportunities are provided for Samantha and her family
Whether or not the child participates, the follow- to speak to others who have had similar experiences, and
ing considerations should bear upon a treatment deci- staff are given the opportunity to voice their concerns.
sion concerning that child: Ultimately, a decision is reached to discontinue
chemotherapy and the goal of treatment shifts from "cure"
• the potential benefits to the child; to "care." Samantha's caregivers assure her and her family
• the potential harmful consequences to the child, that they are not "giving up" but are directing their efforts
including physical suffering, psychological or toward Samantha's physical comfort and her spiritual and
spiritual distress, and death; and psychological needs. Samantha returns home, supported
• the moral, spiritual, and cultural values of the by a community palliative care program, and is allowed
child's family. to have a new kitten. She dies peacefully.
Notes
1. Ruddick, W. 1979. "Parents and Life Prospects" in Having 2. Nelson, J.L. 1992. "Taking Families Seriously" in Hastings
Children: Philosophical and Legal Reflections on Parenthood, Center Report 22: 6.
ed. 0. O'Neill and W. Ruddick. Oxford University Press: 3. Hardwig, J. 1990. "What about the Family?," in Hastings
New York. 124. Center Report 20 (2): 5-10.
I I ;ill 111111
C
4. Leikin, S. 1989. "A Proposal Concerning Decisions to Forgo 17. Lewis, M.A., and Lewis, C.E. 1990. "Consequences
Life-Sustaining Treatment for Young People," in Journal of of Empowering Children to Care for Themselves," in
Pediatrics 115: 17-22. Pediatrician 17: 63-7.
5. Mahowald, M. 1993. Women and Children in Health Care. 18. Yoos, H.L. 1994. "Children's Illness Concepts: Old and
Oxford University Press: New York. 187,189. New Paradigms," in Pediatric Nursing 20: 134-45.
6. Hellmann, J. 1996. "In Pursuit of Harmonized Values: 19. Broome, M.E., and Stieglitz, K.A. 1992. "The Consent
Patient/Parent-Pediatrician Relationships" in The "Good" Process and Children," in Research in Nursing and Health
Pediatrician: An Ethics Curriculum for Use in Canadian Pediatrics 15: 147-52.
Residency Programs, ed. A. Lynch. Pediatric Ethics Network: 20. Erlen, J.A. 1987. "The Child's Choice: An Essential
Toronto. Component in Treatment Decisions," in Child Health Care
7. Pellegrino, E.D. 1979. 'Toward a Reconstruction of Medical 15: 156-60.
Morality. The Primacy of the Act of Profession and the Fact of 21. Baylis, F 1993. "The Moral Weight of a Child's Dissent," in
Illness," in Journal of Medical Philosophy 4: 47. Ethics in Medical Practice 3 (1): 2-3.
8. Malette v. Shulman 119901,67 DLR (4th) (Ont CA). 22. Weithorn, L.A., and Campbell, S.B. 1982. "The
9. Art 11 CCQ. Competency of Children and Adolescents to Make
10. Rozovsky, L.E., and Rozovsky, F.A. 1992. The Canadian Informed Treatment Decisions," in Child Development 53:
Law of Consent to Treatment. Butterworths: Toronto. 53-7. 1589-98.
11. Etchells, E., Sharpe, G., Elliott, C., and Singer, PA. 1996. 23. Lewis, C.C. 1981. "How Adolescents Approach Decisions:
"Bioethics for Clinicians 3: Capacity," in Canadian Medical Changes over Grades Seven to Twelve and Policy
Association Journal 155: 657-61. Implications," in Child Development 52: 538-44.
12. R.B. v. Children's Aid Society of Metropolitan Toronto, (1995] 24. Brock, D.W. 1989. "Children's Competence for Health
1 SCR 315 (SCC). Care Decision-Making," in Children and Health Care: Moral
13. American Academy of Pediatrics. 1995. "Informed and Social Issues, ed. L.M. Kopelman and J.C. Moskop.
Consent, Parental Permission and Assent in Pediatric Kluwer Academic Publishers: Dordrecht (Holland).
Practice," in Pediatrics 95: 314-17. 181-212.
14. Matthews, G.R. "Children's Conceptions of Illness and 25. Adams, PL., and Fras, I. 1988. Beginning Child Psychiatry.
Death," in Children and Health Care: Moral and Social Issues, Bruner/Mazel: New York.
ed. L.M. Kopelman and J.C. Moskop. Kluwer Academic 26. Kestenbaum, C.J., and Williams, D., eds. 1988. Handbook
Publishers: Dordrecht (Holland). 133-46. of Clinical Assessment of Children and Adolescents. University
15. Koocher, G.P, and DeMaso, D.R. 1990. "Children's Press: New York
Competence to Consent to Medical Procedures," in 27. Drane, J.F. 1985. "The Many Faces of Competency," in
Pediatrician 17: 68-73. Hastings Center Report 15 (2): 17-21.
16. King, N.M.P, and Cross, A.W. 1989. "Children as 28. Freyer, D.R. 1992. "Children with Cancer: Special
Decision-Makers: Guidelines for Pediatricians," in Journal Considerations in the Discontinuation of Life-Sustaining
of Pediatrics 115: 10-16. Treatment," in Medical and Pediatric Oncology 20: 136-42.
Position Statement: Treatment may occur when there is consensus that there is a high
degree of probability that:
Decisions Regarding Infants,
Children, and Adolescents • there is irreversible progression to imminent death;
• treatment is clearly ineffective or harmful;
Christine Harrison; Canadian Paediatric Society,
• life will be severely shortened regardless of treat-
Bioethics Committee
ment, and the limitation or withdrawal of interven-
tions will allow greater palliative and comfort care;
• lives will be filled with intolerable distress and
[...] Withholding or Withdrawing Treatment
suffering that cannot be prevented or alleviated.
Some of the principal goals of health care are to main-
tain life and prevent pain and suffering and not to To withhold or withdraw life-sustaining treatment
unthinkingly prolong the dying process. Exceptions may be ethically appropriate when it is clear that this
to the general duty to provide life-sustaining treatment treatment will not benefit the child or adolescent. Such
Source: From Paediatric Child Health 9 (2004) no. 2,99-103. Posted Feb 1, 2004; reaffirmed Feb 1,2016. Reprinted with permis-
sion of the Canadian Paediatric Society (CPS).
Harrison/Canadian Paediatric Society: Position Statement
treatments include aggressive measures aimed at cure, re- or consultant. All discussions and decisions to withhold
suscitation, mechanical ventilation, and so on. Decisions or withdraw treatment should be well documented and
to withhold or withdraw antibiotics or artificial nutrition reviewed after the child or adolescent's death.
and hydration are more controversial but may also be Some families will be comfortable with decisions to
considered if they are detrimental to the comfort of the withhold or withdraw treatment, while others may take
dying child or adolescent.',2 These difficult and often con- longer to reach this decision. The religious, spiritual,
troversial decisions should not be made in isolation and cultural, and moral background of families should be
should include members of the interdisciplinary team recognized in these situations because they frequently
and, when possible and desired, an ethics committee influence families and their decisions. F. . .1
Notes
1. Miraie ED. Withholding nutrition from seriously ill 2. Nelson LJ, Rushton CH, Cranford RE, Nelson, RM,
newborn infants: A parentS perspective. J Pediatr Glover JJ, Truog RD. Forgoing medically provided nutri-
1988;113:262-5. tion and hydration in pediatric patients. J Law Med Ethics
1995;23:33-46.
Withholding or Withdrawing Life- with the outside world. It may be appropriate to

withdraw or withhold life-sustaining treatment.
Sustaining Treatment in Children* 3. The "No Chance" Situation. The child has such
Royal College of Paediatrics and Child Health severe disease that life-sustaining treatment simply
delays death without significant alleviation of
suffering. Treatment to sustain life is inappropriate.
[. . .1 The RCPCH acknowledges that all members of 4. The "No Purpose" Situation. Although the patient
the child health team, in partnership with parents, may be able to survive with treatment, the degree
have a duty to act in the best interests of the child. of physical or mental impairment will be so great
This includes sustaining life and restoring health to an that it is unreasonable to expect them to bear it.
acceptable standard. However, there are circumstances 5. The "Unbearable" Situation. The child and/or
in which treatments that merely sustain "life" neither family feel that in the face of progressive and ir-
restore health nor confer other benefit and hence are reversible illness further treatment is more than
no longer in the child's best interests. can be borne. They wish to have a particular treat-
There are five situations where it may be ethical ment withdrawn or to refuse further treatment ir-
and legal to consider withholding or withdrawal of life- respective of the medical opinion that it may be of
sustaining medical treatment': some benefit.
1. The "Brain Dead" Child.' In the older child3 where In situations that do not fit with these five categories,
criteria of brain-stem death are agreed by two prac- or where there is uncertainty about the degree of future
titioners in the usual way4 it may still be technically impairment or disagreement, the child's life should
feasible to provide basal cardio-respiratory support always be safeguarded in the best way possible by all in
by means of ventilation and intensive care. It is the Health Care Team, until these issues are resolved.
agreed within the profession that treatment in such Decisions must never be rushed and must always
circumstances is futile and the withdrawal of cur- be made by the team with all evidence available. In
rent medical treatment is appropriate. emergencies it is often doctors in training who are called
2. The "Permanent Vegetative" State.5,6 The child to resuscitate. Rigid rules, even for conditions which
who develops a permanent vegetative state follow- seem hopeless, should be avoided, and life-sustaining
ing insults, such as trauma or hypoxia, is reliant treatment should be administered and continued until
on others for all care and does not react or relate a senior or more experienced doctor arrives.
Source: From Withholding or Withdrawing life Sustaining Treatment in Children, 2nd edn (2004), 10-11. Reprinted with permission of
the Royal College of Paediatrics and Child Health.
• LI
II 111111
The decision to withhold or withdraw life- terminal care needs. These include symptom allevia-
sustaining therapy should always be associated tion and care, which maintains human dignity and
with consideration of the child's overall palliative or comfort. [. .1
Notes
*In 2015 the Royal College published a revised edition of determination of death must be made in accordance with
this framework, in a document titled Malting decisions to limit accepted medical standards.
treatment in life-limiting and life-threatening conditions in children: 3. Original definitions of brain death were not applied to
A framework for practice (Larcher V, Craig F., Bhogal K. et neonates as criteria were thought to be affected by brain
al., Arch Dis Child 2015; 100(Suppl 2): sl-s23). In the 2015 immaturity.
edition, the five situations presented in the 2004 framework 4. Task force for the determination of brain death in
are replaced by a more formal classification based on quality children. "Guidelines for the determination of brain death
or quantity of life (see Sec. 3 in the 2015 edition), but there is in children." Annals of Neurology. (1987)21:616-617.
no indication that the Royal College has changed its view on Pediatrics. (1987)80:298-299.
this matter. We use the 2004 statement in this book because it 5. The vegetative state—guidance on diagnosis and management
identifies more clearly the situations in which the limitation of A Report of a working party of the Royal College of Clinical
treatment can be plausibly considered. Of particular interest Medicine (2003)3:249-254. Defines the vegetative state
and importance is situation 5: The "Unbearable" Situation. and uses the terms "persistent" to mean a vegetative state
1. Withdrawal of curative medical treatment should signal that has persisted for four weeks or more and "permanent"
the initiation of palliative care if this has not already been when the vegetative state is deemed to be permanent and it
introduced. is predicted that awareness will never recover.
2. Definition—Brain death occurs when a child has 6. "The persistent vegetative state." Conference of Medical
sustained either (i) irreversible cessation of circulatory and Royal Colleges and Their Faculties of the United
respiratory functions or (ii) irreversible cessation of all Kingdom. Journal of the Royal College of Physicians, London.
functions of the entire brain including the brain stem. A (1996)30:119-121.
Deciding to Forego Life-Sustaining removing life supporting means is ethical provided

that the normal care given an individual who is ill is
Treatment not discontinued.
Judicial Council, American Medical Association In desperate situations involving newborns, the
advice and judgment of the physician should be readily
available, but the decision whether to exert maximal
[...] 2.10 Quality of Life efforts to sustain life should be the choice of the par-
ents. The parents should be told the options, expected
In the making of decisions for the treatment of benefits, risks, and limits of any proposed care; how
seriously deformed newborns or persons who are the potential for human relationships is affected by
severely deteriorated victims of injury, illness, or the infant's condition; and relevant information and
advanced age, the primary consideration should be answers to their questions. The presumption is that the
what is best for the individual patient and not the love which parents usually have for their children will
avoidance of a burden to the family or to society. be dominant in the decisions which they make in de-
Quality of life is a factor to be considered in deter- termining what is in the best interest of their children.
mining what is best for the individual. Life should be It is to be expected that parents will act unselfishly,
cherished despite disabilities and handicaps, except particularly where life itself is at stake. Unless there is
when prolongation would be inhumane and uncon- convincing evidence to the contrary, parental authority
scionable. Under these circumstances, withholding or should be respected. [. . .1
Source: From Current Opinions of the Judicial Council of the American Medical Association (1982).
Macklin: Ethical Relativism in a Multicultural Society 83
Ethical Relativism in a swing in the direction of families, with urgings to "take

families seriously" (Nelson 1992) and even to consider
Multicultural Society the interests of family members equal to those of the
Ruth Macklin competent patient (Hardwig 1990).
The predominant norm in the United States of dis-
closing a diagnosis of serious illness to the patient is not
Cultural pluralism poses a challenge to physicians and universally accepted even among longstanding citizens
patients alike in the multicultural United States [and comprising ethnic or religious subcultures. . . .
Canada], where immigrants from many nations and di-
verse religious groups visit the same hospitals and doc- Perspectives of Health Care
tors. Multiculturalism is defined as "a social-intellectual Workers and Patients
movement that promotes the value of diversity as a core
principle and insists that all cultural groups be treated A circumstance that arises frequently in multicultural
with respect and as equals" (Fowers and Richardson urban settings is one that medical students bring to
1996: 609). This sounds like a value that few enlight- ethics teaching conferences. The patient and family are
ened people could fault, but it produces dilemmas and recent immigrants from a culture in which physicians
leads to results that are, at the least, problematic if not normally inform the family rather than the patient of
counterintuitive. a diagnosis of cancer. The medical students wonder
Critics of mainstream bioethics . . . have complained whether they are obligated to follow the family's wish,
about the narrow focus on autonomy and individual thereby respecting their cultural custom, or whether
rights. Such critics argue that much—if not most—of to abide by the ethical requirement at least to explore
the world embraces a value system that places the with patients their desire to receive information and to
family the community, or the society as a whole above be a participant in their medical care. When medical
that of the individual person. The prominent American students presented such a case in one of the confer-
sociologist Renee Fox is a prime example of such crit- ences I co-direct with a physician, the dilemma was
ics: "From the outset, the conceptual framework of heightened by the demographic picture of the med-
bioethics has accorded paramount status to the value- ical students themselves. Among the 14 students,
complex of individualism, underscoring the principles 11 different countries of origin were represented.
of individual rights, autonomy, self-determination, and Those students either had come to the United States
their legal expression in the jurisprudential notion of themselves to study or their parents had immigrated
privacy" (Fox 1990: 206). from countries in Asia, Latin America, Europe, and the
The emphasis on autonomy at least in the early Middle East.
days of bioethics in the United States, was never in- The students began their comments with remarks
tended to cut patients off from their families by focus- like, "Where I come from, doctors never tell the patient
ing monistically on the patient. Instead, the intent was a diagnosis of cancer," or "In my country, the doctor
to counteract the predominant and longstanding pater- always asks the patient's family and abides by their
nalism on the part of the medical profession. In fact, wishes." The discussion centred on the question of
there was little discussion of where the family entered whether the physician's obligation is to act in accord-
in and no presumption that a family-centred approach ance with what contemporary medical ethics dictates in
to sick patients was somehow a violation of the patientb the United States or to respect the cultural difference of
autonomy Most patients want and need the support their patients and act according to the family's wishes.
of their families, regardless of whether they seek to be Not surprisingly, the medical students were divided on
autonomous agents regarding their own care. Respect the answer to this question.
for autonomy is perfectly consistent with recognition of Medical students and residents are understandably
the important role that families play when a loved one confused about their obligation to disclose information
is ill. Autonomy has fallen into such disfavour among to a patient when the patient comes from a culture in
some bioethicists that the pendulum has begun to which telling a patient she has cancer is rare or unheard
Source: From The Kennedy Institute of Ethics Journal 9 (1998) no. 1,1-22. Copyright CO The Johns Hopkins University Press.
Reprinted with permission of The Johns Hopkins University Press.
II 1111 1111;111
of. They ask: "Should I adhere to the American custom be told the truth about the diagnosis, 65 per cent of
of disclosure or the Argentine custom of withholding Mexican Americans held that belief, 87 per cent of
the diagnosis?" That question is miscast, since there European Americans believed patients should be told
are some South Americans who want to know if they the truth, and 89 per cent of African Americans held
have cancer and some North Americans who do not. that belief.
It is not, therefore, the cultural tradition that should It is worth noting that the people surveyed were all
determine whether disclosure to a patient is ethically 65 years old or older. Not surprisingly, the Korean and
appropriate, but rather the patient's wish to communi- Mexican American senior citizens had values closer
cate directly with the physician, to leave communica- to the cultures of their origin than did the African
tions to the family or something in between. It would Americans and European Americans who were born
be a simplistic, if not unethical, response on the part in the United States. Another finding was that among
of doctors to reason that "This is the United States, we the Korean American and Mexican American groups,
adhere to the tradition of patient autonomy, therefore older subjects and those with lower socioeconomic
I must disclose to this immigrant from the Dominican status tended to be opposed to truth-telling and pa-
Republic that he has cancer." tient decision-making more strongly than the younger,
Most patients in the United States do want to wealthier, and more highly educated members of these
know their diagnosis and prognosis, and it has been same groups. The authors of the study draw the con-
amply demonstrated that they can emotionally and clusion that physicians should ask patients if they want
psychologically handle a diagnosis of cancer. The same to receive information and make decisions regarding
may not be true, however, for recent immigrants from treatment or whether they prefer that their families
other countries, and it may be manifestly untrue in cer- handle such matters.
tain cultures. Although this, too, may change in time, Far from being at odds with the "autonomy
several studies point to a cross-cultural difference in model," this conclusion supports it. To ask patients
beliefs and practice regarding disclosure of diagnosis how much they wish to be involved in decision-making
and informed consent to treatment. does show respect for their autonomy: patients can
One survey examined differences in the attitudes then make the autonomous choice about who should
of elderly subjects from different ethnic groups toward be the recipient of information or the decision-maker
disclosure of the diagnosis and prognosis of a terminal about their illness. What would fail to show respect
illness and regarding decision-making at the end of life for autonomy is for physicians to make these deci-
(Blackhall et al. 1995). This study found marked dif- sions without consulting the patient at all. If doctors
ferences in attitudes between Korean Americans and spoke only to the families but not to the elderly Korean
Mexican Americans, on the one hand, and African American or Mexican American patients without first
Americans and Americans of European descent, on the approaching the patients to ascertain their wishes,
other. The Korean Americans and Mexican Americans they would be acting in the paternalistic manner of
were less likely than the other two groups to believe the past in America, and in accordance with the way
that patients should be told of a prognosis of terminal many physicians continue to act in other parts of the
illness and also less likely to believe that the patient world today. Furthermore, if physicians automatic-
should make decisions about the use of life-support ally withheld the diagnosis from Korean Americans
technology. The Korean and Mexican Americans sur- because the majority of people in that ethnic group
veyed were also more likely than the other groups to did not want to be told, they would be making an
have a family-centred attitude toward these matters; assumption that would result in a mistake almost 50
they believed that the family and not the patient per cent of the time.
should be told the truth about the patient's diagnosis
and prognosis. The authors of the study cite data from Intolerance and Overtolerance
other countries that bear out a similar gap between the
predominant "autonomy model" in the United States A medical resident in a New York hospital questioned a
and the family-centred model prevalent in European patient's ability to understand the medical treatment he
countries as well as in Asia and Africa. had proposed and doubted whether the patient could
The study cited was conducted at 31 senior citizen grant truly informed consent. The patient, an immi-
centres in Los Angeles. In no ethnic group did 100 per grant from the Caribbean islands, believed in voodoo
cent of its members favour disclosure or non-disclosure and sought to employ voodoo rituals in addition to the
to the patient. Forty-seven per cent of Korean Americans medical treatment she was receiving. "How can anyone
believed that a patient with metastatic cancer should who believes in that stuff be competent to consent
to the treatment we offer?" the resident mused. The There are rarely good grounds for failing to respect
medical resident was an observant Jew who did not the wishes of people based on their traditional religious
work, drive a car, or handle money on the Sabbath and or cultural beliefs. But when beliefs issue in actions that
adhered to Kosher dietary laws. Both the Caribbean cause harm to others, attempts to prevent those harm-
patient and the Orthodox Jew were devout believers in ful consequences are justifiable. An example that raises
their respective faiths and practised the accepted rituals public health concerns is a ritual practised among ad-
of their religions. herents of the religion known as Santeria, practised by
The patient's voodoo rituals were not harmful to people from Puerto Rico and other groups of Caribbean
herself or to others. If the resident had tried to bypass origin. The ritual involves scattering mercury around
or override the patient's decision regarding treatment, the household to ward off bad spirits. Mercury is a
the case would have posed an ethical problem requir- highly toxic substance that can harm adults and causes
ing resolution. Intolerance of another's religious or grave harm to children. Shops called "botanicas" sell
traditional practices that pose no threat of harm is, at mercury as well as herbs and other potions to Carib-
least, discourteous and at worst, a prejudicial attitude. bean immigrants who use them in their healing rituals.
And it does fail to show respect for persons and their The public health rationale that justifies placing
diverse religious and cultural practices. But it does not limitations on people's behaviour in order to protect
(yet) involve a failure to respect persons at a more fun- others from harm can justify prohibition of the sale of
damental level, which would occur if the doctor were mercury and penalties for its domestic use for ritual
to deny the patient her right to exercise her autonomy purposes. Yet the Caribbean immigrants could object:
in the consent procedures. "You are interfering with our religious practices, based
At times, however, it is the family that interferes on your form of scientific medicine. This is our form
with the patient's autonomous decisions. Two brothers of religious healing and you have no right to interfere
of a Haitian immigrant were conducting a conventional with our beliefs and practices." It would not convince
Catholic prayer vigil for their dying brother at his this group if a doctor or public health official were to
hospital bedside. The patient, suffering from terminal reply: "But ours is a well-confirmed, scientific practice
cancer and in extreme pain, had initially been given the while yours is but an ignorant, unscientific ritual." It
pain medication he requested. Sometime later a nurse may very well appear to the Caribbean group as an act
came in and found the patient alert, awake, and in of cultural imperialism: "These American doctors with
excruciating pain from being undermedicated. When their Anglo brand of medicine are trying to impose it
questioned, another nurse who had been responsible on us." This raises the difficult question of how to im-
for the patient's care said that she had not continued plement public health measures when the rationale is
to administer the pain medication because the pa- sufficiently compelling to prohibit religious or cultural
tient's brothers had forbidden her to do so. Under the rituals. Efforts to eradicate mercury sprinkling should
influence of the heavy dose of pain medication, the enlist members of the community who agree with
patient had become delirious and mumbled incoher- the public health position but who are also respected
ently. The brothers took this as an indication that evil members of the cultural or religious group.
spirits had entered the patient's body and, according
to the voodoo religion of their native culture, unless Belief System of a Subculture
the spirit was exorcised it would stay with the family
forever, and the entire family would suffer bad conse- Some widely held ethical practices have been trans-
quences. The patient manifested the signs of delirium formed into law, such as disclosure of risks during
only when he was on the medication, so the broth- an informed consent discussion and offering to pa-
ers asked the nurse to withhold the pain medication, tients the opportunity to make advanced directives in
which they believed was responsible for the entry of the form of a living will or appointing a health care
the evil spirit. The nurse sincerely believed that respect agent. Yet these can pose problems for adherents of
for the family's religion required her to comply with traditional cultural beliefs. In the traditional culture
the patient's brothers' request, even if it contradicted of Navajo Native Americans, a deeply rooted cultural
the patient's own expressed wish. The person in charge belief underlies a wish not to convey or receive negative
of pain management called an ethics consultation, and information. A study conducted on a Navajo Indian
the clinical ethicist said that the brothers' request, even reservation in Arizona demonstrated how Western
if based on their traditional religious beliefs, could not biomedical and bioethical concepts and principles can
override the patient's own request for pain medication come into conflict with traditional Navajo values and
that would relieve his suffering. ways of thinking (Carrese and Rhodes 1995). In March
H kill 111111
1992, the Indian Health Service adopted the require- opportunity to make a living will or appoint a health
ments of the Patient Self-Determination Act, but the care agent. A physician or nurse working for the Indian
Indian Health Service policy also contains the follow- Health Service could easily fulfill this requirement by
ing proviso: "Tribal customs and traditional beliefs that asking Navajo patients if they wish to discuss their
relate to death and dying will be respected to the extent future care or options, without introducing any of
possible when providing information to patients on the negative thinking. This approach resolves one of
these issues" (Carrese and Rhodes 1995: 828). the limitations of the published study. As the authors
The relevant Navajo belief in this context is the acknowledge, the findings reflect a more traditional
notion that thought and language have the power to perspective and the full range of Navajo views is not
shape reality and to control events. The central con- represented. So it is possible that some patients who
cern posed by discussions about future contingencies use the Indian Health Service may be willing or even
is that traditional beliefs require people to "think and eager to have frank discussions about risks of treatment
speak in a positive way" When doctors disclose risks and future possibilities, even negative ones, if offered
of a treatment in an informed consent discussion, they the opportunity.
speak "in a negative way," thereby violating the Navajo It is more difficult, however, to justify withhold-
prohibition. The traditional Navajo belief is that health ing from patients the risks of proposed treatment in
is maintained and restored through positive ritual an informed consent discussion. The article about the
language. This presumably militates against disclosing Navajo beliefs recounts an episode told by a Navajo
risks of treatment as well as avoiding mention of future woman who is also a nurse. Her father was a candidate
illness or incapacitation in a discussion about advance for bypass surgery. When the surgeon informed the pa-
care planning. Western-trained doctors working with tient of the risks of surgery, including the possibility that
the traditional Navajo population are thus caught in a he might not wake up, the elderly Navajo man refused
dilemma. Should they adhere to the ethical and legal the surgery altogether. If the patient did indeed require
standards pertaining to informed consent now in force the surgery and refused because he believed that telling
in the rest of the United States and risk harming their him of the risk of not waking up would bring about
patients by "talking in a negative way"? Or should they that result, then it would be justifiable to withhold that
adhere to the Navajo belief system with the aim of risk of surgery. Should not that possibility be routinely
avoiding harm to the patients but at the same time vio- withheld from all patients, then, since the prospect of
lating the ethical requirement of disclosure to patients not waking up could lead other people—Navajos and
of potential risks and future contingencies? non-Navajos alike—to refuse the surgery? The answer
The authors of the published study draw several is no, but it requires further analysis.
conclusions. One is that hospital policies complying Respect for autonomy grants patients who have
with the Patient Self-Determination Act are ethically been properly informed the right to refuse a proposed
troublesome for the traditional Navajo patients. Since medical treatment. An honest and appropriate disclo-
physicians who work with that population must decide sure of the purpose, procedures, risks, benefits, and
how to act, this problem requires a solution. A second available alternatives, provided in terms the patient can
conclusion is that "the concepts and principles of West- understand, puts the ultimate decision in the hands of
ern bioethics are not universally held" (Carrese and the patient. This is the ethical standard according to
Rhodes 1995: 829). This comes as no surprise. It is a Western bioethics. A clear exception exists in the case
straightforward statement of the thesis of descriptive of patients who lack decisional capacity altogether, and
ethical relativism, the evident truth that a wide variety debate continues regarding the ethics of paternalistic-
of cultural beliefs about morality exist in the world. ally overriding the refusal of marginally competent pa-
The question for normative ethics endures: What fol- tients. This picture relies on a key feature that is lacking
lows from these particular facts of cultural relativity? in the Navajo case: a certain metaphysical account of
A third conclusion the authors draw, in light of their the way the world works. Western doctors and their
findings, is that health care providers and institutions patients generally do not believe that talking about
caring for Navajo patients should reevaluate their poli- risks of harm will produce those harms (although there
cies and procedures regarding advance care planning. have been accounts that document the "dark side" of
This situation is not difficult to resolve, ethically the placebo effect). It is not really the Navajo values
or practically. The Patient Self-Determination Act does that create the cross-cultural problem but rather their
not mandate patients to actually make an advance metaphysical belief system holding that thought and
directive; it requires only that health care institutions language have the power to shape reality and control
provide information to patients and give them the events. In fact, the Navajo values are quite the same as
Macklin: Ethical Relativism in a Multicultural Society
the standard Western ones: fear of death and avoidance Western medical practice. According to one descrip-
of harmful side effects. To understand the relationship tion, "always disclose the risks as well as the bene-
between cultural variation and ethical relativism, it is fits of treatment to patients," the conclusion points
essential to distinguish between cultural relativity that to ethical relativism. But a more general description,
stems from a difference in values and that which can be one that heeds today's call for cultural awareness and
traced to an underlying metaphysics or epistemology. sensitivity, would be: "Carry out an informed consent
Against this background, only two choices are discussion in a manner appropriate to the patient's
apparent: insist on disclosing to Navajo patients the beliefs and understanding." That obligation is framed
risks of treatment and thereby inflict unwanted nega- in a non-relative way. A heart surgeon would describe
tive thoughts on them; or withhold information about the procedures, risks, and benefits of bypass surgery
the risks and state only the anticipated benefits of the in one way to a patient who is another physician, in a
proposed treatment. Between those two choices, there different way to a mathematician ignorant of medical
is no contest. The second is clearly ethically preferable. science, in yet another way to a skilled craftsman with
It is true that withholding information about the risks an eighth grade education, and still differently to a
of treatment or potential adverse events in the future traditional Navajo. The ethical principle is the same;
radically changes what is required by the doctrine the procedures differ.
of informed consent. It essentially removes the "in-
formed" aspect, while leaving in place the notion that Obligations of Physicians
the patient should decide. The physician will still pro-
vide some information to the Navajo patient, but only The problem for physicians is how to respond when
the type of information that is acceptable to the Nav- an immigrant to the United States acts according to the
ajos who adhere to this particular belief system. True, cultural values of her native country values that differ
withholding certain information that would typically widely from accepted practices in American medicine.
be disclosed to patients departs from the ethical ideal Suppose an African immigrant asks an obstetrician to
of informed consent, but it does so in order to achieve perform genital surgery on her baby girl. Or imagine
the ethically appropriate goal of beneficence in the care that a Laotian immigrant from the Iu Mien culture
of patients. brings her four-month-old baby to the pediatrician for
The principle of beneficence supports the with- a routine visit and the doctor discovers burns on the
holding of information about risks of treatment from baby's stomach. The African mother seeks to comply
Navajos who hold the traditional belief system. But so, with the tradition in her native country, Somalia, where
too, does the principle of respect for autonomy. Nav- the vast majority of women have had clitoridectomies.
ajos holding traditional beliefs can act autonomously The Iu Mien woman admits that she has used a trad-
only when they are not thinking in a negative way. If itional folk remedy to treat what she suspected was her
doctors tell them about bad contingencies, that will infant's case of a rare folk illness.
lead to negative thinking, which in their view will What is the obligation of physicians . . . when they
fail to maintain and restore health. The value of both encounter patients in such situations? At one extreme
doctor and patient is to maintain and restore health. is the reply that . . . physicians are obligated to follow
A change in the procedures regarding the informed the ethical and cultural practices accepted here and
consent discussion is justifiable based on a distinctive have no obligation to comply with patients' requests
background condition: the Navajo belief system about that embody entirely different cultural values. At the
the causal efficacy of thinking and talking in a certain other extreme is the view that cultural sensitivity re-
way. The less-than-ideal version of informed consent quires physicians to adhere to the traditional beliefs
does constitute a "lower" standard than that which and practices of patients who have emigrated from
is usually appropriate in today's medical practice. other cultures.
But the use of a "lower" standard is justified by the A growing concern on the part of doctors and
background assumption that that is what the Navajo public health officials is the increasing number of re-
patient prefers. quests for genital cutting and defence of the practice
What is relative and what is non-relative in this by immigrants to the United States and European
situation? There is a clear divergence between the countries. A Somalian immigrant living in Houston
Navajo belief system and that of Western science. said he believed his Muslim faith required him to have
That divergence leads to a difference in what sort his daughters undergo the procedure; he also stated
of discussion is appropriate for traditional Navajos his belief that it would preserve their virginity. He was
in the medical setting and that which is standard in quoted as saying, "It's my responsibility. If I don't do
II ;III IIIIi
it, I will have failed my children" (Dugger 1996: 1). Some Africans living in the United States have said
Another African immigrant, living in Houston, sought they are offended that Congress passed a law prohibit-
a milder form of the cutting she had undergone for her ing female genital mutilation that appears to be directed
daughter. The woman said she believed it was neces- specifically at Africans. France has also passed legisla-
sary so her daughter would not run off with boys and tion, but its law relies on general statutes that prohibit
have babies before marriage. She was disappointed that violence against children (Dugger 1996). In a recent
Medicaid would not cover the procedure, and planned landmark case, a French court sent a Gambian woman
to go to Africa to have the procedure done there. A to jail for having had the genitals of her two baby daugh-
New York City physician was asked by a father for a ters mutilated by a midwife. French doctors report an
referral to a doctor who would do the procedure on increasing number of cases of infants who are brought
his three-year-old daughter. When the physician told to clinics hemorrhaging or with severe infections.
him this was not done in America, the man accused the Views on what constitutes the appropriate re-
doctor of not understanding what he wanted (Dugger sponse to requests to health professionals for advice
1996: 1, 9). or referrals regarding the genital mutilation of their
However, others in our multicultural society con- daughters vary considerably. Three commentators gave
sider it a requirement of "cultural sensitivity" to ac- their opinions on a case vignette in which several Af-
commodate in some way to such requests of African rican families living in a US city planned to have the
immigrants. Harborview Medical Center in Seattle ritual performed on their daughters. If the procedure
sought just such a solution A group of doctors agreed could not be done in the US, the families planned to
to consider making a ritual nick in the fold of skin that have it done in Africa. One of the parents sought advice
covers the clitoris, but without removing any tissue. from health professionals.
However, the hospital later abandoned the plan after One commentator, a child psychiatrist, com-
being flooded with letters, postcards, and telephone mented that professional ethical practice requires her
calls in protest (Dugger 1996). to respect and try to understand the cultural and reli-
A physician who conducted research with East gious practices of the group making the request (Brant
African women living in Seattle held the same view as 1995). She then cited another ethical requirement of
the doctors who sought a culturally sensitive solution. clinical practice: her need to promote the physical and
In a talk she gave to my medical school department, psychological well-being of the child and refusal to
she argued that Western physicians must curb their condone parenting practices that constitute child abuse
tendency to judge cultural practices different from their according to the social values and laws of her city and
own as "rational" or "irrational." Ritual genital cutting country. Most of what this child psychiatrist would do
is an "inalienable" part of some cultures, and it does with the mother who comes to her involves discussion,
a disservice to people from those cultures to view it mutual understanding, education, and the warning
as a human rights violation. She pointed out that in that in this location performing the genital cutting
the countries where female genital mutilation (FGM) ritual would probably be considered child abuse.
is practised, circumcised women are "normal." Like The psychiatrist would remain available for a
some anthropologists who argue for a "softer" linguistic continuing dialogue with the woman and others in
approach (Lane and Rubinstein 1996), this researcher her community, but would stop short of making a
preferred the terminology of "circumcision" to that of child-abuse report since the woman was apparently
"female genital mutilation." only considering carrying out the ritual. However, the
One can understand and even have some sympa- psychiatrist would make the report if she had know-
thy for the women who believe they must adhere to ledge that the mother was actually planning to carry
a cultural ritual even when they no longer live in the out the ritual or if it had already been performed. She
society where it is widely practised. But it does not would make the child-abuse report reluctantly, how-
follow that the ritual is an "inalienable" part of that cul- ever, and only if she believed the child to be at risk
ture, since every culture undergoes changes over time. and if there were no other option. She concluded by
Furthermore, to contend that in the countries where observing that the mother is attempting to act in the
FGM is practised, circumcised women are "normal" is best interest of her child and does not intend to harm
like saying that malaria or malnutrition is "normal" in her. The psychiatrist's analysis demonstrates the pos-
parts of Africa. That a human condition is statistically sible ambiguities of the concept of child abuse. Is abuse
normal implies nothing whatever about whether an determined solely by the intention of the adult? Should
obligation exists to seek to alter the statistical norm for child abuse be judged by the harmful consequences to
the betterment of those who are affected. the child, regardless of the adult's intention? Of course,
Macklin: Ethical Relativism in a Multicultural Society
if a law defines the performance of female genital muti- that since US law prohibits female genital mutilation,
lation as child abuse, then it is child abuse, from a legal which has been likened to child abuse, a health profes-
point of view, and physicians are obligated to report sional would be obligated to inform relevant authorities
any case for which there is a reasonable suspicion. of an intention to commit child abuse. Conceivably, US
Legal definitions aside, intentions are relevant for judg- authorities could prevent immigrants from returning to
ing the moral worth of people, but not for the actions this country if they have gone to Africa to have a pro-
they perform. This means that the good intentions of cedure performed that would be illegal if done within
parents could exonerate them from blame if their ac- the United States. But this is a matter of law, not ethics,
tions cause harm to their children, but the harmful ac- and would involve a gross invasion of privacy since to
tions nevertheless remain morally wrong. enforce the ruling it would be necessary to examine
The second commentator, a clinical psychologist the genitals of the adolescent girls when these families
and licensed sex therapist, would do many of the same sought re-entry into the United States. That would be
things as the child psychiatrist, but would go a bit fur- going too far and probably deserves condemnation as
ther in finding others from the woman's community "ethical imperialism." Since the cutting would already
and possibly another support network (Wyatt 1995). have been done, punitive action toward the family
Like most other commentators on female genital muti- could not succeed in preventing the harm.
lation, this discussant remarked that "agents of change Another case vignette describes a Laotian woman
must come from within a culture" (Wyatt 1995: 289). from the Mien culture who immigrated to the United
The third commentator on this case vignette was States and married a Mien man. When she visited her
the most reluctant to be critical. A British historian and child's pediatrician for a routine four-month immun-
barrister, he began with the observation that "a people's ization, the doctor was horrified to see five red and
culture demands the highest respect" (Martin 1995). blistered quarter-inch round markings on the child's
On the one hand, he noted that custom, tradition, abdomen (Case Study: Culture, Healing, and Profes-
and religion are not easily uprooted. But on the other sional Obligations 1993). The mother explained that
hand, he pointed out that no human practice is beyond she used a traditional Mien "cure" for pain, since she
questioning. He contended that the debate over the thought the infant was experiencing a rare folk illness
nature and impact of female circumcision is a "genu- among Mien babies characterized by incessant crying
ine debate," and the ritual probably had practical util- and loss of appetite, in addition to other symptoms.
ity when it was introduced into the societies that still The "cure" involves dipping a reed in pork fat, light-
engage in it. Of the three commentators, he voiced the ing the reed, and passing the burning substance over
strongest opposition to invoking the child abuse laws the skin, raising a blister that "pops like popcorn."
because it "would be an unwarranted criminalization The popping indicates that the illness is not related to
of parents grappling in good faith with a practice that spiritual causes; if no blisters appear, then a shaman
is legal and customary in their home country" (Martin may have to be summoned to conduct a spiritual ritual
1995: 291). In the end, this discussant would approach for a cure. As many as 11 burns might be needed before
the parents "much as a lawyer would address a jury," the end of the "treatment." The burns are then covered
leaving the parents (like a jury) to deliberate and come with a mentholated cream.
to an informed decision. He would also involve the The Mien woman told the pediatrician that infec-
girls in this process, since they are adolescents, and tion is rare and the burns heal in a week or so. Scars
should have input into the deliberations. sometimes remain but are not considered disfiguring.
It is tempting to wonder whether the involve- She also told the doctor that the procedure must be
ment of adolescent girls in deliberations of their done by someone skilled in burning, since if a burn is
parents would, in traditional Gambian culture, be placed too near the line between the baby's mouth and
even remotely considered, much less accepted. The navel, the baby could become mute or even retarded.
"lawyer-jury-adolescent involvement" solution looks The mother considered the cure to have been success-
to be very Western. If these families living in the ful in the case of her baby since the child had stopped
United States still wish to adhere to their cultural crying and regained her appetite. Strangely enough, the
tradition of genital mutilation, is it likely that they pediatrician did not say anything to the mother about
will appreciate the reasoned, deliberative approach her practice of burning the baby, no doubt from the
this last commentator proposed? need to show "cultural sensitivity" She did, however,
Exactly where to draw the line in such cases is a dif- wonder later whether she should have said something
ficult matter. Presumably, one could go farther than any since she thought the practice was dangerous and also
of these commentators and inform the African families cruel to babies.
II 311 II 1111
One commentator who wrote about this case from other cultures who employ painful and poten-
proposed using "an ethnographic approach" to ethics tially harmful rituals that have no scientific basis. This
in the cross-cultural setting (Carrese 1993). This ap- attitude of tolerance is appropriate against the back-
proach need not result in a strict ethical relativism, ground knowledge that the parents do not intend to
however, since one can be respectful of cultural dif- harm the child and are simply using a folk remedy
ferences and at the same time acknowledge that there widely accepted in their own culture. But tolerance
are limits. What is critical is the perceived degree of of these circumstances must be distinguished from a
harm; some cultural practices may constitute atroci- judgment that the actions harmful to children should
ties and violations of fundamental human rights. The be permitted to continue. What puzzles me is the
commentator argued that the pediatrician must first notion that "cultural sensitivity" must extend so far
seek to understand the Mien woman in the context of as to refrain from providing a solid education to these
her world before trying to educate her in the ways of parents about the potential harms and the infliction of
Western medicine. The commentator stopped short gratuitous pain. . . . We ought to be able to respect
of providing a solution, but noted that many possible cultural diversity without having to accept every single
resolutions can be found for cross-cultural ethical con- feature embedded in traditional beliefs and rituals.
flicts. Be that as it may, we still need to determine which The reluctance to impose modern medicine on
of the pediatrician's obligations should take preced- immigrants from a fear that it constitutes yet another
ence: to seek to protect her infant patient (and possibly instance of "cultural imperialism" is misplaced. Is it not
also the Mien woman's other children) from harmful possible to accept non-Western cultural practices side
rituals or to exhibit cultural sensitivity and refrain from by side with Western ones, yet condemn those that are
attempts at re-education or critical admonitions. manifestly harmful and have no compensating benefit
A second pair of commentators assumed a non- except for the cultural belief that they are beneficial?
judgmental stance. These commentators urged respect The commentators who urged respect for the Mien
for cultural diversity and defended the Mien woman's woman's burning treatment on the grounds that it is
belief system as entirely rational: "It is well grounded in practised widely, the reasons for it are widely under-
her culture it is practised widely; the reasons for it are stood among the Mien, and the procedure works, from
widely understood among the Iu Mien; the procedure, a Mien point of view, seemed to be placing that practice
from a Mien point of view, works" (Brown and Jameton on a par with practices that "work" from the point of
1993: 17). This is a culturally relative view of rational- view of Western medicine. Recall that if the skin does
ity. The same argument could just as well be used to not blister, the Mien belief holds that the illness may
justify female genital mutilation. Nevertheless, the com- be related to spiritual causes and a shaman might have
mentators rejected what they said was the worst choice: to be called. Should the pediatrician stand by and do
simply to tolerate the practice as a primitive cultural nothing, if the child has a fever of 104 degrees and the
artifact and do nothing more. They also rejected the op- parent calls a shaman because the skin did not blister?
posite extreme: a referral of child abuse to the appropri- Recall also that the Mien woman told the pediatrician
ate authorities. The mother's actions did not constitute that if the burns are not done in the right place, the
intentional abuse, since she actually believed she was baby could become mute or even retarded. Must we
helping the child by providing a traditional remedy. reject the beliefs of Western medicine regarding causal-
Here I think the commentators are correct in rejecting ity and grant equal status to the Mien beliefs? To refrain
a referral to the child-abuse authorities, since a charge from seeking to educate such parents and to not exhort
of child abuse can have serious consequences that may them to alter their traditional practices is unjust, as it
ultimately run counter to the best interests of the child. exposes the immigrant children to health risks that are
What did these commentators recommend? Not to not borne by children from the majority culture.
try to prohibit the practice directly, which could alien- It is heresy in today's postmodern climate of re-
ate the parent. Instead, the pediatrician could discuss spect for the belief systems of all cultures to entertain
the risk of infection and suggest safer pain remedies. the notion that some beliefs are demonstrably false and
The doctor should also learn more about the rationale others, whether true or false, lead to manifestly harmful
for and technique of the traditional burning "cure." The actions. We are not supposed to talk about the evolu-
most she should do, according to these commentators, tion of scientific ideas or about progress in the Western
is consider sharing her concerns with the local Mien world, since that is a colonialist way of thinking. If it is
community, but not with the mother alone. simply "the white man's burden, medicalized" (Morsy
There is in these commentaries a great reluctance 1991) to urge African families living in the United
to criticize, scold, or take legal action against parents States not to genitally mutilate their daughters, or to
attempt to educate Mien mothers about the harms of people with and people without disabilities. Are the
burning their babies, then we are doomed to permit cultural groups that discriminate against women or
ethical relativism to overwhelm common sense. people with disabilities to be respected equally with
Multiculturalism, as defined at the beginning of those that do not?
this paper, appears to embrace ethical relativism and What multiculturalism does not say is whether all
yet is logically inconsistent with relativism. The second of the beliefs and practices of all cultural groups must
half of the definition states that multiculturalism "in- be equally respected. It is one thing to require that cul-
sists that all cultural groups be treated with respect tural, religious, and ethnic groups be treated as equals;
and as equals." What does this imply with regard to that conforms to the principle of justice as equality. It
cultural groups that oppress or fail to respect other cul- is quite another thing to say that any cultural practice
tural groups? Must the cultural groups that violate the whatever of any group is to be tolerated and respected
mandate to treat all cultural groups with respect and equally. This latter view is a statement of extreme eth-
as equals be respected themselves? It is impossible to ical relativism. If multiculturalists endorse the principle
insist that all such groups be treated with respect and of justice as equality however, they must recognize
as equals, and at the same time accept any particular that normative ethical relativism entails the illogical
group's attitude toward and treatment of another group consequence of toleration and acceptance of numerous
as inferior. Every cultural group contains subgroups forms of injustice in those cultures that oppress women
within the culture: old and young, women and men, and religious and ethnic minorities.
References
Blackhall, Leslie, Murphy, Sheila T., Frank, Gelya, Michel, Fowers, Blaine J., and Richardson, Frank C. 1996. "Why is Multi-
Vicki, and Azen, Stanley 1995. "Ethnicity and Attitudes culturalism Good?," in American Psychologist 51: 609-21.
Toward Patient Autonomy," in Journal of the American Fox, Renee C. 1990. "The Evolution of American Bioethics: A
Medical Association 274: 820-5. Sociological Perspective," in Social Science Perspectives on
Brant, Renee. 1995. "Child Abuse or Acceptable Cultural Medical Ethics, ed. George Weisz. University of Pennsylva-
Norms: Child Psychiatrist's Response," in Ethics & Behav- nia Press Philadelphia 201-20.
ior 5: 284-7. Hardwig, John. 1990. "What About the Family?," in Hastings
Brown, Kate, and Jameton, Andrew. 1993. "Culture, Healing, Center Report 20 (2): 5-10.
and Professional Obligations: Commentary," in Hastings Lane, Sandra D., and Rubinstein, Robert A. 1996. "Judging
Center Report 23 (4): 17. the Other: Responding to Traditional Female Genital Sur-
Carrese, Joseph. 1993. "Culture, Healing, and Professional geries," in Hastings Center Report 26 (5): 31-40.
Obligations: Commentary," in Hastings Center Report 23 Martin, Tony. 1995. "Cultural Contexts," in Ethics & Behavior
(4): 16. 5: 290-2.
, and Rhodes, Lorna A. 1995. "Western Bioethics on Morsy, Soheir A. 1991. "Safeguarding WomeiA Bodies: The
the Navajo Reservation: Benefit or Harm?," in Journal of White Man's Burden Medicalized," in Medical Anthropol-
the American Medical Association 274: 826-9. ogy Quarterly 5 (1): 19-23.
"Case Study: Culture, Healing, and Professional Obligations." Nelson, James Lindemann. 1992. "Taking Families Seriously,"
1993. In Hastings Center Report 23 (4): 15. in Hastings Center Report 22 (4): 6-12.
Dugger, Celia W. 1996. "Tug of Taboos: African Genital Rite Wyatt, Gail Elizabeth. 1995. "Ethical Issues in Culturally
vs US Law," in New York Times (28 Dec.): 1,9. Relevant Interventions," in Ethics & Behavior 5: 288-90.
2.6 Cases
Case 1 1985 with a history of erratic and threatening behav-

iour. He was variously diagnosed as having bipolar
Scott Starson: Refusing Treatment
disorder and/or schizophrenia. In January 1999, the
while Incompetent Ontario Review Board ordered Mr Starson's manda-
tory 12-month detention at the Centre for Addiction
Physics prodigy Scott Starson had been in and out and Mental Health in Toronto after he was found not
of psychiatric facilities in the US and Canada since criminally responsible by reason of mental illness for
II 'ail 111111
uttering death threats to fellow residents of his room- In a 6-3 split decision handed down in June 2003,
ing house. the Supreme Court overturned the finding of incapacity,
Mr Starson's physicians proposed treatment for stating that the CCB had been guided too much by its
the bipolar disorder that included anti-psychotic, anti- interpretation of Mr Starson's best interests rather than
anxiety, and mood-stabilizing drugs. Mr Starson recog- by a strict interpretation of the law.' According to the
nized that he was mentally ill and that he could not be ruling, patients are not held to a best-interests standard
released until he took the medication, but he ref used when being tested for competence and may therefore
to consent because he said the treatment would slow make decisions that are contrary to their physician's
his thinking and prevent him from engaging in the advice, nor do they have to agree with a physician's pre-
scientific research that gave his life meaning—a state cise diagnosis in order to be deemed competent. The
he described as "worse than death." His physicians court concluded that mentally ill patients are presump-
were convinced that taking the medication would be tively entitled to make their own decisions regarding
beneficial to the patient's mental well-being and would treatment unless a "balance of probabilities" exists to
alleviate any concerns about his behaviour in society. override the decisions.
They also looked to the Canadian Health Care Consent Mr Starson's mother, Jeanne Stevens, believed the
Act, which states that "individuals are considered to Supreme Court decision had ruined her son's life. The
be capable with respect to treatment if they are able medication would have taken away his erratic behav-
to understand the information relevant to making iour, allowing him to work and live in the commun-
decisions about a proposed treatment, and able to ap- ity rather than remaining confined to the Brockville
preciate the reasonably foreseeable consequences of a Psychiatric Hospital with no prospects for release. "I'm
decision or a lack of decision."' It appeared to them devastated and I truthfully believe that the Supreme
that Mr Starson could not adequately appreciate the Court did not have sufficient information. It's the end
value of treatment, so they brought the case before a of his life."2
medical review board, which subsequently determined As it turned out, it was not the end of his life. Mr
that Mr Starson was not competent to make medical Starson developed severe paranoid delusions, refused
decisions for himself. A surrogate decision-maker was food, and fell into a life-threatening situation. His mother
appointed and approved the treatment. then authorized treatment, and his condition improved.
Mr Starson applied to the Ontario Consent and Two years later the board ordered the hospital to assist
Capacity Board (ccB) for a review of the physician's Mr Starson to become established in an apartment in the
decision. The CCB agreed with the physician that Mr community. Was the Supreme Court of Canada right to
Starson was unable to fully appreciate the benefits allow him to refuse his medications? Ontario, Manitoba,
and risks of treatment. Next, the case went to judicial Prince Edward Island, the Northwest Territories, and
review at the Ontario Superior Court of Justice, then to Nunavut are the only jurisdictions that allow capable
the Court of Appeal, which upheld the findings of the patients admitted under the Mental Health Act to refuse
reviewing judge. Finally, the case was appealed to the treatment. Should the mother have been allowed to au-
Supreme Court of Canada. thorize the treatment that he earlier refused?
Notes
1. Starson v. Swayze, 2003 5CC 32, [2003] 1 2. CBC News, "Mentally Ill Man May Refuse Treatment, Court
SCR 722. Rules," 6 June 2003, www.cbc.ca/canada/story/2003/06/06/
starson030606.html.
Case 2 of Eastern Ontario in Ottawa for cancerous tumours

No Chemotherapy for Anael: in his brain and spinal cord. Physicians recommended
chemotherapy, even though the tumours were not then
Surrogate Refusal of Treatment for a life-threatening, but Anael's parents refused. They de-
Minor Child cided instead to feed the boy a vegan diet of organic
vegetables, without sugar or animal products, because
In July 2007, three-year-old Anael IlEsperance- they believed that the body could heal itself if given
Nascimento was being treated at the Children's Hospital the right nutrition. They also were concerned about
Cases 93
the toll prior treatments had taken on the child, who treatment far outweigh the harm.' Chemotherapy
had lost weight and had a greenish complexion: "So we is toxic and highly aggressive, and the evidence to
could not imagine continuing it [the treatment]. . . . We establish that treatment is necessary has so far not
would probably have lost him." been produced. The option remains for hospital of-
Officials at the hospital asked the province's child ficials to petition the court to order treatment at a
protection agency to intervene to force treatment, later date if Anael's condition worsens. "The parents
but the agency declined because the boy's condition are making what is doubtless a loving decision but
was not at that point life-threatening.' According to it's not a rational decision and I think that public
Jean-Pierre Menard, a Quebec medical lawyer, the authorities will override their wishes if there is an ef-
court rarely overrides the right of parents to refuse fective alternative," said Arthur Schafer, a University
treatment for their child unless the advantages of of Manitoba medical ethicist.'
Notes
1. CBC News, "Quebec Parents Refuse Cancer Treatment 2. c-ry News, "Quebec Refuses to Force Chemo on 3-Year-
for Boy, 3," 25 July 2007, wwwcbc.ca/canada/ottawa/ Old Boy," 27 July 2007, www.ctv.ca/servlet/ArticleNews/
story/2007/07/25/ot-chemo-070725.html. story/ctvnews/20070727/cancerkid_treatment_070727/20
070727?hub=TopStories.
Case 3 when her condition inevitably deteriorated; instead,

Do Everything for Mom: Advance she would be offered palliative care in the form of
pain and sedative medications until her death. The
Directives and a Surrogate's Right to physician then sent notice to Ontario's Consent and
Demand Treatment Capacity Board (CCB) that she would be seeking an
order allowing her to withhold life-sustaining treat-
In June 2003, 81-year-old Joyce Holland was living ment from Mrs Holland, specifically cardiopulmonary
with advanced Alzheimer's disease in a Toronto long- resuscitation, ventilation, and adrenaline-type drugs
term care facility when staff noticed that she was swal- to raise blood pressure.
lowing into her lungs rather than into her stomach. The daughters together held Mrs Holland's
After being admitted to Toronto Western Hospital, Mrs power-of-attorney for her personal care and acted as
Holland was diagnosed with aspiration pneumonia, a her proxy decision-maker. They disagreed with Dr
condition that required repeated suction of her lungs. Hawryluck's recommendation on the basis of their
She also could not eat or speak, and her hips and back mother's previously stated wishes, which arose from
were marked with deep, painful bedsores, while her her deeply held religious convictions. At the CCB,
knees, elbows, and fingers were permanently bent. the daughters explained that Mrs Holland "believed
When her condition deteriorated, she was transferred strongly in the sanctity of human life" and argued that
to the intensive care unit, where she was placed on a "She would want to fight for that life, regardless of the
ventilator and given drugs to raise her blood pressure. pain or loss of dignity."' Their mother still watched
This treatment was temporarily successful, and Mrs television and interacted with them, communicating
Holland was released, but she was forced to return one in non-verbal ways—signs, they believed, that her life
week later with an infection in her feeding tube. She was worth saving.
stayed in the ICU for one month before being moved In contrast, Dr Hawryluck explained to the board
to a ward where she continued to develop fevers and that even with heroic efforts in the icu, Mrs Holland
infections and required lung suctioning as frequently as would likely die within six months. The board agreed,
every 45 minutes.' stating that Mrs Holland had not left sufficiently clear
Mrs Holland's medical team, which included instructions to her children about how to be treated
Dr Laura Hawryluck, believed that continued treat- in the current circumstances and that the CCB would
ment would not stop the progress of her disease and decide what was in her best interests. That decision was
would unnecessarily prolong her suffering. The team to permit the withholding-of-treatments request on the
met with Mrs Holland's two daughters, Patricia and grounds that the "pain, discomfort and loss of dignity"
Margaret, to discuss palliative care. Dr Hawryluck sug- Mrs Holland would suffer in the course of further treat-
gested that Mrs Holland not be returned to the ICU ment outweighed any potential benefits.3
II III II ei II 1111 .11111J
The daughters appealed the board's decision to consider inappropriate. He also stated that the CCB
Ontario's Superior Court, claiming that they had acted had not given enough consideration to Mrs Holland's
according to their mother's expressed wishes as the expressed wishes, her desire for and enjoyment of life,
proxies and that the ca had inappropriately usurped and the fact that she wanted everything possible to be
their position as proxy decision-maker for their mother. done.' Medical decision-making was given back to
They further argued that the law that gave the CCB the her daughters; the hospital chose not to appeal. Mrs
ability to override them was unconstitutional, denying Holland remained alive in the icu at Toronto Western
their mother her Charter rights to "life, liberty, and se- Hospital for more than a year, at an estimated cost of
curity of the person." When the case was heard on 20 $1500 per day.'
January 2004, Mrs Holland was back in icu undergoing It should be noted that Ontario is unique among
successful treatment: her lungs cleared, her pneumonia Canadian provinces in legally upholding a patient's
abated, and ventilator support was decreased. Judge previously expressed competent wishes. The CCB is al-
Cullity determined that the Health Care Consent Act lowed to determine a patient's best interests only when
does not directly address whether doctors have to the patientb prior competent wishes were not clearly
seek consent to withhold or withdraw treatment they expressed.'
Notes
1. Andrew Duffy and Pauline Tam, "Patients, Doctors in 3. Scardoni v. Hawtyluch (2004), CanL11 34326 (ON sc).
Ethical 'Grey Zone,— Ottawa Citizen 28 April 2005, www 4. Daniel L. Ambrosini and Anne G. Crocker. "Psychiatric
.canada.com/components/print.aspx?id=13185d20-d6d7- Advance Directives and the Right to Refuse Treatment in
4b33-92fc-a259eec714aftsponsor=. Canada," Canadian Journal of Psychiatry 52, no. 6 (2007):
2. Andrew Duffy and Pauline Tam. "End-of-life Dilemma." Ottawa 397-402.
Citizen 18 April 2005, www.canada.com/ottawacitizen/
story.html?id=b2c44167-a987-44bd-bacb-c293bb53a806.
Case 4 the doctor is unable to provide the patient with the in-
Treatment for Jehovah's Witnesses: formation she would need before making a decision to
withhold consent in this specific emergency situation."'
Adults and Children Mr Justice Robins, in his decision to reject the appeal,
stated that a doctor is not free to disregard a patient's
In June of 1979, Georgette Malette, 57, was involved advance instructions, even if the choice is "contrary to
in a head-on car collision that killed her husband. the mandates of his own conscience and professional
She was unconscious when she arrived at a hospital judgment."' "The principles of self-determination and
in her Kirkland Lake, Ontario, community. Her at- individual autonomy compel the conclusion that the
tending physician, Dr Shulman, determined that she patient may reject blood transfusions even if harmful
was suffering from shock due to blood loss and that consequences may result and even if the decision is
a blood transfusion was necessary to save her life. A generally regarded as foolhardy."3 Although the refusal
nurse going through Mrs Malette's purse discovered a of treatment was not verbally made, since there was
card that identified the patient as a Jehovah's Witness no reason to doubt that the card validly expressed Mrs
and requested that no blood be administered under Malette's desire to withhold consent to blood trans-
any circumstances. After getting a second opinion, fusions, the card was considered by Justice Robins as a
Dr Shulman accepted full responsibility for his ac- legitimate advance directive.
tions and administered the transfusions. Mrs Malette In 2006, a 14-year-old Manitoba girl was hospital-
recovered from her injuries and subsequently success- ized in Winnipeg suffering from Crohn's disease, which
fully sued Dr Shulman for assault and battery, win- can cause gastrointestinal bleeding, therefore requiring
ning an award of $20,000 in damages. Dr Shulman blood transfusions. As a Jehovah's Witness, the youth
appealed to the Ontario Court of Appeal, which heard refused treatment, but Manitoba Child and Family
the case in 1990. Services obtained a court order forcing the transfusion
The basis of Dr Shulman's appeal was that the pres- based on her doctor's recommendation. The province
ence of a card on the patient's body is not an effective followed legislation which sets the threshold age for
informed refusal: "the card cannot be effective when medical decision making at 16. The patient's lawyers
Cases 95
claimed that this decision violated her constitutional interventions are constitutionally sound, striking a
rights and argued that decisions should be made on balance between protecting the child and allowing the
a case-by-case basis according to competence and child to choose. However, they also declared that lower
not some arbitrary age scale. In a 6-1 ruling, the courts must consider minor's maturity and decision-
Manitoba Supreme Court declared that such medical making skills in all future enforced-treatment cases.'
Notes
1. Malette v. Shulman, 1990, Ontario Court of Appeal, (1990) 3. Ibid.

72 O.R. (2d) 417. 4. CBC News, 26 June 2009, http://www.cbc.ca/news/canada/
2. Ibid. story/2009/06/26/supreme-blood026.htm1.
Case 5 To make medical treatment acceptable to our Ab-

Markayla Sault: Caring for an original patients the health care system must earn
their trust by delivering respect. We must ensure
Aboriginal Patient that our Aboriginal patients, their families and
communities feel welcome, are comfortable self-
Markayla Sault was an 11-year-old Aboriginal girl identifying as Aboriginal and do not fear judgment
who had acute lymphoblastic leukemia Doctors gave based on stereotypes E. . .1
Markayla a 72 per cent chance of survival if she continued Medical science is not specific to a single
with chemotherapy. A two-year course of treatment was culture but is shared by Aboriginal and non-
planned, but after 11 weeks Markayla experienced severe Aboriginal people alike. Most Aboriginal people
side effects that put her in the intensive care unit. She seek care from health professionals—but nearly
begged for the chemo to be stopped; the family acceded half also use traditional medicines. Aboriginal
to her wish; Family Services in Ontario did not contest healing traditions are deeply valued ancestral
the decision, and she died in January 2015. practices that emphasize plant-based medicines,
On the day Markayla died, an editorial appeared culture and ceremony, multiple dimensions of
in the Canadian Medical Association Journal (CMAJ) argu- health (physical, mental, emotional and spiritual),
ing that doctors must respect the desire of parents to and relationships between healer, patient, com-
substitute "native" medicine for scientific medicine. In munity and environment. These beliefs create ex-
that editorial, "Caring for Aboriginal Patients Requires pectations that Aboriginal patients bring to their
Trust and Respect Not Courtrooms," Dr Lisa Richard- health care encounters; these must be respected.
son, a specialist in internal medicine and co-leader of Doing so is not political correctness—it is patient-
Toronto's Office of Indigenous Medical Education, and centered care.'
Dr Matthew Stanbrook, deputy editor of CMAJ, argued:
Note
1. Quoted passages taken CMAJ, 187, no. 3 (17 February

2015): 162.
2.7 Study Questions
1. What is meant by "decision-making capacity"?

2. Who decides for those who lack decision-making capacity? According to what stan-
dards or criteria?
3. Anyone who has had a surgery and a flu shot understands that there is a difference
in what risks are disclosed. What reasons can be given for this difference, and do
you think they justify the difference in disclosure? There is likewise a difference
II UM 111111
in what risks are disclosed when surgery is recommended and a medication pre-
scribed. Do you think that standards of disclosure should be different for surgery and
medication?
4. How do instructional advance directives differ from proxy directives? What are the
advantages and risks of instructional advance directives? Under what circumstances, if
any, would you make an instructional advance directive?
5. Under what conditions does the Canadian Paediatric Society think that life-sustaining
treatment can be withdrawn or withheld from infants? How does this differ from the
view of the American Medical Association? Which view do you prefer? Do you think
that families should be able to WLST from their child when suffering is severe, as the
Royal College allows? Would you qualify this in any way, and if so, how?
6. How much should health care providers respect the cultural or religious views of
patients and families? Should female circumcision be allowed? Should male circum-
cision? To what extent should the views of Jehovah's Witnesses or Christian Scientists
be taken into account in making medical decisions?
2.8 Suggested Further Reading
Medical Decision-Making
Brock, Dan W. "What Is the Moral Authority of Family Members t o Act a s Surrogates for
Incompetent Patients?" Millbank Quarterly 74 (1996): 599-618.
Gawande, Atul. "Whose Body Is It Anyway?" The New Yorker 4 October 1999,84-91.
Groopman, Jerome. How Doctors Think (New York: Houghton Mifflin Harcourt, 2007).
Jonsen, Albert R., Mark Siegler, and William J. Winslade. Clinical Ethics: A Practical Ap-
proach to Ethical Decisions in Clinical Medicine. Updated in many subsequent editions
(New York: McGraw-Hill, 1992).
Marway, Herjeet, and Heather Widdows. 2015. "Philosophical Feminist Bioethics: Past,
Present, and Future." Cambridge Quarterly of Healthcare Ethics 24, no. 2 (April): 165-83.
Ross, Lainie Friedman. 1998. Children, Families, and Health Care Decision-Making (New
York: Oxford University Press, 1998).
Sullivan, Mark D., and Stuart J. Youngner. "Depression, Competence, and the Right to
Refuse Livesaving Medical Treatment." American Journal of Psychiatry 151 (July 1994):
971-8.
Weir, Robert F, and Charles Peters. "Affirming the Decisions Adolescents Make about Life
and Death." Hastings Center Report 27 (November/December 1997): 29-40.
Advance Directives
Brock, Dan W. "A Proposal for the Use of Advance Directives in the Treatment of Incompe-
tent Mentally Ill Persons." Bioethics 7 (April 1993): 247-56.
Browne, Alister, and Bill Sullivan. 2006. "Advance Directives in Canada." Cambridge Quar-
terly of Healthcare Ethics 19, no. 3 (summer): 256-60.
Dresser, Rebecca. 1994. "Confronting the 'Near Irrelevance' of Advance Directives." Journal
of Clinical Ethics 5: 55-6.
King, Nancy. Making Sense of Advance Directives (Dordrecht: Kluwer Academic Publishers,
1991)
Lynn, Joanne. 1991. "Why I Don't Have a Living Will." Law, Medicine & Health Care 19,
no. 1-2: 101-4.
Olick, Robert S. 2001. Taking Advance Directives Seriously: Prospective Autonomy and Deci-
sions Near the End of Life (Washington: Georgetown University Press, 2001).
Suggested Further Reading
Teno, Joan M., Joanne Lynn, et al. 1994. "Do Formal Advance Directives Affect Resuscita-
tion Decisions and the Use of Resources for Seriously Ill Patients?" Journal of Clinical
Ethics 5 (1994): 23-30 (with following commentary).
Informed Consent
Bok, Sissela. "Shading the Truth in Seeking Informed Consent." Kennedy Institute of Ethics
Journal 5 (1995): 1-17.
Buchanan, Allen E., and Dan W. Brock. Deciding for Others: The Ethics of Surrogate Decision
Making (Cambridge, MA: Cambridge University Press: 1989).
Dolgin, Janet L. "The Legal Development of the Informed Consent Doctrine: Past and
Present." Cambridge Quarterly of Healthcare Ethics 19, no. 1 (January 2010): 97-101.
Faden, Ruth R. "Informed Consent and Clinical Research." Kennedy Institute of Ethics Jour-
nal 6 (1996): 356-9.
and Tom L Beauchamp. A History and Theory of Informed Consent (New York:
Oxford University Press, 1986).
Gostin, Lawrence 0. "Informed Consent, Cultural Sensitivity, and Respect for Persons."
Journal of the American Medical Association 274 (13 September 1995): 844-5.
Howe, Edmund G. "Leaving Laputa: What Doctors Aren't Taught about Informed Con-
sent." The Journal of Clinical Ethics 11 (2000): 3-13.
Kondo, Douglas G., E Marian Bishop, and Jay A. Jacobson. "Residents' and Patients' Per-
spectives on Informed Consent in Primary Care Clinics." Journal of Clinical Ethics 11
(2000): 39-48.
Meisel, Alan, and Mark Kuczewski. "Legal and Ethical Myths about Informed Consent."
Archives of Internal Medicine 156 (December 1996): 2521-6.
Veatch, Robert M. "Abandoning Informed Consent." Hastings Center Report 25 (March-
April 1995): 477-99.
111
?Nfranagernent of viledka[
[Inforrinatjon
3.1 Introduction
Patients and their health care providers have a special relationship. Health care providers
should deliver health benefits to patients. They should do this while respecting patient au-
tonomy, avoiding harm, and acting justly to all those concerned with a patient's well-being.
But maximizing health benefits sometimes conflicts with these other values. For example,
telling a patient that he or she has terminal cancer may have harmful effects on the patient's
general well-being. Telling a patient his or her diagnosis may also be something frowned
on in the patient's culture of origin. The patient's family, concerned with the patient's well-
being, may say that they and not the patient are the ones to be given information about the
diagnosis and possible treatments. Another patient may insist that information about their
medical condition is private and not to be shared with others, even other family members.
What should a health care provider do in these circumstances? Questions like this are
about the management of patient medical information and are the topic of this chapter.
Access to Information
Therapeutic Privilege
As we are told in Chapter 2, there are two models or conceptions of how information is
to be controlled in the relationship between health care providers and their patients. The
therapeutic privilege traditional idea, known as therapeutic privilege, is that everything is in the hands of
the idea that a medical medical professionals. This conception denies that a patient has a right to complete and
professional may deceive or
correct information about the diagnosis or treatments available or expectations for recov-
withhold the truth from a
patient when providing the
ery after treatment. Complete disclosure might worry the patient unduly or even cause
truth will produce harm. serious harm. Telling only part of the truth or even lying can sometimes be good medicine.
A health care professional should decide what information is to be made available to the
patient based on a judgment of what will advance good care.
A more recent conception sees a health care professional as someone with professional
skills and judgment but also a duty to make full disclosure to the patient. A medical professional
on this model may not lie or withhold information from a competent patient. As Chapter 2
emphasizes, this is essential for a patient making an informed decision about the risks and
benefits of proposed treatments. Patients with a terminal illness not only need accurate infor-
mation to make choices about medical care but also how to use the remaining part of their lives.
Access to Information
Lying and Withholding the Truth

We should begin with a distinction.
• A lies to B = A communicates in some way something to B that A believes to be false,

with the goal of deceiving someone (usually B) and in circumstances in which A could
have communicated what A believes to be true.
• A withholds the truth from B = A does not communicate something to B that A believes
to be true, either with the goal of deceiving someone (usually B) or with the goal of
leaving someone (usually B) in ignorance and in circumstances in which A could have
communicated what A believes to be true.
Are there ever instances when a physician or other medical professional does one of
the above without doing the other? The most commented on instances involve the use of
placebos in treatment. Suppose a doctor prescribes something for a patient that she knows placebo a treatment all of
is just a placebo. The doctor might tell the patient that the treatment is not a placebo; the whose effects are placebo
effects; placebo effects are
doctor lies. But the doctor might instead not say anything one way or the other or might
a good or bad effect of a
say something vague like "This treatment has proved to be effective in your kind of case." treatment that occurs because
If the patient is left with the belief that the treatment is not a placebo, then the doctor has the recipient believes the
withheld the truth but not lied. treatment will have the effect.
Does the difference between one way of deceiving a patient (lying) and the other
(withholding the truth) make for a moral difference? It is easy to think that lying is mor-
ally worse than withholding the truth. One is actively doing something while the other is
merely passively letting something happen. To return to the example of a placebo treat-
ment, a doctor who tells a patient that a treatment is not a placebo when it really is seems
to be behaving in a morally worse way than one who manages to allow a patient to think it
is not a placebo without actually saying so. But a minor change in James Rachels's famous
argument against the idea that active killing is morally worse than passively letting some-
one die in "Active and Passive Euthanasia" (see Chapter 6, page 301) should lead us to
doubt that these forms of deception—one active and the other passive—are ethically that
different. Returning to the example of a placebo treatment, in both cases the patient agree-
ing to the treatment is not fully informed by his or her doctor to the extent required by the
standards outlined in Chapter 2.
LA. C
/
The Ethics of Lying
Is it ever proper for a medical professional to lie to a patient, at least when the intent is -to
do good for the patient, as the traditional model of the patient—professional relationship
says it is? The simplest answer to this question comes from philosopher Immanuel Kant
in the reading "On the Supposed Right to Lie from Benevolent Motives" (see page 106).
According to a common (though not universally agreed upon) interpretation of Kant's
essay, it is always wrong to lie to anyone about anything, no matter what the conse-
quences and no matter what the ultimate goal of the deception.' On this understanding
of what Kant is saying, there is nothing special about the relationship between a doctor
and patient or the truth about a patient's medical condition when it comes to this issue.
Why is it always wrong to deceive anyone or at least a patient about his or her diagnosis maxim Kant's name for a
or prognosis and leave them not fully informed? Kant seems to be saying that lying to personal rule for acting r a
anyone even once about the truth would make it impossible to have any confidence that certain way. On some ideas
aboutwhat Kant means, this
anyone is ever being honest rather than deceitful. In Kant's own words, by telling a lie,
must include a statement of
"I cause that declarations in general find no credit" (see page 106). This statement seems the goal of acting and the
an exaggeration. If one person tells a lie to a patient, why would that by itself alter the kind of action that will achieve
believability of every other person? Kant might have in mind his idea that the maxim, the goal.
YU I ai II 'al iuw
3 Management of Medical Information
or personal rule, governing a choice must be one that the person telling the lie is able to
successfully require everyone to follow.
For example, the maxim or rule that doctors who keep the truth from a patient might
be following is "If it would produce good results or avoid bad ones, then a patient (and
anyone who might tell the patient) should be deceived about his or her true medical
condition." Could the doctors require that everyone follow this rule? Would it be self-
defeating or impossible in some way for medical professionals and everyone in contact
with a patient to follow this rule all the time? Well, if every one of them all the time
followed this maxim or rule, no patient would be successfully deceived. The attempts at
deception would be discovered frequently enough that no patient would give any "credit,"
to use Kant's word, to what they are told by the medical staff treating them or to what
anyone else tells them about their medical condition. So the maxim is not universalizable,
and deception is wrong.
This thinking, however, tells us directly only about deception by lying. It does not tell
us about deception achieved by withholding the truth. On the other hand, if these two
types of deception are morally equal as suggested above, Kant's reasoning does say that
withholding the truth as a way of deceiving a patient is also wrong.
Are there any arguments for the conclusion that lying is sometimes the right thing
for a medical professional to do? David Thomasma in "Telling the Truth to Patients:
A Clinical Ethics Exploration" (see page 107) and Robert Stewart in "Telling the Patients
the Truth" (see page 110) list many cases in which medical professionals may be doing
the right thing when they deceive a patient or keep a patient ignorant of his or her true
medical state. One case listed as one in which physicians are allowed or even morally
required to deceive involves the harm that would result from the patient being informed.
As mentioned above, deception is allowed on the traditional conception of the patient—
professional relationship as a therapeutic privilege. The harm need not be medical harm;
it could be psychological or emotional. But recent research (briefly mentioned in "Ethical
Relativism in a Multicultural Society" by Ruth Macklin in Chapter 2; see page 83) con-
cludes that harms of this kind really don't result from patients being told the truth. So
this argument for deceiving a patient or even keeping a patient in the dark (and not fully
informed about their diagnosis or the chances of their medical condition being success-
fully treated) won't work.
Macklin also reports that doctors who have trained in countries other than the US or
Canada say that in their country, medical professionals are expected to provide the truth
to members of the patient's family but not to the patient himself or herself—at least when
the diagnosis or prognosis is bad (as one for terminal lung cancer might be). A reason for
this stance in these other countries and cultures, however, is that it may be understood
and accepted that members of the family become the substitute decision-makers when a
patient is seriously ill, even if the patient is still competent enough to understand what is
happening and to make his or her own decisions about treatments.
Access to Medical Records

The traditional model of the relationship between a medical professional and a patient
denies a patient access to his or her medical records or chart. Sometimes these records
contain statements of ideas considered by a professional as part of the process of diagnos-
ing an illness. These statements can be a source of worry to a layperson patient. They
also often contain technical jargon not everyone understands. But as Arthur Schafer, in
"Medical Secrecy: Patients' Right of Access to Medical Records" (see page 113) points out,
there are many good reasons to allow a patient access to these records. For example, such
access is essential if a patient is to make an informed decision about authorizing the release
of confidential information in the records to third parties.
Confidentiality
Confidentiality
The Duty of Confidentiality

Dave Unger's article, "Breaching Confidentiality," clearly explains (see page 128) that the
duty of confidentiality generally means a physician or other health care professional cannot
reveal any medical information about their patients to others without the patient's consent.
The professional duty of confidentiality covers not only what patients reveal to doctors and
other medical professionals during the course of treatment but also what they conclude
based on their examination or assessment of a patient. It covers all medical records (includ-
ing x-rays, lab reports, and so on) as well as direct communications between a patient and
a doctor or others. Keeping confidences is one of the oldest principles of medical ethics.
In the ancient Greek Hippocratic Oath, new physicians promise this: Hippocratic Oath this
ancient Greek text requires
a new physician to swear to
About whatever I may see or hear in treatment, or even outside of treatment, in
the healing gods to follow
the life of men and women that ought not to be spoken abroad, I will remain specific ethical standards.
silent, holding such things to be unutterable.2
A newer version of the code, written by the dean of the Tufts University medical
school, has new doctors promise this:
I will respect the privacy of my patients, for their problems are not disclosed to
me that the world may known
Exceptions to the Duty

But almost all professional medical ethics codes as well as Canadian laws say that this duty
to keep patient medical information secret and out of the hands of others is not absolute.
According to the Canadian Nurses Association code of ethics, for example, nurses are
acting properly when they disclose patient health information even without the patient's
consent to those who need to know it.3 (The code does not say how to decide who needs
to know) The code of ethics for Canadian psychologists says that information learned in
confidence from a patient must be passed on when keeping it confidential would lead
to serious harm to others or loss of their lives.4 The Canadian Medical Association tells
its members that they should generally keep their patients' personal health information
confidential; nevertheless, they may disclose it to others when the law requires them to do
so.' The code also says that doctors should make sure that their patients understand that
there are these exceptions to complete confidentiality.
But what is the best policy here? Should the duty to respect confidences be absolute—
without exceptions—or not? There are several possibilities, and some are considered in the
readings in this chapter. Here is one:
Unqualified confidentiality: Patients should understand that no information about

them obtained during treatment will be passed on to anyone else under any cir-
cumstances without the consent of the patient.
This policy is supported by Kenneth Kipnis in "A Defense of Unqualified Medical Confi-
dentiality" (see page 117). Kipnis sometimes gives it the label "absolute confidentiality."
He thinks better consequences for everyone will result from following this rule about
confidentiality than by following any other possible rule. Kipnis calls other possible
rules about confidentiality "qualified." Here is a statement of a general rule that qualifies
confidentiality.
II ill 11111,1
Qualified confidentiality: Patients should understand that some information about

them obtained during treatment will be passed on to others under some circum-
stances even without the consent of the patient.
In theory, there are many qualifications or exceptions that fill in the details of this general
rule. Here are three examples:
Example 1: Patients should understand that some information about them ob-
tained during treatment will be passed on to others in some circumstances even
without their consent. These circumstances are ones in which passing on the
information will prevent serious harms to others.
without their consent. These circumstances include ones in which the law re-
quires that the information be passed on.
without their consent. These circumstances include ones in which passing on the
information will benefit the public.
The first two of these qualified rules can be found in Canadian law and professional
codes as well as in law and professional codes in many other countries. The third is
found in British law:6 Supporters of all three rules believe that the best, consequences
will follow from their favoured rule. Consider the last qualified rule (example 3). What
are the benefits to the public that are supposed to come about when information ac-
quired during the treatment of patients is passed on without their consent? An import-
ant source of data for medical research will be opened up; surely this is a significant
benefit, as Bonnie Kaplan notes in "How Should Health Data Be Used?" (see page 134).
Other benefits include the production of reliable health statistics, such as how long the
average Canadian might be expected to live or what percentage of people living in a
certain area are HIV+.
Big Data
A final issue about the confidentiality of medical information is a particularly modern
one. It arises when patient medical information is stored in a way that can be made
easily available for what are sometimes called "secondary uses." The primary purpose
or use for medical information is to benefit the patient, to help in his or her treatment.
Pretty much anything else is a secondary use. Many of these secondary uses of medical
data are realistically available only when medical records are stored electronically and
can be processed and combined with other information stored by computers. Prac-
tically speaking, it would be impossible for such big amounts of data to be processed
manually, so the issues arising from storing medical information electronically is a
modern concern.
Secondary uses of information from a patient's medical record include medical re-
search (as mentioned above), the collection of health statistics (as mentioned above),
and the management of health care costs. These uses are forbidden under the absolute
confidentiality rule without the patient's explicit consent and are the sorts of things
allowed by some of the qualified rules even without this consent. Another second-
ary use is allowing someone or a private company not primarily concerned with the
health of the patient to make money. This is what happened in the two legal cases,
one from the US and the other from the UK, mentioned by Kaplan in her article. In
both cases, pharmacists sold the prescription information they had about patients to
what are called "data mining" companies. These companies mine data in the same way
Confidentiality
that other companies mine the earth. They summarize the data or put it together with
other information they have access to. In these two cases, the data mining companies
weren't interested in what the prescription records said about the patients. (In fact,
the prescription information sold to them by the pharmacists hid patient information
in a way described in the following paragraphs.) Rather, the data mining companies
were interested in the behaviour of the medical professionals writing the prescriptions.
The records contained the names and other identifying information about these med-
ical professionals. The companies searched the records for data about w.bia_doctors
prescribed which drugs and how often. Then the data was sold to drug manufacturing
companies. In_turn, the_ drug manufacturers used the information to help target their
marketing and sales efforts.
Kaplan points out that confidentiality practices commonly require medical records
used in these secondary ways to be anonymized or de-identified. The idea is that when
medical information is stripped of personally identifying information, any secondary
use no longer breaks even the most absolute standard for confidentiality. Unfortu-
nately, this anonymization is not always effective, given all the information available
in public or other records about almost everyone in Canada. Imagine that a hospital
or doctor's office has a number of electronic medical records that look like those in
Table 3.1 below.
These records (rows in the table) contain confidential medical information in the last
column and the person's name in column 1. Table 3.2 shows the result of anonymiza-
tion. The table still has information useful for medical and social science researchers and
administrators.
What do we now need to learn from Table 3.2 that it was John who had the symptoms
of wheezing? If we know from elsewhere—for example, from public census data or the
kind of information people provide every day when they complete forms on the Web—that
there is someone who was born on 17 April 1994 and is male and lives in the postal code
forward sortation area V6T, then the chance is better than 87 per cent that this is John.'
If Table 3.2 contained the full Canadian postal code, then we could determine that it is
John who is N3 with almost complete certainty.' Table 3.2 tells us the medical complaint
of this individual. So we know—with a very high probability—that this was John's medical
problem. Confidentiality of medical records is therefore not always completely protected
by the common practice of anonymization.
Table 3.1 Data with Identifying Details

Name Birth Date Sex Visit Date Postal Code Medical Issue
Sean Phillips 21 Mar 1995 Male 14 Apr 2016 V6T1Z1 Fever
May Ong 10 Nov1994 Female 15 Apr 2016 V6Y2Z6 Chest pain
John Scatopolus 17 Apr 1994 Male 15 Apr 2016 V6T1Z1 Wheezing
Lisa Friedrich 01 Dec 1993 Female 16 Apr 2016 V6Y2Z6 Aching joints
Table 3.2 Data with Identifying Details Removed

Name Birth Date Sex Postal Code FSA Medical Issue
N1 21 Mar 1995 Male V6T Fever
N2 10 Nov 1994 Female V6Y Chest pain
N3 17 Apr 1994 Male V6T Wheezing
N4 01 Dec 1993 Female V6Y Aching joints
I II all 1111.1J
Genetic Information
Here are two examples of genetic tests that provide medically relevant information:
The BRCA test. Using a blood sample, the test analyzes DNA to identify harmful changes
(mutations) in either one of two genes, BRCA1 and BRCA2. Men and women with these
harmful changes have a significantly greater risk of developing breast cancer and, in
women, ovarian cancer compared to the average. Inherited BRCA gene mutations are
responsible for 5 to 10 per cent of breast cancers and about 15 per cent of ovarian
cancers.
The Huntington's Disease (HD) test. Using a blood sample, the test analyzes DNA to iden-
tify harmful numbers of CAG repeats in parts of the Huntington gene. Individuals with
28 or fewer repeats usually will be free of the disease. Those with 40 or more repeats
usually will have the disease, with more repeats meaning that the disease symptoms
will start at a younger age.
In the US there are special laws giving greater protections to genetic health information
like the results of the BRCA or HD tests than to other kinds of health information,8 a practice
genetic exceptionalism the known as genetic exceptionalism. Some of these laws forbid discrimination by insur-
view that genetic information ance companies or employers based on the results of these tests. For example, someone
is a special type of medical
with a higher risk of breast cancer because of a BRCA mutation cannot be refused medical
information and must
therefore be treated with insurance or fired from a job because of this. Some of these laws in the US require that
greater care—for example, hospitals, doctors' offices, the providers of genetic testing services, and other places where
with special laws--than other this information might be created or stored make special efforts to keep it confidential.
health care information. Canada currently does not have any laws that treat the confidentiality of genetic informa-
tion like results of the BRCA or HD or other genetic tests differently from other personal
health information.
Are the results of genetic tests like the BRCA or HD test so different that they should
be given special treatment compared to the results of, for example, a pregnancy test
or a prostate cancer screening test? This is the topic of Thomas H. Murray's "Genetic
Exceptionalism and 'Future Diaries'" (see page 147). Genetic test results can have what
seem like unique and almost paradoxical consequences. Consider A and her genetically
identical twin sister B. (Even identical twins are not completely genetically identical,
but we'll suppose A and B are in this scenario.) A would like to have a test for the BRCA
mutation done. Her family—and of course B's family—have a history of breast and ovar-
ian cancer. Her cancer doctor thinks having this test is a good choice. It would also seem
that A is perfectly within her rights to acquire this kind of medical information about
herself. Now, if A has the test, she can figure out medical information about her genetic-
ally identical twin sister B. She can do this because she knows that B is her genetically
identical twin and not because she peeks at B's confidential medical records. But this is
private information about B's health state, information requiring B's consent to find out.
So A may not have the test providing her with information about her own medical status
without B's consent. Most of the time, however, another person like B will not be As exact
genetic twin. But often A will be able to determine the chance that B has a particular
genetic medical issue. Suppose A had the test for the Huntington's Disease mutation long
before she has or might have any symptoms. If A finds she has a harmful number of CAG
repeats, then she can figure out that her biological full brothers and sisters have at least
a 50 per cent chance of having the harmful mutation as well. If this is information that
any one of her brothers or sisters might not want anyone else to know, then the concern
for the confidentiality and privacy of their medical state forbids A from finding this out
about herself.
Genetic Information
Is any of this unique to medical information that comes from the analysis of someone's
genetic makeup? Not really, says Murray in his article. Suppose A has a test for an HIV infec-
tion. She tests positive. A has a partner, C. Since A and C do not use protection methods
that would prevent C from being infected by A, it is almost certain that C is infected too.
But this is information about C's health that C might want to keep private. So again A may
not find out information about her own HIV status, at least not without C's permission. Or
suppose A and her sister B grew up in Waltham, Massachusetts, a suburb of Boston, in the
1980s. They both drank lots of water supplied by the city. This water was heavily con-
taminated with chemicals that significantly increased the risk of leukemia. Since A knows
that B drank this city water, A knows something about B's risk for this kind of cancer.
Is A allowed to figure this out about B or even herself without B's consent?
One reason why health information based on a person's DNA is said to be special
is its prophetic character, as Murray emphasizes: it tells us something about a person's
likely future health. Someone with a harmfully large number of CAG repeats in his or
her Huntington gene will get Huntington's Disease. But not all genetic information is like
this. The presence of a harmful BRCA mutation doesn't mean that breast cancer is certain.
It does mean, though, that the chance of getting breast cancer is much greater than average.
Anyone who has one copy of the E4 mutation of the APOE gene—something else that can
be found through DNA analysis—has a three times greater risk of developing Alzheimer's
disease than someone with no copies of the mutation. Having two copies means that the
risk is 15 times greater. But these predictions are not special to genetic information. Before
treatments were available to keep multiplication of the virus that causes AIDS in check,
anyone infected with HIV was almost certainly destined to have AIDS in the near future.
It seems that genetic medical information is often not that different from other kinds
of medical information, and there is no reason, as Murray concludes, to give it special
legal protections.
Notes
1. Helga Varden, "Kant and Lying to the Murderer at the Door," Journal of Social Philosophy 41 (2010),
403-21; Christine Korsgaard, "The Right to Lie: Kant on Dealing with Evil," Philosophy and Public
Affairs 15 (1986), 325-49; Thomas L. Carson, Lying and Deception: Theory and Practice (New York:
Oxford University Press, 2010), ch. 3.
2. Steven H. Miles, The Hippocratic Oath and the Ethics of Medicine (New York Oxford University
Press, 2004).
3. Canadian Nurses Association, Code of Ethics for Registered Nurses (Ottawa: Canadian Nurses
Association, 2008).
4. Canadian Psychological Association, Canadian Code of Ethics for Psychologists, 3r d edn. (Ottawa:
Canadian Psychological Association, 2000).
5. Canadian Medical Association, cma Code of Ethics, update 2004 (Ottawa: Canadian Medical
Association, 2015).
6. J. O'Brien and 0. Chantler, "Confidentiality and the Duties of Care," Journal of Medical Ethics 29
(2003), 36-40.
7. The estimate is based on a study by privacy researcher Latanya Sweeney for the US. Latanya
Sweeney, "Simple Demographics Often Identify People Uniquely," Carnegie Mellon University Data
Privacy Working Paper 3 (Pittsburgh, 2000). The study estimated that there is a slightly better than
87 per cent chance of identifying a person in the US using his or her birthdate, sex, and zip code.
A US zip code has about the same average number of residents as a Canadian postal code forward
sortation area. A full Canadian postal code has a much smaller population (less than 1 per cent of
a forward sortation area). Using Sweeney's calculations, identification of a person is nearly certain
given just that person's birthdate, sex, and full Canadian postal code.
8. Department of Health and Human Services, "Genetic Information Nondiscrimination Act of 2008"
(Washington: Department of Health and Human Services, 2009).
I I IA 111111
106 3 Management of Medical Information
3.2 Patient Access to Information
On the Supposed Right to Lie him or someone else, he is not actually bound to be
untruthful in a certain statement to which an unjust
from I3enevolent Motives compulsion forces him.
Immanuel Kant Truth in utterances that cannot be avoided is the
formal duty of a man to everyone, however great the
In the work called France, for the year 1797, Part VI. disadvantage that may arise from it to him or any
No. 1, on Political Reactions, by Benjamin Constant, the other; and although by making a false statement I do
following passage occurs, p. 123:— no wrong to him who unjustly compels me to speak,
yet I do wrong to men in general in the most essential
"The moral principle that it is one's duty to speak point of duty, so that it may be called a lie (though
the truth, if it were taken singly and uncondition- not in the jurist's sense), that is, so far as in me lies
ally, would make all society impossible. We have I cause that declarations in general find no credit,
the proof of this in the very direct consequences and hence that all rights founded on contract should
whi:h have been drawn from this principle by a lose their force; and this is a wrong which is done to
German philosopher, who goes so far as to affirm mankind.
that to tell a falsehood to a murderer who asked If then, we define a lie merely as an intention-
us whether our friend, of whom he was in pur- ally false declaration towards another man, we need
suit, had not taken refuge in our house, would not add that it must injure another, as the jurists
be a crime." think proper to put in their definition (mendacium
est falsiloquium in prajudicium alterius). For it always
The French philosopher opposes this principle in the injures another, if not another individual, yet man-
following manner, p. 124:— kind generally, since it vitiates the source of justice.
This benevolent lie may, however, by accident (casus)
"It is a duty to tell the truth. The notion of duty is become punishable even by civil laws; and that which
inseparable from the notion of right. A duty is what escapes liability to punishment only by accident may
in one being corresponds to the right of another. be condemned as a wrong even by external laws. For
Where there are no rights there are no duties. To instance, if you have by a lie hindered a man who is
tell the truth then is a duty, but only towards him even now planning a murder, you are legally respon-
who has a right to the truth. But no man has a sible for all the consequences. But if you have strictly
right to a truth that injures others." The 7rpofitru adhered to the truth, public justice can find no fault
cpaic6os here lies in the statement that "To tell the with you, be the unforeseen consequence what it may.
truth is a duty, but only towards him who has a right It is possible th4 whilst you have honestly answered
to the truth." Yes to the murderer's question, whether his intended
victim is in the house, the latter may have gone out
It is to be remarked, first, that the expression "to unobserved, and so not have come in the way of the
have a right to the truth" is unmeaning. We should murderer, and the deed therefore have not been done;
rather say, a man has a right to his own truthfulness (ver- whereas, if you lied and said he was not in the house,
acitas), that is, to subjective truth in his own person. and he had really gone out (though unknown to you)
For to have a right objectively to truth would mean so that the murderer met him as he went, and exe-
that, as in meum and tuum generally, it depends on his cuted his purpose on him, then you might with justice
will whether a given statement shall be true or false, be accused as the cause of his death. For, if you had
which would produce a singular logic. spoken the truth as well as you knew it, perhaps the
Now, the first question is whether a man—in murderer while seeking for his enemy in the house
cases where he cannot avoid answering Yes or No— might have been caught by neighbours coming up
has the right to be untruthful. The second question is and the deed been prevented. Whoever then tells a lie,
whether, in order to prevent a misdeed that threatens however good his intentions may be, must answer for
Source: From Kant's Critique of Practical Reason and Other Works on the Theory of Ethics, trans. T.K. Abbott (London: Longmans,
Green and Co., 1898). © SophiaOmni 2012.
Kant: On the Supposed Right to Lie from Benevolent Motives
the consequences of it, even before the civil tribunal, rendered uncertain and useless if even the least excep-
and must pay the penalty for them, however unfore- tion to them were admitted.
seen they may have been; because truthfulness is a To be truthful (honest) in all declarations is there-
duty that must be regarded as the basis of all duties fore a sacred unconditional command of reason, and
founded on contract, the laws of which would be not to be limited by any expediency.
Telling the Truth to Patients: In General Relationships

A Clinical Ethics Exploration In each of the three main reasons why the truth must
be told, as a right, a utility, and a kindness, lurk values
David C. Thomasma that may from time to time become more important
than the truth. When this occurs, the rule of truth-
telling is trumped, that is, overridden by a temporar-
Reasons for Telling the Truth
ily more important principle. The ultimate value in
In all human relationships, the truth is told for a myriad all instances is the survival of the community and/or
of reasons. A summary of the prominent reasons is that the well-being of the individual. Does this mean for
it is a right, a utility, and a kindness. paternalistic reasons, without the person's consent,
It is a right to be told the truth because respect for the right to the truth, the utility and the kindness, can
the person demands it. As Kant argued, human society be shunted aside? The answer is "yes." The truth in a
would soon collapse without truth-telling, because it is relationship responds to a multivariate complexity of
the basis of interpersonal trust, covenants, contracts, values, the context for which helps determine which
and promises. values in that relationship should nredominate.
The truth is a utility as well, because persons need Nothing I have said thus far suggests that the truth
to make informed judgments about their actions. It is may be treated in a cavalier fashion or that it can be with-
a mark of maturity that individuals advance and grow held from those who deserve it for frivolous reasons. The
morally by becoming more and more self-aware of their only values that can trump the truth are recipient survival,
needs, their motives, and their limitations. All these community survival, and the ability to absorb the full
steps toward maturity require honest and forthright impact of the truth at a particular time. All these are only
communication, first from parents and later also from temporary trump cards in any event. They can only be
siblings, friends, lovers, spouses, children, colleagues, played under certain limited conditions because respect
co-workers, and caregivers.' for persons is a foundational value in all relationships.
Finally, it is a kindness to be told the truth, a kind-
ness rooted in virtue precisely because persons to whom In Health Care Relationships
lies are told will of necessity withdraw from important, It is time to look more carefully at one particular form
sometimes life-sustaining and life-saving relationships. of human relationship, the relationship between the
Similarly, those who tell lies poison not only their rela- doctor and the patient or sometimes between other
tionships but themselves, rendering themselves incap- health care providers and the patient.
able of virtue and moral growth.2 . . . Early in the 1960s, studies were done that revealed
the majority of physicians would not disclose a diagno-
Overriding the Truth sis of cancer to a patient. Reasons cited were mostly
those that derived from nonmaleficence. Physicians
. . . Not all of us act rationally and autonomously at all were concerned that such a diagnosis might disturb
times. Sometimes we are under sufficient stress that the equanimity of a patient and might lead to desper-
others must act to protect us from harm. This is called ate acts. Primarily physicians did not want to destroy
necessary patemalism.,Should we become seriously ill, their patients' hope. By the middle 1970s, however,
others must step in and rescue us if we are incapable of repeat studies brought to light a radical shift in phys-
doing it ourselves. . . . ician attitudes. Unlike earlier views, physicians now
Source: From Cambridge Quarterly of Healthcare Ethics 3 (1994), 375-82. Copyright 1994 Cambridge University Press. Reprinted
with permission of the Cambridge University Press.
II 111413
emphasized patient autonomy and informed consent temporarily in favour of more important long-term
over paternalism. In the doctor—patient relation, this values, which are known in the relationship.
meant the majority of physicians stressed the patient's Finally, the goal of health care relations is treatment
right to full disclosure of diagnosis and prognosis. of an illness. An illness is far broader than its subset, dis-
One might be tempted to ascribe this shift of atti- ease. Illness can be viewed as a disturbance in the life of
tudes to the growing patients' rights and autonomy move- an individual, perhaps due to many non-medical factors.
ments in the philosophy of medicine and in public affairs. A disease, by contrast, is a medically caused event that
No doubt some of the change can be attributed to this may respond to more interventionist strategies.5
movement. But also, treatment interventions for cancer Helping one through an illness is a far greater per-
led to greater optimism about modalities that could offer sonal task than doing so for a disease. A greater, more
some hope to patients. Thus, to offer them full disclosure enduring bond is formed. The strength of this bond
of their diagnosis no longer was equivalent to a death sen- may justify withholding the truth as well, although in
tence. Former powerlessness of the healer was supplanted the end "the truth will always out."
with technological and pharmaceutical potentialities.
A more philosophical analysis of the reasons for a Clinical Case Categories
shift conies from a consideration of the goal of medi-
cine. The goal of all healthcare relations is to receive/ The general principles about truth-telling have been
provide help for an illness such that no further harm reviewed, as well as possible modifications formed
is done to the patient, especially in that patient's vul- from the particularities of the health care professional—
nerable state.3 The vulnerability arises because of in- patient relationship. Now I turn to some contemporary
creased dependency. Presumably, the doctor will not examples of how clinical ethics might analyze the_ bier-
take advantage of this vulnerable condition by adding archy of values surrounding truth-telling.
to it through inappropriate use of power or the lack of There are at least five clinical case categories in
compassion. Instead, the vulnerable person should be which truth-telling becomes problematic: intervention
assisted back to a state of human equality, if possible, cases, long-term care cases, cases of dying patients, pre-
fr,ee from the prior dependency.' vention cases, and non-intervention cases.
1 First, the goal of the health care giver—patient re-
lation is essentially to restore the patient's autonomy. Intervention Cases
Thus, respect for the right of the patient to the truth is Of all clinically difficult times to tell the truth, two typ-
measured against this goal. If nothing toward that goal ical cases stand out. The first usually involves a mother
can be gained by telling the truth at a particular time, of advanced age with cancer. The family might beg the
still it must be told for other reasons. Yet, if the truth surgeon not to tell her what has been discovered for
would impair the restoration of autonomy, then it may fear that "Mom might just go off the deep end." The
be withheld on grounds of potential harm. Thus, the movie Dad, starring Jack Lemmon, had as its centre-
goal of the healing relationship enters into the calculus piece the notion that Dad could not tolerate the idea of
_ of values that are to be protected. cancer. Once told, he went into a psychotic shock that
, Second, most healthcare relationships of an inter- ruptured standard relationships with the doctors, the
ventionis: character are temporary, whereas relationships hospital, and the family. However, because this diagno-
involving primary care, prevention, and chronic or dying sis requires patient participation for chemotherapeutic
care are more permanent. These differences also have a interventions and the time is short, the truth must be
bearing cn truth-telling. During a short encounter with faced directly. Only if there is not to be intervention
health care strangers, patients and health care provid- might one withhold the truth from the patient for a
ers will of necessity require the truth more readily than while, at the family's request, until the patient is able
during a long-term relation among near friends. In the to cope with the reality. A contract about the time al-
short term, decisions, often dramatically important ones, lowed before telling the truth might be a good idea.
need to be made in a compressed period. There is less op- The second case is that of ambiguous genitalia.
portunity to manoeuvre or delay for other reasons, even if A woman, 19-years-old, comes for a checkup because she
there are concerns about the truth's impact on the person. plans to get married and has not yet had a period. She is
Over a longer period, the truth may be withheld very mildly retarded. It turns out that she has no vagina,
for compassionate reasons more readily. Here, the pa- uterus, or ovaries but does have an undescended testicle
tient and physician or nurse know one another. They in her abdomen. She is actually a he. Should she be told
are more likely to have shared some of their values. In this fundamental truth about herself? Those who argue
this context, it is more justifiable to withhold the truth for the truth do so on grounds that she will eventually
Thomasma: Telling the Truth to Patients: A Clinical Ethics Exploration
find out, and more of her subsequent life will have been terminally ill cancer patient by telling her that she did
ruined by the lies anddisingenuousness of others. Those not have to worry about her insurance running out!
who argue against the truth usually prevail. National stan- Yet in every case of dying patients, the truth can
dards exist in this regard. The young woman is told that be gradually revealed such that the patient learns about
she has something like a "gonadal mass" in her abdomen dying even before the family or others who are resisting
that might turn into cancer if not removed, and an oper- telling the truth. Sometimes, without- directly saying
ation is performed. She is assisted to remain a female. "you are dying," we are able to use interpretative truth
More complicated still is a case of a young Hispanic and comfort the patient. If a car driver who has been in
woman, a trauma accident victim, who is gradually an accident and is dying asks about other family mem-
coming out of a coma. She responds only to commands bers in the car who are already dead, there is no neces-
such as "move your toes." Because she is now incompe- sity to tell him the truth. Instead, he can be told that
tent, her mother and father are making all care decisions "they are being cared for" and that the important thing
in her case. Her boyfriend is a welcome addition to the right now is that he be comfortable and not in pain.
large, extended family. However, the physicians discover One avoids the awful truth because he may feel respon-
that she is pregnant. The fetus is about five weeks old. sible and guilt-ridden during his own dying hours if he
Eventually, if she does not recover, her surrogate decision- knew that the rest of his family were already dead.
makers will have to be told about the pregnancy, because
they will be involved in the terrible decisions about con- Prevention Cases
tinuing the life of the fetus even if it is a risk to the moth- A good example of problems associated with truth-
er's recovery from the coma. This revelation will almost telling in preventive medicine might come from
certainly disrupt current family relationships and the role screening. The high prevalence of prostate cancer
of the boyfriend. Further, if the mother is incompetent among men over 50-years-old may suggest the utility
to decide, should not the boyfriend, as presumed father, of cancer screening. An annual checkup for men over
have a say in the decision about his own child? 40-years-old is recommended. Latent and asymptom-
In this case, revelation of the truth must be care- atic prostate cancer is often clinically unsuspected and
fully managed. The pregnancy should be revealed only is present in approximately 30 per cent of men over
on a "need to know" basis, that is, only when the sur- 50 years of age. If screening were to take place, about
vival of the young woman becomes critical. She is still 16.5 million men in the United States alone would be
progressing moderately toward a stable state. diagnosed with prostate cancer, or about 2.4 million
men each year. As of now, only 120,000 cases are newly
Long Term Cases
diagnosed each year. Thus, as Timothy Moon noted in
a recent sketch of the disease, "a majority of patients
Rehabilitation medicine provides one problem of truth- with prostate cancer that is not clinically diagnosed will
telling in this category If a young man has been para- experience a benign course throughout their lifetime."6
lyzed by a football accident, his recovery to some level The high incidence of prostate cancer coupled with
of function will depend upon holding out hope. As he a very low malignant potential would entail a whole
struggles to strengthen himself, the motivation might host of problems if subjected to screening. Detection
be a hope that caregivers know to be false, that he may would force patients and physicians to make very diffi-
someday be able to walk again. Yet, this falsehood is not cult and life-altering treatment decisions. Among them
corrected, lest he slip into despair. Hence, because this are removal of the gland (with impotence a possible
is a long-term relationship, the truth will be gradually outcome), radiation treatment, and most effective of
discovered by the patient under the aegis of encourage- all, surgical removal of the gonads (orchiectomy). But
ment by his physical therapists, nurses, and physicians, why consider these rather violent interventions if the
who enter his life as near friends. probable outcome of neglect will overwhelmingly be
benign? For this reason the US Preventive Services Task
Cases of Dying Patients Force does not recommen,c1 either for or against screen-
Sometimes, during the dying process, the patient asks ing for prostate cancer.! Quality-of-life issues would
directly "Doctor, am I dying?" Physicians are frequently take precedence over the need to know.
reluctant to "play God" and tell the patient how many
days or months or years they have left. This reluctance Non-Intervention Cases
sometimes bleeds over into a less-than-forthright This last example more closely approximates the kind of
answer to the question just asked. A surgeon with whom information one might receive as a result of gene map-
I make rounds once answered this question posed by a ping. This information could tell you of the likelihood
II 1111i1
or probability of encountering a number of diseases First, truth is contextual. Its revelation depends
through genetic heritage, for example, adult onset or upon the nature of the relationship between the doctor
type II diabetes, but could not offer major interventions and patient and the duration of that relationship.
for most of them (unlike a probability for diabetes). Second, truth is a secondary good. Although im-
Some evidence exists from recent studies that the portant, other primary values take precedence over the
principle of truth-telling now predominates in the doctor— truth. The most important of these values is survival
patient relationship. Doctors were asked about revealing of the individual and the community. A close second
diagnosis for Huntington's disease and multiple sclerosis, would be preservation of the relationship itself.
neither of which is subject to a cure at present. An over- Third, truth is essential for healing an illness. It
whelming majority would consider full disclosure. This may not be as important for curing a disease. That is
means that, even in the face of diseases for which we have why for example, we might withhold the truth from
no cure, truth-telling seems to take precedence over pro- the woman with ambiguous genitalia, curing her dis-
tecting the patient from imagined harms. ease (having a gonad) in favour of maintaining her
The question of full disclosure acquires greater health (being a woman).
poignancy in today's medicine, especially with respect Fourth, withholding the truth is only a tempor-
to Alzheimer's disease and genetic disorders that may ary measure. In vino, veritas, it is said. The truth will
be diagnosed in utero. There are times when our own eventually come out, even if in a slip of the tongue. Its
scientific endeavours lack a sufficient conceptual and revelation, if it is to be controlled, must always aim at
cultural framework around which to assemble facts. the good of the patient for the moment.
The facts can overwhelm us without such conceptual At all times, the default mode should be that the
frameworks. The future of genetics poses just such a truth is told. If for some important reason, it is not to
problem. In consideration of the new genetics, this be immediately revealed in a particular case, a truth-
might be the time to stress values over the truth. management protocol should be instituted so that all
caregivers on the team understand how the truth will
Conclusion eventually be revealed.
Truth in the clinical relationship is factored in with

knowledge and values.
Notes
1. Bok, S. 1989. Lying: Moral Choice in Public and Personal 4. See Nordenfelt, L., issue editor. 1993. "Concepts of Health
Life. Vintage Books: New York. and Their Consequences for Health Care," in Theoretical
2. Pellegrino, E.D., and Thomasma, D.C. 1993. The Virtues Medicine 14 (4).
in Medical Practice. Oxford University Press: New York. 5. Moon, T.D. 1992. "Prostate Cancer," inJournal of the
3. Cassell, E. 1982. "The Nature of Suffering and the Goals American Geriatrics Society 40: 622-7 (quote from 626).
of Medicine," in New England Journal of Medicine 306 (11): 6. Ibid.
639-45.
p;
Telling Patients the Truth Respect for Persons
Robert Scott Stewart
According to Immanuel Kant, respecting human dig-
nity follows from the fact that humans have a special
Introduction
capacity to live autonomously. That is, humans live
David Thomasma (1994) suggests that there are three not only by laws that are imposed upon them, whether
reasons why, in general, individuals ought to tell the by nature or by legislatures, but humans also have
truth and/or be told the truth: respect for persons, util- the unique capacity to decide what laws or rules will
ity, and kindness. This research paper follows him in govern them, which is the literal meaning of auton-
his assessment and discusses the issue of truth telling omy. To make decisions for others, then, is to _failto
to patients within his model. respect a person's autonomy. Over the past number of
Source: From Online Journal of Health Ethics 6 (2010), 1. Reprinted with permission of the Open Access Journal.
Stewart: Telling Patients the Truth
years in North America, society has seen the impact beneficence, which obligate health care workers to "do
of this idea in many ways. In terms of models for the no harm" on the one hand and to pursue positive out-
physician—patient relationship, society has moved from comes on the other (Beauchamp & Childress, 2001).
a paternalistic model to an informative one (Emanuel Under the paternalistic model, physicians often used
& Emanuel, 1992). In the paternalistic model, on the this principle in order to justify withholding seriously
basis of his/her particular expertise, the physician bad diagnoses from patients, and patients' families con-
was empowered to act in the best interests of his/her tinue to use this rationale when urging physicians and
patients. As such, it was not uncommon for phys- other health care workers to do so. Writing in 1979,
icians, when working under this model, to withhold physician Mack Lipkin employs this sort of reasoning in
information from patients in some situations, such his defense of deceiving patients. In the first place, he
as when there was a bad prognoses, for the patienth argues, it is impossible for patients to be told the "whole
"own good." Indeed, as Braddock notes (1998), "In truth" because they do not have the medical expertise
1961 only 10% of physicians surveyed believed it was to understand it. One result of this is that patients can
correct to tell a patient of a fatal cancer diagnosis. . . . actually be misinformed by information. "Cancer" or
[By] 1979, 97% felt that such disclosure was correct." "heart trouble," which can range from curable and non-
On the one hand, this is in keeping with what patients serious to incurable and fatal, can mean a vastly different
want. "For instance, 90% of patients surveyed said array of things to different patients. Because patients are
they would want to be told of a diagnosis of cancer or already anxiety ridden and vulnerable, they are likely
Alzheimer's disease" (Braddock, 1998). to misunderstand the doctor's diagnosis and prognosis,
As health care has moved away from a paternalis- and indeed, "[t] he news of serious illness drives some
tic model toward an informative model over the past patients to irrational and destructive behavior" (Lipkin,
thirty years, the principle of patient autonomy has been 1979). This fact, he says, justifies withholding-informa-
taken more seriously (Emanuel & Emanuel, 1992). In tion. The fact that there is a demonstrable-placebo effect
this model, patients are conceived as having the right that can have incredibly positive effects on a patient's
to be involved in decisions about their health care and health justifies a great deal of what society would now
indeed ultimately to provide informed consent for any call spin, if not outright lying (Lipkin, 1979).
treatment option. Despite disagreement regarding how As Cullen and Klein (2000) point out, however,
much information is necessary for a person to have in there are several problems with Lipkin's argument. First,
order to provide such informed consent (Randall & his claim about patients not possessing sufficient med-
Downie, 2006), clearly withholding information that ical knowledge to understand fully their diagnosis and
the patient has a serious illness such as cancer, or that prognosis is disingenuous. There are many occupations
they are in fact palliative, is far beyond any reasonable where the expert has much more knowledge than the
standard of informed consent. Hence, if health profes- customer—from lawyers to car mechanics and computer
sionals assume that cancer needs to be treated—even if software specialists. This does not give these profession-
this treatment is limited to palliative care—then morally als the right to withhold information or "spin" the truth:
and legally, informed consent must be provided, and it rather, it obligates them to be honest and to find ways
is impossible to achieve this when the patient's basic in which to articulate points in ways the individual will
diagnosis and prognosis are withheld from him or her. understand. Most importantly, however, in attempting
(.-Th Hence, under the principle of respect for persons to promote a patient's good, arguments such as Lipkin's
it is clear that at some point within the course of some completely ignore the respect owed to patients as au-
process, which this paper discusses further below, tonomous beings. People faced with imminent death
competent patients must be informed of both their have a right to that information, and, moreover, it may
diagnosis and prognosis. actually be good for them to know their prognosis in at
least two ways. First, "most treatments for serious dis-
Utility eases require the full cooperation of the patient" (Cullen
& Klein, 2000). Second, in cases where no cure is at
Utility concerns weighing costs and benefits and acting all likely, "good" for a patient may consist in planning
in such a way as to maximize benefits over costs. for his or her death. This might include a variety of ac-
Traditional utilitarian theory equated costs with pain tivities from entering palliative care, securing child care
or unhappiness and benefits with pleasure or happi- for dying patients who are also young parents, recon-
ness (Bentham, 1789/2007; Mill, 1863/2007), and in necting with lost loved ones, to making out a final will.
contemporary bioethical theory utility is typically as- Hence, Cullen and Klein argue, telling patients the truth
sociated with the principles of non-maleficence and ought to be considered the "default position," which
I I !III llllll
can be overridden only in special cases and/or for a lim- focus of concern is on the character of the persons in-
ited amount of time. cAt times, patients themselves may volved and of the relationship between various actors
explicitly request thdt they not be told the truth about (as opposed to deontology and utilitarianism where
their condition if it is very dire. Withholding the truth the focus is on specific judgments such as does this
in such a case as this is actually an instance of respecting action produce the most utility or would this action
the patient, not overriding it for some other purpose. At promote respect for persons). In virtue ethics, then,
other times, deceiving a patient may be justified, in the individuals want to consider truth telling within the
short term, in order to have the best chance to restore a context of relationships. Clearly, it would be hard to
patient to health. This would require that the physician develop a real friendship, for example, with a person
know his/her patient's personality well enough that he who often deceived the other person in one way or
or she know that the patient cannot handle bad news another. Similarly, it would hard to imagine a trusting
and that telling him or her such bad news will lessen relationship being built and sustained between a health
the chances of patient recovery Physicians ought to be care worker and a patient if the patient could not trust
cautious in using such a justification for failing to tell the health care worker to tell him or her the truth. Of
the truth. In the first place, health professionals rarely course, this leaves open the possibility that the rela-
know other people that well, and, in general, they tend tionship would not be negatively affected if the truth
to underestimate's people's ability to handle bad news. never came to light. Thomasma offers an example of
Secondly, this argument is at best a defense of telling the such a case where a dying victim of a car crash is told
truth in a process over a period of time, not for within his last minutes of life that his family in the car with
holding information indefinitely or lying. Indeed, Freed- him are being taken care of even though all of them
man (1993) refers to such a process as "offering truth." are in fact dead (Thomasma, 1994). Typically, however,
Unlike the more typical situation where the phys- the truth of the situation will come out, even in the
ician, having ascertained that the patient is competent, typical palliative care case. The prominent and extenu-
proceeds to tell the patient everything about their diagno- ating features of the car crash case are that the person
sis and prognosis in one sitting, "offering truth" conceives came to the hospital as an emergency patient and will
of truth telling to patients as a process over a period of die almost immediately. These are important because
time. As Freedman says: "A patient's knowledge of diag- (1) emergency cases do not typically demand the sort
nosis and prognosis is not all-or-nothing. It exists on a of informed consent that competent patients do and
continuum, anchored at one end by the purely theoretical (2) the patient will die before any further measures have
àbsolute ignorance' and at the other by the unattainable to be taken that would demand his consent. Even pal-
`total enlightenment.' Actual patients are to be found liative patients typically do not present in this fashion.
along this continuum that vary in response to external That is, it would be quite uncommon that an undiag-
factors (verbal information, non-verbal clues, etc.) as well nosed cancer patient arrived at the hospital minutes or
as internal dynamics such as denial" (Freedman, 1993). hours before his/her death from his/her cancer. There
The best way to begin this process, Freedman suggests, is almost always a much longer period of time between
is to find out where the patient is on the continuum by diagnosis and death. Even if that period is only a few
hearing "from the patient himself or herself, so that [the days or weeks, there will be decisions that have to be
health care team] can confirm what he or she knows or made—such as the one whether to transfer the patient
clear up any misunderstanding that may have arisen" into a palliative care unit. In such cases, it is very un-
(Freedman, 1993). Beginning here, the health care team likely that the patient will not somehow, perhaps un-
will not only know what the patient him/herself knows intentionally, discover their diagnosis and prognosis. In
(which is often much more than the family thinks the such cases as these, health professionals are presented
patient knows), they can also get a sense of how much with a worst case scenario. Whatever benefit was to
the patient wants to know. Surely knowing this can only come of concealment is not met, and they have hence
help in promoting the patient's good. disrespected the patient for no compensating benefit.
Moreover, the health professionals have been unkind
Kindness and jeopardized the possibility of a fruitful and whole-
some relationship between health care team members
Thomasma (1994) looks upon kindness as a kind and the patient. Moreover, a further point can be made.
of virtue, and virtue ethics are typically thought to Asking health care professionals to withhold informa-
differ in an important respect from deontology and tion or lie to patients puts them in a moral dilemma
utilitarianism discussed above under the rubric of and can lead to moral distress and residue. Over time,
respect for persons and utility. In virtue ethics, the this can even threaten the very moral identity of health
Stewart: Telling Patients the Truth
care workers and cause them to leave their profession begins by getting the competent patient to tell his/her
or to stop caring for patients in the way they had previ- own account of what he/she understand about his/her
ously (Kelly, 1998; Rubin & Baylis, 2000). condition and how much he/she wants to be told. Given
that patients have a right rather than a duty to be told
Conclusion the truth, patients can decide for themselves that they
do not want to be told the truth and that their proxies
This article has argued that the default position for should make decisions for them. This may be especially
physicians and other health care workers is to tell their true for those cultures that do not value truth telling and
competent patients the truth. The basis for this are: re- autonomy as much as is currently the case for the major-
spect for persons, utility, and kindness. Telling compe- ity in North America (Buken, 2003; Asai, 1995). Cases
tent patients the truth need not be done, however, "all at where the truth will be withheld for long periods of time
once." Indeed, in many situations, it is better to think of will be extraordinarily rare, and the burden of proof will
telling the truth to patients as a process, which typically fall squarely on those who want to withhold the truth.
References
Asai, A. (1995). "Should physicians tell patients the truth." Kelly, B. (1998). "Moral identity: A follow-up study of Eng-
The Western Journal of Medicine 163, 36-39 lish graduate nurses. Journal of Advanced Nursing 24:
Buken, 0. (2003). 'Truth-telling information and communi- 1063-1069.
cation with cancer patients in Turkey."Journal of the Inter- Kenny, N. (2004). "What's fair? Ethical decision-making in an
national Society for the History of Islamic Medicine 3: 31-36 aging society" Canadian Policy Research Networks Research
Beauchamp, T & Childress, J. (2001). Principles of biomedical Report.
ethics. 5th ed. Oxford: Oxford University Press. Lipkin, M. (1979, 4 June). "On telling patients the truth."
Bentham, J. (1789/2007). Introduction to the principles of Newsweek 13. In R Munson, ed. (2004). Intervention and
morals and legislation. Dover. reflection: Basic issues in medical ethics, 7th ed. (Belmont,
Braddock, C.H. (1998). "Truth-telling and withholding infor- CA: Wadsworth), 154-156.
mation." Ethics in Medicine. http://depts.washington.edu/ McDonald, M. (2001). A framework for ethical decision-
bioethx/topics/truth.html. making: version 6.0-Ethics Shareware January www
Cullen, S & Klein, M. (2000). "Respect for patients, phys- .ethics.ubc.ca/people/mcdonald/decisions.htm.
icians, and the truth." In R. Munson, ed. (2004). Inter- Mill, JS. (1863/2007). Utilitarianism. Dover.
vention and reflection: Basic issues in medical ethics, 7th ed. Randall, F & Downie, RS. (2006). The philosophy of palliative
(Belmont, CA: Wadsworth), 155-163. care. Oxford: Oxford University Press.
Ells, C. (2006). "Healthcare ethics committees: Contribu- Thomasma, D. (1994). Telling the truth to patients: A clinical
tion to review of institutional policy" HEC Forum 18 ethics exploration." Cambridge Quarterly of Healthcare
265-275. Ethics 3: 375-382. In T. Beauchamp & L Walters, eds.
Emanuel, EJ & Emanuel, L. (1992). "Four models of the (2003). Contemporary issues in bioethics, 6th ed, 128-132.
physician-patient relationship." Journal of the American Thomson/Wadsworth.
Medical Association 267, 16: 2212-2226. Webster, GC & Baylis, E (2000). "Moral residue." In Mar gin of
Freedman, B. (1993). "Offering truth." Archives of Internal Error: The ethics of mistakes in the practice of medicine (Hager-
Medicine 153, 3: 572-576. stown, MD: University Publishing Group), 172-175.
Medical Secrecy: Patients' Right of Subsequently, the arrogance of the physician who took
this action was much criticized in the Soviet press.
Access to Medical Records At almost the same time, in Toronto, a young woman
Arthur Schafer was forced to go to court to obtain access to her father's
medical records. The family wished to determine the
In Moscow, recently, a woman dying of a brain tumour quality of the medical care he received before he died.
was thrown out of a Soviet hospital. Her offence? She Both of these cases raise interesting and important
looked, without authorization, at her medical record. questions: why are physicians, in general, so reluctant
Source: From Humane Medicine 1, 2 (October 1985), 91-9. Reprinted with permission of the author.
LE I I ILII 111111
114 3 Management of Medical Information
to allow the individual's access to their own records? nursing, respiratory therapy, and clinical pharmacy;
And, why are patients increasingly insistent upon their four unit secretaries; four hospital financial officers;
right of access? and four chart reviewers.
It seems obvious that the risk of serious errors and
The Dignity of the Patient omissions is significant when the information in a pa-
tient's medical file is generated by such a large number
Many people will have had an experience somewhat of persons. Partly, this is because physicians and others
like the following You are in the doctor's office. He has vary in the level of skill and care they exercise in re-
been taking notes as you describe your problems and cording the information. The 1980 Krever Commission
symptoms. Then he is called from the room. The med- on confidentiality of health records cited many cases in
ical file is left, open, on the desk. which incorrect information was entered into patient
Who has not felt tempted, in such a situation, to files, with disastrous consequences. If patients have
take a surreptitious look at what the doctor has written no right of access to the information in their files, they
in that file? Whether you succumb to this temptation or may be denied the opportunity to correct errors and
resist, you may feel belittled by the lack of open access omissions.
to your record. As one patient put it, "It is not that Moreover, in a variety of circumstances, people
everyone would want to know, but it is the assumption will be asked to waive the confidentiality of their health
that you are not one of the team that is so undignified." records. When you apply for life insurance, or welfare,
It is important to stress that the issue is not simply or worker's compensation, or credit, or employment,
one of indulging idle curiosity. The issue, rather, con- you may be asked to release such information. How
cerns the basic nature of the doctor—patient relation- can patients make intelligent, informed decisions as to
ship. When patients' rights advocates insist upon the whether to give such permission for release of informa-
need for legislation to guarantee a patient's right of tion unless they have access to their files and an oppor-
access to information in the medical record, they are tunity to ensure that the information contained therein
rejecting, implicitly or explicitly, the physician's trad- is accurate and non-prejudicial?
itional paternalistic role as benevolent sovereign. Given the large number of persons who have au-
thorized access to the patient's personal medical record,
Correcl:ing Mistakes and Protecting and given the number of persons who may seek and
Confidentiality gain illegitimate access to the computers in which
such information is stored (not always very securely),
Most patients are aware only vaguely that many people it seems ironic that the one person who lacks access
have access to their health records; and even fewer are is often the patient. You may not agree entirely with
aware that their record may contain serious errors and Siegler when he declares that "medical confidentiality,
omissions. as traditionally understood by patients and doctors, no
Prompted by a patient's complaint, Siegler, an• longer exists,' but doubtless, there is some cause for
American physician, took the trouble to enumerate the concern.
number of persons who had legitimate access to the
patient's hospital record.' He was surprised to find that The Doctors' Dilemma
"at least 25 and possibly as many as 100 health profes-
sionals and administrative personnel at (his) university The case in favor of patient access to personal medical
hospital had access to the patient's record and that all records is so strong that one may wonder that so many
of them had a legitimate need . . . to open and use physicians continue to resist. Some of this resistance
that chart." The patient who complained had a med- may be attributed to the physician's desire to retain
ical condition—mild chronic obstructive pulmonary power over the patients. Monopoly of knowledge con-
disease—that was relatively straightforward and rou- fers power. Those physicians who enjoy the status of
tine. Despite this, those with access to his file included "priest" and expect their patients to accept, without
six attending physicians—the primary physician, the question, decisions made on their behalf, naturally will
surgeon, the pulmonary consultant, and others; 12 hour wish to retain exclusive control of access to medical
officers--medical, surgical, intensive-care unit, and records. This monopoly provides the ancillary "benefit"
"covering" house staff; 20 nurses (on three shifts); six of making it difficult (if not impossible) for patients to
respiratory therapists; three nutritionists; two clin- use the legal system against incompetent or unethical
ical pharmacologists; 15 students—from medicine, physicians.
Schafer: Medical Secrecy: Patients' Right of Access to Medical Records
It would be a mistake, however, to view all oppos- recordkeeping practices; for example, they might at-
ition to patient access to medical records as a conspir- tempt to remember potentially frightening diagnoses
acy of professionals against the public. As the American rather than entering them in the record. Or they might
Medical Association has declared: "It is our position that become reluctant to enter into the patient's record such
the right of a patient to medical information from his unflattering comments as "working class background-??
physician is based upon the fiduciary relationship which Poor self-image." The physician would omit such com-
imposes a duty to act in the best interest of the patient." ments in order to avoid mutual embarrassment and to
It is frequently claimed that physicians have both a preserve the patient's confidence, but such omissions
right and a duty to withhold information from patients might lead to less efficient clinical management.
when they judge this to be in the patients' best interests.
Complete disclosure is opposed on the grounds that it Resolving the Dilemma
is likely to alarm and upset patients needlessly. (Imagine
the anxiety of the patient who reads his physician's ten- Fortunately, there are ways of resolving or "dissolving"
tative speculations, e.g., "query, possibility of cancer.") most of these objections. If medical records were writ-
According to this view of the doctor—patient relation- ten on the assumption that patients might see them,
ship, patients should be willing to trust the good will the physician would be obliged to replace technical
and sound judgement of their physicians. jargon by good clear English, without prejudice to
In an editorial published by the Journal of the Royal medical efficiency. It is conceded that, in the absence
Society of Medicine,' J.M.A. Northover, a British consult- of supplementary explanation, the patient would find
ant surgeon, argues that patient trust in doctors would the information in the file often more confusing than
be undermined by giving patients access to their medical edifying. However, the conclusion to be drawn from
record: "Part of that trust is based on the understand- this is not that patients should be denied access to
ing between the parties that the doctor regards help- their files, but that the physician should be present
ing the patient to come to terms with the disease and when the patient receives the file, to provide neces-
its implications as an important part of his or her job; sary explanation and, where necessary, reassurance.
many doctors feel that this responsibility requires the Patients are entitled to know the truth about their
careful 'rationing' of information, the best interests of medical status, and to have an opportunity to correct
the patient determining just how that information is di- errors and omissions. It seems reasonable to assume
vulged." Northover concedes that patients frequently are that those who wish to protect themselves from
underinformed, but he insists that doctors should "resist such information generally will not request access.
the endeavours of those who advocate patient access to The process of explaining and interpreting medical
medical files," principally because "information manage- records is likely to require some additional investment
ment is an important aspect of proper medical care." of the physician's time, but this would not be time
Northover buttresses his argument against giving wasted. Surely patients are entitled to receive both
patients the right of access to their medical files with the information and emotional reassurance.
argument that these records usually are written in a lan- The experience of having the physician explain the
guage that for technical and other reasons would be of significance of the information in his history is likely
little practical use to patients. "Every doctor," he declares, to enhance the patient's trust and thereby contribute
". . . would be horrified by the suggestion that the best to good therapy. A trust based on shared knowledge
way to inform a patient is simply to hand over a fairly is more valuable (because better founded) than one
incomprehensible, untidy and potentially frightening based on blind faith. As for the mental "distress" or
document." Patients are likely to be misled, and alarmed "alarm" some physicians fear will result from free pa-
unnecessarily by the "dispassionately pessimistic" style tient access, we need to see some evidence of this. Such
preferred by many physicians in their notetaking, and fears may be exaggerated or misplaced. A randomized
by the sinister diagnostic possibilities included as part of clinical trial to assess the comparative therapeutic ef-
a conscientious differential diagnosis. Northover confers fects of access and secrecy might show that patients
the label "filophile" on those who favor giving patients experience greater fear and anxiety when they are
a legal right of access to their files and accuses them of denied access to their files than when they gain access
favoring "do-it-yourself' communication. in propitious circumstances. Ignorance has a way of
Another potentially harmful consequence of generating unwarranted anxieties and fears that can be
giving patients the right to see their files would be more corrosive of patient morale and trust than would
that many physicians might change deleteriously their be produced by genuine knowledge.3
I I al 111111
As For the danger that the patients morale may be In the rare cases where this does occur and a patient is
undermined when he discovers in the record comments harmed by the policy of openness, one may regret the
which are unflattering, we can respond that such com- harm and yet still insist that patients have a right to
ments have no place in the record in any event, because know what is in their records and, in any event, that
they have no medical justification. Thus, instead of writ- more harm is caused by the present policy of secrecy.
ing "hypochondriac," the doctor would show a more
professional attitude by recording that the patient seems Records of Psychiatric Patients
unduly worried. This problem then could be discussed
openly. [f the physician cannot persuade his patient that The case for complete openness and disclosure is least
this assessment is appropriate, the patient may choose strong with respect to some kinds of psychiatric pa-
to seek other advice. This outcome is consistent with a tient. In Sweden, where patients are entitled to ask for
patient's right to autonomy and should not necessarily their medical records, doctors have the right to refuse
be viewed as an unfortunate consequence. access if they believe that such access might endanger a
Northover concedes that "it is a sad reflection patient's life. This seems to be a reasonable comprom-
on the medical profession that many members of the ise, although some patients' rights advocates fear that
public are sufficiently dissatisfied by their doctor's physicians, including psychiatrists and other mental
inclination and ability to talk to them that they feel a health professionals, often misuse their discretionary
need to seek access to their medical records to find out power over mental patients.
what is happening." But his opposition to legislation Sociologists have frequently noted that once a pa-
that would open medical files at the patient's request tient has been labelled as "mentally ill," mental health
rests on his belief that Information management is an professionals tend to interpret all the patient's subse-
important aspect of proper medical care" and his belief quent behaviour as evidence confirming the original
that the Hippocratic principle of non-maleficence (not diagnosis.4 Thus, innocent behaviour may be inter-
doing harm) requires that physicians carefully ration preted as pathological because of the context in which
the information they give their patients. it occurs. Once a patient has been so labeled, and the
Dr. Northover illustrates his position with the fol- label is recorded in the patient's file, it exerts a powerful
lowing example: "In a state of frightened desperation, a influence upon all future decisions made with respect
patient might ask 'Have I got cancer?', whilst wanting to to that patient. If the patient has been mislabeled,
hear only one answer: 'No.' The good doctor will recog- much harm may result.
nize the situation and will choose his or her words care- However, if mental patients (and/or their family
fully and compassionately." The crucial question (ignored, or a court-appointed guardian) had a right of access
however, by Northover) is this: "How can the doctor be to their records, psychiatrists might be encouraged
sure that the patient does not want to know the truth?" to develop a more careful and scrupulous attitude
Perhaps the doctor has misinterpreted the patient's real towards recordkeeping. Cohen has offered some evi-
wishes; perhaps the doctor has projected his/her anxieties dence that, for many psychiatric patients, access to
about cancer onto the patient. Failure to inform, however information about themselves can be therapeutically
well motivated, will often produce more harm than good. beneficial.5 Of course, there are some instances in
Whether a patient suspects deception or is successfully which the risk of serious harm to the patient from
kept in a state of complete ignorance, the lack of know- access is so great that access ought to be denied. It
ledge will often produce more harm than would have seems to me that such cases are likely to be uncom-
been produced by the "painful" truth. I believe the risks mon. Hence, a sensible policy would be to legislate a
associated with medical paternalism are too great to be right to access but at the same time to incorporate in
brushed aside. This is not to deny that physicians ought the law a set of criteria for non-disclosure. This would
to communicate painful information to patients with tact provide a safety buffer for such exceptiona cases.
and sensitivity. The communication skills Northover calls
for (pp. 93 and 95) will be no less important once pa- Canadian Law
tients have won the legal right to access their files.
This position does not deny that exceptional At present, only a few Canadian provinces have enacted
cases may require the withholding of certain specific legislation which confers upon patients the legal right of
information. But those of us who are not physicians access to the information contained in their medical files.
will find it difficult to accept that many patients who In Alberta, for example, the medical record in the doctor's
genuinely wish to remain in ignorance of their diag- office belongs to the physician, and in the hospital it be-
nosis or prognosis will insist upon seeing their record. longs to the hospital The patient is entitled, upon request,
Schafer: Medical Secrecy: Patients' Right of Access to Medical Records
to have access to the information contained in the record Conclusion

but is not entitled to physical possession of the records.
In Quebec, medical records are regarded asii6p- Evidence from jurisdictions in which patients have
erty to be shared between physician and patients. gained the legal right of access to the information in
Hospital records are shared between the institution, their medical files suggests that only a minority will
the physician, and the patient. In neither case is the choose to exercise their rights. Does this mean that pa-
patient entitled to take the original record and amend tients regard the right of access as of little importance?
it. Patients may be denied access to their records only The minority who seek such access seem to see the
when such access is deemed to be prejudicial to the right as highly important. It is probable that the consumer's
patients' health. rights philosophy which has increased in strength during
Section 4.02 of the regulations under the Quebec the last few decades will continue to spread and to influ-
Medical Act reads: "Except when it is prejudicial to the ence "consumers" of health care, that is, patients. Whether
patient's health, the physician must respect the patientb or not the number of patients demanding access to their
right to information in the records which concern him medical files continues to grow, it should be emphasized
and to obtain a copy of such records." With respect to that the right of access is important even to those who
hospital records, the Quebec legislation reads: "A re- do not wish to see their files. That is, it is important that
cipient. to whom an establishment refuses access to his patients know that they have the right to see what is in
record or ref uses to give written or verbal communica- their files, even if they do not choose to exercise this right.
tion of it may, on summary motion, apply to a judge of The attitudes and values of both physicians and
the Superior Court, Provincial Court, Court of the Ses- patients have changed significantly over the past decade.
sions or Youth Court or to the Commission, to obtain Patients now expect to take more responsibility for their
access to or communication of it as the case may be. The own health. Physicians are significantly less paternalistic
judge shall order such establishment to give such recipi- than they used to be. This change in attitude has been
ent access to his record, or communication of it, as the beneficial for both physicians and patients. It seems
case may be, unless he is of the opinion that it would be likely that when patients across Canada win the legal
seriously prejudicial to the health of such recipient to right to see their medical files, this change will enhance
examine his record." the values of patient autonomy and patient responsibility.
References
1. Siegler M. NE JM 1982;307: 1518-1521. diagnosed but undisclosed malignant disease." The Lancet
2. Ibid. 1985. pp. 93-95. 1976;1: 300-303.
3. See, e.g., Aphidi R. "Informed consent: A study of pa- 4. Garman E. Asylums. Penguin Books, Harmondsworth, 1968.
tient reaction." JAMA 1971;216: 1325-1329; Mcintosh J. 5. Cohen RN. Whose file is it, anyway? National Council for
"Patient's awareness and desire for information about Civil Liberties, 1983.
3.3 Confidentiality of Information
A Defense of Unqualified Medical 1982: After moving to Honolulu, Wilma and

Andrew Long visit your office and ask you to be
Confidentiality their family physician. They have been your pa-
Kenneth Kipnis tients ever since.
1988: Six years later the two decide to separate.
The Case of the Infected Spouse Wilma leaves for the Mainland, occasionally send-
ing you a postcard. Though you do not see her
The following fictionalized case is based on an actual professionally, you still think of yourself as her
incident. doctor.
Source: From American Journal of Bioethics 6 (2006) (2), 7-18. Reprinted with permission of Taylor & Francis Journals.
II 111111
1990: Andrew comes in and says that he has em- It is not enough that a warning would discharge a
barked upon a more sophisticated social life. He duty to Wilma, merely so she could make an informed
has been hearing about some new sexually trans- choice. Plainly, the paramount concern has to be to save
mitted diseases and wants to be tested. Testing Wilma's life. Finally, Wilma is not a mere stranger. In-
reveals that he is positive for the AIDS virus, and he stead she has an important relationship with you—her
receives appropriate counseling. doctor—that serves as a basis for special obligations:
You have a special duty to look out for her health.
1991: Visiting your office for a checkup, Andrew In the light of these five considerations, it should
tells you Wilma is returning to Hawaii for recon- not be a surprise that the conventional wisdom in med-
ciliation with him. She arrives that afternoon and ical ethics overwhelmingly supports either an ethical
will be staying at the Moana Hotel. Despite your obligation to breach confidentiality in cases like this
best efforts to persuade him, Andrew leaves with- one or, occasionally and less stringently, the ethical per-
out giving you assurance that he will tell Wilma missibility of doing so (Lo 1995). Notwithstanding this
about his infection or protect her against becoming consensus, it is my intention to challenge the received
infected. view. I will argue in what follows that confidentiality
in clinical medicine is far closer to an absolute obli-
Do you take steps to see that Wilma is warned? If you gation than it has generally been taken to be; doctors
decide to warn Wilma, what do you say to Andrew when, should honor confidentiality even in cases like this
two days later; he shows up at your office asking how you Although the focus here is on the Case of the Infected
could reveal his confidential test results? Spouse, the background idea is that, if it can be dem-
If you decide not to warn Wilma, what do you say to onstrated that confidentiality should be scrupulously
her when, two years later in 1993, she shows up at your honored in this one case where so many considerations
office asking how you, her doctor; could possibly stand idly support breaching it, the duty of confidentiality should
by as her husband infected her with a deadly virus. She now be taken as unqualified in virtually all other cases as
knows she is positive for the virus, that she was infected by well (Kottow 1986). I shall not, however, defend that
her husband, and that you—her doctor—knew, before they broader conclusion here.
reconciled, that her husband would probably infect her. Although this essay specifically addresses the obli-
The ethical challenges here emerge from an ap- gations of doctors, its approach applies more broadly
parent head-on collision between medical confidenti- to all professions that take seriously the responsibility
ality and the duty to protect imperiled third parties. to provide distressed practitioners with authoritative
Notwithstanding Andrew's expectation of privacy and guidance (Kipnis 1986, 63-79; Wicclair 1986). With
the professional duty to remain silent, it can seem its focus narrowly on "professional obligations," the
unforgivable for anyone to withhold vital assistance in methodology used below also represents something of
such a crisis, let alone a doctor. The case for breaching a challenge to much of the conventional thinking in
confidentiality is supported by at least five considera- medical ethics.
tions: First, the doctor knows, to a medical certainty,
that Andrew is both infected with HIV and infectious. Clearing the Ground: What Professional
Second, knowing Wilma as a patient, let us suppose Obligations Are Not
the doctor reasonably believes that she is not infected.
(Wilma cannot be at risk of contracting the disease if Among philosophers, it is commonplace that if people
she is infected already.) Third, Wilma's vulnerability is are not asking the same questions, they are unlikely
both serious and real. HIV infection is both debilitating to arrive at the same answers. It may be that the main
and, during those years, invariably fatal: The couple's reason doctors have difficulty reaching consensus in
sexuality makes eventual infection highly likely. Fourth, ethics is that, in general, systematic discussion about
assuming that preventing Wilma's death is the goal, it professional responsibility is commonly confused with
is probable that, were Wilma to be told of Andrew's at least three other types of conversation. When one
infection, she would avoid exposing herself to the risk. asks whether one should call the hotel to warn Wilma,
This is not a trivial condition: many people knowingly one can be asking: 1) what the law requires 2) what
risk illness and injury out of love and other honorable one's personal morality requires (e.g., as an Orthodox
motivations. Molokai's Father Damien contracted and Jew, a Roman Catholic, etc.); or 3) what is required by
died from Hansen's disease while caring for patients one's most deeply held personal values (e.g., preventing
he knew might infect him. Soldiers, police, and fire- deaths or scrupulously honoring other obligations).
fighters commonly expose themselves to grave risk. Discussions can meander mindlessly over all three
Kipnis: A Defense of Unqualified Medical Confidentiality
areas without attending to boundary crossings. More Focusing on the legal standard, it is useful to dis-
to the point, effective deliberation about professional tinguish between "special" and "general" legal duties.
obligations, as I will try to show, differs importantly Special duties can apply to individuals occupying cer-
from all three of these discussions. Accordingly, it is ne- tain roles. A parent, but not a bystander, has a special
cessary to identify and bracket these other perspectives duty to rescue a drowning daughter, firefighters and
in order to mark off the intellectual space within which police officers have special duties to take certain oc-
practitioners can productively reflect on questions of cupational risks, and doctors have many special duties
professional responsibility. Let us examine these differ- toward their patients: confidentiality is a good example.
ent conversations. In contrast, virtually everyone has a general duty to be
scrupulously careful when handling explosives, to pay
Law taxes on income, to respect others' property, and so on.
The conventional wisdom on the ethics of medical It is notable that the duty to warn in Tarasoff is a special
confidentiality has been largely shaped by a single legal duty, applicable only to those occupying roles.
case: Tarasoff v. Regents of the University of California So if my neighbor casually assures me he is going to
(Supreme Court of California; 529 p. 2d 553, Cal. kill his girlfriend tomorrow, the Tarasoff ruling does not
1974). In 1969, Prosenjit Poddar, a student at U.C. require me to warn her:
Berkeley told a university psychologist he intended to It is surprising to many that the default standard
kill a Ms. Tatiana Tarasoff, a young woman who had in Anglo-American jurisprudence is that there is no
spurned his affections. The psychologist dutifully re- general duty to improve the prospects of the precar- ,
ported him to the campus police, who held him briefly iously placed, no legal obligation to undertake even
and then set him free. Shortly afterwards, Poddar did an easy rescue. As first-year law students discover, one
as he said he would, stabbing the young woman to can stand on a pier with a lifeline in hand and, with
death. The Tarasoff family sued the University of complete impunity, allow a stranger to drown nearby.
California for their daughter's death, finally prevail- Although we will pass over it, it is notable that, in
ing in their contention that the psychologist (and, by general, the parties who are legally obligated to warn
implication, the University) had failed in their duty are those who are otherwise ethically obligated not to
to protect, since neither Tatiana nor those able to ap- disclose. One should reflect on the absence of a general
prise her of danger were warned. The University was duty to warn.
found liable and had to compensate the family for its The easy transition from law to ethics reflects a
loss. Today it is hard to find discussions of the ethics common error. The mistake is to move from the prem-
of confidentiality that do not appeal to this legal par- ise that some action is legally required (what the Tara-
able and, occasionally, to its California Supreme Court soff opinion establishes in the jurisdictions that have
moral: "The protective privilege ends where the public followed it) to the conclusion that the same action is
peril begins." ethically required. But ethical obligations can conflict
Taking its cue from Tarasoff, the prevailing stan- with legal ones. Journalists, for example, are sometimes
dard in medical ethics now holds that the obligation ordered by the courts to reveal the identities of their
of confidentiality will give way when a doctor is aware confidential sources. Although law demands disclo-
that a patient will seriously injure some identified other sure, professional ethics requires silence. Reporters
person. (One might ask why disclosure is not required famously go to jail rather than betray sources. Jour-
when a patient will seriously injure many unidentified nalists can find themselves in a quandary: while good
persons. Under the narrower standard, there is no duty citizens obey the law and good professionals honor
to alert others about an wv-infected prostitute who nei- their professional codes, laws requiring journalists to
ther informs nor protects a large number of anonymous violate their duties to confidential sources force a tragic
at-risk clients.) We assume that the physician knows choice between acting illegally and acting unethically.
Andrew is seropositive, that Wilma is likely seronega- Conscientious persons should not have to face such
tive, that the two will likely engage in activities that decisions.
transmit the virus, and that breaching confidentiality Similarly in pediatrics, statutes may require doctors
will probably result in those activities not occurring to report suspicions of child abuse. But where protect-
and Wilma's not becoming infected. Thus, a physician's ive agencies are inept and overworked and foster care
warning in the Case of the Infected Spouse will mean that is dangerous or unavailable, a doctor's report is more
Wilma is very likely to remain infection-free, and a fail- likely to result in termination of therapy and further
ure to warn her is very likely to result in her eventual injury to the child instead of protection and care. To
death from AIDS. obey the law under these appalling, but too common,
II al III Lid
circumstances is most likely to abandon and even cause offered here to discern much of what I would want to
harm to the minor patient, both of which are ethically say about those other cases.
prohibited in medicine. i To assume that legal obliga- Though I will not discuss them, institutional poli-
tions al ways trump or settle ethical,, ones is to blind cies (hospital rules, for example) function very much
oneself to the possibility of conflict. Professions have like laws. Both involve standards that can be imposed
to face these dilemmas head-on instead of masking externally upon practitioners. Both can be formulated
them with language that conflates legal standards and knowledgeably and wisely or with a disregard for es-
ethical ones. They must conceive professional ethics as sential professional responsibilities.
separate from the law's mandate. When law requires
what professional responsibility prohibits (or prohibits Personal Morality
what professional responsibility requires), professional We will understand a "morality" as a set of beliefs about
organizations must press the public, legislatures, and obligations. There are plainly many such sets of beliefs:
the courts to cease demanding that conscientious prac- the morality of Confucius has little in common with the
titioners dishonor the duties of their craft. This is an moralities of George W. Bush and Thomas Aquinas. For
important responsibility of professional organizations. most of us, morality is uncritically absorbed in child-
It is a mistake to configure professional , obligations hood, coming to consciousness when we encounter
merely to mirror the law's requirements. Rather, the others whose moral beliefs differ.
law's requirements must be configured so that they do There are still parts of the world in which virtu-
not cpnflict with well-considered professional obliga- ally all members of a community are participants in a
tions! Law is a human artifact that can be crafted well common morality. But moral pluralism now seems a
or badly In a well-ordered society no one will have to permanent part of the social order. Consider a Jeho-
choose between illegality and immorality. Since the law vah's Witness physician who is opposed, on religious
can require conduct that violates ethical standards (and grounds, to administering blood transfusions. If this
ethical standards can require conduct that violates the doctor were the only physician on duty when his
law), it cannot be the case that legal obligations auto- patient needed an immediate transfusion, a choice
matically create ethical obligations. As the tradition of would have to be made between being a good Jeho-
civil disobedience shows, it can be ethically permissible vah's Witness and being a good doctor. The doctor's
or obligatory (though not legal) to violate an unjust law. personal moral convictions are here inconsistent with
Even though laws cannot create ethical obligations professional obligations. It follows that clarity about
by fiat, professions need to distinguish between the personal morality is not the same as clarity about
state's reasonable interests in the work of doctors (e.g., medical ethics. Professionalism can require that one
preventing serious harm to children) and the specific set aside one's personal morality or carefully limit one's
legal mandates a state imposes (e.g., requiring doctors exposure to certain professional responsibilities. Here
to report suspicion of child abuse to an incompetent the rule has to be that doctors will not take on re-
state agency). Just as patients can make ill-considered sponsibilities that might conflict with their personal
demands that should not be satisfied, so too can the morality. Problems could be sidestepped if the Jeho-
state and its courts. vah's Witness doctor specialized in a field that didn't
Accordingly, it is assumed that the state has a involve transfusion (e.g., dermatology) or always
legitimate interest in preventing harm to people and worked with colleagues who could administer them. If
that doctors have an ethical obligation to further that I am morally against the death penalty, I shouldn't take
important public objective. The focus in this essay is on on work as an executioner. If I am deeply opposed to
itfie shape of the resulting ethical obligation as it applies the morning-after pill, I shouldn't counsel patients at
I narrowly to cases like those involving Wilma Long and a rape treatment center. To teach medical ethics in a
Tatiana Tarasoff. Because they introduce complexities pluralistic professional community is to try to create
that will carry us far afield, we set aside cases involv- an intellectual space within which persons from varied
ing: (a) children brought in by parents (Kipnis 2004); backgrounds can agree upon responsible standards for
(b) patients referred for independent medical evalua- professional conduct. Participants in such a conversa-
tion; (c) mentally ill or retarded patients in the custody tion may have to leave personal morality at the door.
of health care institutions; (d) health care that is the For some, it may be a mistake to choose a career in
subject of litigation; (e) gunshot, knife wounds, and medicine.
the like; (f) workers' compensation cases; and a few If ethics is a critical reflection on our moralities,
others. While a much longer discussion could cover then the hope implicit in the field of medical ethics is
these areas, many readers can extend the analysis that we might some day reach a responsible consensus
on doctors' obligations. While medicine has dozens of To summarize the argument so far, discussion
codes, it is not hard to observe commonalities: the stan- about professional obligations in medicine is not the
dards for informed consent, for example. At a deeper same as discussion about legal and institutional obli-
level, there can also be consensus on the justifications gations, personal morality, or personal values. If a
for those standards. One role for the philosopher is, as responsible ethical consensus is to be achieved by a
in this essay, to assess carefully the soundness of those profession, it is necessary for physicians to learn to
arguments. A major task for professions is to move bracket their personal moral and value commitments
beyond the various personal moralities embraced by and to set aside, at least temporarily, their considera-
practitioners and to reach a responsible consensus on tion of legal or institutional rules and policies. The
common professional standards. practical task is to create an intellectual space within
which responsible consensus can be achieved on how
Personal Values physicians, as professionals, ought to act. I will now
Values are commonly a part of an explanation of per- describe one way in which this might be done.
sonal conduct. It is always reasonable to ask of any
rational action: what good was it intended to pro- The Concept of a Professional Obligation
mote? While some wear shoes to avoid hurting their
feet (embracing the value of comfort), others think Professional ethics involves disciplined discussion
they look better in shoes (embracing aesthetic values). about the obligations of professionals. One place to
Where we have to make personal decisions, often we begin is with a distinction between personal values,
consider how each option can further or frustrate our already discussed, and what can be called "core profes-
values and try to decide among the good and bad sional values." A physician can prefer (1) pistachios to
consequences. Brazil nuts and (2) confidentiality to universal canaoi.)
This strategy can serve when the question is "What While the preference for pistachios is merely personal;
should I do?" But the question "What should a good the preference for confidentiality is a value all doctors
doctor do?" calls for a different type of inquiry. For ought tii_possess. The distinction between personal
while I have many personal values, the "good doctor" values and "core professional values" is critical here.
is an abstraction. She is neither Protestant nor Bud- There is what this flesh-and-blood doctor happens to
dhist, doesn't prefer chocolate to vanilla, and doesn't care about personally and what the good doctor ought
care about money more than leisure time. questions to care about. This idea of a "good doctor" is a social
about professional ethics cannot be answered in terms construction, an aspect of a determinate social role,
of personal values. an integral element of medical professionalism. Our
A second difficulty appears when we consider that idea of a good doctor includes a certain technical/in-
one can give perfect expression to one's most deeply tellectual mastery coupled with a certain commitment
held personal values and still act unethically. Hannibal to specific professional values. As with the Jehovah's
Lecter in Silence of the Lambs and Mozart's Don Gio- Witness doctor, personal and professional values may
vanni are despicable villains who give vigorous effect be in conflict. As part of an appreciation of the ethical
to deeply held if contemptible personal values. While claims of professionalism, physicians must be prepared
personal values can determine action, they do not guar- to set aside their personal values and morality, to set
antee that the favored actions are ethical. aside what the legal system and their employers want
Accordingly, we cannot appeal to our personal them to care about, and to take up instead the ques-
values to inquire about what physicians in general tion of what the responsible physician ought to care
ought to do. Medicine has no personal values, only in- about. The profession's core values inform those pur-
dividual physicians do. When a physician must decide poses that each medical professional should have in
whether or not to resuscitate a patient, personal values common with colleagues. In discussing the profession-
should have nothing to do with the issue. Whether you ally favored resolution of ethically problematic cases
like the patient or detest him, whether you are an athe: (the Case of the Infected Spouse, for example), physicians
ist or a fundamentalist believer in a joyous hereafter, can ask—together—how medicine's core professional
should not weigh in the balance. A key part of pro- values ought to be respected in those circumstances.
fessionalism involves being able to set personal values We have alluded to some of these core professional
aside. While medical students have much to gain by values. Trustworthiness needs to be on the list. Benefi-
becoming clear about their personal values, that clarity cence toward the patient's health needs is essential.
is not the same as responsible certainty about profes- Respect for patient autonomy is a third. Others might
sional obligations. be collegiality (duties to colleagues) and perhaps a few
11 I II 3 111111
others: nondiscrimination and a certain deference to Second, ethical problems can also arise when it
families are among the most commonly mentioned is unclear what some core professional value requires
candidates. If we were to leave out that doctors should one to do. Though we can all agree that doctors should
care about the well-being of the public, the argument avoid harming their patients, there is no professional
for confidentiality would be easy. But it too properly consensus on whether deliberately causing the deaths
goes on the list. Anyone seeing no point in furthering of certain unfortunate patients—those experiencing ir-
and securing these values would be ill-suited for the remediable and intense suffering—is always a betrayal
practice of medicine. of beneficence. Likewise, although doctors may be in a
Each of these professional values has two dimen- position to prevent harm to third parties, it is not well
sions. Along one vector, they define the shared aspira- understood what they must do out of respect for that
tion of a profession. At any time, medicine's ability value. When core values conflict, what is required is a
to benefit patients will be limited. But it is a part of priority rule. When they are unclear, what is required is
the profession's commitment to push its envelope, to removal of ambiguity: what philosophers call "disam-
enlarge its collective competency and draw upon its biguation." These two tasks—prioritizing and disam-
knowledge and skill. Those who master and extend the biguating core professional values—need to be carried
profession's broadest capabilities are exemplary con- out with a high degree of intellectual responsibility.
tributors, but practitioners do not discredit themselves The above list of medicine's core values is not con-
by failing to serve in this estimable way troversial. Propose a toast to them at an assemblage of
Along the second vector, values define a bottom physicians, and all can likely drink with enthusiasm.
line beneath which practitioners shall not sink. Para- What is less clear is why such a consensus should
phrasing Hippocrates, although you may not always be obligate professionals. A criminal organization can
able to benefit your patients, it is far more important celebrate its shared commitment to the oath of silence.
that you take care not to harm them. Knowingly to But it doesn't follow that those who cooperate with
harm a patient (on balance) is not merely a failure to the police are unethical. In addition to organizational
realize the value of beneficence. It is a culpable betrayal "celebratability," three additional elements are required
of that value, a far more serious matter. . to establish a professional obligation.
All the values above can be understood in this The first element is that attention to core values
second way. Trustworthiness entails that I not lie to pa- has to be a part of professional education. Most med-
tients or deliberately withhold information they have an ical education is aimed at beneficence. The procedures
interest in knowing. Respect for patient autonomy can used in informed consent express a commitment of re-
require that I not use force or fraud upon them. And the spect for patient autonomy and trustworthiness. If the
concern for the well-being of the public requires that profession wholly fails to equip its novices to further
that interest somehow appear prominently upon every its core values, it can be argued that it is not serious
practitioner's radar screen, that doctors not stand idly about those professed values. Its public commitments
by in the face of perils the profession can help to avert will begin to look like they are intended to convey an
and, as a lower limit, that they not do anything to in- illusion of concerned attention. In replicating itself, a
crease public peril. Consider that the overutilization of profession must replicate its commitment. Students
antibiotics, resulting in drug-resistant infectious agents, of medicine must come to care about the goods that
is professional misconduct that increases public peril. doctors ought to care about. Because justice is rarely
Ethical problems can arise, first, when core values explored as a topic in medical education, I do not
appear to be in conflict, as with the Case of the Infected think it can be counted as a core professional value.
Spouse. At issue are trustworthiness toward Andrew on However some parts of justice—nondiscrimination, for
one side and beneficence toward Wilma and a concern example—are routinely covered.
for the well-being of the public on the other. If the The second element is critical. The core values are
conflict is real, what is required is a priority rule. For not just goods that doctors care about and that doctors
example, the concept of decisional capacity is part of a want other doctors to care about. They are also goods
priority rule resolving the well-studied conflict between that the rest of us want our doctors to care about. I
beneficence and autonomy when do physicians have want my doctor to be trustworthy, to be intent on bene-
to respect a patient's refusal of life-saving treatment? fiting my health, to take my informed refusals seriously,
There is what the patient wants and what the patient and so on And we want our doctors to look out for the
needs. But when a patient is decisionally capacitated well-being of the public. The core professional values
and informed, his or her refusal trumps the doctor's are also social values. (Consider that it is not reasonable
recommendation. to want our mobsters to respect their oaths of silence.)
Kipnis: A Defense of Unqualified Medical Confidentiality 123
The third element flows from the second: an exclu- circumstances? Each doctor is ethically required to
sive social reliance upon the profession as the means by do what a responsible doctor ought to do: in order
which certain matters are to receive due attention. We to properly respect the core values of the profession.
mostly respect medical competence. But it is precisely To become a doctor without a proper commitment to
because, as a community, we have also come to accept respect the profession's values is to be unfit for the prac-
that doctors are reliably committed to their values (our tice of medicine. So how are trustworthiness and confi-
values) that we have, through state legislatures, granted dentiality to be understood in relationship to medicine's
the medical profession an exclusive monopoly on the commitment to diminish risks to third parties?
delivery of medical services. The unauthorized practice In the Case of the Infected Spouse, the ethical ques-
of medicine is a punishable crime. If, like the medical tion is posed in 1991, after the doctor—family relation-
profession, one were to make a public claim that, be- ship has been in place for a decade. The dilemma arises
cause of unique skills and dedication, some important during and immediately after a single office visit, forcing
social concern ought to be exclusively entrusted to you, a choice between calling Wilma either you will have
and the public believes you and entrusts those import- to explain to Andrew, in two days, why you disclosed
ant matters to you, incidentally prohibiting all others his infection to his wife or you will have to explain to
from encroachment upon what is now your privilege, Wilma, in two years, why you did not disclose his in-
you would have thereby assumed an ethical obligation fection to her. Each option has a bad outcome: the be-
to give those important matters due attention. Collect- trayal of Andrew's trust or the fatal infection of Wilma.
ively, the medical profession has done exactly this in Either way, you will need to account for yourself.
securing its monopoly on the delivery of certain types Infection seems a far worse consequence for Wilma
of health care. Accordingly, the profession has a col- than betrayal is for Andrew. Much of the literature on
lective obligation to organize itself so that the shared confidentiality has been shaped by this fact, and per-
responsibilities it has assumed in the political process haps the standard strategy for resolving the problem
of professionalization are properly discharged by its calls attention to the magnitude and probability of
membership. the bad outcomes associated with each option. While
A sound code of ethics consists of a set of standards predictions of harm can sometimes be wrong, it can
that, if adhered to broadly by the profession's member- be evident that Tatiana Tarasoff and Wilma Long are
ship, will result in the profession as a whole discharging at grave risk and, accordingly, it can seem honorable
its responsibilities. Where physician behavior brings to diminish the danger to vulnerable parties like them.
about a public loss of that essential trust, society may Justice Tobriner appeals to a version of this consequen-
have to withdraw the monopolistic privilege and seek tialist argument in Tarasoff:
a better way of organizing health care. Professionaliza-
tion is but one way of organizing an essential service. Weighing the uncertain and conjectural character
There are others. of the alleged damage done the patient by such
In summary, the medical profession has ethical a warning against the peril to the victim's life, we
obligations toward patients, families, and the com- conclude that professional inaccuracy in predicting
munity because of its public commitment to secure violence [or deadly infection] cannot negate the
and further certain critical social values and because therapist's duty to protect the threatened victim.
of society's exclusive reliance on the profession as its
means of delivering certain forms of health care. With Beauchamp and Childress, in their widely read
the professional privilege comes a reciprocal collective Principles of Biomedical Ethics (2001, 309), urge clin-
responsibility (Kipnis 1986, 1-14). We can now turn icians to take into account "the probability that a harm
our attention to medicine's responsibility to diminish will materialize and the magnitude of that harm" in
public perils. any decision to breach confidentiality. (While they
also urge that clinicians take into account the potential
The Duty to Diminish Risks to Third Parties impact of disclosure on policies and laws regarding
confidentiality, they are not very clear about how this
There is an implication for the way in which we must assessment is to be carried out.) In brief, the very bad
now understand the problem in the Case of the Infected consequences of not disclosing risk to Wilma—disease
Spouse. The opening question "Do you take steps and death and the betrayal of her trust—outweigh the
to warn Wilma?" has to be understood as a question not-all-that bad consequence of breached confidential-
about medical ethics and not about "you." We want to ity to Andrew. Your explanation to Andrew could cover
know what the "good doctor" should do under those those points.
II III I ai II IA 111111
The preferred argument would go something like passed along, when, to whom, and what could happen
this: The state's interest in preventing harm is weighty. then. They should never encourage or accept trust only
Medicine has an obligation to protect the well-being of to betray their patients afterwards. To do so is to betray
the community. Because the seriousness of threatened the value of trustworthiness.
grave injury to another outweighs the damage done to But now a second problem emerges. If prospect-
a patier t by breaching confidentiality, the obligation ive patients must understand in advance that a doctor
of confidentiality must give way to a duty to prevent will report evidence of a threat to others, they will only
serious -num to others. Accordingly, despite confiden- be willing to disclose such evidence to the doctor if
tiality, warning or reporting is obligatory when it will they are willing to accept that those others will come
likely avert very bad outcomes in this way. Of course to know. If it is important to them that the evidence
clinicians should try to obtain waivers of confidenti- not be reported, they will have a weighty reason not to
ality before disclosure, thereby avoiding the need to disclose it to those who are obligated to report it.
breach a duty. But the failure to obtain a waiver does Some have questioned this proposition, arguing
not, on this argument, affect the overriding obligation that there is no empirical evidence that prospective pa-
to report. tients will avoid or delay seeking medical attention or
conceal medically relevant information if confidential-
A Defense of Unqualified Confidentiality ity is qualified in this way. Despite widespread report-
ing practices, waiting rooms have not emptied, and no
As powerful as the above justification is, there are prob- one really knows if people stop talking openly to their
lems with it. Go back to 1990, when Andrew comes in doctors when confidentiality is breached.
to be tested for sexually transmitted diseases. Suppose Three responses are possible regarding this claim.
he asks: "If I am infected, can I trust you not to disclose First, there is a serious difficulty doing empirical re-
this to others?" If, following the arguments set out in the search in this area. How, for example, do we determine
previous paragraphs, we are clear that confidentiality the number of abusive parents who have not brought
must be breached to protect third parties like Wilma, their injured children to doctors out of a fear that they
then the only truthful answer to Andrew's question is will get into trouble with the authorities? How many
"No. You cant trust me." If the profession accepts that HIV+ patients avoid telling their doctors all about their
its broad promise of confidentiality must sometimes be unsafe sexual practices? How many of us would volun-
broken, then any unqualified assurances are fraudulent, teer unflattering truthful answers to direct questions on
and the profession should stop making them. If there these and other shameful matters? It is notoriously dif-
are exceptions, clinicians have a duty to be forthcom- ficult to gather reliable data on the embarrassing, crim-
ing about what they are and how they work Patients inal, irresponsible things people do and the steps they
should know up front when they can trust doctors and take to avoid exposure, especially if those are wrongful
when they can't. To withhold this important informa- too. I don't want to suggest that these problems are
tion is tc betray the value of trustworthiness. insurmountable (Reddy et al. 2002), but they are de-
Accordingly, the argument for breaching confiden- cidedly there and they often make it hard to study the
tiality has to be modified to support a qualified confi- effects of these betrayals.
dentiality rule, one that carves out an exception from Second, despite the problems, certain types of
the very beginning, acknowledging an overriding duty indirect evidence can occasionally emerge. Here are
to report under defined circumstances. (In contrast, an two anecdotal examples from Honolulu. There was a
unqualified confidentiality rule contemplates no excep- time, not long ago, when military enlistees who were
tions.) Instead of undertaking duties of confidentiality troubled by their sexual orientation knew that military
and then violating them, doctors must qualify their ex- doctors and psychologists would report these prob-
pressed obligations so they will be able to honor them. lems to their officers. Many of these troubled soldiers
Commentators who have walked through the issues therefore obtained the services of private psychologists
surrounding confidentiality have long understood the and psychiatrists in Honolulu, despite the fact that free
ethical r_ecessity of "Miranda warnings" (Bok 1983; services were available in military clinics. The second
Goldman 1980): A clinician would have to say early example emerged from the failure of the Japanese
on, "Certain things that I learn from you may have to be medical system to keep diagnoses of HIv infection
disclosed to . . . under the following circumstances . . .; confidential. Many Japanese who could afford it trav-
and the following things might occur to you as a result eled to Honolulu for diagnosis and treatment, avoid-
of my disclosure: . . ." If doctors are ethically obligated ing clinics in Japan. At the same time, Japanese data
to report, they need to say in advance what will be on the prevalence of HIv infection were unrealistically
low, especially considering the popularity of Japanese the doctor should be able to obtain a waiver of confi-
sex tours to the wv-infected brothels of Thailand. Evi- dentiality, and Wilma could then be informed. Once
dence of this sort can confirm that the failure to respect permission .to report is given, the ethical dilemma
Confidentiality can impair the ability of doctors to do disappears. 'Notice that for this group of patients, an
their job. exceptionless confidentiality rule works j ust as well as a
And third, there is an argument based on the rule requiring doctors to override confidentiality when
motivational principle that if one strongly desires that necessary to protect endangered third parties. At-risk
event E does not occur and one knows that doing act parties will be warned just the same but with appropri-
A will bring about event E, then one has a weighty ate permission from patients. In these cases there is no
reason not to do act A. The criminal justice system need to trim back the obligation of confidentiality since
is based on this idea. We attach artificial and broadly patients in this first group are, by definition, willing to
unwelcome consequences (imprisonment and other have a report made.
forms of punishment) to wrongful, harmful conduct Difficulties arise with the second type of patient:
with the expectation that, even if inclined, most people those who will not want credible threats reported.
will decide against the conduct in order to avoid the Notice that these prospective patients are in control
unwelcome consequence. If I don't want to go to prison of the evidence doctors need to secure protection
and a career in burglary will likely result in my going for parties at risk. If a patient cannot be drawn into
to prison, then I have a weighty reason to choose a dif- a therapeutic alliance—a relationship of trust and
ferent career. Likewise, if I don't want my marriage to confidence—then doctors will not receive the infor-
be destroyed by my wife's discovery that I am HIV+ and mation they need to protect imperiled third parties
I know that telling my doctor about reconciliation will (at least so long as patients have options). As a result,
result in her discovering just that, then I have a weighty doctors will not be able to mobilize protection. When
reason not to tell my doctor. The presumption must one traces out the implications of a reporting rule on
be in favor of the truth of this seemingly self-evident what needs to be said in 1990 (when Andrew asked to
principle. If critics allege that it is false or otherwise be tested and the doctor disclosed the limits to con-
unworthy of endorsement, it seems the burden of dis- fidentiality), it becomes evident that Wilma will not
proof belongs to them. It is their responsibility to come be protected if Andrew (a) does not want her to know
up with disconfirming evidence. and (b) understands that disclosure to his doctor will
It can be argued, in rebuttal, that people still result in her knowing. Depending on his options and
commit burglary and, despite reporting laws, people the strength of his preferences, he will be careful about
still go to doctors for HIV testing, even knowing that what he discloses to his doctor, or will go without med-
confidentiality has its limits. But no one would main- ical advice and care, or will find another physician who
tain that punishing convicted criminals totally prevents can be kept in ignorance about his personal life.
crime and that breaching confidentiality results in We began by characterizing the Case of the Infected
all people avoiding or delaying medical treatment or Spouse as an apparent head-on collision between the
concealing aspects of their lives. The situation is more doctor's duty of confidentiality and the duty to protect
complicated. imperiled third parties. But if the argument above is
Consider that Andrew belongs to one of two groups sound, there is no collision. The obligation to warn
of prospective patients. Members of the first group are third parties does not provide added protection to
willing enough to have reports made to others. Mem- those at-risk In particular, a no-exceptions confiden-
bers of the second are deterred from disclosure by the tiality rule has a better chance of getting the facts on
fear of a report. Of course we can't know in advance the table, at least to the extent that honest promises of
which type of patient Andrew is, but if both groups are confidentiality can make it so. To be sure, clinicians
treated alike, uncertainty will not be a problem. (While would have to set aside the vexing "Should I report?'
this division into two groups may be oversimplified, conundrum and search for creative solutions instead.
working through the qualifications would take us too These strategies will not always prevent harm, but they
far afield.) will sometimes. The nub of the matter is that these
Consider the first group: patients who would be strategies can never work if they can't be implemented.
willing to have a report made. Recall that the phys- And they can't be implemented if the fear of reporting
ician in the Case of the Infected Spouse tried to obtain deters patients from disclosure. Accordingly, there is no
assurance that Wilma would be protected. Under an justification for trimming back the obligation of confi-
unqualified confidentiality rule—no exceptions—if the dentiality since doing so actually reduces protection to
patient were willing to have reports made to others, endangered third parties, increasing public peril.
II al liIIlI
The argument advanced here is that—paradoxically— if this can be done best, as I have tried to show, by hon-
ethical and legal duties to report make it less likely that oring confidentiality and doing one's best to protect im-
endangered parties will be protected. Depending on the periled third parties within that framework, then what
prospective patient, these duties are either unnecessary you must say to both Wilma and Andrew, when they
(when waivers can be obtained) or counterproductive enter your office in 1982, should be something like this:
(when disclosure to the doctor is deterred and interven-
tions other than disclosure are prevented). There is an ethical problem physicians sometimes
In part, the conventional wisdom on confidentiality face in taking on a married couple as patients.
errs in focusing on the decision of the individual clin- It can happen that one partner becomes infected
ician at the point when the choice has to be made to with a transmissible disease, potentially endanger-
disclose or not. The decision to violate confidentiality ing the other. If the infected partner won't share
reaches backwards to the HIv test administered years information with me because he or she fears I will
earlier and, as we shall see, even before. Perhaps little warn the other, there will be no protection at all for
will be lost if one doctor betrays a single patient one the partner at risk There may, however, be things I
time or if betrayals are extremely rare. But medical can do if I can talk with the infected partner. What
ethics is not about a single decision by an individual I promise both of you is, if that were to happen,
clinician. The consequences and implications of a rule I will do everything I possibly can to protect the
governing professional practice may be quite different endangered partner, except for violating confi-
from those of a single act. Better to ask, what if every dentiality, which I will not do. You both need to
doctor did that? remember that you should not count on me to
While it is accepted here that doctors have an guarantee the wholesomeness of your spouse, if
overriding obligation to prevent public peril, it has doing this means betrayal.
been argued that they do not honor that obligation
by breaching or chipping away at confidentiality. This It is in these words that the final explanation to
is because the protective purpose to be furthered by Wilma can be found. If Wilma understands from the
reporting is defeated by the practice of reporting. The beginning that medical confidentiality will not be
best public protection is achieved where doctors do breached, if she (and the public generally) understand
their best work and, there, trustworthiness is probably that the precariously placed are safer under unquali-
the most important prerequisite. Physicians damage fied confidentiality, she will understand she has final
both their professional capabilities and their commun- responsibility for her choices. If you are clear enough
ities when they compromise their trustworthiness. about it, she will grasp that she can't depend on you
If the argument above is sound and confidential- to protect her at the cost of betrayal and that she is
ity must be respected in this case, we must now return better off because of that. Both the doctor and the
to the question of what the doctor must say to Wilma medical profession collectively need to work through
when, now infected, she returns to the office two years these issues and fully disclose the favored standard to
after the reconciliation. Though this question has finally prospective patients before the occasion arises when
to be faced in 1993, it is on the table before her return to a doctor must appeal to it. The view defended here is
Honolulu. It is there even before Andrew asks to be tested that the profession should continue to make an un-
in 1990, and you then have to decide whether to live out qualified pledge of confidentiality and mean it.
the trust he has placed in you or disabuse him of it. In It is also appropriate to consider what should be
fact, the problem is on the table in 1982, when the couple said to Andrew as he is about to leave your office in
first enters your office and asks you to be their physician. 1991 to prepare for a romantic dinner with Wilma
As a doctor, you have obligations of beneficence and con- I once spent part of an afternoon with a healthcare pro-
fidentiality, and you owe both to each. But now—having fessional who had served in Vietnam. He had counseled
read this far—you are aware that something can happen married enlistees who had returned from visits with
that you cannot control; and, if it does happen, you will their wives and had been diagnosed with a venereal
face those apparently conflicting obligations. You can disease that was probably contracted before they left
only provide what you owe to one if you betray your Vietnam. It is likely that these men may have infected
obligation to the other. That is the choice you will have their wives. This clinician had learned how to persuade
to make in 1993, unless you (and the medical profession) these men to agree to disclosure. He stressed that their
contour professional responsibilities now. wives would likely find out eventually and that the
in choosing a governing ethical principle, the emotional and medical consequences would be far more
end-in-view is to protect vulnerable third parties, and severe because of the delay. More importantly—given
the soldiers' tentative decisions not to let their at-risk to risk killing patients can disqualify one for the
spouses know—he would ask whether this was a mar- practice of medicine. While personal morality can
riage they really wanted to preserve? I recall that he play a decisive role in career choice, it shouldn't play
claimed a near perfect record in obtaining permission a decisive role within medical ethics. Many enter
to notify the at-risk spouses. It would be useful if there medicine believing that good citizens must prevent
were skilled allied caregivers, bound by confidentiality, serious injury to others, even if that means violat-
who could routinely conduct these specialized counseling other obligations, But the task of professional
ing sessions. While this is not the place to set out the ethics in medicine is to set out principles that, if
full range of options for a profession reliably committed broadly followed, will allow the profession to disc
to trustworthiness, it will suffice to point out a direction charge its collective responsibilities to patients and
for professional and institutional development. §-0-ciaY, Confidentiality, I have argued, is effective
at getting more patients into therapeutic alliances
Concluding Remarks more quickly, it is more effective in bringing about
better outcomes for more of them, and—counter-
Even if the forgoing is accepted, what may trouble intuitively—it is most likely to prevent serious harm
doctors still is a fear that they will learn about an en- to the largest number of at-risk third parties. Now
dangered person and be barred by this no-exceptions it is ethically praiseworthy for honorable people to
confidentiality rule from doing anything. Actually, belong to a profession that, on balance, diminishes
there is only one thing they cannot do: disclose. All the amount of harm to others, even though these
other paths remain open. Even if a reporting rule keeps same professionals must sometimes knowingly allow
many prospective patients out of the office or silences (and sometimes even cause) harm to occur. Although
them while they are there, the rule protects doctors doctors may feel guilty about these foreseeable con-
from the moral risk of having to allow injury to third sequences of their actions and inactions, they are not
_
parties when a simple disclosure would prevent it. This guilty of anything. They are acting exactly as it is
distress is significant and has to be faced. reasonable to want doctors to act.
Here we must return to an error discussed earlier: It is hard enough to create therapeutic alliances
the conflation of personal morality and professional that meet patients' needs. But if doctors take on the
ethics. Like law, personal morality can also conflict added duty to mobilize protective responses without
with professional responsibility. We considered a waivers of confidentiality, their work may become im-
Jehovah's Witness surgeon, morally prohibited from possible in too many important cases. And all of us will
administering blood transfusions to patients needing be the worse for that. The thinking that places the moral
them. Likewise a Catholic doctor may be unable to comfort of clinicians above the well-being of patients
discuss certain reproduction-related options. And and their victims is in conflict with the requirements of
despite understandable moral misgivings, doctors professional responsibility properly understood. While
everywhere must be prepared to administer high-risk it will be a challenge for many honorable physicians to
treatments they know will cause the deaths of some measure up to this standard, no one ever said it was
of their patients. Paradoxically, a personal inability easy to be a good doctor.
References
Beauchamp, T. L., and J. E Childress. 2001. Principles of bio- Lo, B. 1995. Resolving ethical dilemmas: A guide for clinicians.
medical ethics. New York: Oxford University Press. Baltimore Williams and Wilkins.
Bok, S. 1983. Secrets. New York: Pantheon Books. Reddy, D. M., R. Fleming, and C. Swain. 2002. Effect of man-
Goldman, A. 1980. The philosophical foundations of professional datory parental notification on adolescent girls' use of
ethics. Totowa, NJ: Rowman & Littlefield. sexual health care services. Journal of the American Medical
Kipnis, K.1986. Legal ethics. Englewood Cliffs, NJ: Prentice-HalL Association. 288: 710-714.
Kipnis, K. 2004. Gender, sex, and professional ethics in child Wicclair, M. 1986. A shield right for reporters vs. the admin-
and adolescent psychiatry. Child and Adolescent Psychiatric istration of justice and the right to a fair trial: Is there a
Clinics of North America. 13(3): 695-708. conflict? Business & Professional Ethics Journal. 4(2).
Kottow, M. 1986. Medical confidentiality. An intransigent and
absolute obligation. Journal of Medical Ethics. 12: 117-22.
II ,111 1111.111
Breaching Confidentiality However children suffering abuse or neglect at the hands

of their parents, or when the parents are making unwise
Dave Unger decisions for them, each province has a mechanism by
which children can be protected. Discussion in this sec-
E. . .1 Information that is revealed to someone such as tion is brief not because this topic is unimportant but
a health care provider can be potentially compromis- rather the contrary, it is so important there is little dis-
ing (embarrassing, financially compromising, legally agreement (in principle) about the proper governance
compromising, etc.) to that individual. Confidentiality and enforcement of this area: there is a duty to breach
is the concept that this information can be kept safe privacy and confidentiality considerations in favour of
and only be revealed to authorized persons—persons the welfare of children in every region of Canada.
authorized by the patient. In general, the scope of that This is one area where "Good Samaritanism" is not
authority is very narrow and very carefully defined only morally required, but legally required. Section
and scrupulously defended. This is fundamental to 215 of the, Canada Criminal Code outlines the duty
a notion of respect . . . for autonomy. Put respect for parents or guardians owe to children:
autonomy is only one of the principles in bioethics,
and where autonomy competes with the beneficence Everyone is under a legal duty as a parent, foster
(and or non-maleficence) of the greater society or of parent, guardian or head of a family to provide ne-
another individual, that is the place where contemplat- cessaries of life for a child under the age of sixteen
ing breaking confidentiality becomes a necessary con- years, to provide necessaries of life to their spouse
sideration. .1. . [C] onfidentiality (as an individual right) or common-law partner, and to provide necessaries
is carefully protected, but in the interest of the compet- of life to a person under his charge if that person is
ing collective concerns of the population or society, it unable, by reason of detention, age, illness, mental
is sometimes necessary to breach it. What follows is a disorder or other cause, to withdraw himself from
discussion of some of the situations in which this can that charge, and is unable to provide himself with
and must occur, the ethical rationale for warranting necessaries of life.'
these occurrences, and an examination of the relevant
guidelines policy and law in this arena. Breach of this duty is an offence under the Criminal
The discussion starts with a look at notification: the Code, and hence, is a federal concern. But in child protec-
mandatory notification and reporting of such things as tion the prosecution of errant parents under criminal law
child abuse and communicable diseases. Next the dis- is not the issue of primacy, child protection is carried
cussion turns to the notion of a duty to self report, with out under civil law. The courts have inherent jurisdic-
a focus on the archetypal example, HIV infection. Keep- tion over incapacitated persons including children,
ing with the HIV theme, the notion of a "duty to warn" under the general and ancient common law provision of
is introduced, that is, the duty of a health care profes- parens patriae—the legal concept by which the state is
sional to warn a third party who is at risk of harm from a given power to intervene and protect those who are being
patient under the care of that professional. This concept abused or injured and are unable to protect themselves.
is further developed with a look at genetic diseases and In effect, parens patriae is the legal instrument by which
finally this discussion funnels into a discussion of duty the state can effectively become the "parent" when a real
to warn in general and the relevant law and policy that parent or caregiver is effectively lacking or negligent. This
guides health care providers in this difficult area. common law mechanism is not often invoked as there
is almost universal statutory law across Canada that
1) Reporting and Notification makes the means of child protection more explicit. Each
province has a child protection agency that is run by the
First, to a relatively non-contentious topic: the duty ministry responsible for family and child development.
to report child abuse. A just society provides for its These ministerial departments have different names in
vulnerable and its voiceless members. Sanctity of the different provinces (it is called the Ministry of Children
family is acknowledged in Canada. The respect for the and Youth Services in Ontario, and the Ministry of Chil-
autonomy of the family in governing its own affairs and dren and Family Development in BC) and they refer to
directing the course of childrearing according to family this issue variably as child welfare or child protection.
(or cultural) values is fundamental to Canadian society. Their mandate is established by provincial legislation:
Source: From The Canadian Bioethics Companion, Chapter 2 (2011). Reprinted with permission of the author.
Unger: Breaching Confidentiality
in BC, this is the Child, Family and Community Service There is a duty to report impaired physicians and a
Act; in Ontario, thkChild and Family Services Act; and in duty to report physicians suspected of sexually assault-
Alberta, the Child, Youth and Family Enhancement Act. ing patients and to report impaired flight personnel—
It is mandatory in Canada to report abuse or neglect these serve also as a means of professional regulation.
of children to the ministry. It is important to note that it is There is a duty to report impaired or unfit drivers
mandatory for anyone to report abuse that is witnessed (such as those with seizures or dementia). This list is
or suspected, not just health care workers. Children are not comprehensive but serves to point out the extent of
deemed to need protection if they are physically or sex- the paternalism and the range of harm reduction tactics
ually abused, emotionally harmed, deprived of necessary that are invoked in the name of the greater good. It also
health care,2 or if parents are absent from home or have highlights the degree to which breaches of confidential-
otherwise abandoned their children and are unwilling to ity occur within a nation that holds autonomy so dear.
provide parenting. The legislation typically protects the With an appeal to a (legitimate) public health
identity of the reporting persons as well. agenda, Canada has several mechanisms in place for
Children can be apprehended by law enforcement, the protection of its citizens against certain communic-
hospital personnel etc., and are usually given over to able diseases. This is one of the few legislated (albeit
social workers or similar agents. Immediate health very limited) breaches of confidential information of
concerns are managed before the children are ultim- patients per se. It is mandatory to report diseases (both
ately found safe living conditions—usually this is in new and emerging, as well as old and now rare) that
the home of another family member or a foster home. pose the threat of rapid spread, difficult containment,
The case of a child needing protection will be decided or that pose threat of significant morbidity. Reports
in a court. Sometimes this means several hearings and are made to provincial health authorities (usually to a
navigating a complicated legal process before a defin- "Provincial Health Officer" or "Chief Medical Officer"
itive outcome is found. Ultimately, the goal is to return of the various provinces), but these authorities differ in
children to their home of origin or their family once the their scope. For the most part, tracking and reporting is
potential for abuse is mitigated or the situation has been automatic and there are robust federally or provincially
dealt with. The child protection laws give authority to mandated agencies at the helm. Most of this reporting
child protection workers to follow up on the safety of is not done with the immediate knowledge or consent
the children long after they are returned to their parents of the patients involved; and for the most part this is
or families and sometimes the children are supervised understood as acceptable by citizens in the society.
or returned gradually to live with their parents.3 Communicable diseases which are reportable, or
Moving from the concerns of the vulnerable indi- notifiable, are listed in the Notifiable Diseases Summary
viduals to the public health concerns of a population, at the Public Health Agency of Canada (PHAC).4 There
there are other forms of mandatory reporting that are not are provincial and territorial statutes in each region
predicated on "good Samaritan" principles but are none- stipulating the reporting of these diseases provincially
theless good for society and overstep confidentiality of and ultimately nationally and a report is published as
patient information. In Canada there are a whole array of the Canada Communicable Disease Report (ccDR) to
situations that require reporting under the various health monitor trends and keep surveillance.5 These diseases
acts, motor vehicle acts, and vital statistics acts (etc.) of include sexually transmitted diseases such as chla-
the provinces and territories. These include unexpected mydia and gonorrhoea and the re-emerging syphilis.
deaths and births and many, many matters in between. All Reporting also includes now rare diseases that were
deaths must be recorded as vital statistics. Deaths must be once common but are carefully monitored in light of
reported to the coroner's office if they are unexpected, attempts at vaccine eradication. Examples include
self-induced or accidental, or if they occur outside a such things as measles, mumps and rubella. Significant
hospital for a patient that is not known to be terminally chronic infections such as Hepatitis C and HIV are re-
ill. Births are also required to be recorded. Accidents that portable. Syphilis is an example of why it is important
occur at a worksite are required to be reported and man- to report infections inasmuch as monitoring has re-
aged through the appropriate Worker's Compensation vealed new outbreaks and vulnerable populations that
Board (or similar agency) in each province. In addition are affected by this new visitation of an old scourge.'
most provinces have registries of certain medical condi- Other infections are reportable not only to monitor
tions, such as cancer registries that compile statistics on demographics and aid prevention, but also to monitor
and monitor follow-up of cancer patients. treatment and follow-up; the example of tuberculosis
There_are_times __when confidentiality can be (TB) will be discussed in the next section It should be
breached to protect members of the public from harm. noted that infectious disease reporting while effective
uW I 11 II 111111
within a province, and while effective in conveying result.' If the US is any indication, this is a trend that
statistics to national agencies, are not perfect inasmuch will likely continue to grow: in the US as of 2005 there
as contact tracing (wherein the public health agency re- were 45 states that offered anonymous testing.8
sponsible will track down contacts of the index case or
infected patient) between provinces often breaks down. 2) Disclosing and Warning
Patients have no choice over the reporting of these
diseases; their confidentiality is implicitly waived. It is the Reporting is akin to disclosing when the information is
responsibility of health care providers to report this fact transmitted, not to a government agency, but to other
to patients before the testing is ordered so that patients personal stakeholders. In discussing the obligation to self
can make an informed decision about their rights and, report (disclose), HIV is a prime example. There are sev-
in effect, voluntarily waive their confidentiality. All this eral aspects to this issue. First: consider the obligations
is not to say that their confidentiality is entirely waived, that a health care provider has to his or her patients. Much
while these diseases are reportable, it simply means that of the above discussion involves the respect for the auton-
persons other than the health care provider and the pa- omy and the confidentiality of patients, but one other way
tient have access to the results. These results do not go in which the interests of the autonomous person must
beyond the disease control authorities and any follow-up yield to the needs of the many is in the doctor patient
contact tracing done by these authorities is anonymous— relationship where there exists the potential for harm to
patients' information is not public, the confidentiality patients. In this respect, the old adage, "first do no harm"
seal is broken and the envelope is simply expanded. is immediately apparent as operational—non-maleficence
There are diseases that lie in ethical grey zones. It is must come before autonomy. As discussed above, in every
mandatory to report HIV in all provinces and anonymous province there are professional colleges responsible for
contact tracing is provided by the various disease control the licensing and the discipline of physicians, dentists,
agencies in the provinces (again, this is sometimes ineffec- midwives, nurses, podiatrists and other specialties that
tual between provinces). This arrangement has driven the provide patient care (for example the College of Phys-
emergence of a new issue and a hot topic among those in- icians and Surgeons of each province). Health care pro-
terested in public health ethics, one that is especially acute viders that deliver this care have an ethical responsibility
in the management of HIV. Breaches of confidentiality are to report to their professional college-the presence of a
allowed in the reporting of HIV so it has been questioned diagnosis that could compromise their patients—the clas-
whether people will forego testing on these grounds sic example is the surgeon who has HIV. The college can
alone. That is, it has been a growing fear that people will then, at its discretion, restrict or alter the licensing of these
remain intentionally ignorant of their infection to spare professionals, but it is usually left to the professional to
themselves the stigma and embarrassment of notification self report in the first place, and hence restrict his or her
and contact tracing, and this situation, of course, puts the own practice so as to not compromise his or her patients.
broader society in harm's way as possible infections go un- The next aspect of the duty to report is the duty a
checked and unreported. One theoretical way to resolve person has to report his or her own HIV status to a sexual
this issue would be to allow anonymous testing of individ- partner. (This can be viewed as a transition to the issue
uals, they (theoretically) could spare themselves the pain of "duty to warn" dealt with immediately below) While
of shame and stigma, and society could then rely on them there are several specific American laws on this count,
to be good stewards of this personal information and use Canadians rely on sections of the Criminal Code for suf-
appropriate discretion in their sexual and drug use habits. ficient jurisprudence in this real& Although not strictly
A complete discussion of this cannot be had here, speaking the first case, the most prominent case and the
nor are there agreed upon resolutions of the argument one that has set the stage for future precedents was that
in either the empirical or theoretical literature. None- of R v Cuerrier wherein Henry Cuerrier was charged and
theless, there remains a loud cry for the availability ultimately convicted of assault for having sex with two
of anonymous testing in Canada (and the US). There separate partners.w Here it was found that because, al-
is no true means of anonymous testing at hospitals though the women consented, they were not told of his
and labs in most provinces as positive tests will be HIV status and therefore (taking a page from the bioethics
traced back to identify and record the patient and the playbook) could not be considered to have given proper
disease and contact tracing will be attempted. There informed consent. Since then, there have been over
are, however, specialized facilities in many Canadian 70 such cases in Canada, and some HIV advocates would
cities that attempt to provide anonymous (so called suggest that they are tendering an ever increasingly perni-
"non-nominal") testing whereby a patient's identifiable cious attitude in law against persons with HiwAiDs.11 In
information (such as name) is not attached to the lab April 2009, Johnson Aziga was convicted in Ontario for
two counts of first degree murder (the first conviction on Much is made in the bioethical literature on defining
this charge in Canada) for infecting two women who died the full extent of this issue, and much ink spilled also on
and several counts of aggravated assault.12 the implications of legislation and policy in this realm.
Genetic issues are simultaneously individual and familial
in nature, the one facet merges into the other, and obli-
3) Duty to Warn
gations to the patient sometimes collide with obligations
As was mentioned above, there are duties to report to patients' families, and this, in turn, places a health
many diseases and circumstances, but where this care provider on very thin ice. Part of the problem in
crosses over to a duty to warn the public or a specific this realm is in the determination of risk and imminence,
individual is a significant departure from the confi- and whether this information would be certain to benefit
dentiality issues that have been discussed so far, and the third party—these issues are considered below. The
a significant area of contention in the study of public uncertainty of these factors gives duty to warn a dubious
health ethics. Where a duty to report becomes a Duty- legitimacy, yet the magnitude of the diseases considered
to-Warn is another area in which confidentiality is strikes fear that failing to warn could be catastrophic. As
waived and represents an ethical minefield that has not the risk of injury stemming from ignorance is amplified,
been entirely mapped(The basic logic here follows the the case for a duty to warn becomes more compelling,
"Harm Principle": that. individual rights and autonomy In Canada there is no specific statute or law which
must give way when there are risks of harm to other governs this problem, nor has there been a case to test
Carrying on in the theme of HIV, there is an obligation the matter of whether it is ethically justifiable or obliga-
to report HIV to provincial health officers, and then a tory to inform family members of a serious genetic ill-
duty for said officers to attempt contact tracing of pos- ness diagnosed in one's patient.16,17 Canada can look to
sible sexual contacts and give information to contacts significant cases internationally for guidance, as these
about the need to be tested. (This is standard with any issues have been forced upon other countries and where
STD or reportable disease, and the health officers' duty is they have been decided in courts there are precedents for
usually dispatched through its mandated delegate, the a duty to warn, and where professional advisory bodies
provincial centre for disease control.) All of this is done have been forced to scrutinize the issue, policy has con-
while maintaining the anonymity of the index case. densed into guidelines that favour a duty to warn under
But precisely when it becomes a duty to broadcast the specific circumstances." In the US there have been three
status and identity of a potentially dangerous infectious cases that have found a physician has a duty to warn the
person to the general public is a topic which is long on relatives of a patient with an identifiable genetic disease,
speculation, yet short on empirical evidence and legal or at the very least, warn the patient that he or she has
backing. Police forces have been criticized for releasing the duty to warn the family members.19-21
media advisories on known HIV positive persons, and This area remains unclear but it is possible that some
on the other end, governments have been criticized for of the law and policy recommendations discussed below
not warning about potentially dangerous HIV positive may, in theory or in practice, come to bear on this issue.
persons. Thus far, there have been no legal cases specif- What is certain is that in the rapid advance of medical and
ically involving either a duty to warn, or the breach of genetic technology, this is sure to be an ethically charged
duty to warn, for a health care provider to warn about area that will continue to be confronted and shaped in the
an HIV infection or HIV infected person.1" The years years to come. At present, the current ethical guidelines
to come will be fraught with challenging cases and seem to accommodate these issues at least to some extent.
evocative new policy as this disease pushes these ethical Current good clinical practice guidelines of any Canadian
issues to new extremes and with a new urgency." professional body would certainly accommodate general
While much of the foregoing discussion surrounds recommendations. Before a diagnosis is even made, gen-
HIV—as it exemplifies the ethical problems and defines the etic counselling should take place: counselling that sets a
contour of the law and policy of Canada in these ethical clear course for the patient regardless of the diagnosis, and
realms—there are further, broader concerns about a duty a clear plan of action for the disposition and disclosure of
to warn that move beyond this archetypal disease situation. the information once it is available. In this manner, the
Before moving to a general discussion on duty to warn and issue of a duty to warn is likely to be stemmed before it
protect, a digression is warranted into the realm of genetic becomes an issue. Once a diagnosis is made, it is advisable
diseases and the obligations that flow from their diagnoses. that physicians (or more likely and more appropriately,
Just as serious socially transmissible diseases evoke ques- a genetic counsellor) first ask the diagnosed patient to
tions of duty and obligation, so too do serious genetically disclose of their own accord (even facilitating a family
transmissible diseases evoke such questions. conference or helping to draft family letters to enhance
II ;A 1111.11.1
understanding). Failing this, the physician (or counsellor) codes of ethics and conduct, and more importantly are
shouldnext try to gain the consent of the index case before sanctioned by the provincial health acts to license and
disclosing to family members.17 Beyond this there are no govern their members. Furthermore, regulatory bodies
firm guidelines but the good judgment of the clinician are to ensure that "best practices" are implemented by
and decisions should be made after consideration of the their members. It is this arrangement that gives the
general circumstances under which a duty to warn is war- legal instruments their teeth. Enforcing a duty to warn
ranted, and discussion will now turn to this. or penalizing a health care provider for not carrying out
Generally and conceptually, the notion of a duty this duty is an issue fraught with controversy and dif-
to warn jibes with basic ethical considerations; there ficulty. Patient advocacy groups have in some instances
is a point at which rights to individual autonomy must pointed out that the law does not require a health care
yield to the collective concerns of the broader society. provider to breach confidentiality but rather pqrmits
Precisely where this threshold is located is a challenge him or her to do so. Stemming from concerns noted
for bioethicists (and public health ethicists) and for above, this is particularly the case with HIV notifica-
health care providers on the front line. The problem is tions; HIV patient advocates are quick to point out that
fairly straightforward (and a detailed discussion of this the common law does not oblige, but rather grants dis-
will be left to the bioethical and public health ethics cretion in disclosure. It is difficult to define just how far
theorists and to other texts): the relationship between to go in carrying out the obligations. Many health care
a care provider and patient is often very intimate, and providers are poor assessors of risk or predictors of the
trust and confidentiality are integral to actually making actions of their patients, and this imposes a rather open-
this relationship work. Betraying confidences, even ended obligation on them to protect or warn. Legal an-
while advancing the needs of the broader public, may alysis has suggested that the liability that providers have
utterly destroy the provider—patient relationship and, is in the assessment of the degree of violence or harm
in turn, erode the foundation of the care giver—patient that can be done and not in prediction of probability—
enterprise. And this enterprise, the Canadian health the magnitude, not the likelihood, takes precedence."
care and social welfare system, is sacred, dearly cher- There is a rich history of case law in this realm.
ished, fiercely defended, and scrupulously maintained The American case of Tarasoff v Regents of the University
and paid for by the Canadian public—threatening it is of California is the case that is often first cited (even in
something Canadians do not have much appetite for. Canadian law) in the development of an understand-
Where a dearth of concrete empirical data and a ing of the duty to warn. In this case a psychologist was
paucity of incontrovertible ethical doctrine create a deemed to have an obligation to warn and protect third
vacuum, the space is often filled by legislation. Patients' parties (in this case Tatiana Tarasoff) threatened by one
confidentiality is protected by statutory law and policy of his patients.25 Canadian case law evolved more slowly
as was noted above, but there are no specific laws to in- and points of law established in increments. It was estab-
struct just when to break these confidences aside from lished that physicians were responsible to act reasonably
the situations already noted regarding driving, flying, to warn and protect the general public from harm, both
birthing or dying. There is no statutory legislation on identifiable and non-specific non-identifiable victims.26."
obligations placed on health care providers to warn or Furthermore, a patient's chart and information could be
protect their patients. Rather, this obligation is defined commandeered for the purposes of preventing injustices
in Canada by case law. These concepts called "duty to to others.28 But it was the landmark case of Smith v Jones
warn" (which was introduced above) and "duty to pro- that set the bar in this topic." In this case a man (Jones)
tect" have a fairly rich case law evolution in Canada, was charged with assaulting a prostitute. He was assessed,
but before a discussion of this, it is necessary to make at the request o f the defence, by a psychiatrist (Smith) who
some distinctions between these terms. Duty to warn determined he was very likely to repeat his offence and kill
means an obligation of the health care provider to warn or maim prostitutes in the future. While Jones ultimately
a Third party (or potential third party) of danger from a pleaded guilty, the issue became one of whether the psych-
patient of that health care provider. Duty to protect goes iatrist could disclose the information about his dangerous
beyond duty to warn and includes the duty to protect a patient to the appeal judge. Indeed Dr. Smith earnestly
third party—this can include actions performed on the wanted to disclose this information as he feared for the
threatening patient, such as holding them against their greater public. Ultimately, the ruling in the Supreme Court
will under a provincial Mental Health Act.22-23 of Canada held that the information was to be revealed
As discussed above, in the Canadian framework to the court, and that the counsellor—client privilege (be-
most health professions are self-regulatory; that is, tween a lawyer and his or her client), and by extension the
they are governed by professional colleges that have physician—patient privilege, should be sacrificed for the
greater good of_protecting_the public. The opinion of the were adopted by the Royal College of Physicians and
court outlined several factors that were to be considered Surgeons, the Canadian Medical Protective Associa-
in breaking confidential relationships and these continue tion (CMPA), and Ontario professional colleges. A year
to provide basic guidance in this realm. later the Smith case developed a notion of which fac-
So after all this theoretical and legal consideration, tors should be considered in adhering to these recom-
there are a few practical things that can be said to guide mendations including: what the specific threat is, how
health care providers in their duty to warn and they were serious the threat is, how imminent the threat is and
derived in the following forums. In 1998, a working how likely to be carried out (whether the patient ac-
group of medical experts in Ontario established several tually has the means, etc.), whether the intended victim
guidelines for the duty to warn. These included the gen- is specific or general, whether there is a hist ry of this
eral recommendations that a physician (and this can be type of violence in the patient's past. Th CMPA which is
broadened to include other care providers) must warn the institution which oversees the medico- egal affairs of
if a patient threatens a third party with serious harm physicians, still adheres to these general principles and
and if that harm is "more-likely-than-not" to be carried uses the Smith case for illumination.30
out. This was true of threats on identifiable individuals In sum, in dispatching these duties to warn and
("I am going to kill Mary Smith"), or non-specific yet protect, the recommendations backed by law have been
credible and well planned threats ("I am going to plant variously interpreted, but in the broadest of terms in-
a bomb in this place . . ." or "I am going to continue to clude some of the following guidance. If a threat is likely
molest 8 year old boys near my home . . ."). Further rec- and imminent, and if the potential for serious harm
ommendations were that there should be some standard (physical or otherwise) is great, care givers should warn
by which to evaluate these threats and their risks, and the local law enforcement and even, in some cases, the
in the face of uncertainty that an evaluation by another third party. It is also included in the duty to protect
colleague, as to the extent and seriousness of the threats, that the care giver could consider detaining the patient
should be sought. Finally, there were recommendations under the provincial mental health act against his or her
that all threats should be seriously considered by health will. Further, almost every professional society recom-
care providers and that it was better to err on the side mends that in cases like these the care giver should seek
of caution in notifying authorities.24 These general rules legal counsel of his or her own as soon as possible. F. . .1
Note
The copyright holder for this article would like it noted

that the book from which this reading is derived is not
a peer-reviewed publication.
References
1. Government of Canada. Criminal Code of Canada. 6. For the record. CMAJ. August 12, 2008;179(4):316.
R.S.C. 1985, c C-46, as amended. Section 215 (1). 7. Information on these sites can be obtained through the
2. Necessary health care is, of course, a very open-ended Public Health Agency of Canada website: http://www
term, and can include such things as refusing to give .phac-aspc.gc.ca/aids-sida/info/4-eng.php.
consent for an operation, blood transfusion, cancer 8. According to research by the Henry J Kaiser Family
treatment, etc. The parents can refuse to consent to Foundation. http://www.kfr.org/hivaids/upload/Up-
these things for themselves but cannot refuse to allow dated-Fact-Sheet-HIV-Testing-in-the-United-States.pdf.
an acceptable standard treatment for their dependent 9. Grant I. The boundaries of the criminal law. the crim-
children. inalization of the non-disclosure of HIV. Dalhousie Law
3. Children are not always taken out of or returned to fam- Journal. Spring 2008(31).
ilies, these laws pertain to any person that is the legal 10. R v. Cuenier; [1998] 2 SCR 371 (Supreme Court of Canada
guardian of children, family or otherwise. 1998).
4. The Notifiable Diseases Summary can be viewed at the 11. Symington A. Criminalization confusion and concerns:
Public Health Agency of Canada website: http://www.phac- the decade since the Cuerrier decision. HIV AIDS Policy
aspc.gc.cdpublicat/ccdr-rmtc/04vol30/dr3021b-eng.php. Law Rev. May 2009;14(1):1, 5-10.
5. The Canada Communicable Disease Report can be 12. R v Aziga, at the time of this writing this case is still open
found on the Public Health Agency website: http://www to appeal, and it is certain the final page has not been
.phac-aspc.gc.ca/index-eng.php. turned on this case.
II Lill IIIIJJ
13. One would be remiss to fail to mention a key concept 17. Lacroix M, Nycum G, Godard B, Knoppers BM. Should
in public health ethics that concerns such as these hinge physicians warn patients' relatives of genetic risks?
on; and that is the "Harm Principle." This fundamental CMA J. Feb 26 2008;178(5):593-595.
concept is attributed to John Stuart Mill: "That the only 18. Godard B, Hurlimann T, Letendre M, Egalite N. Guidelines
purpose for which power can be rightfully exercised for disclosing genetic information to family members: from
over any member of a civilized community, against his development to use. Fam Cancer. 2006;5(1):103-116.
will, is to prevent harm to others" (Mill J. On Liberty, 19. Pate v Threlhel, 661 So.2d 278 (Florida 1995).
1859). This is, in essence, a justification for intervention 20. Safer v. Estate of Pack, 677 A.2d (NJ Superior Court App
by the state that warrants infringements on personal au- Div 1996).
tonomy in the name of harm prevention or reduction. In 21. Molloy v, Meier 679 NW.2d 711 (Minnesota 2004).
public health practice this is most commonly considered 22. This distinction is important inasmuch as it can mean
in the context of a duty to protect the public from harm. that confidentiality is not breached at all, and the
14. A notable exception may be found in Pittman Estate v. remedy is to simply detain the patient.
Bain (1994) 112 DLR (4th) 258 Ontario Central Div- 23. Canadian Psychiatric Association. The Duty to Protect. CPA
ision, (Ontario Supreme Court 1994) wherein Dr Bain Position Paper 2002. http://publications.cpa-apcorWmedia
failed to notify his patient that he could be HIV positive .php?mid=154&xwm=true. Accessed 28 February 2011.
through contaminated blood. In this case the patient 24. Ferris LE, Barkun H, Carlisle J, Hoffman B, Katz C,
was HIV positive, and the doctor was found liable for Silverman M. Defining the physician's duty to warn: con-
damages to his patient but also, by extension to the pa- sensus statement of Ontario's Medical Expert Panel on
tient's wife whom he also, indirectly, failed to warn. Duty to Inform. CMAJ. Jun 2 1998;158(11):1473-1479.
15. HIV is being used as an exemplary and archetypal case 25. Tarasoff v Regents of the University of California, 17 Cal. 3d
more for the purposes of demonstration of ethical issues 425, 551 P2d 334, 131 Cal. Rptr 14 (California Supreme
and the contour of law and policy in Canada, but the util- Court 1976).
ity of this has its limits as HIV becomes imbued with more 26. Carver PJ. Chapter 10: Mental health law in Canada.
peculiarities that conjure new ethical problems. Some of In: Downie JG, Caulfield TA, Flood CM, eds. Can-
these problems have to do with the changing nature of adian health law and policy. 3rd ed. Toronto: LexisNexis
the disease which, on one hand, represents dangers of Canada; 2007:399-436.
transmission if untreated, and on the other hand, rela- 27. Gray JE, O'Reilly RL. Clinically significant differences
tively minimal risk of transmission if well treated. One among Canadian mental health acts. Can J Psychiatry.
new issue is now becoming just where to draw the line in May 2001 ;46(4):315-321.
a duty to disclose or warn: does one have an obligation to 28. R v Ross (1993), 119 NSR (2d) 177 (CA) (Nova Scotia
disclose if one's virus is uncontrolled but not if one's virus Court of Appeal 1993).
is well treated (undetectable in the blood)? There are a 29. Smith v Jones, (1999] 1 S.C.R. 455 (Supreme Court of
host of new problems that attach to HIV specifically that Canada 1999).
are beyond the scope of this text. 30. Evans KG, Gowling Lafleur Henderson LLP (General
16. Gold J. To warn or not to warn? Genetic information, Legal Counsel). A medico-legal handbook for Canadian
families, and physician liability. McGill Medical Journal. physicians (Sixth edition). Ottawa, Ontario: Canadian
2004;8(1):72-78. Medical Protective Association; 2005.
How Should Health Data Be Used? contribute to new concerns over privacy, confidential-
ity, and data protection. Two court cases, one in the
Privacy, Secondary Use, and United Kingdom and one in the United States, provide
Big Data Sales opportunities for thinking through ethical issues re-
lated to these developments. Each case involved selling
Bonnie Kaplan
data for marketing prescription drugs, and in each case
the court decided in favor of selling the data. However,
Introduction the cases were decided on different grounds, raising
Electronic health records, data sharing, big data, and more general issues of secondary use of health data and
secondary use of health data enable exciting oppor- the growth of health-related databases, data sharing,
tunities for improving health and healthcare. They also data aggregation, and biometric identification.
Source: From Cambridge Quarterly of Health Care Ethics 25 (2016) (2), 312-29. Reprinted with permission of Cambridge University
Press Journals.
Kaplan: How Should Health Data Be Used? Privacy, Secondary Use, and Big Data Sales
Significant health data protection, policy, and clinicians, and for society as a whole.3 Intimacies
ethical considerations are inherent in these cases. The are revealed in the interest of good healthcare, so
cases call into question just what constitutes "privacy" clinicians' professional and fiduciary duties include
and "public interest," and considerations for balancing a duty of confidentiality. Therefore, health informa-
them. They provide an opportunity to weigh privacy tion is given special protection internationally, though
against the numerous beneficial uses of data: for in- specific ways of achieving it differ. Lifestyle choices,
dividual patient care, public health, research, biosur- reproductive abilities, and stigmatizing conditions
veillance, and marketing. The cases prompt ethical are considered highly sensitive. But what is included
questions of commodifying medical information and in these categories differs with cultural background,
of harmonizing policy across jurisdictional bound- from place to place, and from time to time. What is
aries. They raise concerns of how health data can, considered very private, embarrassing, stigmatizing,
and should, be used. Their consequences may affect or grounds for discrimination varies among individ-
biomedical informatics, patient and provider privacy, uals and groups.4 Countries, likewise, vary in what
and regulation in ways this article explores, both in the personal information is treated as needing restricted
United States and elsewhere. collection, use, and disclosure.5,6 They also balance
How health data can, and should, be used is at the privacy and other considerations differently; thus
intersection of public health, research, care, privacy, privacy protection is more lax in some places than in
and ethics. This article provides an ethical analysis of others. In India, for example, the judiciary considers
these interwoven ethical issues involving appropriate, privacy on a case-by-case basis, as an exception to the
judicious, and ethical secondary data use, reflecting a rule that permits government interference in private
more general discussion of big data benefits and harms, life. Unlike in Europe and the United States, public
and touching on common dualities of the individual interest, welfare, and safety take precedence over in-
versus the aggregate or the public interest, research dividual rights, liberty, and autonomy.'
(or, more broadly, outside the health field, innovation) Yet, as discussed subsequently, even if individual
versus privacy, individual versus institutional power, clinicians scrupulously meet the professional obliga-
identification versus identity, identification versus au- tion of confidentiality, confidentiality can be comprom-
thentication, and virtual versus real individuals and ised by legal requirements to collect, document, and
contextualized information.' disseminate personal health information, especially
I start by discussing what makes health data spe- when maintained in computer databases that can be
cial, including international consensus on the import- combined easily with other sources of information
ance of the clinician's duty of confidentiality and on about the person.8 What patients reveal for the pur-
health data privacy or protection. Next I summarize the pose of healthcare may then be used in ways they never
court cases. Then I consider who benefits from data intended. Privacy practices have not caught up to these
disclosure and aggregation, and secondary use for data trends.
mining, research, and sale. Throughout, I highlight
potential benefits and harms and argue that transpar- Fair Information Practices and De-identification
ency flexibility, and accountability is needed. Ethical The same Fair Information Practices (FIPs) underpin
and policy analysis should assess data uses and users, privacy policies in both the European Union and the
as some are more compatible with societal norms and United States. The European Union and the United
values than others. States each protect personal data, including data con-
Considering how health data should be used in cerning health, albeit differently.
light of these issues suggests policy opportunities con- The United States approaches privacy by sector;
cerning patient data and privacy protection. As the use separate laws address confidentiality in distinct do-
of electronic health records, electronic medical devices, mains, such as finance and healthcare. Health data pri-
mobile and e-health applications, and biometric, social vacy collected in the course of clinical care is governed
and behavioral, and genomic data spreads, these con- by the U.S. Health Insurance Portability and Account-
siderations are becoming more relevant worldwide. ability Act (HIPAA) of 1996, extended by the HIPPA
Privacy Rule in 2001 and again in 2013 by changes
What's Special about Health Data?— mandated by the 2009 Health Information Technology
International Principles for Economic and Clinical Health (HITEcH) Act (part of
the American Recovery and Reinvestment Act [ARRA] of
All countries recognize confidentiality as a pa- 2009) and the Genetic Information Non-Discrimination
tient's right2 that is good for individual patients and Act (GINA) of 2008.9'1°,11
.11111.1
The European Union takes a more comprehensive Duty of Confidentiality

general approach to privacy; Article 8 of the European Health data privacy relates not only to expectations
Convention on Human Rights includes the right to about privacy in general but also to norms involving
data protection. This right is embodied in the 1995 professional practice, privilege, autonomy, paternalism,
Directive 95/46/EC on the Protection of Individuals and protected communication and the duty of confi-
with Regard to the Processing of Personal Data and dentiality, as well as to requirements for data collection,
on the Free Movement of Such Data.12 Member states dissemination, and retention.
implement directives differently, but the EU Data Physicians and nurses have duties both to their
Protection Regulation establishes a single set of rules individual patients and to the health of their commun-
for data protection across the EU; the final texts are ities.26 At least since the time of the Hippocratic Oath,
expected to be adopted by the European Parliament it is believed, societal norms and common law have
at the beginning 2016 and the new rules to become recognized that clinicians' duty to patients includes
applicable two years after.13,14 maintaining confidentiality, except where protecting
Despite their differences, both the United States the public interest or other individuals requires over-
and the EU construe privacy as control and protec- riding it. The International Code of Ethics of the World
tion of data rather than other conceptions of privacy15 Medical Association (wMA) makes respecting the
Both the United States and the EU also make special right to confidentiality a duty integral to a physicians'
note of health information, and both rely on stripping responsibility to patients.27 The WMA Declaration of
data of content presumed to identify the individual Helsinki—Ethical Principles for Medical Research In-
represented by the data. As Paul Ohm points out: "In volving Human Subjects (revised 2013) places a duty
addition to HIPAA and the EU Data Protection Directive, on physicians "to protect the life, health, dignity, integ-
almost every single privacy statute and regulation ever rity, right to self-determination, privacy, and confiden-
written in the U.S. and the EU embraces—implicitly tiality of personal information of research subjects . . .
or explicitly, pervasively or only incidentally—the as- even though they have given consent."28
sumption that anonymization protects privacy, most Recognizing that this personal information,
often by extending safe harbors from penalty to those whether collected for research or clinical practice,
who anonymize their data."16 increasingly is held in databases, in 2002 the WMA
As these safe harbors stipulate, neither HIPAA nor adopted the Declaration on Ethical Considerations
the EU Data Protection Directive applies after data are Regarding Health Databases: "Confidentiality is at the
de-identified. However, relying on de-identification heart of medical practice and is essential for main-
contributes to what is considered an inadequate and taining trust and integrity in the patient—physician re-
problematic legal framework for data protection.17 lationship. Knowing that their privacy will be respected
Addressing concerns over de-identification "would gives patients the freedom to share sensitive personal
require a significant shift in approach towards data information with their physician."29
protection across Europe."18 Similar deficiencies plague In this 2002 declaration, the WMA reaffirmed that
the United States.'9,2° violating this duty could "inhibit patients from confid-
Privacy protection, then, depends not merely on ing information for their own health care needs, ex-
de-identification but on expectations, transparency, ploit their vulnerability or inappropriately borrow on
and how data are used. De-identification, or anonym- the trust that patients invest in their physicians" while
ization, presumes that it is possible to identify and at the same time recognizing the value of secondary
enumerate the kinds of data that might contribute to health data use for quality assurance, risk management,
privacy risks and to specify how to prevent harms,21 and retrospective study3°
that such a list is static and sufficient in all contexts,22 Thus, a key reason for treating health data as re-
and that there are no privacy harms if the individual quiring special protection is to maintain trust between
is not identified, even though individuals may object clinician and patient in the interest of both social and
to uses of their personal data even if they themselves public order as well as better care for each individual
are anonymous.23 Furthermore, HIPAA permits sec- patient. In recognition of this ethical duty, confidenti-
ondary uses of data for research, public health, law ality is seen worldwide as a health professional's legal
enforcement, judicial proceedings, and other "public duty, one that protects the professional from giving
interest and benefit activities," without individual legal testimony, thereby serving the interests of pa-
authorization, thereby assuming that "public inter- tient and public by maintaining trust during medical
est" is clearly understood.24,25 All are questionable encounters. Nowhere can private data about a patient
assumptions. rightly be passed to a third party without that patient's
permission, except as required by law. French criminal was a plaintiff in one of the court cases) unsurprisingly
law makes this universal spirit apparent by criminaliz- takes the position that analyzing doctors' prescrib-
ing a physician's breach of confidentiality even in court ing habits contributes to patients becoming informed
testimony even if the patient would allow it.31 consumers.36
How people value and respond to concerns about Yet patients can be harmed when data about them
health data privacy is affected by context and common are used to violate privacy: to deny employment,
expectations of privacy.32 Many recognize that clin- credit, housing, or insurance, and for identity theft
icians need highly personal information in order to care and other unsavory purposes. Some fear that patients
for them, and, because of the long-standing history of who are insecure about the confidentiality of prescrip-
trust in professional confidentiality, such patients will- tion or other health record information may withhold
ingly share sensitive information with those who treat information, refuse diagnostic and genetic testing,
them. As Deryck Beyleveld and Elise Histed eloquently or decline electronic prescriptions.37,38,39 People do
point out: change their behavior and withhold information in
order to protect their health information privacy.4°
Information that patients provide for their treat- Even before the widespread use of electronic health
ment is about very personal and sensitive areas of records, a 2000 Gallup poll indicated that the vast
their lives. Indeed, it relates to their very existence, majority of people in the United States opposed third-
both physically and symbolically. As such, it is party access to medical data without a patient's per-
not information that they may be presumed to be mission, and, furthermore, that 67 percent of those
prepared to disclose or have used freely. It is their polled opposed the release of data to medical research-
vulnerability, constituted by pain and distress, or ers.41 Similarly, the Pew Internet and American Life
fears about their health and lives, that leads them Project reported that, to protect privacy, according
to disclose this information to health professionals. to a 1999 survey, nearly one in six patients withheld
At the same time, people are apt to attach great im- information, provided inaccurate information, doctor-
portance to intimate information about themselves hopped, paid out of pocket instead of using insurance,
and their bodies, and this can be associated with or even avoided care. Eighty-five percent feared that
mystical and religious beliefs, which by their very seeking health information on the Internet would
nature can be idiosyncratic.33 result in changes in insurance coverage or otherwise
reveal their information.42
Patient Benefits and Harms
Individuals also may provide health information freely Transparency and Consent
via health-related social networking, web postings, As information resources become more ubiquitous
and searches; or because it is required legally, as for and information sharing becomes more profitable,
prescriptions. Such information could be consolidated more thought is needed concerning which data uses
and linked with other data for beneficial or nefarious are acceptable and what control individuals should
purposes, sometimes without individuals' knowledge. have over data about themselves. Privacy violations
Patients benefit from having their record information may compromise patient care, the information in
available from previous clinical visits, whether or not patients' records, and patient—clinician relationships.
those visits were with the same clinician or in the same The principles of data protection—transparency, legit-
facility, because clinicians can make better care deci- imacy, and proportionality—embodied in the EU Data
sions in light of fuller understanding of their patients' Protection Directive, therefore, specify that the person
past clinical histories. Patients also benefit from public from whom data are obtained should be informed
health surveillance and research that depends on com- of what will be done with this information and to
bining health information from individual patients to whom it will be disclosed. This allows the individual
improve public health and develop better treatments. to consent or object and to withdraw or correct the
Patients may benefit from making identifiable infor- data. Also, according to the directive, the data should
mation concerning adverse drug events available to be kept only as long as necessary for the specified
pharmaceutical companies so that those companies purpose,43 even though that could compromise later
can follow up with patients and improve drug safety, as retrospective research.
Source Informatics argued in the U.K. court case dis- Patients' privacy concerns are exacerbated when
cussed subsequently, and as the International Pharma- patients, and even clinicians, have little idea of what
ceutical Privacy Consortium argues more generally.34,35 becomes of their data, or just what is protected and
Data aggregator ims Health Canada (outs Health, Inc., what is not.44 Withholding information from one's
clinician is neither in the public interest nor beneficial aggregators perform a valuable service in collecting,
to that patient's individual interest in proper healthcare. cleaning, and combining these and other data into
Yet, removing identifying information from patient re- useful resources, though the value does not accrue dir-
cords may not alleviate concerns, especially in light of ectly to those who are the original source of the data
increasing public awareness of privacy violations sur- Data aggregators should be compensated for the value
rounding big data and the ease with which data sets added, but the sources deserve some benefit as well.
that were meant to be kept apart now are combined Currently, they primarily bear costs, both financially
and used for reidentification.45,46,47,48,49 Further, with- and in privacy.
out transparency, consent is meaningless. The combination of required disclosure of per-
sonal data and the ease with which data can be col-
Two Court Cases lected and disseminated is not unique to pharmacies.
It is a cost of healthcare to collect and store patient re-
Two court cases provide occasion for thinking about the cords, a cost passed on to patients and payers, whether
ethical implications of data sale and secondary use in private or governmental. The organizations provid-
light of international principles of health data privacy and ing these data obtain it from those legally required
protection. Each case involves selling prescription data to provide it—from individuals who pay directly, or
for pharmaceutical marketing. In both the United States indirectly through their private or public insurers, for
and the United Kingdom, data aggregators successfully its collection and maintenance. These individuals gain
challenged restrictions on such data use and sale. little direct benefit from the aggregation and sale of data
The 2011 U.S. Supreme Court case Sorrell v IMS about them, and they may be harmed by it. It mostly
Health Inc. et al.50 was decided on free speech grounds. occurs without their knowledge or permission. Even in
Although the legalities involve unique features of U.S. light of arguments that patients should be required as
constitutional law, a similar case in the U.K. in 2000, a condition of treatment to allow data about them to
R v Department of Health, Ex Parte Source Informatics be used for research—a requirement counter to profes-
Ltd.," points to the international nature of the ethical sional norms to provide care—it seems improper to re-
issues. That case was decided on the grounds that sell- quire either patients or clinicians to disclose data they
ing anonymized (de-identified) data did not violate would otherwise choose to keep private so that others
pharmacists' duty of confidentiality. may financially profit from them, whether or not the
The decision in each case runs counter to public data are de-identified.
expectations of health data confidentiality. The public Secondary use and big data analytics also are af-
is hardly aware that aggregating and selling prescription fected by the costs of collecting, storing, and organ-
and other health data are an international enterprise. izing data, as well as by the costs of meeting regulatory
Thus, the Sorrell and Source cases raise more general requirements. To reduce costs, health data processing
global concerns of privacy and data protection, on the is outsourced from countries with stronger privacy
one hand, and appropriate use and secondary use of protections to countries with weaker ones, despite its
data for data mining, marketing, research, public health, sensitive nature and consequent privacy risks." Also
and healthcare, on the other. Elsewhere I address data to reduce costs, U.S. marketing organizations oppose
de-identification, prescription and other health data opt-in consenting on the grounds that it would increase
aggregation and sale, and issues more specific to these the cost of doing business."
two cases.52 This article explores other ethical issues But costs must be paid somehow. Both the Source
related to the cases—the benefits and harms of data and Sorrell cases were fought to protect the commercial
sale; the trade-offs among privacy, individual health, value of health information. One way of recovering
and public health; and the need for transparency—so costs is by selling these data Though some sources
ethical dimensions of responsible and ethical health provide some data sets at little or no cost to researchers,
data collection and use can be assessed. cost could make it easier for pharmaceutical companies
and other commercial enterprises than for researchers
to access data.",56 Some fear that the trend toward
Who Benefits?
treating data as private property could make it more
Clinical data include data that patients are required to difficult to develop comprehensive databases crucial
provide to receive care. In both the Sorrell and Source for public health and research.37
cases, prescription data was aggregated and sold. Pa- Research, trade, and innovation, as well as the
tients, prescribers, and pharmacies are required by globalized healthcare industry, provide considerable
law to collect information related to prescribing. Data public benefit. There are ethical as well as economic
costs to privileging privacy, but economic value may ready for compliance. Research organizations were
not be more important than privacy or other considera- among those concerned about its impact. It is espe-
tions. Law and common ethical practice prevent releas- cially relevant here that the regulation defined personal
ing medical information without a patient's permission, data as any information about an individual, whether
but U.S. law does not prevent selling or transferring it relates to his or her private, professional, or public
rights to records.' Data that can be sold, can be sold life; and thus such data includes medical information.
and replicated anywhere and, once sold, may be used Much of these personal data—a name, a photo, an
for good or ill. Tracing the chain of data sales and use email address, bank details, posts on social networking
is difficult, making transparency and consent nearly websites, or a computer's IP address61—too, are part
impossible the further data are transferred from the of medical records. The original proposed regulation,
original source. therefore, increased health data protection and would
have made illegal the NHS mass database of citizens'
Health Data Uses: Big Data, Data Mining, health information, which could provide a valuable
Research, and Biosurveillance resource for improving care.62,63,64 Opposition from
the NHS and other research organizations contributed
Electronic health records and health information net- to changes put forward by the EU justice ministers in
works provide a wealth of data for public health, health March 2015 to improve data sharing across healthcare
outcome improvements, and research. Data could be services. They also tabled amendments regarding how
used for a range of beneficial purposes, from outcomes to manage such special forms of data as health and gen-
and comparative effectiveness research to designing etic data, and when patient consent is needed.65 The
clinical trials and monitoring drug safety. The benefits European Parliament, the Council, and the Commis-
of these data for public health, marketing, research, sion agreed on the new regulation late in 2015 and it is
drug development, identifying adverse effects, and expected to be adopted by the European Parliament at
biosurveillance; for reducing costs and overprescrib- the beginning of 2016.66
ing; and for regulating devices and software all are This NHS database also provokes privacy concerns
intertwined with privacy concerns. For some of these while providing financial benefit, as the NHS sells the
purposes, it is crucial to be able to identify individuals data.67 Individuals can opt out of the new care.gov
and link together an individual's records, so a require- database, which was to contain, for the first time, re-
ment for de-identification may further impair research. cords from primary care (GP) practices. Privacy con-
However, there also could be harms. Patients may cerns delayed including those GP records." Although
withhold sensitive information if they fear it will be used other rules allow greater third-party access to other
against them, even though it may be useful for other pur- NHS databases,69 insurers, pharmaceutical companies,
poses. Studies based on analyzing large data sets could be and other private commercial enterprises will receive
compromised if individual prescribers or patients with- "pseudoanonymized" records that the NHS claims "will
hold information or their consent for data use.59 not contain information that identifies you," but that
Privacy advocates, researchers, and public health instead include NHS numbers, birth dates, postcodes,
officials can be at odds over how to achieve benefits and ethnicity and gender information.70 The database
while protecting privacy, their disagreements may stem was created, according to the NHS England website, to
from different values and historical legacies. For ex- improve NHS services,n and to "drive economic growth
ample, the U.K.'s National Health Service (NHS), Royal by making England the default location for world-class
College of Physicians, and the Wellcome Trust led a health services research."72
coalition of leading medical research organizations In the United States, too, researchers and bioethi-
opposed to the proposed European General Data Pro- cists recognize that privacy protections can impede
tection Regulation, which, unlike the Data Protection research and healthcare quality improvement, with
Directive it would replace, would bind all 28 member calls from such influential agencies as the Institute of
countries. The proposal was acceptable to most EU na- Medicine to change the HIPAA Privacy Rule to allow
tions; the European Parliament approved the commit- for information-based research—that is, research using
tee report in full in 2014.60 medical records or stored biological samples.73
The regulation affects any organization that gath- Some innovative approaches to meeting privacy re-
ers, processes, and stores data, whether operating search, and commercial needs for data sharing include
within the EU, doing business with organizations the new international Open Humans Network, which
within the EU, or storing data in EU-member coun- "attempts to break down health data silos through an
tries. As of this writing, most organizations were not online portal that will connect participants willing to
■ II I Al II dill 111141
share data about themselves publicly with researchers clinicians would not anticipate. Data sold by both U.S.
who are interested in using that public data and con- state and federal agencies can be linked to individuals
tributing their analyses and insight to it,"74 and busi- by using publicly available information, even if some of
nesses based on similar ideas, such as PatientsLikeMe. the data are de-identified:92,93
Using the data people post, PatientsLikeMe produces Some may consider what is done with these data
publishable material on patient outcomes and compara- as harmful to some of the individuals who have pro-
tive effectiveness, which is valuable for effectiveness vided the data and, at the same time, as beneficial to
research. Epidemiologic trends also can be identified other individuals, depending on what the data reveal.
through social media postings.75'76'77 Those engaging in This combination of benefits and harms is evident
this social networking presumably feel it is beneficial to in a variety of examples in which one's records affect
them. Even so, it would be better if they were aware of one's services and costs. In the United States, where
what is done with their data, instead of being surprised private medical insurance is the norm, private insurers
if they have not read subscription agreements carefully use prescription and other claims data to deny insur-
enough to know that PatientsLikeMe sells data to phar- ance, charge differential premiums, or exclude some
maceutical and other companies and that sites such as conditions." Businesses often check the MIB (Medical
Face book are not private places.'" Information Bureau) for job applicants' underwriting
data." Aggregators purchase and combine data from
Who Sells and Uses Data? One Man's the states as well as from pharmacies." Credit agen-
Bread Is Another Man's Poison cies are the most frequent buyers of multistate health
profiles, though ims Health also purchases data from
As is evident from the multiplicity of uses, health data the states.97 Government fusion centers, designed to
are valuable. Internationally, the idea of "liberating" promote data sharing among federal agencies and state
data for secondary use is recognized as beneficial for and local governments, combine data from multiple
individual and public benefit, research, entrepreneur- sources—including health record information—for
ship, and policy. Though transborder data flow is regu- law enforcement, immigration control, and homeland
lated by international agreements, such as the EU Data security."'"
Protection Directive, presumably health data could be Organizations, too, may benefit financially while
sold worldwide, to anyone, for any purpose. Balancing providing social benefit through data sales. The Amer-
this with privacy concerns is fraught.8° Strong privacy ican Medical Association and the U.S. Centers for
protection, such as the rights-centric approach of the Medicare and Medicaid sell provider data, whereas
European Court of Human Rights, could adversely state Health Information Exchanges (Ins) sell second-
affect the globalized healthcare industry and innova- ary data. loo,1o1,1oz The U.K.'s National Health Service,
tion and trade.8182'83.84 too, sells data.103 Insurance companies or health infor-
Entire patient records are among the many possible mation technology vendors might aggregate and sell
sources of data for which there is a lucrative market, provider-identified data on performance and quality
for laudable as well as unsavory purposes. Incidents of measures, the number of procedures performed, U.S.
medical identity theft increased by more than 20 per- meaningful use criteria, data security breaches, and
cent in 2014 compared to 2013.85 In the active black other useful compilations. Cash-strapped community
market in identifiable medical record information, health organizations, state Regional Extension Centers
health information is more valuable than U.S. Social (REcs), county hospitals, the U.S. Veterans Adminis-
Security numbers for identity the ft.86'87 Though prices tration, the Indian Health Service, the Joint Commis-
vary, such information sells for about ten times more sion, or hospital associations also could sell data for
than credit card numbers (which typically sell for no similarly beneficent purposes. Hospitals routinely sell
more than a few U.S. dollars) because it can be monet- birth records.'"
ized by getting treatment paid for via identity theft or to Genetic data are also double-edged. Such data
extort money from hacked corporations.88'89 are needed for research, personalized medicine, and
Electronic records also make it possible for com- biobanking but also can make individuals and com-
puter or software vendors, intermediaries, or newly munities vulnerable. For example, in 2000, Iceland's
created organizations to bundle and sell rights and parliament sold exclusive rights to all the genetic and
data," a practice useful for research, policy, marketing, genealogical data from each of its 275,000 citizens to
and business. In the United States, there is an exhaust- the U.S. company deCODE Genetics. Soon thereafter,
ive list of organizations that can use and legally sell deCODE signed a $200 million contract with Hoffman
health information,91 some for purposes patients and LaRoche to search for several common human genetic
diseases. Iceland had an opt-out policy, and the data ventures or medical tourism facilities in another. Some
were encrypted to de-identify individuals. Neverthe- may welcome learning of such opportunities, whereas
less, the Icelandic Supreme Court later ruled that creat- others may feel harassed or violated. Individuals in one
ing the database was unconstitutional because it did country may experience salutary or salacious effects
not adequately protect personal privacy105 from having (identified or possibly re-identified) data
Clearly, provider or patient information is valu- available elsewhere. But without transparency, there is
able. Hospitals could purchase data about competitors, little chance of gaining individual consent or, on both
providers could identify populations for treatment, individual and societal levels, assessing harm or benefit.
researchers could conduct studies involving healthcare
and public health practices, and government agencies Ownership, Commodification,
could identify and influence health trends. If such sales and De-contextualization
were restricted, some fear, the data would not be col-
lected or maintained at all, which could compromise The right to sell data is muddied by lack of clarity over
research and new drug development.106,107 The Iceland the legalities of data ownership. Law in and outside the
genetic database sale, for example, led to identification United States does not address health data ownership
of genes linked to disease,1°8,1°9 though capitalizing on clearly, it is not clear who the owner should be, or
these kinds of discoveries was limited to the company whether ownership is better than current or alterna-
with exclusive rights to this gene discovery. DeCODE's tive approaches.115,116 It also is not clear where those
2009 bankruptcy and the consequent database owner- who sell data analytics services obtain the data, or how
ship change from a scientific research company to Saga they might use them.117 Well-known electronic health
Investments LLC, and the subsequent sale of the data- record vendors have sold de-identified copies of their
base in 2012 to biotech pioneer Amgen, again raised patient databases to pharmaceutical companies, med-
questions about data privacy and use.110,111 ical device makers, and health services researchers.118
Countries as different as Canada, Estonia, Sweden, Vendor contracts are unusual in that some vendors lay
Singapore, and the Kingdom of Tonga have developed claim to patient record data, whereas businesses and
various models for protecting privacy and differing financial institutions typically do not give up their data
policies regarding commercial involvement and rights to their software vendors.119 Regardless of whether
to samples for gene banks, all with the goal of improv- the data themselves or the means of access to them
ing the public health of the studied population, and, are owned by electronic health records vendors, some
in some cases, to generate revenue for national health- academic medical centers pay to get data from their
care budgets. Though all these policies are intended to own patients' records. Vendors often consider their
maintain confidentiality, all of the data uses require per- contracts intellectual property and do not reveal these
sonal identifiers so as to link individuals' records from and other contract provisions, a practice the American
genetic, medical, genealogical, and lifestyle databases. Medical Informatics Association considers unethical.120
International controversy over such databases, there- If health data are property, presumably, whoever
fore, centers around confidentiality, consent, to what owns the data can sell them. Some advocate clearly
extent commercial interests should influence policy, and defined property rights in medical information, giving
whether commercial ownership facilitates or impedes patients the right to monetize their access and control
research,112,113,114 all of which are concerns related to rights, as a way for individuals to control and benefit
collecting and selling healthcare data in general. from what happens to data about them.121 Others argue
As a way of raising additional considerations, I pose against property rights in patient data and advocate
possibilities that might occur were there unrestricted instead for public ownership akin to a data commons
selling of health data. Abortion opponents presum- so that data from multiple sources can be de-identified
ably could buy aggregated prescription information and combined population-wide for public benefit.'22
for medications that cause abortions, or animal rights Commodifying medical information strikes still others
activists could buy information about researchers' as anathema to professional values and the special
animal or veterinary medicine purchases. Depending relationship between doctor and patient. Privacy is
on who buys it and their purpose, such information valued because it facilitates ideals of personhood in-
could threaten or protect researchers', clinicians', and volving autonomy, individuality, respect, dignity, and
patients' safety and might have adverse effects on worth as a human being.123 Therefore, the idea of sell-
research and clinical practice or might open new av- ing personal health data also disturbs those who think
enues. Physicians, patients, hospitals, and so on, in one the practice commodifies the self and sullies ideas of
country may be targeted for marketing by commercial personhood.124,125 Compromising of personhood is
uw I ai I1..11 IIILIJ
compounded because data in databases necessarily are is in the public interest for personal health data to be
decontextualized. De-identification is an attempt to made available, just what that "public interest" is,131,132
remove any connection with the person, but even iden- and, for that matter, just what "privacy" comprises and
tifiable health record data typically do not include all entails as norms evolve.133 The issues include consid-
information a person may consider central to the self. ering, in a healthcare context, the dualities playing out
with respect to big data in domains other than health-
Conclusions care: the individual versus the aggregate, research
versus privacy, individual versus institutional power,
Widespread use o f electronic patient record systems en- identification versus identity, identification versus
ables opportunities to improve healthcare through data authentication, and virtual people versus real people
sharing, secondary use, and big data analytics. Multiple and contextualized information. They involve big data
healthcare professionals, payers, researchers, and com- harms and benefits related to innovation and economic
mercial enterprises can access data and reduce costs advancement, power shifts, access to knowledge, and
by eliminating duplication of services and conducting freedom of communication.
research on effective care. However, widespread use of Societies and governments need to grapple with
electronic patient records systems also creates more op- these ethical issues, tensions between privacy and
portunities for privacy violations, data breaches, and other considerations, and shifting norms. The numer-
inappropriate uses. ous cross-cutting issues suggest that other areas of law,
Ethical and policy analysis related to health data ethics, and social policy also can inform related ethical
and informatics should consider benefits and harms, and legal considerations. For some time, the legal, bio-
taking into account both the uses and users of the in- ethics, and informatics communities have been con-
formation.126,127 Embarrassing an estranged spouse by sidering issues such as appropriate secondary use of
publishing his or her mental health records is more dis- data; patient and clinician relationships in light of the
tasteful than using those records combined with others' growth of electronic health records and health infor-
to study and improve mental health. As this example mation technologies;134,135,136 reliance on increasingly
suggests, some users (the researcher) are more appeal- untenable de-identification; burgeoning electronic data
ing than others (the spouse). Moreover, an uncontrover- collection, sharing, transmission, and aggregation; data
sial use may be morally offensive if the user is unsavory use for public health, research, and innovation; and the
or controversial.128 How should distinctions be made so privacy and security of health data.
that some data uses and users are permissible and some As health information exchanges and health tour-
not? On what grounds? And who is best placed to make ism develops; as lifetime electronic health records that
such decisions: the courts or legislators, clinicians and follow patients across governmental and institutional
researchers who are most familiar with their data needs, boundaries are used more widely; as databases grow
companies that develop and market new medications, and biobanks become digital; as biometric identification
or patients and prescribers, who are most affected by becomes more common; as radio frequency identifica-
privacy violations and can best weigh the relative im- tion devices (RFips) are embedded in medical devices,
portance of various values to themselves.'29 smart-pills, and patients; as home sensors and monitors
Those most familiar with, closest to, and affected are increasingly used; as mobile, wearable, and e-health
by the potential use should have a strong say. They need applications expand; and as health information exchan-
to know about those uses, though, to express their pref- ges develop,137,138,139,140 informaticians can add to the
erences in an informed, thoughtful way. Many patients conversation among governments, courts, regulatory
do not know what is, or can be, done with data about agencies, professional societies, and other organizations
them, but keeping them ignorant is not the way to ad- to consider responses to issues involving health-related
dress concerns. Lack of accountability and transpar- data. Combining legal and ethics scholarship with in-
ency about health data uses feeds the public's privacy formaticians' expertise concerning judicious and ethical
concems,"° undermines the possibility of informed data collection and use, together with their technical
consent, and impairs research, care, and public health. knowledge of data aggregation and identification, can
Ethical considerations over data use will, and contribute to more informed policies.
should, evolve as the public becomes more aware of The Source and Sorrell court cases can provoke an
the value and pitfalls of data sharing, data aggregation, initial reaction of outrage over privacy violations and
and data mining. Cases like Source and Sorrell encour- data use without consent. Consequently, they call into
age debate over propriety and values related to different question just what constitutes "privacy" and "public
kinds of data use. They also lead to examining when it interest" and stimulate considerations as to how to
balance them. They provide an opportunity to weigh especially i f the public is unaware of the reuse. Doing so
privacy against numerous beneficial uses for data. without individuals' permission violates international
Transparency and accountability are needed so that principles of data privacy.141,142,143,144,145 The court cases
harms and benefits can be judged through public dis- prompt ethical questions about commodifying medical
cussion and so that individual as well as societal deci- information and harmonizing policy across jurisdic-
sions can be made on more informed and thoughtful tional boundaries. Their consequences may affect bio-
grounds. Using data collected for one purpose (such medical informatics, patient and clinician privacy, and
as prescriptions) for another purpose (such as pharma- regulation in ways this article explores, in the United
ceutical marketing) can undermine public confidence, States, United Kingdom, and elsewhere.
Notes
1. Laura Wexler's comments as a respondent at "The Critical available at http://www.gpo.gov/fdsys/pkg/FR-2013-01-
Life of Information," a conference at Yale University, April 11, 25/pdf/2013-01073.pdf (last accessed 2 July 2014).
2014, outlined dualities related to big data; see http://wgss 12. European Union. EU Directive 95/46/EC- The Data
.yale.edu/sites/default/files/files/Critical%20Life%20of%20 Protection Directive; available at http://www.dataprotection
Information%20Program%20spreads.pdf (last accessed 19 .ie/docs/EUDirective-95-46-EC--Chapter-2/93.htm (last
Aug 2014) for conference information. accessed 23 Mar 2014).
2. Jost TS. Readings in Comparative Health Law and Bioethics. 2nd 13. European Commission, Directorate General for Justice
ed. Durham, NC: Carolina Academic Press; 2007. and Consumers. Agreement on Commission's EU Data
3. Institute of Medicine (tom). Beyond the HIPAA Privacy Protection Reform Will Boost Digital Single Market 2015 Dec
Rule: Enhancing Privacy, Improving Health through Research. 15; available at http://europa.eu/rapid/pressrelease_IP-
Washington, DC The National Academies Press; 2009, at 78. 15-6321_en.htm (last accessed 5 Jan 2016). See also
4. See note 3, tom 2009, at 79. European Commission, Directorate General for Justice and
5. Jones P Permission-based marketingunder Canada's new Consumers. Reform of EU Data Protection Rules; available
privacy laws. Franchise LawJournal 2004; 24 (2): 267-303. at http://ec.europa.eu/justice/ data-protection/reform/
6. Walden I. Anonymising personal data. InternationalJournal of index_en.htm (last accessed 5 Jan 2016).
Law and Information Technology 2002; 10 (2): 224-37. 14. Rossi B. Countdown to the EU General Data Protection
7. Srinivas N, Biswas A. Protecting patient information in India: Regulation: 5 steps to prepare. Information Age 2015
Data privacy law and its challenges. NUJS Law Review 2012; 5 Mar 24; available at http://www.information-age.
(3): 411-24. com/itmanagement/risk-and-compliance/123459219/
8. Kaplan B. Selling health data: De-identification, privacy, countdown-eu-general-dataprotection-regulation-5-steps-
and speech. Cambridge Quarterly of Healthcare Ethics 2015; prepare (last accessed 13 May 2015).
24 (3): 256-71. 15. Solove DJ. A taxonomy of privacy. University of
9. United States Government, Department of Health and Human Pennsylvania Law Review 2006; 154 (3): 477-560.
Services, Office for Civil Rights. Summary of the HIPAA Privacy 16. Ohm P Broken promises of privacy. Responding to the
Rule; available at http://www.hhs.gov/ccr/privacy/hipaa/ surprising failure of anonymization. UCLA Law Review
understanding/summary/ (last accessed 30 June 2013). 2010; 57: 1701-77, at 270.
10. United States Government, Department of Health and 17. Taylor MI Health research, data protection, and the public
Human Services, Office for Civil Rights. Standards for interest in notification. Medical Law Review 2011; 19 (2):
Privacy of Individually Identifiable Health Information; 267-303.
available at http://aspe.hhs.gov/admnsimp/ final/ 18. See note 17, Taylor 2011, at 303.
pvcguidel .htm (last accessed 19 Jan 2014). 19. Kaplan B. Patient health data privacy. In: Yanisky-Ravid
11. United States Government, Department of Health and S, ed. The Challenges of the Digital Era: Privacy, Information
Human Services, HSS Press Office, New rule protects and More. New York: Fordham University Press;
patient privacy, secures health information 2013 forthcoming.
Jan 17; available a t http://wwwhhs.gov/about/news/ 20. See note 8, Kaplan 2015.
2013/01/17/new-rule-protects-patientprivacy-secures- 21. See note 16, Ohm 2010.
health-information.html (last accessed 1 Jan 2016). See 22. See note 19, Kaplan forthcoming.
also United States Government, Department of Health 23. Beyleveld D, Histed E Betrayal of confidence in the Court
and Human Services, Office of the Secretary 45 CFR of Appeal. Medical Law International 2000; 4: 277-311.
Parts 160 and 164: Modifications to the HIPAA Privacy, 24. Koontz L. What is privacy? In: Koontz L, ed. Information
Security, Enforcement, and Breach Notification Rules Privacy in the Evolving Healthcare Environment. Chicago:
under the Health Information Technology for Economic Healthcare Information and Management Society (timitsS);
and Clinical Health Act and the Genetic Information 2013: 1-20.
Nondiscrimination Act; other modifications to the HIPAA 25. See note 19, Kaplan forthcoming.
Rules; final rule. Federal Register 2013 Jan 25: 5565-702; 26. See note 8, Kaplan 2015.
mu I H 11 MILLI
27. World Medical Association. International Code of 47. Benitez K, Malin B. Evaluating re-identification risks with
Medical Ethics; available at http://wwwwma.net/ respect to the HIPAA Privacy Rule. JAMIA (Journal of the
en/30publications/lOpolicies/c8/index.html (last accessed American Medical Informatics Association) 2010; 17 (2):
2 May 2014). 169-77.
28. World Medical Association. Declaration of Helsinki- 48. See note 16, Ohm 2010.
Ethical Principles f or Medical Research Involving 49. See note 8, Kaplan 2015.
Human Subjects; available at http://www.wma.net/ 50. Sorrell v IMS Health, Inc., et al., 131 S. Ct. 2653 (2011).
en/30publications/lOpolicies/b3 Oast accessed 2 May 2014). 51. R v Department of Health, Ex Parte Source Informatics Ltd.
29. World Medical Association. Declaration on Ethical [C.A. 2000] 1 All ER 786. See also R v Department of
Considerations Regarding Health Databases; available at Health, Ex Parte Source Informatics Ltd. European Law Report
http://www.wma.net/en/30publications/lOpolicies/d1 (last 2000; 4: 397-414.
accessed 2 May 2014). 52. See note 8, Kaplan 2015.
30. See note 29, WMA 2014. 53. See note 7, Srinivas, Biswas 2012.
31. See note 2, Jost 2007. 54. See note 5, Jones 2004.
32. Malin BA, El Emam K, O'Keefe CM. Biomedical data 55. Baxter AD. ims Health v. Ayotte: A new direction on
privacy: Problems, perspectives, and recent advances. commercial speech cases. Berkeley Technology Law Journal
JAMIA (Journal of the American Medical Informatics 2010; 25: 649-70.
Association) 2013; 20 (1): 2-6. 56. Pasquale F Restoring transparency to automated authority.
33. See note 23, Beyleveld, Histed 2000, at 296. Journal on Telecommunications and High Technology Law
34. Dunkel YE Medical privacy rights in anonymous data: 2011; 9: 235-54.
Discussion of rights in the United Kingdom and the 57. Rodwin MA. Patient data: Property, privacy, and the public
United States in light of the Source Informatics cases. Loyola interest. American Journal of Law and Medicine 2010; 36:
of Los Angeles International and Comparative Law Review 586-618, at 589.
2001; 23 (1): 41-80. 58. Hall MA, Schulman KA Ownership of medical
35. See note 7, Srinivas, Biswas 2012. information. JAMA 2009; 301 (12): 1282-4.
36. See note 5, Jones 2004. 59. Gooch GR, Rohack JJ, Finley M. The moral from Sorrell:
37. Powell J, Fitton R, Fitton C. Sharing electronic health Educate, don't legislate. Health Matrix 2013; 23 (1):
records: The patient view: Informatics in Primary Care 237-77.
2006; 14 (1): 55-7. 60. N-is European Office. Data Protection; 2015 Mar 24;
38. Schers H, van den Hoogen H, Grol R, van den Bosch W available at http://www.nhsconfed.org/regions-and-eu/nhs-
Continuity of information in general practice: Patient european-office/influencing-eu-policy/data-protection (last
views on confidentiality. Scandinavian Journal of Primary accessed 15 May 2015).
Health Care 2003; 21 (1):21-6. 61. See note 14, Rossi 2015.
39. See note 23, Beyleveld, Histed 2000. 62. O'Donoghue C. EU research group condemns EU regulation
40. See note 32, Malin et al. 2013. for restricting growth in life sciences sector; 2014; available
41. See note 34, Dunkel 2001, at 70. at http://www.globalregulatoryenforcementlawblog.com/
42. Choy C, Hudson Z, Pritts J, Goldman J. Exposed Online: 2014/02/articles/data-security/eu-research-group-
Why the New Federal Health Privacy Regulation Doesn't condemns-eu-regulationfor-restricting-growth-in-life-
Offer Much Protection to Internet Users. Health Privacy sciences-sector (last accessed 23 Mar 2014).
Project, Institute for Healthcare Research and Policy, 63. Farrar]. Sharing NHS data saves lives; EU obstruction will not.
Georgetown University: Pew Internet and American The Telegraph 2014 Jan 14; available at http://wwwtelegraph
Life Project; 2001, at 4; available at http://www. .co.uk/ health/nhs/10569467/Sharing-NHS-datasaves-
pewinternet.org/files/oldmedia/Files/Reports/2001/ lives-EU-obstruction-will-nochtml Oast accessed 23 Mar
PIP_HPP_ HealthPriv_report.pdf.pdf (last accessed 11 2014).
May 2015). 64. European Public Health Alliance. [Update] General Data
43. See note 12, EU 2014. Protection Regulation; available at http://www.epha.org/
44. McGraw D. Building public trust in uses of Health 5926 (last accessed 23 Mar 2014).
Insurance Portability and Accountability Act de-identified 65. NHS Confederation. EU ministers table changes to data
data. JAMIA (Journal of the American Medical Informatics privacy; 2015 Mar 13; available at http://nhsconfed.org/
Association) 2013; 20 (1): 29-34. news/2015/03/eu-ministers-table-changes-to-dataprivacy-
45. Curfman GD, Morrissey S, Drazen JM. Prescriptions, laws (last accessed 14 May 2015).
privacy, and the First Amendment. New England Journal of 66. See note 13, European Commission 2015.
Medicine 2011; 364 (21): 2053-5. 67. Doctorow C. UK set to sell sensitive NHS records to
46. Tien L. Online behavioral tracking and the identification commercial companies with no meaningful privacy
of Internet users. Paper presented at: From Mad Men to protections-UPDATED; 2014 Feb 4; available at http://
Mad Bots: Advertising in the Digital Age [conference]. The boingboing.net/2014/02/04/uk-set-to-sell-sensitivenhs-r
Information Society Project at the Yale Law School. New .html (last accessed 5 Feb 2014).
Haven, CT; 2011.
68. Donnelly L Hospital records of all NHS patients sold to 86. Avila J, Marshall S. Your medical records may not be private
insurers. The Telegraph 2014 Feb 23; available at http:// ABC News Investigation. ABC News 2012 Sept 13; available
www.telegraph.co.ukihealth/healthnews/10656893/ at http://abcnews.go.com/Health/medical-recordsprivate-
Hospital-records-of-all-NHS-patients-soldto-insurers.html abc-news-investigation/story?id=17228986&page=2 (last
(last accessed 24 July 2014). accessed 22 Mar 2014).
69. See note 68, Donnelly 2014. 87. Nguyen V, Nious K, Carroll J. Your medical records
70. NHS Choices. Your records: Better information means could be sold on black market: NBC Investigative Unit
better care; available at http://www.nhs.uk/nhsengland/ surprises strangers with private medical details. NBC Bay
thenhs/records/healthrecords/pages/care-data.aspx (last Area 2013 June 18; available at http://www.nbcbayarea
accessed 24 July 2014). .com/news/local/Medical-Records-Could-Be-Sold-on-
71. See note 70, NHS Choices 2014. .Market-212040241.html (last accessed 22 Mar
72. Ramesh R. NHS patient data to be made available for sale B210a1c4k)
to drug and insurance firms. The Guardian 2014 Jan 19; 88. Lawrence D. End of Windows XP support means added
available at http://www.theguardian.com/society/2014/ opportunity for hackers. Businessweek 2014 Apr 4; available
jan/19/nhs-patient-data-availablecompanies-buy (last at httpi/ www.businessweek.com/articles/2014-04-04/
accessed 24 July 2014). end-of-windows-xp-support-meansadded-opportunity-for-
73. Institute of Medicine. Beyond the HIPAA Privacy Rule: hackers (last accesced 1 July 2014).
Enhancing Privacy, Improving Health through Research. 89. Shahani A. The black market for stolen health care data.
Washington, DC: National Academies; 2009; available NPR; 2015 Feb 13; available at http://wwwnpr.org/
at http://www.iom.edu/-/media/Files/Report%20 sections/alltechconsidered/2015/02/13/385901377/
Files/2009/Beyond-the-HIPAAPrivacy-Rule-Enhancing- theblack-market-for-stolen-health-care-data (last accessed
Privacy-Improving-Health-Through-Research/HIPAA%20 14 May 2015).
report%20brief%20FINAL.pdf (last accessed 22 Jan 2014). 90. See note 58, Hall, Schulman 2009.
74. Open Humans Network. Open Humans Network wins 91. See note 34, Dunkel 2001.
Knight News Challenge: Health Award; available at http:// 92. See note 47, Benitez, Malin 2010.
openhumans.org (last accessed 1 July 2014). 93. Roberston J. States' hospital data for sale puts privacy in
75. Christakis NA, Fowler J H. Social network visualization jeopardy. Health Leaders Media; 2013; available at httpi/
in epidemiology. Norwegian Journal of Epidemiology 2009; www.healthleadersmedia.com/content/QUA-292963/States-
19 (1): 5-16. hospital-data-for-sale-putsprivacy-in-jeopardy (last accessed
76. Christakis NA, Fowler JH. Social network sensors for early 14 June 2013).
detection of contagious outbreaks. PLoS ONE 2010; 5 (9): 94. Brief for the New England Journal of Medicine, the
e12948. Massachusetts Medical Society, the National Physicians
77. Velasco E, Agheneza T, Denecke K, Kirchner G, Eckmanns Alliance, and the American Medical Students Association as
T. Social media and Internet-based data in global systems Amici Curiae Supporting Petitioners, William H. Sorrell v IMS
for public health surveillance: A systematic review. The Health, Inc. et al., 2010 U.S. Briefs 779 (No. 10-779), 2011
Milbank Quarterly 2014; 93 (1):7-33. U.S.S. Ct. Briefs LEXIS 299.
78. Andrews L. I Know Who You Are and I Saw What You Did: 95. Holtzman DH. Privacy Lost: How Technology Is
Social Networks and the Death of Data Privacy. New York Endangering Your Privacy. San Francisco: Jossey-Bass;
Free Press; 2011, at 1-3. 2006, at 195.
79. Angwin J. Dragnet Nation: A Quest for Privacy, Security, and 96. See, for example 12PC Health Data Store CMS MedPAR
Freedom in a World of Relentless Surveillance. New York: Hospital Data File; available at httpi/www.healthdatastore.com/
Times Books, Henry Holt; 2014, at 33-4. . par-hospital-data-filerspx (last accessed 13 Sept
80. Geissbuhler A, Safran C, Buchan I, Bellazzi R, Labkoff 2c111
01s3m)ed
S, Eilenberg K, et al. Trustworthy reuse of health data: A 97. [Winston JS]. States' hospital data for sale puts patient
transnational perspective. International Journal of Medical privacy in jeopardy; 2013 June 7; available at haps://
Informatics 2013; 83 (1): 1-9. www.annualmedicalreport.com/states-hospital-data-for-
81. See note 7, Srinivas, Biswas 2012. s2ale
o1rtspatient-privacy-in-jeopardy (last accessed 19 Jan
82. See note 17, Taylor 2011.
83. Bambauer JR. Is data speech? Stanford Law Review 2014; 98. Bady A. World without walls-privacy laws should be
66: 57-120. recrafted for the data fusion age Technology Review 2011;
84. Zarsky TZ. The privacy/innovation conundrum. Lewis & 114 (6):66-71.
Clark Law Review 2015; 19(1); available at http://ssrn.com/ 99. United States Government, Department of Justice. Fusion
abstract=2596822 (last accessed 19 May 2015). Center Guidelines: Developing and Sharing Information and
85. Dvorak K. Med identity theft continues t o rise; 2015 Feb Intelligence in a New Era; 2006; available at http://www.
23; available at http://www.fiercehealthit.com/story/med- it.ojp.gov/documents/fusion_center_guidelines.pdf (last
identitytheft-continues-rise/2015-02-23?utm_medium= accessed Mar 2012).
nl&utm_source=intemal (last accessed 14 May 2015). 100. See note 45, Curfman et al. 2011.
101. United States Government, Department of Health and http://thehealthcareblog.com/blog/2013/11/20/

Human Services, Centers for Medicare and Medicaid whose-data-is-it-anyway-2 (last accessed 3 Feb 2014).
Services. Agreement for Use of Centers for Medicare & Medicaid 120. Goodman KW, Berner E, Dente MA, Kaplan B, Koppel
Services (CMS) Data Containing Unique Identifier s, Form CMS- R, Rucker D, et al. Challenges in ethics, safety, best
R-0235, OMB No. 0938-0734; available at http://www.cms practices, and oversight regarding HIT vendors, their
.gov/Medicare/CMS-Forms/CMS-Forms/downloads//cms-r- customers, and patients: A report of an AMIA special task
0235.pdf (last accessed 13 Sept 2013). force. JAMIA (Journal of the American Medical Informatics
102. Hebda T, Czar P Handbook of Informatics for Nurses and Association) 2011; 18 (1): 77-81.
Healthcare Professionals. 4th ed. Upper Saddle River, NJ: 121. Hall MA. Property, privacy, and the pursuit of
Pearson/Prentice Hall; 2009, at 321. interconnected electronic health records. Iowa Law Review
103. See note 68, Donnelly 2014. 2010; 95: 631-63.
104. See note 95, Holtzman 2006, at 192. 122. See note 57, Rodwin 2010.
105. McGraw Hill General and Human Biology Case Studies. 123. See note 3, IOM 2009, at 77.
Gene Banks versus Privacy Invasion; available at http://www 124. See note 58, Hall, Schulman 2009.
mhhe com/biosci/genbio/casestudies/sellinggenes.mhtml 125. Atherley G. The public-private partnership between
(last accessed 2 May 2014). ims Health and the Canada Pension Plan. Fraser Forum
106. Brief for the Association of Clinical Research 2011: 5-7.
Organizations as Amici Curiae Supporting Respondents, 126. Miller RA, Schaffner KF, Meisel A. Ethical and legal
William H. Sorrell v IMS Health, Inc., et al., 2011 WL issues related to the use of computer programs in
2647130 (2011) (No. 10-779), (2011). clinical medicine. Annals of Internal Medicine 1985;
107. See note 59, Gooch et al. 2013. 102: 529-36.
108. See note 105, McGraw Hill 2014. 127. Goodman KW. Health information technology:
109. Austin MA, Harding 5, McElroy C. Genebanks: A Challenges in ethics, science and uncertainty. In: Himma
comparison of eight proposed international genetic K, Tavani H, eds. The Handbook of Information and
databases. Community Genetics 2003; 6 (1):37-45. Computer Ethics. Hoboken, NJ: Wiley, 2008: 293-309.
110. Gillham WW. Genes, Chromosomes, and Disease: From 128. See note 127, Goodman 2008.
Simple Traits, to Complex Traits, to Personalized Medicine. 129. Data mining case tests boundaries of medical privacy.
Upper Saddle River, NJ: Pearson Education, published as CMAJ 2011; 183 (9): E509-10.
FT Press Science; 2011, at 18-19. 130. See note 44, McGraw 2013.
111. Amgen. Amgen to acquire deCODE Genetics, a global 131. See note 17, Taylor 2011.
leader in human genetics; available at wwwamgen.com/ 132. See note 57, Rodwin 2010, at 617-18.
media/media_pr_detail.jsp?releaselD=1765710 (last 133. See note 15, Solove 2006.
accessed 2 May 2014). 134. Goodman KW. Ethics, information technology, and
112. See note 109, Austin et al. 2003. public health New challenges for the clinician-patient
113. Annas GI Rules for research on human genetic relationship. Journal of Law, Medicine and Ethics 2010
variation-lessons from Iceland. New England Journal of Spring: 58-63.
Medicine 2000; 342 (24):1830-3. 135. Kaplan B, Litewka S. Ethical challenges of telemedicine
114. Gulcher JR, Stefansson K. The Icelandic Healthcare and telehealth Cambridge Quarterly of Healthcare Ethics
Database and informed consent. New England Journal of 2008; 17(4):401-16.
Medicine 2000; 342 (24): 1827-9. 136. See note 19, Kaplan forthcoming.
115. See note 19, Kaplan forthcoming 137. See note 134, Goodman 2010.
116. Evans BJ. Much ado about data ownership. Harvard 138. See note 135, Kaplan, Litewka 2008.
Journal of Law & Technology 2011; 25 (1): 69-130. 139. See note 19, Kaplan forthcoming.
117. For example, GE Data Visualization uses information 140. Roland D. UK to get 200 high-tech factory jobs
"based on 7.2 million patient records from GE's proprietary making "swallowable sensors." The Telegraph 2014
database"; available at http//visualization.geblogs.com/ Mar 10; available at http://www.telegraph.co.uk/
visualization/network (last accessed 27 Sept 2013). finance/10687395/UK-toget-200-high-tech-factory-jobs-
GE Healthcare's healthcare IT solutions-available at makingswallowable-sensors.html (last accessed 17 July
http://www3.gehealthcare com/en/Products/Categories/ 2014).
Healthcare_IT?gclid=CIKQ4Z6P7LkCFcE7OgodTDI 141. See note 24, Koontz 2013.
APQ and http://www3.gehealthcare.com/en/Products/ 142. See note 44, McGraw 2013.
Categories/Healthcare_IT/Knowledge_Center (last 143. See note 23, Beyleveld, Histed 2000.
accessed 27 Sept 2013)-includes patient records and 144. See note 12, EU 2014.
patient portals 145. Rodrigues RJ, Wilson P Schanz SJ. The Regulation of
118. Sittig DE Singh H. Legal, ethical, and financial dilemmas Privacy and Data Protection in the Use of Electronic Health
in electronic health record adoption and use. Pediatrics Information: An International Perspective and Reference
2011 Apr; 127 (4): e1042-7. Source on Regulatory and Legal Issues Related to Person-
119. Moore J, Tholemeier R. Whose data is it anyway? Identifiable Health Databases. Washington, DC: World
The Health Care Blog, 2013 Nov 20; available at Health Organisation (wHo); 2001.
Murray: Genetic Exceptionalism and "Future Diaries" 147
3.4 Genetic Information
Genetic Exceptionalism and Group on the Ethical, Legal, and Social Implications
of the Human Genome Project. The task force used
"Future Diaries": Is Genetic the term genetic exceptionalism to mean roughly the
Information Different from Other claim that genetic information is sufficiently different
Medical Information? from other kinds of health-related information that it
deserves special protection or other exceptional meas-
Thomas H. Murray ures. After many attempts to make the case for genetic
exceptionalism, the task force abandoned the effort. At
A few years ago, at a meeting to discuss the larger im- least for the purpose of deciding who should receive
plications of the Human Genome Project, the head of health care coverage, we concluded that genetic infor-
the Federal Bureau of Investigation's laboratories leaned mation did not differ substantially from other kinds of
over to me and related this story. The FBI lab had con- health-related information.' For many reasons, we were
ducted analyses of samples connected to the bombing reluctant to reach this conclusion. Our expertise lay in
of the World Trade Center in Manhattan. They had an genetics, our mandate came from the Human Genome
envelope in which an incriminating document had been Project, and the problem of genetics and health insur-
sent. With a technology known as PCR, polymerase ance, as big as it was, was nevertheless tiny compared
chain reaction, they were able to amplify enough DNA to the prospect of trying to comprehend—much less
from the back of the stamp to link it with one of the change—the entire American health care system. I will
chief suspects through a genetic fingerprint. They also explain why we abandoned genetic exceptionalism
learned that someone else had licked the envelope itself. shortly. But first I want to explore why it is that so many
In other words, if someone cares enough to go to thoughtful and knowledgeable commentators think it
the trouble, it is possible to get substantial information is important to treat genetic information as special.
about a person's genetic makeup from the tiny bits of In an article on genetic privacy, Lawrence Gostin
genetic material we scatter around us without much predicts a move to longitudinal, electronic clinical re-
thought. The cells mixed in with our saliva and the cords for each individual that would "contain all data
bulbs at the base of the hairs we continuously shed are relevant to the individual's health collected over a life-
two widely distributed sources of raw material for cre- time . . . continually expanded from prebirth to death,
ating genetic information about each of us. and accessible to a wide range of individuals and insti-
Of course, at least some other medically relevant tutions."' George Annas, Leonard Glantz, and Patricia
information about us is just as readily available. We Roche, authors of a model Genetic Privacy Act, con-
may be pallid and cachectic, morbidly obese, rosy- clude a rousing defense of genetic exceptionalism with
cheeked, or missing limbs. We may be depressed or this claim: "To the extent that we accord special status
manic, delusional or paranoid, sharp as a tack or for- to our genes and what they reveal, genetic information
getful and confused. So the mere fact that people can is uniquely powerful and uniquely personal, and thus
learn things about us that are medically relevant would merits unique privacy protection."3 The model Genetic
not seem to distinguish genetic data from other kinds Privacy Act, like other legislation intended to protect
of health-related information. against genetic discrimination, insists both that genetic
What, if anything, makes genetic information dif- information can be distinguished from other medical
ferent from other health-related information? Can it, in information and that it ought to be so distinguished. Just
concept and in practice, be singled out? Regardless of why is genetic information, in these authors' words,
whether it really is different from medical information, "uniquely powerful and uniquely personal"? Annas,
are there characteristics of genetic information or of the Glantz, and Roche offer three reasons.
society into which it will flow that should lead us to act First, they liken a person's genetic profile to a
as if it were different? "future diary" and claim that genetic information "can
I was chair of the Task Force on Genetic Infor- predict an individual's likely medical future for a var-
mation and Insurance of the NIH-DOE Joint Working iety of conditions." They go on to argue that we should
Source: From Genetic Secrets: Protecting Privacy and Confidentiality in the Genetic Era, ed. Mark A. Rothstein (Yale University Press,
1997). Reprinted with permission.
lb I
think of DNA as a "coded probabilistic future diary and unalterable. Looking toward the future, people
because it describes an important part of a person's might be concerned that genetic information about
unique future and, as such, can affect and undermine race and ethnicity could fall into the hands of groups or
an individual's view of his/her life's possibilities. Unlike governments with hostile or totalitarian ambitions. The
ordinary diaries that are created by the writer, the in- recent experience of ethnic conflict in Eastern Europe
formation contained in one's DNA, which is stable and and Africa, as well as enduring racism in the United
can be stored for long periods of time, is in code and States, reinforces that concern.
is largely unknown to the person Most of the code Gostin third and fourth factors—the "unique
cannot now be broken, but parts are being deciphered quality" that permits identification of the individual
almost daily."4 Let us call this, borrowing a phrase from and the possible future uses of stable DNA—require
Nancy Wexler, the concern about genetic prophecy. some explanation. The "unique quality" is what en-
The second reason Annas, Glantz, and Roche cite abled the FBI in the World Trade Center bombing
for regarding genetic information as unique is that "it case to confirm who licked the stamp and who licked
divulges personal information about one's parents, the envelope. Gostin notes an interesting feature of
siblings, and children."5 The biology of the claim is genomic segue nee information. Genetic fingerprinting
straightforward enough: we get half of our genes from operates by detecting differences in the DNA sequences
each biological parent, and we pass half of ours on to of individuals in a number of regions of the genome
each of our biological children. Genetic information known to be highly variable. If a database contained
about each of us, then, is also to some extent informa- sufficient information about the sequence, even if the
tion about our ancestors, descendants, and other such person's name were not attached to the file, it might
biological relations as sisters and brothers. Call this the be possible to identify the individual whose sequence
concern for kin. it is, in a manner similar to the method of genetic
As their third reason, the authors cite a history fingerprinting. So, although the practice of removing
of genetics being used to stigmatize and victimize identifying information is usually thought to confer
people. Whether they have in mind eugenics programs, anonymity by making the records impossible to trace
ill-conceived genetic screening, or the use of genetic in- to an individual, that may not be the case with records
formation by employers, insurers, or others, they do not containing significant chunks of DNA sequence data.
say. Call this the concern about genetic discrimination. (Of course, this problem is not limited in principle
Other authors have offered their own lists of reasons to genomic information. It is not hard to conceive of
supporting genetic exceptionalism. Gostin, for example, databases that, while containing no names, still contain
offers these "compelling justifications" for protecting enough information—e.g., community, neighborhood,
the privacy of genetic information: "the sheer breadth age, marital status, number and ages of children, oc-
of information discoverable; the potential to unlock cupation, make, model and year of cars—to allow indi-
secrets that are currently unknown about the person; viduals to be identified.)
the unique quality of the information enabling certain It is also useful to distinguish between two types
identification of the individual; the stability of DNA ren- of genetic collections. One form is databases of gen-
dering distant future applications possible; and the genetic information that either exist in electronic form
eralizability of the data to families, genetically related or are readily translatable into a computer-searchable
communities, and ethnic and racial populations.' The database. Such databases might, in the future, include
first two factors—breadth and the potential to unlock DNA sequence information about identifiable indi-
secrets—seem to be expansions of the concern about viduals; they are unlikely to do so now. Researchers
genetic prophecy. The last concern mentioned, the gen- and others are less interested in raw sequence than in
eralizability of the data, is an interesting expansion of variants of important genes—mutations and benign
the concern for kin. In Gostin's view, the concern does polymorphisms—or in short stretches of DNA that
not stop at the family's door but extends to those larger are useful for genetic fingerprinting. The other type
groupings of people who share a certain genetic herit- is collections of tissue samples containing DNA. Each
age. I agree. A focus on genetics emphasizes racial and sample could yield, in principle, an entire sequence
ethnic differences, a very sensitive matter both now and for each individual. Gostin asserts correctly that such
possibly in the future. Stressing the genetics of race has tissue samples contain sufficient information to study
the potential to intensify those divisions, while reinfor- any and all genes, as well as the DNA needed to do a
cing the view that perceived differences are not mere genetic fingerprint, which could then be linked to the
accidents of culture and circumstance but are grounded individual if his or her matching genetic fingerprint
in biology, which is itself seen as somehow fundamental were on file. But it is crucial to understand that the
Murray: Genetic Exceptionalism and "Future Diaries"
information in such collections of genetic materials is The second argument for genetic exceptionalism is
opaque. Considerable work and expense are needed concern about kin—not merely our immediate family
to get any of the latent information from a sample.' but perhaps also the larger ethnic community. Again,
Gostin's worry about the durability of DNA is correct it is difficult to claim uniqueness, or even special im-
on the bare facts. In the World Trade Center bombing portance and sensitivity, for genetic information. That
case, the FBI, after all, was able to take microscopic bits one member of a family has tuberculosis is certainly
of DNA and identify two individuals. But without the relevant to the rest of the household, all of whom are
motivation and resources required to analyze the DNA, in danger of infection, along with everyone who works
we do not have genetic information per se, merely the with or goes to school with the infected individual.
raw materials from which genetic information could be Likewise, if one partner in a marriage has a sexually
derived. (But, then, we would not need a fancy tissue transmitted disease, that information is important for
bank either. One of the most massive collections would the other partner. Or suppose the main wage earner
be in the hands of the Publisher's Clearinghouse—all in the household showed early signs of heart disease
those stamps and envelopes, with return addresses yet!) that could bring disability and death. Wouldn't the
Are the arguments in favor of genetic excep- other family members have a profoundly important
tionalism persuasive? The strong form of genetic stake in knowing this? They might make very different
exceptionalism claims that genetic information is career training or employment choices if they knew the
unique—"uniquely powerful and uniquely personal," breadwinner was likely to be struck down soon.
as Annas, Glantz, and Roche have written. We could Although the concern for kin could amplify the
admit to hyperbole in that statement and look for a sensitivity of genetic information, it does not render
weaker form of genetic exceptionalism, one that claims that information unique. It can be very important, to
that genetic information is sufficiently distinctive from be sure, but whether genetic information is important
other information that it ought to receive greater pri- and sensitive enough to distinguish it from other sorts
vacy protection. The policy question, after all, is the of information is not yet clear.
one we are most concerned with. Genetic information The fear of discrimination is the third of the candi-
does not have to be unique in order to warrant special dates for the unique character of genetic information.
protection, but it does have to be distinctive and espe- Here again, genetics is not alone. Institutions and in-
cially sensitive. In evaluating the arguments, we should dividuals can and have used all sorts of information,
consider the case that could be made if only the weaker both visible and occult, as the basis for discrimination.
form of genetic exceptionalism is supported. In underwriting for health insurance, for example,
The argument from genetic prophecy is not com- insurers use evidence of current disease or future dis-
pelling. Genetic information is neither unique nor disease risk—whether it is genetic or nongenetic doesn't
tinctive in its ability to offer probabilistic peeks into our matter—to decide who gets a policy, what that policy
future health. Many other things afford equally inter- covers, and how much it costs. Whether this discrimin-
esting predictions. Some of them would be impossible ation should be regarded as fair or unfair is debated.
to conceal and so fall outside the concern of privacy— But it is difficult to make the argument that it is fair to
some people, for example, are avid skydivers or para- discriminate on nongenetic factors but unfair to dis-
sailers. Other types of information would be hidden, criminate on genetic ones.
just like most genetic information; examples include Some nongenetic factors, to be sure, can be
asymptomatic hepatitis B infection, early HIV infection, thought of as a matter of choice, in contrast with one's
and even one cholesterol level. These have implica- genes. Such risky behaviors as smoking, thrill-seeking,
tions for future health that are every bit as cogent and and the like do seem different from being struck by
sensitive as genetic predispositions. Huntington disease, over which we have no control.
Perhaps genetic risks ought to be treated differ- But there are plenty of nongenetic risks that no more
ently because there are potentially so many of them, reflect genuine individual choice than genetic risks.
and such a massive prospective difference in quantity If the air we breathe and the water we drink are pol-
effectively makes a qualitative difference. Or perhaps luted, if our parents or co-workers are heavy smokers,
what is most worrisome is that our genetic risks are if we are reasonably prudent but injured in an acci-
occult, hidden; attributes that others could know even dent nonetheless, it is hard to say that we bear any
as we remain ignorant about them. Here lurks an image significant measure of responsibility for the resulting
of genetic information as a mysterious, powerful, and illnesses. Likewise, most links between genes and dis-
inexorable force that will dominate and control our ease are likely to be very different from the link in Hun-
futures. Keep this image in mind. We will return to it. tington disease. Most will be probabilistic associations
II Al 111111
rather than straightforward causal connections. And choose to share what we have recorded in our diary,
many may be modifiable by the individual's actions. No but we want to be able to say when and with whom
simple equation exists between genetic factors and in- we share it.
exorable fate, or nongenetic factors and being open to The future diary metaphor captures several sources
individual choice and action. If we are less inclined to of our uneasiness about genetic privacy. Our "genetic
worry about discrimination on the basis of health risk diary" was written not by us but by an agency completely
factors that are open to modification and individual outside our control. It can be read not only by others but
choice, then let us recognize that as the relevant differ- without our permission or even knowledge. Last, it is
ence, and not confuse it with the distinction between possible for strangers, whose purposes are not ours, to
genetic and nongenetic factors. use those genetic "secrets" to harm us. Little wonder that
The concern for genetic information and dis- we want to protect whatever is written in that "diary."
crimination may help explain some of the interest But the metaphor also misleads. It implies that the
in genetic privacy because it broadens and sharpens contents of that future diary reflect what is most intim-
important perceptions. First, it broadens the pool ate, central, and important about us—that it reveals, in
of possible factors that might be used to discrimin- some fundamental way, our social and personal iden-
ate against an individual, and it likewise expands the tity, our loves and interests, and our actions. In fact,
pool of individuals who might become the subjects of our genomes have little or nothing to say about any
discrimination. Second, it sharpens the widely held of these crucial matters. The metaphor also promotes
moral intuition that we should not be punished for genetic determinism. In complex disorders with many
things beyond our control. (This does not contradict contributing factors, such as many cancers and heart
the warning just stated about oversimplifying the con- disease, genetic information may indicate only a rough
nection between genes and illness. Health insurers, for range of probabilities, something that falls far short of a
example, work on probabilities. It might be enough "probabilistic future."
that a woman has a mutated form of the breast cancer So much for the case in favor of genetic excep-
I gene [BRCAI], hence an approximately 85% lifetime tionalism. What about the case against it? When the
risk of breast cancer, for an insurer to deny her cover- Task Force on Genetic Information and Insurance
age. Whose "fault" it might be is irrelevant.) rejected genetic exceptionalism, it did so with reluc-
Perhaps what really frightens and galls us about tance. At least for the purpose of deciding who should
discrimination on the basis of genetic information is its have access to health care coverage, the task force
reliance on information about us over which we have could find no sound way to distinguish genetic from
no control and may not even know ourselves. Here nongenetic diseases and risks. Genetic exceptionalism
again is the hidden and mysterious nature of genetic in- depends on what we have come to call the "two-bucket
formation, joined with its aura of power and ubiquity, theory" of disease. According to this model, there are
lurking close beneath the surface of our discomfort. two buckets—one labeled "genetic," the other labeled
George Annas's metaphor of our genes as our future "nongenetic"—and we should be able to toss every
diary captures the power and ambiguity of the personal disease and risk factor into one of the two.9 So, Hun-
significance of genetic information. tington disease goes into the "genetic" bucket and get-
Recall the description of an individual's genome as ting run over by a truck goes into the "nongenetic" one.
his or her "coded probabilistic future diary because it But many diseases and risks don't fit neatly into either
describes an important part of a person's unique future bucket. Take breast cancer. Some cases of breast cancer
and, as such, can affect and undermine an individ- have strong genetic roots, but others have no clear
ual's view of his/her life's possibilities. Unlike ordinary genetic connection. For that matter, not every woman
diaries that are created by the writer, the information with a mutated BRCAI gene will develop breast cancer.
contained in one's DNA . . . is in code and is largely And some apparent risk factors have little or no link
unknown to the person."8 The metaphor is ingenious, to genetics. Similar complexity exists for heart disease:
powerful, and provocative. What features of genetic cholesterol is a risk factor, and one's cholesterol level
information does it capture? What is the source of its can be modified by diet, exercise, and other factors; but
power? Does it mislead in any way? our genes have as much or more to do with the level of
Diaries are intimate, private places where we con- cholesterol circulating in our blood as our environment
fide what is most important about who we are, who or behavior. Into which bucket, then, should we toss
and what we love, and what we do. Those same con- breast cancer? Heart disease? Cholesterol level?
fidences are also just those things we may be most re- On conceptual grounds, the task force agreed
luctant to broadcast. There may come a time when we that the argument for distinguishing genetic from
Murray: Genetic Exceptional ism and "Future Diaries"
nongenetic information was dubious. Certainly, there create enormous practical problems in record keeping
were cases of relatively unambiguous and pure genetic and information flow.13
or nongenetic diseases and risk factors. But in many The task force had another reason for abandoning
more cases, including many of the most interesting genetic exceptionalism: we concluded that there was
and important ones, the two-bucket theory was hope- no good moral justification for treating genetic in-
lessly inadequate. formation, genetic diseases, or genetic risk factors as
Efforts to legislate genetic privacy have confronted categorically different from other medical information,
the two-bucket problem by confining themselves more diseases, or risk factors.14 If someone genuinely needed
or less to whatever fit neatly into the genetic bucket. health care, it did not matter whether one could find a
Karen Rothenberg notes this in her study of state laws, genetic root for the disease or whether it was the prod-
which typically limit the definition of genetic informa- uct of nongenetic bad luck or accident. Some people
tion to the results of DNA testing.w The draft Genetic regard so-called self-inflicted or lifestyle-related mal-
Privacy Act itself began with a conceptually rich, inclu- adies as exceptions. The argument here turns on both
sive definition but moved to a much narrower one. This moral and policy premises—roughly, that to the extent
is how the act's authors describe the decision they faced: that people are responsible for their own misfortune,
"either create a definition for genetic information that is there is no moral obligation to share the cost of their
consistent from the viewpoint of theory and principle, care, and that the point of access to health insurance is
but not of much practical value, or design provisions that the proper place to extract that cost or to inflict punish-
are capable of practical application that would have the ment. Whether everyone finds this position persuasive
effect of protecting the most private and potentially stig- is, however, not the crucial point. Our need for health
matizing genetic information." Their definition reads: care in most cases will be the product of a complex
"The term private genetic information means any infor- mix of factors, genetic and nongenetic, both within the
mation about an identifiable individual that is derived scope of our responsibility and outside of that scope.
from the presence, absence, alteration, or mutation of a The distinction between genetic and nongenetic fac-
gene or genes, or the presence or absence of a specific tors is not the crucial one. We are inclined to regard
DNA marker or markers, and that has been obtained: it as crucial only when we fall for the overselling of
(1) from an analysis of the individual's DNA; or (2) from genetics—when we fall prey, that is, to genetic reduc-
an analysis of the DNA of a person to whom the individ- tionism and genetic determinism.
ual is related."u They acknowledge that this definition Resounding statements about the significance of
leaves out biochemical tests for genes—potentially a genetic information are not hard to find. Lawrence
massive gap, because what we will often be interested Gostin, for example, claims that "the features of a
in clinically are the products of the genes, not the genes person revealed by genetic information are fixed—
themselves. But they observe ruefully that amending the unchanging and unchangeable."" George Annas,
definition of private genetic information to include test- Leonard Glantz, and Patricia Roche write: 'To the
ing for gene products "makes the distinction between extent that we accord special status to our genes and
genetic information and medical information generally what they reveal, genetic information is uniquely
more difficult to justify."" powerful and uniquely personal, and thus merits
This conceptual problem has fed another difficulty unique privacy protection.""
with genetic exceptionalism: how to identify and keep Some scholars worry that the significance of gen-
separate genetic from nongenetic information in the etics has been oversold. In their book The DNA Mys-
medical record. If it could not be determined whether tique: The Gene as a Cultural Icon, Dorothy Nelkin and
some piece of information belonged in one category Susan Lindee bemoan what they see as genetic hype:
or the other, it seemed practically infeasible to divide
medical records into those portions that were genetic As the science of genetics has moved from the lab-
(hence offlimits), from those portions that were non- oratory to mass culture, from professional journals
genetic (therefore available to prospective insurers). to the television screen, the gene has been trans-
More recently, Lawrence Gostin argued that laws offer- formed. Instead of a piece of hereditary informa-
ing special protection to genetic information may be tion, it has become the key to human relationships
problematic because "different standards would apply and the basis of family cohesion. Instead of a string
to data held by the same entity" depending on whether of purines and pyramidines, it has become the es-
it met the definition of genetic information; because sence of identity and the source of social difference.
"other health conditions raise similar sensitivity issues" Instead of an important molecule, it has become
such as HIV and mental illness; and because it could the secular equivalent of the human soul."
II 1111.1.1
Other scholars have voiced similar concerns. In a however, relative to the other components of that func-
study of factors contributing to the social power of gen- tion and almost always relative to particular environ-
etics, Eric Juengst identifies genetic determinism and mental conditions." Keller warns of the consequences
genetic reductionism, along with their familial implica- of a too-hasty acceptance of genetic reductionism: "a
tions, as the sources of the anxiety caused by genetics. far more radical depersonalization of medicine than
He notes that the link of genetics to disease in the public that initiated by the earlier and more general march of
mind is defined by examples such as Huntington disease, medicine from art to science." This step would invite the
in which having the mutated gene effectively determines effacement of not only the patient's environment and his-
the outcome: if one lives long enough, one will get the tory but also the patient him or herself.2°
disease. "This history still affects the way many people The paths I have explored lead to one destination.
think about genetic risk information, by leading them to Genetic information is special because we are inclined
assume that genetic diagnostics of any kind have more to treat it as mysterious, as having exceptional potency
predictive power than other kinds of health risk assess- or significance, not because it differs in some funda-
ments."18 Juengst notes that images of crystal balls and mental way from all other sorts of information about
"future diaries" bolster this deterministic image of gen- us. Portions of that mystery and power come from the
etic information. He also remarks that such images invite opaqueness of genetic information, the possibility that
fatalism and social stigmatization. Because genetic risks others will know things about the individual that he or
encourage an explanation of disease in biological terms, she does not know, and how genetic information con-
they are open to reductionistic accounts of both diseases nects individuals to immediate family and more distant
and persons. That is, genetic risks may come to be seen kin. The more genetic information is treated as special,
as the explanation for complex multifactorial diseases. the more special treatment will be necessary. Yet none
They may also be seen as fundamental, defining charac- of these factors is unique to genetic information.
teristics of the persons who have such risks, essentially I propose that genetic exceptionalism—the plea to
reducing those persons to their genetic propensities. treat genetic information as different from other health-
Carl E Cranor considers the concept of causation in related information—is an overly dramatic view of the
genetics and looks critically at both those who under- significance of genetic information in our lives. It is a
emphasize and those who overemphasize the causal reflection of genetic determinism and genetic reduc-
role of genes. He acknowledges the concerns of those tionism at least as much as the product of genuinely
worried about possible social misuses of genetic explan- distinctive features of genetic information. There is a
ations but worries that those critics place too much im- disturbing corollary to this thesis: there is a vicious
portance on minimizing the causal significance of genes, circularity in insisting that genetic information is differ-
when they should be criticizing the misuses they fear. ent and must be given special treatment. The more we
Cranor urges us to "recognize that genes, for at least repeat that genetic information is fundamentally unlike
some single-gene diseases (even if rare), appear to make other kinds of medical information, the more support
discernible causal contributions to some diseases, but to we implicitly provide for genetic determinism, for the
deny that this settles all questions concerning scientific notion that genetics exerts special power over our lives.
understanding, recommendations about research agen- I therefore suggest a revision of the "future diary"
das, and the important normative and practical issues as metaphor for genetic information. Why not regard our
to how such diseases should be treated or èngineered,' genes as a list of the obstacles we are likely to encounter
if at all."19 That is, we do not have to pretend that genes and perhaps as a somewhat better prediction of how
are unimportant to avoid determinism or reductionism. long we will have to do what matters to us, to be with
We should give genes their due, but no more than that. the people we love, and to accomplish the tasks we
Evelyn Fox Keller points out the simplistic premises have set for ourselves? Our genes no more dictate what
of genetic determinism "The idea of 'a gene for X' al- is significant about our lives than the covers and pages
ready presupposes the existence of an organism capable of a blank diary dictate the content of what is written
of identifying, translating, interpreting, and making within. Our genes might be regarded metaphorically as
productive use of a particular gene." As she goes on to the physical, but blank, volume in which we will create
explain, "All biological functions are composite func- our diary. Some volumes have fewer pages in which to
tions, involving the correct `reading' of many genes. The write, some more. Certain pages, often toward the back
more complex the function, the more genes are likely to of the volume, may be more difficult to write on. And
be involved. What genetics can and often does enable us some leaves may require great skill and effort to open
to do is to identify aberrations in some component part at all. But the physical volume is not the content of the
that lead to failure of the composite function—always, diary. The content we must write ourselves.
Murray: Genetic Exceptionalism and "Future Diaries"
Notes
1. Task Force Report: Genetic Information and Insurance 10. Karen H. Rothenberg, "Genetic Information and Health
(Bethesda, Md.: Genetic Information and Health Insurance: State Legislative Approaches," Journal of Law,
Insurance, National Institutes of Health, National Center Medicine and Ethics 23 (1995): 312-19.
for Human Genome Research, 1993). 11. Robert J. Pokorski, "The Potential Role of Genetic Testing
2. Lawrence 0. Gostin, "Genetic Privacy," Journal of Law, in Risk Classification," Report of the Genetic Testing
Medicine and Ethics 23 (1995): 320, 321. Committee to the Medical Section of the American Council
3. George J. Annas, Leonard H. Glantz, and Patricia A. of Life Insurance, Hilton Head, S.C., June 10, 1989.
Roche, "Drafting the Genetic Privacy Act: Science, Policy, 12. Annas, Glantz, and Roche, "Drafting the Genetic Privacy
and Practical Considerations,"Journal of Law, Medicine and Act," 326.
Ethics 23 (1995): 360, 365. 13. Gostin, "Genetic Privacy," 324.
4. George J. Annas, "Privacy Rules for DNA Databanks: 14. Thomas H. Murray, "Genetics and the Moral Mission of Health
Protecting Coded 'Future Diaries,'" Journal of the American Insurance," Hastings Center Report 22, no. 6 (1992): 12-17.
Medical Association 270 (1993): 2346-50; Annas, Glantz, 15. Gostin, "Genetic Privacy," 324.
and Roche, "Drafting the Genetic Privacy Act," quotation on 16. Annas, Glantz, and Roche, "Drafting the Genetic Privacy
p. 360. Act," 365.
5. Annas, Glantz, and Roche, "Drafting the Genetic Privacy 17. Dorothy Nelkin and M. Susan Lindee, The DNA Mystique:
Act." The Gene as a Cultural Icon (New York: W. H. Freeman,
6. Gostin, "Genetic Privacy," 326. 1995), 198.
7. Thomas H. Murray and Norman T. Mendel, "Introduction: 18. Eric T. Juengst, "The Ethics of Prediction: Genetic Risk and
The Genome Imperative," Journal of Law, Medicine and the Physician-Patient Relationship," Genome Science and
Ethics 23 (1995): 309-11. Technology i, no. I (1995): 21, 30.
8. Annas, Glantz, and Roche, "Drafting the Genetic Privacy 19. Carl E Cranor, "Genetic Causation," in Are Genes Us? The
Act," 360. Social Consequences of the New Genetics, Carl E Cranor, ed.
9. Thomas H. Murray, "Assessing Genetic Technologies: (New Brunswick, NJ.: Rutgers University Press, 1994).
Two Ethical Issues," International Journal of Technology 20. Evelyn Fox Keller, "Masters Molecules," in ibid.,
Assessment in Health Care 10 (1994): 573-82. quotations on pp. 90, 91, 97.
3.5 Cases
Case 1 nothing was said about lung cancer.'•2 Surprisingly,

the prime minister (Winston Churchill) had access
George VI's Cancer to the king's medical records. With the help of his
own doctor, Churchill was able to figure out from this
In September 1951, a cancer tumour was found in record the true diagnosis. What he learned was soon
the left lung of George VI, the king of Great Britain.'•z confirmed by a member of the king's staff (and this
George had been a heavy cigarette smoker for much of information was in turn passed on to some members
his life, and the lung cancer was probably a bad side of the cabinet by the prime minister)! They kept this
effect of this.3 His lung and the tumour were removed, information a secret from everyone else.
and George recovered enough to continue with Why did the treating physicians not tell the king or
some of his work as king until his death in February even other members of his family the truth?' Back in the
1952.1,4,5 Elizabeth, his daughter and now the queen early 1950s, the presence of a cancer tumour probably
of Great Britain, took over some of the other parts of meant that George had not long to live, even if the main
his j ob.1'6 Everyone-the king himself, his family, the cancer tumour were removed successfully. The doctors
government, and people in general-knew at the time knew that George would want to carry on with his royal
that George was not in the best of health.2,5.6 They all duties for as long as possible.' They also thought that
also knew some details of his health problems. But George would be better able to continue if he believed that
George and his immediate family did not learn until he was recovering and would live for more than a short
much later that he had lung cancer.',4 They were all while longer. They exercised their therapeutic privilege-
told by the treating physicians that the operation was choosing what they thought would be best for the patient
done to free a bronchial tube or airway in his left lung; (and Britain) while keeping the patient in the dark On the
■m I II II IIILL1
traditional model of the doctor—patient relationship, de- (see page 128). But almost no aspect of the life of members
ceiving King George in this way was perfectly proper. On of the British royal family is ordinary. By tradition and law,
some views, however—for example, that of Kant in "On almost everything they do and everything that happens to
a Supposed Right to Lie from Benevolent Motives" (see them—induding matters of life and death and marriage
page 106)—this was immoral, no matter what the motive. and birth that would for ordinary people be personal and
Why did the prime minister have access to the kin0 private—is under the control of others.2.539 The death of
medical records? Of course, for most ordinary people in a king, for example, is a matter that will affect the public
Britain or Canada, these medical records would not usu- interest.10 If George might die soon, then the government
ally be available to others except with the patient's consent, needed to prepare for this and for a new person (Elizabeth,
as Dave Unger points out in "Breaching Confidentiality" the king's eldest child) to take over these royal duties.
Notes
1. Philip Ziegler, George VI The Dutiful King (London: Penguin 6. Rosie Waites, 'The Moment a Princess Became a Queen,"
Allen Lane, 2014), ch 6. BBC News Magazine 6 February 2012.
2. Sarah Bradford, George VI (New York Viking, 2011), 7. The presentation of these events in the Netflix series
ch 13. "The Crown," Episode 1, closely follows the account in Sarah
3. I.D. Canacher, "The King's Anaesthetic," Journal of Medical Bradford, George VI.
Biography 23 (2015), 139-45. 8. Michael Billig, Talking of the Royal Family (London and
4. Kenneth Weisbrode, Churchill and the King (New York New York Routledge, 1998), ch 4.
Penguin Books, 2015), p. 180. 9. Vernon Bogdanor, The Monarchy and the Constitution
5. Kate Samuelson, "Everything to Know about the (New York Oxford University Press, 1995).
Death of King George VI, as Seen on The Crown," 10. Emily Jackson, Medical Law Text Cases and Materials, 3rd
Time 9 Nov. 2016. edn (Oxford: Oxford University Press, 2013), ch 7.
Case 2 been given the information, they said, she would have
Arndt v Smith terminated the pregnancy and there would not have been
these problems.14 This failure on the part of the doctor
In early May 1986 Carole Arndt became infected with was negligent and responsible for the existence of a child
chicken pox in the 12th or 13th week of her preg- with medical issues and the extra costs and other care
nancy. t•2.3 She talked to her doctor, Dr Margaret Smith, burdens Ms Arndt and her partner had as a consequence.
by telephone about her pregnancy and the chicken pox This case was first heard in the Supreme Court of
infection in mid-May. Once she recovered from the illness British Columbia in front of Mr Justice Hutchinson in
she saw her doctor in person. This happened in late May March 1994 when Miranda was seven years old.' He
1986.' Dr Smith told Ms Arndt that her illness might have agreed with Ms Arndt's lawyers that Dr Smith failed to
affected the fetus she was carrying. As Dr Smith admit- tell Ms Arndt as much about the harms her chicken pox
ted in testimony, she did not tell Ms Arndt all she knew infection might cause to her unborn child as she should
about the effects a chicken pox infection could have on have. She failed to fulfill her responsibilities as Ms Ar-
the development of a fetus.' Dr Smith did tell her about endt's physician.' However, Justice Hutchinson did not
the more frequent risk of skin scars and limb defects but agree that the doctoit failure was to blame for the birth
not about rarer but still possible and more damaging of a child with Miranda's health care issues.'•5 He noted
risks of eye and central nervous system abnormalities.' that Ms Arndt's pregnancy was carefully planned and
Ms Arndt continued the pregnancy to term. Unfortu- that a child was much wanted.'•2 He also noted that Ms
nately, the child, a daughter named Miranda, was born Arndt had expressed strong doubts about the validity of
with medical problems caused by her mother's chicken the kind of conventional medical and health care know-
pox infection. She required continuous tube-feeding be- ledge that Dr Smith would have relied on to tell her
cause she was unable to swallow and she needed many about the risks to a fetus from maternal chicken pox.
operations. Ms Arndt and her partner sued Dr Smith for Given these two items, the judge doubted that Ms Arndt
the extra costs of raising a child with these issues. Her would have chosen to terminate her pregnancy even if
lawyers argued that Dr Smith should have told her about she had been given all the information she should have
all or at least more of the harms she knew could result been given.' So Dr Smith's withholding of information
from a maternal chicken pox infection. If Ms Arndt had about the risks from Ms Arndt was not responsible for
Cases
the birth of a child with health care problems, the costs "modified objective" one.6 This was a test devised in
of the medical care, or the extra care burdens. an earlier landmark case, Reibl v Hughes.t•2.3.6 On this
The case was next heard in the Court of Appeal for standard or rule, what should have been found out was
British Columbia in March 1995.2 The appeal was made what a reasonable person in Ms Arndt's circumstances
by the lawyers for Ms Arndt. Mr Justice Lambert wrote with a reasonable set of concerns and beliefs would have
the judgment. He also said the issue was not whether done if the doctor had provided all the necessary infor-
Dr Smith had failed to tell Ms Arndt as much as she mation.' Because in the court's view the wrong standard
should have but whether the failure was to blame for the or test was used, it ordered a new trial.
harm that followed.2.5 Justice Lambert's view was that The matter was again heard at court in 1997, this
Justice Hutchinson had applied the wrong rule or stan- time in the Supreme Court of Canada.3 The lawyers
dard to settle this issue of responsibility." The standard for Dr Smith had appealed the decision of the Court
Justice Hutchinson had used was what legal writings of Appeal for British Columbia; they did not want the
call the "subjective test" for causation or blame.° Justice matter to be retried.3 The majority opinion was given by
Hutchinson had asked what Ms Arndt would have done Chief Justice Lamer. It confirmed the Court of Appeal's
if she in particular—not someone else with other beliefs view that the issue was not whether Dr Smith had pro-
or values or powers of reasoning—had been given all vided all the information she should have to her patient
the information her doctor should have provided her but whether the doctor's failure to provide information
about the risks to her fetus from the chicken pox in- she should have was to blame for the existence of a
fection. He had concluded she would not have had an child with the issues Miranda presented to her parents.
abortion. So Dr Smith was not responsible for the harms Furthermore, the majority agreed that the correct test to
that followed. According to Justice Lambert, however, apply to answer this question is the modified objective
the proper test or standard to use to prove that the doc- test.3•7 While there were dissenting opinions, the major-
tor's failure caused or was to blame for the harm is the ity opinion was that the case should be retried.
Notes
1. Arndt v Smith, British Columbia Law Reports vol. 93 2nd that follow. The two issues are theoretically separate. On
series (1994), p 220. the other hand, as L.E. and F.A. Rosovsky point out in
2. Arndt v Smith, British Columbia Law Reports vol. 6 3rd "The Nuts and Bolts of Obtaining Consent to Treatment" in
series (1995), p 201. Chapter 2 (page 45), physicians are sometimes concerned
3. Arndt v Smith, Supreme Court Reports vol. 2 (26 Jun about the first—providing all the information to their
1997), p 539. patients that they should—because they wish to avoid being
4. At the time of Ms Arendt's pregnancy, Canadian abortion blamed and sued for any harms or damage that may follow
law would have required her to get the approval of if they fail to provide all the information they should.
a therapeutic abortion committee to terminate her 6. Gerald Robertson, "Informed Consent Ten Years Later: The
pregnancy. As Mr Justice Hutchinson in the first hearing of Impact of Reibl v Hughes," Canadian Bar Review 70 (1991), 423.
the case at the Supreme Court of British Columbia pointed 7. Karen Capen, "Supreme Court Reaffirms Landmark
out, this committee might have declined to allow Ms Informed-Consent Ruling in Chickenpox Case," Canadian
Arndt to have an abortion. Medical Association Journal 157 (1 September 1997),
5. Chapter 2 (page 45) considers what a doctor or other 553-4. This comment both separates out and runs
medical professional should tell a patient or the patient's together the issue of failing to disclose information that
substitute decision-maker in order to provide informed should have been provided by the treating medical
consent. A doctor may fail to tell her patient everything she professional and the issue of responsibility for harms or
should but not always be responsible for the bad effects bad effects that follow. See note 5 above.
Case 3 weeks, a picture of the flight's final moments emerged.

The Crash of Germanwings Flight 9525 Soon after the plane reached cruising altitude, the first
officer and co-pilot, Andreas Lubitz,4 took control of
the plane, locked the captain out of the cockpit, and
On 24 March 2015, Germanwings Flight 9525, on its deliberately flew it into the side of a mountain.
way from Barcelona to Dusseldorf, crashed into a ravine Lubitz had a long history of mental health issues 2.5
in the French Alps. All 144 passengers and six crew After two months of pilot school in Germany in late 2008,
members were killed. t.2.3 In the following days and for example, he withdrew with depression because the
11 111 1 II II 1,11 11141
demands made on him by his training were too great. authorities to predict that he would do what he did and
He had 10 months of psychotherapy and antidepressant stop him? As Donald Bersoff pointed out in a presiden-
medical treatments before returning to pilot school and tial address to the American Psychological Association,
completing his training program. In December 2014, he psychologists are expert at diagnosing mental illnesses,
started seeing psychologists and other medical profes- but they are not expert at predicting whether someone
sionals again. He complained of stress due to uncertain- with a particular illness will do the kind of thing Lubitz
ties about his job at Germanwings. He could not sleep, did."'8 Of course, doctors treating him did say in let-
and his eyesight was deteriorating. An MRI scan showed ters to Lubitz that he was not fit to return to work as a
no physical damage that would explain the problems pilot. However, under Germany's medical confidential-
with his eyes; they may have been psychosomatic—a ity laws, neither his employer nor aviation authorities
real disability but one caused by his psychological issues. were allowed to see these letters without Lubitz's express
He was prescribed antidepressants, and at least two of permission.2'9 Medical confidentiality laws are not so
the doctors Lubitz saw wrote letters to him stating that strict in other countries such as Canada. As Dave Unger
he was unfit to continue work as a pilot. Lubitz ignored in "Breaching Confidentiality" (see page 128) explains,
these letters and went back to work.5 Canadian rules allow a medical professional to break
Could this disaster have been prevented? If the pilot patient confidentiality to avoid harm to innocent others,
had not been locked out of the cabin by Lubitz, then like the passengers on an airplane. But according to phil-
Lubitz would not have had control of the plane or at osopher Kenneth Kipnis in "A Defense of Unqualified
least would not have been able to deliberately crash it. Medical Confidentiality" (see page 117), the require-
But the pilot was locked out by a mechanism designed ment in German law of absolute confidentiality is just as
to stop terrorists or others from entering the cockpit and things should be. Even in a case where a disaster like this
taking control of the plane.4 Would knowing about Lu- one results, there should be no breaking of confidential-
bitz's depression have allowed his employer or aviation ity except with a patient's clearly stated consent.
Notes
1. "What Happened on the Germanwings Flight," New York 6. David Bersoff, "Protecting Victims of Violent Patients
Times 27 March 2015. While Protecting Confidentiality," American Psychologist
2. Jeff Wise, Fatal Descent: Andreas Lubitz and the Crash of July-August 2014.
Germanwings Flight 9525 (Audible Studios on Brilliance 7. Suzanne O'Sullivan, "When the Body Speaks," Psychology
Audio, Feb. 2017). Today January-February 2017.
3. BEA, "Final Report [Germanwings Flight 4U9525]" (March 8. Denise Grady, "Signs May Be Evident in Hindsight, but
2016). Predicting Violent Behavior Is Tough," New York Times 18
4. James Fallows, "Could the Germanwings Crash Have Been September 2013.
Avoided?" The Atlantic 26 March 2015. 9. Editorial Board, "Editorial: Germanwings Crash an
5. SPEIGEL Staff, "The Death Wish of a Germanwings Co- Example of When Medical Privacy and Safety Collide"
pilot," SPIEGEL Online 27 March 2015. Chicago Tribune 18 March 2016.
Case 4 Personal Genome Project Canada at the University

Personal Genome Mapping of Toronto's McLaughlin Centre and the Hospital for
Sick Children in Toronto.5 The DNA tests offered by
In 2013, whole genome sequencing—the mapping of a companies like 23andMe and Ancestry.ca in Canada
person's entire DNA or genetic blueprint—started to be and the US and elsewhere in the world map less than
available in the United States to private individuals."2'3 1 million of the 3 billion base pairs in a person's entire
A company offering this service might require the sig- genome.° These are the services considered by Tim
nature of a physician, but in other ways there wasn't Caulfield in "Direct-to-Consumer Genetics and Health
and still isn't anything to stop anyone with enough Policy" in Chapter 8 (see page 434).
money from having their complete genetic information In August 2015, Samantha Schilit—a researcher
analyzed. Websites for companies like GenomeMe in working on human genetics at Harvard University—
Richmond, BC, claim to offer this service in Canada.4 took advantage of this availability and dropping prices
As well, anyone can have their personal genome com- for whole genome mapping to have hers done. 1.2j Who
pletely sequenced as part of a research project like the studying in this area would not be curious about their
Cases
own genetic makeup and not want to satisfy the curios- What neither sister gave very much thought to until
ity if they had the means to do so? But when Saman- later is whether Arielle might have had veto power over
tha contacted her sister, Arielle Schilit Nitenson, to tell Samantha's satisfying her curiosity about her genome.1'2'9
her about this project, she found that her sister wasn't Arielle is not just Samantha's sister but her identical twin.
as enthusiastic.1,2'7 Arielle was familiar with the issues Identical or monozygotic twins come from the same fer-
concerning the confidentiality of genetic medical infor- tilized egg and are nearly genetically identical. In other
mation from working as a researcher at the US National words, the whole genome sequence for Samantha would
Human Genome Research Institute.'•z She pointed out also be a whole genome sequence for Arielle. Does that
that the map of her genome would become part of Sam- mean Arielle's consent was needed for Samantha to have
antha's medical record. Anyone with access to this record her genome sequenced?'•z•9 According to Laura Her-
would have access to her whole genome sequence and scher, an instructor and writer in the ethics of genetic
whatever it revealed about her risk of inherited health counselling, no one has to get the permission of any
problems. This access might include, for instance, inves- other family member to have their genetic information
tigators in any health care study she agreed to be a sub- analyzed.' But Herscher may be describing what the law
ject in. It would also include insurance companies. The says rather than what ethical considerations require.
US Genetic Information Nondiscrimination Act (GINA) As it turned out, the genome map for Samantha
forbids discrimination based on genetic information by showed something that helped Arielle when Arielle
employers and health insurance providers.'•2 There is became pregnant with her first child in 2016.' The
nothing, however, that prevents life, disability, or long- DNA analysis showed that Samantha—and therefore
term care insurers from using this information when it is Arielle—had a mutation that could cause problems
available to set premiums she would have to pay or even with pain medications and delivery methods com-
to decide not to offer her insurance. (Canada currently monly used during labour. Knowing this, Arielle was
has no law similar to GINA.8) Despite her sisters doubts, able to make arrangements with her obstetrician that
Samantha had her genome sequenced. avoided these medications and techniques.
Notes
1. Emily Mullin, "Do Your Family Members Have a Right to Your 6. Elizabeth Murray, "Inside 23andMe Founder Anne Wojcicki's
Genetic Code?" MIT Technology Review 22 November 2016. $99 DNA Revolution," FastCompany 14 October 2013.
2. Samantha Schilit and Arielle Schilit Nitenson, "My 7. Britt Wray and David Oppenheim, Genomics (National Film
Identical Twin Sequenced Our Genome," Journal of Genetic Board Canada Ontario Studio, in development, 2017).
Counselling Online First Articles 2016. 8. Kerry Gold, "How Genetic Testing Can Be Used against
3. Randy Shore, "Personal Genomics: The Test for You—and How Bill S-201 Could Change That," The Globe
Everything," Vancouver Sun 27 June 2014. and Mail 3 April 2016; Yvonne Bombard, Ronald Cohn, and
4. Pamela Fayerman, "BC Gene Sequencing Laboratory Stephen Scherer, "Why We Need a Law to Prevent Genetic
Retracts Accreditation Claim after News Story Exposes Discrimination," The Globe and Mail 19 September 2016.
Bogus Info," Vancouver Sun 25 August 2015. 9. Anneke Lucassen and Angus Clark, "Should Families Own
5. Carolyn Abraham, "Would You Make Your DNA and Health Genetic Information?" British Medical Journal 335 (7 Jul
Data Public If It May Help Cure Disease?" The Globe and 2007), 22-3.
Mail 7 December 2012.
3.6 Study Questions
1. Is there any morally important difference between a medical professional lying to a

patient and withholding information from the patient about his or her condition or
treatment options?
2. Are there ever any circumstances when it would be right for a doctor to offer a patient
a placebo as treatment? What if there were no better treatment options?
3. Patients are usually told, or at least believe, that their health care information is private
and that hospitals and medical professionals will keep it confidential. Why? Wouldn't
it be better for everyone if this information were not kept confidential in any way?
II al 111111
4. Should doctors and other medical professionals always keep information they acquire
about their patients completely and absolutely confidential—that is, never, ever break
confidentiality?
5. The US Genetic Information Nondiscrimination Act forbids discrimination based on
genetic information by employers and health insurance providers. Should Canada
have a similar law?
6. If medical records are anonymized—that is, all personally identifying information is
removed or disguised—will that always protect the confidentiality of these records?
Beyene, Yewoubdar. "Medical Disclosure and Refugees: Telling Bad news to Ethiopian
Patients." The Western Journal of Medicine 157 (1992): 328-33.
Beyleveld, Deryck. "Conceptualizing Privacy in Relation to Medical Research Values," in
Sheila McLean, ed., First Do No Harm: Law, Ethics and Healthcare (London and New York:
Routledge, 2016), 151-64.
Fry, Sara T. "Confidentiality in Health Care: A Decrepit Concept?" Nursing Economics 2
(1984): 413-18.
Higgs, Roger, "Truth at the Last: A Case of Obstructed Death?" Journal of Medical Ethics 8
(1982): 48-50.
Kermani, Ebrahim J., and Sanford L. Drob. "Tarasoff Decision: A Decade Later Dilemma
Still Faces Psychotherapy" American Journal of Psychotherapy 41 (1987): 271-86.
Lipkin, Mack. "On Telling Patients the Truth." Newsweek (6 April 1979). Reprinted in
Johnna Fisher, ed., Biomedical Ethics: A Canadian Focus, 2nd edn.
Pellegrino, Edmund. "Is Truth Telling to the Patient a Cultural Artifact?" Journal of the
American Medical Association 268 (1992): 1734-5.
Siegler, Mark "Confidentiality in Medicine: A Decrepit Concept." The New England Journal
of Medicine 307 (1982): 1518-21. Reprinted in Johnna Fisher, ed., Biomedical Ethics:
A Canadian Focus, 2nd edn.
Stanard, Rebecca, and Richard Hazler. "Legal and Ethical Implications of HIV and the Duty
to Warn for Counselors: When Does Tarasoff Apply?" Journal of Counseling & Development
73 (1995): 397-400.
Takala, Tuija. "The Right to Genetic Ignorance Confirmed." Bioethics 13 (1999): 288-93.
- cro.lessona0 Ethilcs
4.1 Introduction
The emphasis on patient autonomy that has supplanted the paternalistic Hippocratic
tradition in health care also requires a rethinking of professional ethics. Whereas not
too long ago, health care providers, and physicians in particular, exercised greater,
often absolute, authority over medical decisions, matters are different today. We saw in
Chapters 1-3 that competent patients now have rights to fully informed consent and
to accept or refuse any offered treatment. Those rights are reflected in law and are sup-
ported by moral reasoning. Today, health care providers' autonomy and responsibilities,
and their relationships with patients and each other, must acknowledge and be amended
to fit these circumstances. Professional ethics is broadly encompassed by these issues,
since it is about the responsibilities and standards of conduct that professionals are ex-
pected to live up to.
However, there is substantial controversy about the nature and boundaries of pro-
fessional ethics in health care. Such controversy is undoubtedly due partly to surviving
paternalistic Hippocratic influences and biases that remain to be worked out. But debates
about professional ethical standards also reflect the intimate connection between profes-
sional ethics and a diverse set of genuinely unsettled issues and problems in health care.
The articles in this chapter survey much of this ground. They demonstrate the challenges
in the way of stating a clear account of professional ethics.
Professional Autonomy and Professional Ethics
Because of the acceptance of respect for patient autonomy, it is natural now to ask what
the boundaries are to the autonomy of health care providers, usually referred to as profes-
sional autonomy. An answer to this question will be shaped by a conception of health care
providers' professional responsibilities, or their professional ethics generally. This connec-
tion between professional autonomy and professional responsibility can be demonstrated
simply enough. As a rough approximation, professional autonomy can be described as "a
right to say no" when asked to do things that are in conflict with professional responsibil-
ities. Thus, to have an account of professional autonomy requires that we understand the
professional responsibilities of health care providers. Those responsibilities will include
their duties to respect patient autonomy, alongside an account of other duties to patients,
other health care providers, and their profession. Taken together, these responsibilities
and duties reflect the main elements of a general account of professional ethics for health
care providers.
nni 11 II Al 11110
4 Professional Ethics
Making sense of professional autonomy within an account of professional respon-

professional
autonomy roughly, health
sibility that recognizes patient rights to autonomy is the aim of the first article in this
care providers' right to say chapter, Edmund Pellegrino's classic "Patient and Physician Autonomy: Conflicting Rights
"no" when asked to act in and Obligations in the Physician—Patient Relationship" (see page 165). Pellegrino argues
ways that conflict with their that despite the frequent appearance of conflict, autonomy and beneficence are actually
professional obligations.
complementary principles in health care. He argues that physicians need to be respectful
of patients' own conception of their health and well-being, noting that "some of the most
vital" patient interests are reflected in their choices and their own conception of their "life
story" Thus, patient autonomy cannot be separated from patient well-being or beneficence.
But Pellegrino goes on to say that patient autonomy cannot override physicians' rights to
avoid engaging in activities they honestly believe are not medically required, are harm-
ful, or are "morally repugnant." Recognizing physician autonomy in these circumstances
seems plausible because it preserves a sphere of physician discretion and control over
their practice of health care. To put it in Kantian terms, respect for professional autonomy
prevents health care providers from being used as mere means by patients or society in
ways that violate the health care goals of their profession. But there are problems, and it
is not clear that the attempt to reconcile autonomy and beneficence is entirely successful
or that it provides an adequate direction on other matters related to professional respon-
sibilities and duties. It will be useful to think about these potential limitations as you read
the Pellegrino article.
Consider, in particular, that if we respect a patient's role in determining what is in
their best interests, it will often be unclear when physicians are entitled to have the final
say in determining whether a treatment is not medically indicated or is too risky to be
undertaken. Whose views matter most in these circumstances? At what point does the
threat of harm from a requested treatment entitle a physician to say "no"? Two points
are particularly relevant. The first is that health care providers have to acknowledge that,
although they are typically in the best position to judge the positive and negative effects of
a treatment, they can also be mistaken. They may misjudge the medical efficacy of a treat-
ment, its psychological benefits, and the extent of risk for a particular patient. A second
point is that health care providers should acknowledge that patients are typically in the
best position to judge whether the anticipated positive benefits of a treatment are worth
accepting the risk of harm from treatment, though patients too can make mistakes and
regret decisions. This leaves a large area of uncertainty over when physicians and other
health care providers can legitimately say "no" when they judge that a patient's requested
treatment is not medically indicated or is harmful. This is an unsatisfying position. The
uncertainty over when to say "no" can present a "back door" to paternalism and the old
Hippocratic approach. Or it may lead to physicians feeling obligated to support or carry
out treatments that are inappropriate.
Pellegrino's discussion also does not address several key issues with respect to profes-
sional responsibility. Can health care providers say "no" to patients' requests for access
to scarce resources so that these resources can be used in cases where they have a better
chance of producing medical benefits? What obligations do health care providers have to
say "no" to patients who may pose threats of harm to others, including other patients or
health care providers? These are issues that must be addressed in developing a conception
of professional responsibility and professional autonomy. They are addressed in some of
the remaining articles in this chapter.
Futility
As one practical application of professional autonomy, Pellegrino talks about the import-
ance of recognizing physicians' right to say "no" to "medically useless or futile treatments."
Intuitively, this seems plausible, and health care providers often use the notion of futility
Futility
as a ground for refusing requested treatment. We saw it employed in the Rasouli case in
Chapter 1 to justify withdrawal of life support. Futility is also used to withhold CPR or
organ transplants or chemotherapy for cancer, to name some other instances. But there is
great controversy over whether futility is a useful clinical concept that can be legitimately
used to refuse treatment. Robert D. Truog, Allan S. Brett, and Joel Frader's "The Problem
with Futility" lays out the problems succinctly (see page 176).
Truog et al. point out that the notion of futility is typically laden with, and is used to
obscure, assumptions about values—for example, that short-term survival is not worth
pursuing. In the Rasouli case, this was extended to arguing that a low quality of life was
not worth preserving. Rather than reflecting any notion of futility, the Rasouli case and
others like it seem to reflect a conflict of values between what physicians and what patients
or their families regard as a life that has value. If so, what is at stake is not a clear question
about whether a treatment is futile but what sort of an existence or life a patient could find
meaningful and what risks are worth taking to pursue such a life. If we accept Pellegrino's
argument that patients are typically in a position to know their own best interests, phys-
icians must take very seriously patients' views about the value of survival or low quality
of life and engage them or their families in discussion about them if they feel they are
mistaken. The Truog et al. article raises important questions about whether futility is used
to circumvent such discussions.
The authors of the article also point out that it is rarely true that treatments offer zero
chance of success. If so, it is reasonable to ask whether treatments with a very low prob-
ability of success can be regarded as futile. What often seems to be happening in these cases
is that health care providers are reluctant to treat because it wastes scarce resources or is
likely to cause more harm to the patient than benefit. Are such considerations really about
futility? Aren't they really about whether patients understand the risks of treatment or
whether scarce resources should be used where there is little prospect of medical success?
If the answer is "yes," futility is used to override discussion about what is most directly
relevant and at stake.
Truog et al. suggest that the few cases where a treatment offers no chance whatsoever
of producing a benefit for the patient (what they call "physiological futility") represents an
"ethically coherent ground" for invoking the concept of futility. However, even this limited
defense of futility can be challenged, for patients may still derive psychological benefits
from being treated or may rightly believe that health care providers can be wrong in their
assessment of physiological futility. If there is psychological benefit with no significant or
overriding risks or costs, is that not a reason to treat? Why, for example, should a terminally
ill patient's hope be denied in such circumstances?
This article by Truog et al. has contributed to a substantial and productive literature on
futility Their scepticism about the value of the concept of futility is now widely accepted
among biomedical ethicists, though it remains in common use clinically. An interesting
question not discussed in the article is whether health care could do without the concept
of futility completely.
Gatekeeping
"Gatekeeping" refers to the role that health care providers play in determining who has
access to health care. Gatekeeping can be formulated in negative or positive terms. Nega- gatekeeping the role played
tive gatekeeping refers to obligations health care providers have to prevent patients from by health care providers in
determining who has access
receiving treatments or tests. In this situation, health care providers stop patients from "get-
to health care services.
ting through the gate" into the health care system. Positive gatekeeping refers to obligations
health care providers have to get patients through the gate and into the system.
Obligations of negative and positive gatekeeping are inevitable parts of any health
care system. Avoiding unnecessary tests or treatments are accepted examples of negative
II 131 nurI
162 4 Professional Ethics
gatekeeping. Since treatments and tests have costs and typically involve some risks, this
is usually good medical practice. Getting people who need treatments or tests into the
system poses no significant moral problems in principle either. But both can come into
conflict with other obligations, particularly where resources are scarce. Controversy arises
over what obligations health care providers have to act as gatekeepers to control costs or to
allocate scarce resources to patients who may be in a better position to benefit from them.
Negative gatekeeping is a part of the general problem of fairly allocating scarce health
care resources. This issue is discussed in depth in Chapter 7 in which related issues, such
as refusing scarce treatment on the grounds of responsibility for illness or considerations
of social merit (or lack thereof), are also considered. We introduce discussion of resource
allocation here by focusing on conflicting general duties that allocation decisions can place
on health care providers.
It is often assumed that health care providers owe undivided loyalty to their own pa-
tients. If so, there are no conflicting duties, and faced with a choice they will always decide
to provide a scarce resource to their own patient, regardless of whether this is an efficient
or fair way to allocate scarce resources. In "Should Physicians Be Gatekeepers of Medical
Resources?" Milton Weinstein argues, however, that this approach to health care provider
tragedy of the responsibilities will lead to a sort of "tragedy of the commons." Physicians who act only
commons when each in the best interests of their patients without regard to consequences for the overall health
individual seeks to maximize care system would soon "collectively reach the limits of health care resources," compromis-
personal benefit from a scarce
resource so that demand
ing everyone access to health care, or what he calls the "medical commons." Weinstein
overwhelms supply and the argues that some form of collective action is necessary to limit the power of physicians to
resource is no longer available act solely in their own patients' interests. He rejects proposals for having physicians do
to some or all. cost—benefit analyses at the bedside to determine whether their patients are entitled to
scarce resources and argues that the only practical approach that would preserve relations
of trust between patients and their health care providers is to impose some sort of external
constraints (such as rules of rationing) on the ability of health care providers to act in the
best interests of their patients. The physician would then be free to act in the best interests
of the patient but within the constraints of the system.
Weinstein's proposal puts a substantial burden on health care providers to develop rules
of rationing and to manage the expectations of patients, raising questions about preserving
relationships of trust between health care providers and patients and about the stability of the
system. An alternative not discussed is to have a more publicly accountable system of rationing
to establish constraints. This could involve physicians, elected officials, civil servants, hospital
administrators, and patients' and citizens' groups working together in a transparent way to
solve these problems. This is arguably how a public system of health care like Canada's could
operate. It suggests an advantage of public systems of health care over the mainly private
American system that Weinstein's proposal is largely directed toward. A largely public system
of health care can involve all stakeholders in an open, accountable process for developing a
gatekeeping system of rationing scarce resources and managingcompetingclaims on resources.
Perhaps a primarily public system of health care would have a greater chance of
gaining legitimacy regarding professional obligations and citizens' expectations around
rationing and gatekeeping among both health care providers and citizens. But the problem
with a public system like Canada's, implicitly noted by Weinstein, is that with relatively low
personal insurance costs and little or nothing in the way of user fees, the cost of medical
care is not a factor for patients, which may encourage excessive use of health care at the
expense of the medical commons. This suggests that carefully written rules of rationing
would be necessary or the burden would be back on health care providers to make ra-
tioning decisions on a case-by-case basis. But even carefully written rationing rules may be
at odds at times with physicians' fiduciary obligations to do what is best for their patients.
Such conflicts may be more acute within a system that includes no disincentives to patient
access. As well, politicians who make short-term calculations about what policies give
Gatekeeping 163
them the best chance to be elected, or re-elected, may also be tempted to support or ignore
loose enforcement of rationing.
A possible lesson to be learned from reflecting on Weinstein's discussion is that it is
unrealistic and unfair to expect health care providers, either individually or collectively,
to bear the main burden of gatekeeping or rationing society's scarce health care resources.
Another lesson is that it is one thing to establish publicly agreed upon expectations and
standards around gatekeeping but it is crucial that the standards themselves be stable and
reasonably self-sustaining. Escalating costs of health care, especially cost overruns, suggest
that Canada has not yet found a way to achieve stable and self-sustaining rationing of
scarce health care resources, despite having a largely publicly run and accountable health
care system. A potential solution that is suggested in Weinstein's discussion would be to
have Canadian patients bear more of the direct cost of their care through deductible pay-
ments or user fees, though these are widely prohibited now: Another proposal would be
to offer positive incentives for prudent use of health care resources—for example, through
rebates on insurance fees.
Harm to Others
Patients can request treatments or services that can put other patients or health care provid-
ers at risk of harm. Such risks can include verbally abusive or violent behaviour or risks of
transmission of communicable diseases.
Any admission to hospital puts one at a more than everyday risk, and in consenting
to the admission the patient accepts that. Even so, it is plausible that patients in hospitals
accept only minimal levels of risk. This might be, roughly, the level of risk a reasonable
patient would expect other patients to accept in order to permit that patient to receive
treatment and care. This would include the stipulation that measures have been taken to
minimize risks to patients from other patients. This is one clear professional obligation of
health care providers. But it is less clear what the obligations of health providers are to pro-
vide care to patients who present more than minimal levels of risk to health care providers.
What levels of risk from patients are health care providers required to assume? When are
they at the limits of their professional obligations to treat and care for others?
It is reasonable to believe that health care providers accept greater levels of risk than
do patients. After all, they have consented to become members of professions that involve
caring for patients with communicable diseases and for mentally ill patients who may be
abusive or violent. But it is not clear what the upper levels of professionally required risk-
taking are. It is easiest to argue that high levels of risk have been accepted when health care
providers have specifically consented to work in high-risk areas, such as emergency rooms
and mental health wards. But even when more than minimal risks are accepted by direct
consent, it is reasonable to think that health care providers will have conflicting obligations,
say, to family members—and of course they have conflicting considerations of self-interest.
One attempt to address these issues is given in the Canadian Medical Association's
policy statement "Caring in a Crisis: The Ethical Obligations of Physicians and Society
during a Pandemic" (see page 188). The CMA policy, written in the aftermath of the 2003
SARS epidemic in Toronto, recognizes a qualified obligation to provide care in a pandemic.
It refers to its Code of Ethics as stating that physicians should "provide whatever appropri-
ate assistance you can to any person with an urgent need for medical care." But the policy
also recognizes the role of a principle of consent in assessing what risks physicians can be
expected to undertake. It says "the concept of personal autonomy allows physicians some
discretion in determining where and how they will practice medicine," presumably by
choosing specialties and practices that expose them to lower levels of risk. Arguably, this
means that health care providers cannot be straightforwardly compared to public safety
II al 1111111
workers like police and firefighters, who have less discretion about accepting risk by virtue
of their choice of profession. So there is a conflict that reflects an uncertainty about what
level of risk health care providers are obligated to assume. The CMA policy does not resolve
that uncertainty. Instead, it discusses the "reciprocal obligation" society has "to support and
compensate" physician and health care workers who "will be expected to put themselves
directly in harm's way, and to bear a disproportionate burden of the personal hardships
associated with the pandemic." This societal obligation includes making sure that health
care workers have adequate resources for responding to the circumstances and that com-
pensation is available for health care workers who are injured, including for their families
in the case of death.
The CMA policy is instructive for the way that it recognizes the difficulty of assessing
the limits of health care providers' obligations to assume risks associated with their profes-
sion. It is also useful in clarifying that society must support those who take on such risks if
it expects them to assume them. How strong the professional obligation is to accept those
risks, even when those supports are clearly in place, remains unclear.
Conscientious Objection
Conscience-based objections to providing requested treatment are an accepted, albeit
controversial, feature of the provision of health care. Professional governing bodies today
recognize that health care providers may "conscientiously object" when patients request
treatments that are permitted by the standards of their profession but conflict with a health
care provider's moral values—for example, when a health care provider receives a request
for physician-assisted dying or an abortion. Currently, the main ethical controversy arises
over what should be done to provide care for patients in such circumstances. Does the
conscientiously objecting or refusing health care provider have a duty to refer the patient to
a health care provider who is willing to provide the care? The ethical argument against this
conscientious objection/ is that if conscientious objection or refusal is accepted, it must also be wrong to require
refusal when health objecting health care providers to make such referrals because this makes them complicit
care providers refuse on
in a practice they find morally objectionable. But, equally, it seems inappropriate to leave
moral grounds to provide a
treatment that is permitted
patients to their own devices, since this can add stress, create a sense of abandonment, and
by the standards of their delay treatment. Also, patients are sometimes not able to act on their own to seek help from
profession. other health care providers.
Issues around conscientious objection by health care providers and institutions are
likely to emerge prominently in Canada in the coming years in light of the Supreme Court
of Canada decision in 2015 recognizing a right for some competent adults to physician-
assisted dying (PAD). The ensuing federal legislation codifying this right recognizes that
health care providers can refuse to participate in PAD, but it does not say anything about
a requirement to refer. Health care providers' professional bodies are currently struggling
with this issue, although reasonable accommodation of some rights to conscientious objec-
tion is universally accepted by all health care professions in Canada.
Udo Schuklenk and Ricardo Smalling's "Why Medical Professionals Have No Moral
Claim to Conscientious Objection Accommodation in Liberal Democracies" (see page 191)
reviews and rejects classic defences of conscientious objection and the Canadian federal
government's, the Supreme Court of Canada's, and health care professional associations'
acceptance of some role for conscientious objection in health care. Schuklenk and Small-
ing believe that it is inappropriate to accommodate physicians' conscientious objections to
provide medical services that are uncontroversially legal and that patients have a right to
receive. Their argument is multi-faceted, but the core of it rests on a consideration of phys-
icians' voluntary acceptance of the scope of practice of their profession, which is evidenced
by their decision to join the medical profession. They note as well the arbitrary nature of
which conscientiously held beliefs are or are not thought to be worthy of accommodation
and the potential negative effects on patients from such accommodation. Schuklenk and
Smalling's article is provocative and sure to encourage debate. It is worth asking whether
they are tolerant enough of pluralism in a liberal democracy and how their views would
affect the composition of health care professions.
Professional Ethics Codes

It is natural to look to health care professionals' ethics codes to resolve moral problems like the
ones discussed above. There is general consensus, however, that professional ethics codes are
of little use in these sorts of cases. In "The Functions and Limitations of Professional Codes of
Ethics," Dale Beyerstein explains why. In brief, ethics codes do not solve ethical problems or di-
lemmas because those problems reflect uncertainties about the application of the codes. That is
why they are problems! At this point, we must look to deeper moral and often non-moral con-
siderations to solve them. Beyerstein argues that ethics codes nevertheless fulfill other import-
ant purposes. They give general guidance to health care professionals; they reflect agreed upon
and publicly stated standards of practice and so inform members of the public about expected
standards; and they inform members of health care and other professions roughly about what
to expect the limitations of cooperation will be when working with each other. This latter point
reflects an issue that has not yet been touched upon—namely, that the ethical responsibilities
of health care providers include the duty to say "no" when they are asked by other health care
providers to do things that are in conflict with the standards of their profession.
Beyerstein's article is an apt coda to this chapter. Professional ethics and the duties it
imposes on health care providers are filled with problems and uncertainties. It challenges
us to think deeply about pressing moral problems. An optimistic view is that over time,
many of these challenges will be worked out in policies and practices and incorporated
as assumed elements of professional ethics and ethics codes. But even if such optimism is
warranted, new problems are bound to arise, or boundaries will be pushed back only to
reveal other problems.
4.2 The Nature and Limits of Professional Autonomy and

Professional Responsibility
Patient and Physician Autonomy: of equal or greater weight than beneficence.' Progr-
essively since then, patient autonomy has become
Conflicting Rights and Obligations in the dominant principle shaping physician—patient
the Physician-Patient Relationship relationships.
Three serious moral conflicts have emerged as a
Edmund D. Pellegrino
result: first, beneficence and autonomy have been
polarized against each other when they should be
For centuries, physician beneficence went unchal- complementary; second, the physician's moral claim to
lenged as the first principle of medical ethics. To be autonomy has received little attention; and third, the
sure, some physicians had, at times, violated this "autonomy" of medical ethics, itself, has come under
principle. But no creditable ethical opposition was a serious threat. This essay will examine each of these
mounted until a quarter of a century ago when patient three consequences resulting from the rise of patient
autonomy was asserted as a prima facie moral principle autonomy. It shall do so from the point of view that the
Source From Journal of Contemporary Health Law & Policy 10, (1994) 1,47-67. Reprinted with permission.
II LE 111111
physician—patient relationship is a moral equation with their powers of choice, and choice is a distinctly human
rights and obligations on both sides and that it must activity. To obstruct the capacity for autonomy is to as-
be balanced so that physicians and patients act benefi- sault an essential part of a personb humanity because
cently toward each other while respecting each other's the choices we make are so much an expression of our
autonomy. Effecting this balance is a morally manda- membership in the human community, of who we are
tory and exacting exercise. The compass points that or what we want to be as individual members of that
might guide this balancing are to be found in a reflec- community. Human beings are owed respect for their
tion on the concepts of autonomy and beneficence, the autonomy because they have an inherent dignity. They
way the content of these abstract notions is provided do not have dignity because they are autonomous.
by the clinical encounter, and the way conflicts may be Human beings who lack or have lost the capacity for
resolved in particular clinical situations. autonomous actions are nonetheless humans who
Out of this reflection, five conclusions will retain their inherent dignity. Respect for persons com-
emerge: (1) in concept, autonomy and beneficence are prises more than respect for autonomy.
complementary and not contradictory; (2) in theory Autonomy has taken on a distinctive negative con-
and in practice, autonomy is a positive as well as a nega- notation. Arising, as it did, as a moral claim against in-
tive principle; (3) the actual content of the concepts of vasion of human rights by tyrannous government, it has
beneficence and autonomy is defined in specific actions come to mean a right of self-determination against those
and decisions in the light of which conflicts are best who would usurp that right. In medical ethics, it is con-
understood and resolved; (4) the physician's autonomy ceived largely as a moral and legal defense against phys-
as a person and a professional must be factored into the ician paternalism and against those who would impose
equation; and (5) medical ethics, as an enterprise, must their values—social, moral, or otherwise—on others.
maintain a certain "autonomy" in the face of political But autonomy is also a positive concept. It im-
and socio-economic pressures. plies an obligation to foster the human capacity for
self-determination, to enhance it, and to remove the
I. The Concept of Autonomy obstacles to its full operation. This is especially import-
ant in clinical medical ethics where pathophysiological,
A. Autonomy in General emotional, and social realities complicate the actualiza-
Autonomy is one of those widely applauded concepts tion of patient autonomy. If taken as a strictly negative
which, on closer inspection, turns out to be difficult concept of non-interference, autonomy can be self-
to define with precision.2 This is not the place to defeating for patients and self-serving for physicians.
review the range of construals of the term. Rather, I This positive aspect of autonomy will become clearer as
will limit myself to that construal which centers on the I fill in the content of the concept as it operates in the
etymology of the word itself, which means "self-rule." clinical situation.
What is common to most definitions is the notion that
an autonomous person is one who, in his thoughts, B. Patient Autonomy and Physician Beneficence
words, and actions, is able to follow those norms he Twenty-five years ago, the political and moral notion
chooses as his own without external constraints or of autonomy was appropriated as one of the prima
coercion by others.3 facie principles of medical ethics.' There were good
The history of the concept is complex, and its reasons for the emergence of patient autonomy at that
roots are political as well as moral. Politically, auton- time. The rights of patients to refuse unwanted treat-
omy came into prominence during the Enlightenment ment had been neglected for entirely too long. In the
as an assertion of the individual's right to be free from mid-sixties, these rights could no longer be denied as
tyrannous government—not of law per se, but of unjust participatory democracy, better public education, and
law.4 Morally, autonomy encompasses the right of per- the civil rights movements became realities. All author-
sons to freedom of conscience and to respect as agents ity claims came under suspicion. The abuses of profes-
capable of making their own judgments in accord with sional and bureaucratic power were widely publicized
universal moral principles,' or in accord with freely ar- and no longer tolerable. Moreover, the unprecedented
rived at decisions.6 powers of medicine made the choice of medical treat-
Autonomy gets its status as a moral right of ments a far more significant matter than it had ever
humans from the fact that human beings have the been in the past.
capacity to make rational judgments about their own The principle of patient autonomy was seen
lives, choices, and interests. Self-governance deserves as the patient's protection against usurpation of his
respect because it is the way human beings actualize right to participate in decisions that affected his life.
Pellegrino: Patient and Physician Autonomy
This amounted to a denial of the long tradition of med- As Dworkin points out, autonomy implies a "cap-
ical paternalism (or parentalism), which considered acity to reflect upon one's motivational structure and
the duty of physicians to decide what was best because make changes in that structure."n Without associat-
the patient lacked medical knowledge and might lose ing with others and drawing on their preferences and
hope if he knew the whole truth about his options values, we lack the opportunity to alter or reaffirm our
or prognosis. values because we do not know what alternatives are
Since paternalists acted in the name of benefi- available and why they might be preferable. To move
cence, beneficence was equated with paternalism and from the abstract realm of concept to actual decisions,
thereby came to be interpreted as a counter-principle autonomy needs content, and this comes from reflec-
to autonomy. Morally valid and invalid forms of benefi- tion not only on our own past values but on the values
cence and autonomy were not distinguished from each of others at the moment of choice. It is the physician's
other. The dilemmas of medical decision-making soon obligation to enhance, empower, and enrich the pa-
were reduced to weighing the principles of autonomy tient's capacity to be autonomous. An autonomous
and beneficence against each other.8 choice requires that we fill in, to the extent possible,
Medical paternalism and parentalism, however, the action or choice that maximizes realization of the
are not to be equated with beneficence, conceptually patient's values. Thus, autonomy has a positive as
or in practice. Paternalism does not account for the well as a negative aspect. To become a reality, patient
patient's preferences or values that are part and parcel autonomy requires cooperation and assistance from
of her good or best interests.9 Paternalism makes the the physician. In short, it requires the physician's be-
medical good of the patient the only good and sub- neficent attention to make the patient's autonomy an
verts other goods to that good. Paternalism violates authentic, as well as an independent, reality.
the patient's autonomy in the name of the patient's
best interests while ignoring or overriding some of the C. Physician Autonomy
most vital of those interests. This cannot be a benefi- In all the current discussions about the moral status
cent act because the patient's own choices are so much of patient autonomy, the autonomy of the physician
an expression of his or her own life story or person- is often neglected. This philosophy has serious de-
hood. To violate or ignore the patient's choices is, by fects. The physician—patient relationship is one of
definition, a maleficent act, an injury to the patient's mutual obligation—like any truly ethical relationship.
humanity. Only when the patient's human capacity to The physician as a human being has the same claim
act autonomously is impaired (i.e., when the patient to respect for his or her capacity to make personal
is incompetent) may we resort to paternalism as a be- choices, to follow his or her conscience about what
neficent act to override objections to treatment.10 is good medicine and what is morally acceptable as a
This is the negative aspect of autonomy. Important person. Personal and professional ethics are not fully
as it is, it is a distortion of the idea of autonomy to separable from each other. Therefore, the patient's
equate it with total independence from the physician moral right of autonomy must be balanced with respect
or others in making treatment decisions. The cultural for the physician's autonomy. Autonomy cannot be a
bias against dependence or even the semblance of unilateral moral right for either patients or physicians.
dependence is strong in American life. However, total Physician autonomy may be considered under
independence is unrealistic in any walk of life. Human three headings: (1) autonomy as a person, which gives
beings live in community and personal association, moral status to the physician's personal moral values
especially when they are patients. Patients especially and conscience; (2) autonomy as a physician, which
need the input of others if their own choices are to be gives moral status to the physician's knowledge and
genuine ones. Physicians are needed to provide infor- obligation to use it wisely and well; and (3) autonomy
mation and to discuss this information with patients as a member of a profession, of a moral community
to enable and empower them to use their autonomy with collective obligations to patients and society.
wisely. Patients must compare their values with those I have written elsewheren of the moral obligations and
of others in the context of some community of belief the autonomy of medicine as a moral community, and
which they accept in whole or in part. Patients cannot will confine myself here only to the first two construals
identify with their current choices without reference of the physician's autonomy as an individual.
to some structure of values which they formed in the The autonomy of the physician as a person has its
past and which they reaffirm or reject at the moment of roots in the same ground as the autonomy of any other
choice. This is part of knowing ourselves, and we know person (i.e., the physician's capacity as a person for
ourselves largely in relation to others. rational judgment and expression of preference with
II 11111 Mid
respect to values and choices). The physician, therefore, provides essential factual data on which rational social
cannot be expected to lay aside or ignore his deeper policy should be based. But the actual choices of values
personal beliefs, values, or religious commitments. To are not the prerogative of physicians or any other
be sure, patient autonomy requires that the physician "experts"—politicians, economists, or even ethicists.
not impose his values in his decisions for the patient. Social value questions are a matter of concern for the
But patient autonomy cannot require the physician to whole of society In this respect, the expert is like any
sacrifice his personal moral integrity even for what the other member of society with no authority over the
patient may believe to be a morally good purpose. values of other individual members of that society or
Respect for the physician's autonomy also derives over the society as a whole. The arguments of experts
from the fact that, under normal circumstances, the may have more cogency but no more authority than
physician must write the orders that are carried out by those of others.
others. The physician cannot avoid the fact that she is
the focal point through which harm and benefit of a II. Autonomy: Its Content in the
clinical decision will flow in a majority of cases. The Clinical Context
physician therefore is a de facto moral accomplice in
what happens to her patient. She cannot place respon- Autonomy in general, and physician and patient au-
sibility on others for morally indefensible decisions tonomy in particular, might conceivably be defined
or for cooperation in decisions that violate her con- in the abstract in some general way congenial to a
science. This inescapable fact of the physician—patient large number of people. However, when we begin to
relationship places unavoidable obligation on the give it content in the context of illness, the problems
physician to avoid action she deems harmful to her with absolutism and abstraction become evident, as
patient, even if that action is "required" by state regula- do the implications of autonomy as a solely negative
tion, policy, or law." moral right.
The physician's autonomy as a physician is also First of all, no two persons experience illness the
grounded in the possession of expert knowledge same way. No two persons have the same way of ex-
needed by sick people and society. The power, itself, pressing their capacity for autonomous choice. Very
conferred de facto by the possession of such knowledge, few patients demand only "the facts." Some will seek
demands that the physician be free to use it according a wide variety of opinions before deciding on their
to her best judgment.'4 If the physician is to fulfill the own; some will not. Some patients will prefer to ex-
moral requirement to make her knowledge available to ercise their autonomy by giving it up to a surrogate
those who need it, she must be allowed sufficient dis- (i.e., someone they trust to make the decision for them,
cretionary latitude to apply that knowledge as ration- or perhaps even the physician). The majority will want
ally, efficiently, and safely as possible. This is essential if to express their own way of being autonomous by
physicians are to fulfill their part of the covenant with asking not only for facts but also for the doctor's opin-
society and with individual patients. Physicians enter ion and the opinions of family and friends.
this covenant from the first day in medical school, when Thus, the content of the idea of autonomy, when
they accept the privileges and the obligations that go it is actualized, will vary with the patient's prior values
with the acquisition of medical knowledge and skill.15 and cultural, personal, and social relationships. These,
Clearly, this third sense of physician autonomy can in turn, condition a patient's response to illness. Sick-
never be absolute. If the physician is incompetent, acts ness forces a confrontation with the self and with the
in his own self-interest, or acts paternalistically in the need to adapt to this illness, here and now. Sickness is a
sense I defined earlier, he misuses his expertise and test of our values. For each of us, our response to sick-
violates his covenant with both the patient and society. ness is unique, and thus the way we express our auton-
That covenant is based on trust in the doctor's Oath omy is also unique.'6 Patients will vary in the degree of
which commits him to use his knowledge primarily in dependence or independence they desire depending on
the service of the sick. their relationship with the physician, on their relation-
The physician's autonomy as a physician is also ship to their society or community, and on the degree
limited when she mistakes medical expertise and au- of trust they impute to others.
thority for expertise in questions of values. The phys- Second, no matter what degree and kind of auton-
ician has no standing as an expert in human values omy a patient chooses, the very fact of illness physio-
and no authority to set the goals or priorities of public logically or psychologically compromises the actual
policy relative to the allocation or distribution of health expression of autonomy to some degree. The sick
care resources. To be sure, the physician's knowledge person is dependent on the physician's knowledge and
help; otherwise she would not need or seek medical spectrum are not difficult to identify, no one can draw
help. In addition, in varying degrees, she is in pain, precise lines between advice, persuasion, manipula-
anxious, fearful, and vulnerable. Brain function may tion, and coercion. But the difficulty of drawing a line
be temporarily or permanently compromised by fever, does not justify a presumption in favor of paternalism.
shock, medication, age, or dementia. Rather, it increases the physician's obligation in benefi-
To restore autonomy, physicians must first attend cence to protect autonomy by the most scrupulous self-
to reversing these physiological and psychological examination of his own motives in obtaining consent.
impediments to the optimal exercise of autonomy. In Much, therefore, still depends on the physician's
such cases, medical treatment is essential to restore character and sensitivity and her possession of the
autonomy. This may require temporarily downplaying virtue of benevolence. The physician's character may
or overriding the patient's autonomy until normal turn out to be the last safeguard of the patient's au-
sensorial states are attained and then enhancing and tonomy and well-being. But, ultimately, the physician
empowering it as the capacity for self-determination and patient must decide together what is to be done.
returns. During this transition, beneficence (i.e., acting Only in this way can patient autonomy become a
in the patient's best interest) modulates the physician's cooperative and beneficent enterprise, rather than an
move from "weak" paternalism to enhancement of the adversarial one.
patient's full autonomy. All of this applies with special force to surrogate
Third, no matter what degree of autonomy a pa- decision-making and to advance directives, which
tient may want or in what way he wants to express become operative when a competent patient loses the
it, the patient is vulnerable to deception in the infor- capacity to make his own decisions. Here, the patient's
mation he receives. The patient is dependent on the wishes are represented by others or by a written docu-
physician's disclosure of diagnosis, prognosis, treatment. The surrogate's wishes have the moral status we
ment options, side effects, effectiveness, outcomes, usually attribute to a competent patient and should be
etc. Which facts the physician chooses, which she respected as such. However, family and friends can be
emphasizes, and which she represses are often subtly in a financial or emotional conflict of interest with the
or frankly conditioned by her judgment of what she welfare of an incompetent patient. They may even wish,
thinks is in the patient's best interests. As any clinician consciously or subconsciously, to relieve themselves
knows, she can get almost any decision she wants from of the emotional and physical burdens of caring for a
most patients. Therefore, even the most conscientious chronically ill person. Their representations of what the
physician must exert great care to avoid manipulating patient's autonomous decision would have been were
the patient's choices, even for good reasons. The fact he competent are open to serious question.
that physicians can so easily influence the patient's When "autonomy" is expressed in a living will or
choice makes the full operation of patient autonomy other advance directive, an assessment must be made
problematic. For that very reason, it is morally incum- of whether the decision executed in the past, when a
bent on the physician to protect patient autonomy as person was competent, represents what the patient
scrupulously as possible and to try to help the patient would want now, when the patient is no longer compe-
realize its positive content. tent. Is this person, now in a persistent vegetative state,
This cannot mean, as some erroneously argue, the same person who originally made out that living
that autonomy in a sick person is a fiction, that to try will? Is autonomy, in its full meaning, so absolute that
to enhance it is a sham, and that we should return to it binds us to decisions the benefits and the import of
the Hippocratic tradition of benign authoritarianism. which the patient could not possibly have anticipated
Such a reversal would be an intolerable suppression of and which, in the actual context of a particular deci-
the patient's human right as a rational being to make sion, may not be in his present interests?
uncoerced choices. Physicians and others, therefore, In these circumstances, the patient is in need of a
have an obligation not to take advantage of the patient's beneficent agent—one who can be trusted to protect
vulnerability. Informed consent is an empty notion or a him or her from the autonomous decisions of others,
charade if the information on which it is based is biased even those who might be legally, but not morally, valid
in favor of the physician's preferences. surrogates. This agent may have to be the physician,
None of this means that physicians cannot advise nurse, or other health professional who acts under
or persuade patients to do what they think is right. Not the principle of beneficence. Regardless of whether
to do so is a species of moral abandonment. Patients the surrogate is a family member, friend, or the health
are entitled to know what physicians think is "best," professional, when the patient's autonomy has been
all things considered. Although the extremes of this transferred to others, it must be held in trust. If that
Hal 1111111
trust is violated, the surrogate loses her moral status as health professionals is the psychological burden family
well as her "autonomy" to make choices for the patient. surrogates carry when they must decide whether to
Not enough attention has been given to the lim- discontinue life support measures. Here, the autonomy
itations placed on the "Western" notion of autonomy owed the patient is transferred to the valid surrogate.
when applied to the care of patients from different In such cases, families often feel they are being asked
cultural backgrounds. Even in the "West," as Surbone to sentence a loved one to death or, by their decision,
points out in the case of Italian culture, patients may are actually participating in the death of the patient.
not expect or want to make decisions, preferring to They need reassurance when the medical situation is
leave them to families or the physician.17 Is it a benefi- one which, in the eyes of the physician, is "hopeless."
cent or maleficent act to insist on or offer autonomous Physicians cannot simply leave the entire burden to
decision-making in these circumstances? Is it morally the surrogate or even the patient. They must share that
wrong, or rather a legitimate compromise, to work burden.
within the patient's cultural confines? Are we compelled Thus, detachment is not a beneficent act. Often,
by the fundamental nature of the principle of "respect when families or patients ask that "everything" be
for persons" to impose our view of autonomy, or may done, they are seeking reassurance that everything that
we compromise it in the name of cultural integrity? could be effective or beneficial be done, not that "every-
These nuances in the full expression of patient thing"—irrespective of probabilities of success—be
autonomy in clinical decisions underscore the fact that done. They also want to share their responsibility for
autonomy cannot function in actuality without benefi- cessation of life-support with the doctor. The focus of
cence. Beneficence, properly exercised, is the guarantor ethical concern may well shift, however, from the sub-
of autonomy, rather than its enemy. Enhancing auton- stantive to the procedural when irreconcilable conflict
omy enabling and empowering the patient to make her about what constitutes beneficence or the patient's best
own choices, and helping the patient to understand the interests occurs.
choices before her in terms of her own past values are
all acts of beneficence. These acts enhance the positive Ill. Challenges to Physician Autonomy
content of autonomy and are crucial to any compre-
hensive notion of the patient's welfare. On this view, A seriously neglected facet of the growing dominance
the health professional holds the patient's autonomy in of patient autonomy is its impact on the physician's
trust. While it must be protected, it cannot be divorced autonomy. The physician—patient relationship, like
from beneficence. It requires the physician's involve- any ethical relationship, is a reciprocal one. In the
ment, not her disengagement. In short, if the positive justifiable concern for patient autonomy, it is easy to
content of the concept of autonomy is to be realized, it forget that the physician is a moral agent as well as the
will require beneficent action on the part of the health patient. As such, the physician's autonomy, as well as
professional. the patient's, is deserving of respect. When the two are
I must emphasize this point because the current in conflict, the patient's wish does not automatically
pressure to assure patient independence is eliciting trump the physician's. The physician's autonomy, like
two morally dangerous responses on the part of health the patient's, has its negative and positive construals.
professionals. One response is to emphasize the nega- It may seem paradoxical to worry about physician
tive non-interference dimension of autonomy. This autonomy when it is the patient who is vulnerable and
negative conception of autonomy reduces the ethics of the doctor who holds the knowledge and power the
the physician—patient encounter to procedure rather patient needs. This fact rightly imposes the heavier
than substance. On this view, as long as the procedure moral burden on the physician in the equilibration of
allows for autonomy all is well. Autonomy is absolu- the autonomy relationship. He cannot use his claim
tized in principle and practice. This may lead to the to autonomy to violate the patient's capacity to make
second response, namely, that physicians will accede self-governing choices. But the physician is, like the pa-
to whatever the patient or valid surrogate wants. This tient, a human being, entitled to respect for his capacity
prompts the physician to transfer all responsibility to to reason, judge, and make choices that are authentic-
patients, family, or friends. This occurs with alarming ally "his." He cannot impose his values on the patient,
frequency in the care of infants, the elderly, and de- just as the patient cannot impose hers on the physician.
mented patients, who may be over- or under-treated The physician—patient relationship is a moral equation
because their surrogates demand it. with reciprocal rights and obligations.
Indeed, one of the most important contributing Today, that equation is becoming unbalanced as
factors to disagreement between family surrogates and patient autonomy is elevated to the status of a trumping
principle, morally as well as legally. For some, this even an objective determination. They argue that the idea of
implies or includes overriding the physician's values, futility is so freighted with both the patient's and phys-
his discretionary latitude in clinical decisions, and, in ician's values that it should be abandoned entirely.24
some cases, even his rights of conscience. As patient Others would retain the concept only for obvious situa-
autonomy receives more and more legal sanction, the tions of total brain death, permanent vegetative state, far
problem of preserving the physician's moral integrity advanced malignant disease,25 or when a treatment has
will grow. This danger is accentuated by the deficiency failed in the last 100 cases.26 Still others would institu-
of "conscience clauses," which could provide statutory tionalize the criteria for futility in hospital policies that
protection for physicians who refuse to provide or par- would bind the physician to compliance.27 Underlying
ticipate in procedures they find repugnant on moral or this debate is the challenge to the physician's expertise
religious grounds.18 to determine when a treatment—or all treatment—is
In the United States, these threats to the phys- useless, ineffective, or not indicated because the heal-
ician's autonomy and conscience derive from the ing, caring, or curing ends of medicine can no longer
evolution of autonomy from a negative to a positive be attained. One of the Hippocratic Authors made it an
right. The rights of patients to make decisions consist- ethical obligation for the physician and the patient28
ent with their own values was first noted as a right to to desist from treatment when the limits of medicine's
refuse unwanted treatment in 1914.18 In 1976, it was power had been reached.
extended to a right of valid surrogates to refuse life- This limitation on the clinician's discretionary lati-
saving measures over the physician's objections.2° In tude in the use of medical knowledge and skills is espe-
1983, the President's Commission extended autonomy cially dangerous when dealing with surrogate decisions
to include the right of participation in "Do Not Resusci- for incompetent patients. Do patients or surrogates
tate" orders. This meant that physicians should offer really know what doing "everything possible" means?
treatments (like resuscitation) that patients might want Must we respect orders for "no tubes" or "extraordinary
even if they were not judged medically indicated.21 As measures" when these may well be effective and bene-
a result, the pristine right of refusal of unwanted treat- ficial and might have been desired if the patient were
ment is now becoming, for some, a right to demand now competent? Did the previously competent patient
treatment—even over the doctor's best medical judg- really intend to foreswear such measures? Is it ever pos-
ment.22 Elevating patients' demands for specific kinds sible in a living will or medical directive to anticipate
of care to moral status under the rubric of autonomy what one would wish at the moment of actual decision-
poses several challenges to the physician's right to her making? Must vigorous, ineffective, burdensome, and
own moral integrity. futile treatment be continued because the living will or
First, there is the challenge to the physician's judg- surrogate requires it?
ment of what is good medicine (i.e., medicine that is These difficulties do not vitiate living wills or sur-
rationally sound in diagnosis, prognosis, and thera- rogate decisions by those with a durable power of at-
peutics). For patients to claim a right to any procedures torney for health. They do warrant caution about the
they wish is to challenge a conscientious physician's content of autonomy and its actualization in particular
integrity as a physician. It depreciates his expertise, cases in which there is doubt about what the patient
reduces his discretionary latitude in decision-making, wanted to be done. They alert us to the fact that the
and makes him a technical instrument of another per- physician's discretion can be so restricted by advance
son's wishes. What is more important is that this can directives that the patient's welfare is compromised.
pose a risk to the patient's well-being and subvert the Moreover, a mistaken respect for autonomy or the
healing purpose for which medicine is intended in the physician's fear of violating autonomy becomes an
first place. What is demanded may not be indicated, excuse for moral detachment, which is actually moral
effective, or beneficial. Such demands violate the in- abandonment.
ternal morality of medicine as a practice.23 They can The physician is accorded discretionary latitude
redound to the patient's harm by undermining the in clinical decisions because medical knowledge must
physician's moral obligation to provide sound advice be applied to individual cases. The care of individual
and sound practice and to avoid medically useless or cases is not reduced to a set formula but rather must
futile treatments. be modulated by a host of clinical and personal fac-
This threat is especially pressing today in the tors peculiar to each patient. Without discretionary
debate over medical futility and who defines it. Some latitude, the physician cannot personalize and indi-
would do away with the concept entirely because they vidualize care; she cannot fulfill her obligation to use
consider that defining "futility" is not, and cannot be, her knowledge for the patient's best interests. Without
I I AI MLA
constraints on discretionary latitude, the physicianb cannot be ignored. If legal or societal sanction for a
decisions can violate the patient's values or produce certain procedure becomes widespread, will this war-
physical harm. The balance between too narrow and rant violation of the physician's conscience? Some see
too wide a definition of discretionary space is a deli- these as matters of such societal benefit that the phys-
cate, but increasingly important, one to strike. ician's private moral and religious beliefs should be
Another place where physician autonomy is en- dissociated from his professional life.32 This will pose
dangered is in the sensitive realm of the physicianb an impossible situation for the morally conscientious
religious beliefs. In the future, the secular trend in our Orthodox Jew, Roman Catholic, or Muslim in certain
society and the drive for autonomy may converge to fields of medicine.
place constraints on the physician's religious convic- Another challenge to the physician's moral integ-
tions and values. Current legal literature already re- rity and autonomy is one encountered by physicians
flects instances of subtle, and sometimes not so subtle, in countries with "managed" health care systems. The
coercion of the consciences of nurses and physicians political and economic pressures of health care policy
who oppose or refuse to participate in abortion, ster- and reform already place the physician in a position of
ilization, the use of abortifacient, or to carry out dir- moral conflict. The economic and fiscal drive behind
ectives to withdraw feeding tubes.29 Medical students such programs can make the physician a moral accom-
and residents are under increasing pressure to learn plice in practices he deems injurious to his patientb
and to participate in abortion training by practice. Ap- well-being. No matter what setting he is in—cost
plicants to medical schools are now frequently asked containment, rationing, acting as a gatekeeper, an
about their views on abortion. No solid data are avail- institutional milieu of managed health care or man-
able on whether their answers influence the admission aged competition, a publicly funded clinic adhering to
committee's or interviewers decision to accept or reject clinical guidelines, etc.—all place the physician in the
them. Nevertheless, the question is asked so often position of double or triple agency.
that it seems unlikely to be of only passing interest to The physician's professional commitment to advo-
interviewers. cacy for her patient may put her at odds with an institu-
Fortunately, the right of conscientious refusal on tions or society's well-being. Exigency, expediency, and
grounds of personal beliefs is currently protected.3° economics, not ethics, drive such systems. Does the
However, past statements of official bodies like the "autonomy" of the institution or health policy override
American College of Obstetricians and Gynecologists the autonomy of either, or both, the patient and phys-
with respect to training in abortion techniques as a ician? How are "good" business, economic, political,
condition of residency approval are worrisome, even and fiscal policy and the moral purpose of medicine to
though superseded at the present time.3' be reconciled? Which takes precedence when conflict is
Another possible challenge to physician autonomy unavoidable? Such questions are sure to become more
presents itself in the current debate about voluntary widespread in the future as the zeal for cost contain-
euthanasia and assisted suicide. It seems very likely ment and managerial ideologies, rather than the welfare
in view of the current drift of public and professional of patients, are enshrined in law and public policy.
opinion, that one or both of these practices will become
legal. When this happens, these procedures will also N. Procedural Ethics and Conflict Resolution
very likely become "benefits" or entitlements in our
future health care system. The Clinton Administra- The central moral issues in any attempt to balance pa-
tion is likely to include abortion among "reproductive tient and physician autonomy are substantive. But when
services" in its proposed Health Security Act. Pressures substantive moral issues are unresolvable, procedures
on physicians are then sure to mount to provide abor- for ethically dealing with the conflict are necessary. The
tions as part of the "benefit package." We are promised autonomy of patients, their surrogates, and physicians
that the rights of conscience of those who find abortion all carry moral weight and, on that account, command
morally reprehensible are to be protected. However, in respect. The ethical goal of any procedure aimed at
a climate of moral pluralism, self-determination, and conflict resolution should be to protect each agent's au-
consensus ethics, this could change. tonomy to the extent possible. To this end, a variety of
I cite these examples not to provoke furious de- procedural moves are morally plausible when a conflict
bates about the moral status of the procedures in in moral or professional values reaches an impasse.
question but because whatever one's position may be, To begin with, the patient or patient's valid sur-
the moral problem of the integrity of the physician's rogate can discharge the physician and engage one
autonomy and moral rights of expression of conscience who will take care of the patient on his or her terms.
Alternatively, the physician can withdraw, respectfully participate in state-ordered executions or in coerced
and without recrimination, on grounds of preservation interrogation of war prisoners). Just when and how in-
of his moral integrity. However, these alternatives are dividual physicians should refuse to comply with social
possible only if another physician is willing to under- conventions is not a matter of precise formulation.
take the care of the patient. To withdraw without trans- No one can enter the mind and heart of another and
ferring care to another competent physician is morally untangle the moral psychology of a particular moral
and legally unacceptable and constitutes abandonment. choice, but this fact does not vitiate judgment about
Hence, the question about how the transfer should be the ethical probity of the act in question.
made arises. Some of the most complex and difficult situations
In both situations, patients or surrogates might occur when surrogates are acting for infants whose
claim a right to assistance in selection of another future values cannot be known.34 In such cases, the
physician specifically congenial to their moral values. conflict will often be between the surrogate's and the
In cases that do not involve a fundamental moral physician's estimates of what is "best" for the infant.
principle (e.g., a patient's choice of a "lump" resection The vagaries of "quality of life" estimates complicate
and radiation as against radical mastectomy for breast the issue because of the impossibility of assessing how
cancer or the use of non-standard, but not harmful, the infant would make that evaluation when he or she
medical treatment), such cooperation could be ethic- becomes aware it must live with the infirmities the
ally appropriate. The issue may be more fundamental, physicians prognosticate.
however, when it involves voluntary euthanasia, abor- These infirmities are often complex, may result
tion, physician-assisted suicide, or withdrawal of care in a life of prolonged disability and discomfort, and
from patients in a persistent vegetative state. In such constitute grave emotional, physical, and fiscal bu-
instances, some would argue that the physician who rdens for parents and society. In a society propelled by
withdraws has an obligation to find another physician economic constraint, reluctance to sacrifice material
to undertake the patient's care under the family's or pa- goods even for disabled children, and an obsession
tienth terms. For others, this would constitute an un- with physical beauty, it is not uncommon for even
acceptable degree of moral complicity by cooperation conscientious parents to decide to withhold or with-
in an act one considers morally untenable.33 draw life-sustaining treatment in order to spare the
The most difficult situation, for which there is no infant a "life of suffering" or poor quality. But "quality
totally satisfactory solution, is when physicians and pa- of life" and "value of life" are not synonymous terms.
tients or their surrogates disagree on a serious and fun- The modern tendency to use them synonymously is a
damental ethical issue and the possibilities of physician serious point of discord between secular and religious
withdrawal, or discharge of the physician by patient perspectives on ethical decisions. What is right and
or surrogate, are foreclosed by external circumstances. good in these circumstances is problematic and may
There may be no physician willing to undertake care be interpreted in contradictory ways by physicians,
on the patient's or surrogate's terms. No other physician nurses, parents, and even governmental regulation.35
may be available for reasons of geography or urgency of In these circumstances, we may search for procedures
the clinical situation or lack of the required expertise. which will protect the autonomy of all participants, but
Or, the physician may be employed in an institutional the substantive ethical disagreement remains.
setting (i.e., prisons, the military, certain managed When the physician believes the parents' decision
care plans, or residency training programs) in which would constitute grave harm to the infant, she has
physician choice is limited by virtue of his occupy- several procedural alternatives. One alternative is to
ing a specific, socially defined role. In these settings, withdraw and ask the parents to engage another phys-
failure to perform the expected role (i.e., participation ician. This would be permissible if the physician did
in state-ordered executions) could result in significant not think that withdrawal would result in grave moral
fiscal penalties, discharge from one's job, and legal or and physical harm to the infant. Where the physician
disciplinary action. might judge otherwise, whether treatment is withheld
When there are irreconcilable differences in moral or continued, then that is medically futile.36 At such a
commitments and the physician cannot extricate her- time, the physician has an obligation to take whatever
self by reasons of exigency or limitations imposed by measures are available to avert harm, such as appeal to
patient or society, the physician must still be faithful an ethics committee or, if necessary, to the courts.
to her conscience. This may mean acceptance of the A proponent of absolute parental autonomy might
attendant penalties for refusal to comply with institu- justifiably ask what moral claim a physician can have
tional, legal, or socially defined goals (e.g., refusal to to judge, or even question, a parent's surrogacy rights.
I II 111111
This point of view assumes that parents have absolute medical ethics itself? Why not leave it to be negotiated
dominion over their children and that their decisions between physician and patient? Perhaps medical ethics
will invariably be benevolent and altruistic; it also should be a changing, socially constructed contract
ignores the covenantal trust relationship between the varying from society to society, era to era, and patient to
physician and the patient. For a variety of reasons— patient. Some argue that medicine and its ethics must
pride, shame, or unwillingness to confront the ex- be whatever is negotiated politically between the pro-
pense, financial and emotional, of caring for a disabled fession and government. Other socially and politically
child—parents may decide to undertreat. On the other constructed forms of ethical justification are currently
hand, they may opt for futile overtreatment out of lack popular as well. They imply that there is no such thing
of information, religious conviction, or fear of being in as a universally binding medical ethic, only an ethic of
some way responsible for their infant's death. political expediency or societal convention.
Furthermore, the autonomy imputed to parents Some of us, however, think this would be disas-
cannot be absolute. Physicians and nurses are obliged trous for medicine, the physician, and the patient. The
by virtue of their commitment to the well-being of their autonomy—that is really to say the moral integrity—
patients to act in the interests of the infant. They must, of both physicians and patients must somehow be
of course, appreciate that decisions surrounding the preserved. So, too, must the integrity of the ethics
care of very sick and potentially disabled infants must of medicine itself. Medical knowledge is too power-
involve the family. Indeed, in a real sense, the whole ful a tool to become an instrument of governmental
family becomes a "patient," whose collective interests or social pressures, or private negotiation, however
must be safeguarded. The implications of the decision benign their motives may appear to be. Medicine is
on the future lives of the whole family are, therefore, also too powerful to go wholly unregulated. There are
not to be denied; however, these considerations by too many examples of the subversion of the powers of
themselves do not justify withholding or withdrawing medicine to evil purpose by unjust political regimes
treatment that is effective, beneficial, and not dispro- to make the ethics of medicine a subject for political
portionately burdensome. negotiation.37 There are too many examples of the
When there is obvious and overt conflict between way unregulated medical "entrepreneurs" or morally
the good seen by parents and the medical good of the bankrupt physicians can exploit the vulnerability of
infant, the obligation is greater to the most vulnerable the sick.
person (in this case, the infant). Situations involving Medical ethics must maintain a degree of in-
such irreconcilable conflicts of obligation are some- dependence if it is to protect the sick person. It must
times unavoidable. Still, we are obliged to do as much remain subject to public criticism but not be controlled
as possible to respect the physician's obligations as by social convention. It must also be protected from
physician as well as the autonomy of surrogates or subversion by the profession itself. This requires a
patients. Sometimes both cannot be respected without much firmer philosophical grounding for medical
unacceptable compromises, on one side or the other. ethics than we now possess. The possibility of achiev-
Before such an impasse is reached, all other meth- ing universal approbation for a commonly held ethic of
ods of conflict resolution should be exhausted. Ethics the profession seems to be receding today in the face
committees can serve to clarify the issues and perhaps of the multicultural, morally pluralistic, and morally
suggest a way in which compromise could effectively be relativistic temper of the times. This climate, however,
reached in a manner that preserves the moral integrity cannot justify abandoning the effort. Nothing less is
of all the participants. Appointment of legal guardians called for than a reconstruction of the ethics of the
and appeals to the courts are far less satisfactory. In any relationship between patient and doctor. This will be
case, all of these devices address only the procedural difficult, indeed because the "remarkable solidarity"
resolution of the practical conflict. They certainly do and "singular beneficence," which Osler praised,38 are
not resolve the ethical dilemma of conflicting claims to rapidly disappearing in the worldwide questioning of
autonomous decision-making. the moral values that have traditionally undergirded
medical ethics.
V. The Integrity of Medical Ethics I have purposely said little about the principle
of justice, which must also be factored into the equa-
Some would suggest that the problem is with medical tion. On the whole, this facet of the physician—patient
ethics itself, with the insistence on universal rules of relationship has been underdeveloped. It is now neces-
moral conduct on which physicians base their moral sary to establish the conceptual relationships among
claim to autonomy as physicians. Why not change justice, autonomy, and beneficence, as well as their
actualization in the clinical context. Justice has the 3. Physician autonomy is limited by a competent
interesting facet of being both a principle and a virtue. patient's or valid surrogate's moral right to ref use
The incorporation of justice into the autonomy- proffered treatment. The physician is obliged,
beneficence equation will require a prior clarification however, to help the patient arrive at an autono-
of how principles and virtues are conceptually and mous decision by enhancing or empowering the
practically related. patient's capacity to make authentic, self-governing
Despite the difficulties, the effort to balance choices.
the autonomy equation is not futile. Its importance 4. The patient's autonomy is limited when it becomes
impels us to the effort to try to find the points of bal- a demand for treatment the physician honestly be-
ance. Autonomy and beneficence are two principles lieves is not medically indicated, is injurious to the
so closely tied to the healing ends of medicine that to patient, or is morally repugnant.
violate either is to imperil the moral integrity of both 5. The physician's autonomy is limited on questions
patients and physicians. Nevertheless, any compre- of value, e.g., on questions of the goals or purposes
hensive moral philosophy for the health professions to which medical knowledge may be put for par-
must encompass more than these two principles. ticular individuals or societies.
Justice must be included and account taken of both 6. Societies and institutions must establish mechan-
virtue and the moral psychological insights of non- isms, with only minimal recourse to law, for uni-
principle-based theories. lateral discontinuance of the relationship when
In any case, one step in the larger effort is to try either patient or physician feels personal integrity
to achieve a better balance between the two most is being compromised.
powerful principles shaping physician-patient rela- 7. The first principle of medical ethics is still benefi-
tions today. Several precepts need to be built into the cence. Beneficence is essential if autonomy is to be
current re-examination of the foundations of profes- authentically expressed and actualized.
sional ethics:
In sum, beneficence and autonomy must be mutually
1. Patient autonomy is a moral right of patients, and re-enforcing if the patient's good is to be served, if
it is a duty of physicians to respect it. the physician's ability to serve that good is not to be
2. Integrity of conscience and professional judgment compromised, and if the physician's moral claim to
are moral rights of physicians. Society and patients autonomy and the integrity of the whole enterprise of
have an obligation to respect them. medical ethics are to be respected.
Notes
1. Tom L. Beauchamp & James E Childress, Principles of 11. Dworkin, supra note 2, at 108.
Biomedical Ethics 67-119 (3rd ed. 1989). 12. Edmund D. Pellegrino, The Medical Profession as a Moral
2. Gerald Dworkin, The Theory and Practice of Autonomy Community, 66 Bull. N.Y. Acad. Med. 221 (1990).
3-6,12-20 (1988). 13. Edmund D. Pellegrino, Societal Duty and Moral Complicity:
3. This is a paraphrase of the essentials of Dworkin use of The Physician's Dilemma of Divided Loyalty, 16 Int'l J.L. &
the term. Id. at 7-12. Psychiatry 371 (1993).
4. See John Locke, Locke's Second Treatise on Civil 14. See David M. Mirvis, Physicians' Autonomy-The Relation
Government (Lester DeKoster ed., 1978). between Public and Professional Expectations, 328 New Eng.
5. See Immanuel Kant, Grounding for the Metaphysics of J. Med. 1346 (1993).
Morals (James W. Ellington trans., 1981). 15. Edmund D. Pellegrino, The Ethics of Medical
6. See John Stuart Mill, On Liberty (Elizabeth Rapaport Education, in The Encyclopedia of Bioethics (Warren
ed., 1978). T. Reich ed., 1994).
7. Beauchamp & Childress, supra note 1. 16. George]. Agich, Reassessing Autonomy in Long-Term Care,
8. Tom L. Beauchamp & Lawrence B. Mccullough, Medical Hastings Center Rep., Nov-Dec. 1990, at 12.
Ethics: The Moral Responsibilities of Physicians 22-51 17. See Antonella Surbone, Truth Telling to the Patient, 268
(1984). JAMA 1661 (1992).
9. See Edmund D. Pellegrino & David C. Thomasma, For the 18. See Lynn D. Wardle, Protecting the Rights of Conscience of
Patient's Good: The Restoration of Beneficence in Health Health Care Providers, 14]. Legal Med. 177 (1993).
Care 23-25 (1988). 19. Schloendorff v. Society of N.Y. Hosps., 105 N.E. 92,93
10. James E Childress, Who Should Decide? Paternalism in (N.Y. 1914).
Health Care 102 (1982). 20. In re Quinlan, 355 A.2d 647 (NJ. 1976).
uu I UI
II Jl 1111113
21. President's Commission for the Study of Ethical Problems in 31. See Barbara L. Lindheim & Maureen A. Cotterill, Training
Medicine and Biomedical and Behavioral Research, Deciding in Induced Abortion by Obstetrics and Gynecology Residency
to Forego Life-Sustaining Treatment: A Report of the Ethical, Programs, 10 Fam. Plan. Persp. 24 (1978).
Medical, and Legal Issues in Treatment Decisions 241 (1983). 32. This dissociation is one I have encountered already in
22. In re Wanglie, No. Px-91-283 (D. Minn. July, 1991). private conversation with medical students, colleagues,
23. For a further discussion of the morality issues in medicine, and influential laypeople.
see John Ladd, The Internal Morality of Medicine: An 33. Pellegrino, supra note 12.
Essential Dimension of the Patient-Physician Relationship, in 34. See Arthur E. Kopelman, Dilemmas in the Neonatal Intensive
The Clinical Encounter: The Moral Fabric of the Patient- Care Unit, in Ethics and Mental Retardation 243 (Loretta
Physician Relationship 209 (Earl E Shelp ed., 1983). Kopelman & John C. Moskop eds., 1984).
24. R.D. Truog et al., The Problem with Futility, 326 New Eng. 35. Child Abuse Amendments of 1984, Pub. L. No. 98-457,
J. Med 1560 (1992). 98 Stat. 1749 (1984) (amending 42 U.S.C. § 5101
25. Stuart J. Youngner, Who Defines Futility? 260 JAMA 2094 (1974)).
(1988). 36. See Anne Bannon, The Case of the Bloomington Baby,
26. Nancy S. Jecker & Lawrence J. Schneiderman, Medical Hum. Life Rev., Fall 1982, at 63; Michael McCarthy,
Futility: The Duty Not to Treat, 2 Cambridge Q. Healthcare Anencephalic Baby's Right to life? 342 Lancet 919
Ethics 151 (1993). (1993); John]. Paris et al., Physicians' Refusal of
27. Lance K. Stell, Stopping Treatment on Grounds of Futility: A Role Requested Treatment: The Case of Baby L, 322 New
for Institutional Policy, 11 St. Louis U. Pub. L. Rev. 481 (1992). Eng. J. Med. 1012 (1990).
28. Hippocrates, The Art, in 2 Hippocrates 185-217 (W.H.S. 37. See Edmund D. Pellegrino, Societal Duty and Moral
Jones trans., 1981). Complicity: The Physician's Dilemma of Divided Loyalty, 1
29. Wardle, supra note 18. Int'1].L. & Ethics (June 1994).
30. Executive Board Minutes from the American College of 38. William Osler, Chauvinism in Medicine, in Aequanimitas:
Obstetrics and Gynecology at 12, Item 6.2 (Jan. 1992) (on With Other Addresses to Medical Students, Nurses and
file with organization). Practitioners of Medicine 267 (1943).
The Problem with Futility to obtain consent for a do-not-resuscitate (DNR) order
when cardiopulmonary resuscitation (CPR) is deemed
Robert D. Truog, Allan S. Brett, and Joel Frader futile.8 The fact that this concept has appeared in
law and policy may seem to indicate that it is clearly
understood and widely accepted. In reality, however,
"FUTILITY" is one of the newest additions to the lexi- the notion of futility hides many deep and serious
con of bioethics. Physicians, ethicists, and members of ambiguities that threaten its legitimacy as a rationale
the media are increasingly concerned about patients for limiting treatment.
and families who insist on receiving life-sustaining
treatment that others judge to be futile. A clear under- Paradigms of Futility
standing of futility has proved to be elusive, however.
Many clinicians view futility the way one judge viewed Contemporary discussions of futility have centered pri-
pornography: they may not be able to define it, but marily on cases involving patients in a persistent vege-
they know it when they see it.' tative state and those involving the use of CPR A third
The notion of futile medical treatment may go type of case, involving organ-replacement technology,
back to the time of Hippocrates, who allegedly advised has received little attention but is helpful to our under-
physicians "to refuse to treat those who are over- standing of futility.
mastered by their diseases, realizing that in such cases
medicine is powerless."' More recently, the concept Futility and the Persistent Vegetative State
has appeared frequently in court decisions and policy The first type of scenario involving the question of futil-
statements.3-6 The so-called Baby Doe law exempts ity is represented by the recent Minnesota case of Helga
physicians from providing treatment that would be Wanglie.9 Mrs Wanglie was an 86-year-old woman
"virtually futile."7 The Council on Ethical and Judicial who had been dependent on mechanical ventilation
Affairs of the American Medical Association (AMA) re- and in a persistent vegetative state for more than a year.
cently concluded that physicians have no obligation Her husband insisted that she believed in maintaining
Source: From The New England Journal of Medicine, 326 (1992), 1560-64. Reprinted with permission of the Massachusetts
Medical Society
Truog/Brett/Frader: The Problem with Futility
life at all cost and that "when she was ready to go . . . "futile" extracorporeal membrane oxygenation is inher-
the good Lord would call heel° Her physicians, on the ently different from a decision not to use "futile" CPR
other hand, believed that the continued use of mech- or "futile" intensive care. If all three treatments can be
anical ventilation and intensive care was futile. When characterized as objectively futile, then unilateral deci-
attempts to transfer her elsewhere failed, they sought to sions not to offer them should be equally justified.
have a court appoint an independent conservator with As it is used in these three cases, the concept of
responsibility for making medical decisions on her futility obscures many ambiguities and assumptions.
behalf. The judge denied this petition and reaffirmed These can be usefully grouped into two categories:
the authority of her husband as legal surrogate. problems of value and problems of probability.
Three days later, Mrs Wanglie died.
Cases like that of Mrs Wanglie seldom reach the Futility and Values
courts, but they are probably not rare. A similar case
involving a child with severe brain damage was con- It is meaningless simply to say that an intervention
cluded with a settlement favorable to the family before is futile; one must always ask, "Futile in relation to
a judicial decision.11 what?" The medical literature provides many examples
in which the importance of identifying the goals of
Futility in Cases Involving CPR treatment has not been fully appreciated. The effective-
The second prototypical scenario involves the use ness of CPR, for example, is often discussed in terms of
of DNR orders. Although the techniques of CPR were whether patients who require the procedure can sur-
originally intended only for use after acute, revers- vive long enough to be discharged from the hospital.'
ible cardiac arrests, the current practice is to use CPR This definition of success usually implies that short-
in all situations unless there is a direct order to the term survival is a goal not worth pursuing. Patients or
contrary. Since cardiac arrest is the final event in all family members may value the additional hours of life
terminal illness, everyone is eventually a candidate for differently, however. Indeed, physicians and other care-
this medical procedure. DNR orders were developed to givers have repeatedly been shown to be poor judges of
spare patients from aggressive attempts at revival when patients' preferences with regard to intensive care."-18
imminent death is anticipated and inevitable. Never- Schneiderman and colleagues have argued that
theless, patients or families sometimes request CPR treatments that merely preserve permanent uncon-
even when caregivers believe such attempts would be sciousness or that cannot end dependence on intensive
futile. Some have argued that in these circumstances a medical care should be considered futile.19 Although
physician should be able to enact a DNR order without society may eventually endorse decisions to override the
the consent of the patient or family.12-14 previously expressed wishes of patients or the desires
of surrogates who demand such treatments, it does not
Futility and Organ-Replacement Technology follow that the treatments are futile. Mr Wanglie would
Although the bioethical debate over the question of have rejected this conclusion, and there is no reason to
futility has been most concerned with cases involv- dismiss his view out of hand. The decision that certain
ing CPR and the treatment of patients in a persistent goals are not worth pursuing is best seen as involving a
vegetative state, a third type of futility-related judg- conflict of values rather than a question of futility.
ment has gone essentially unchallenged. It involves the Certainly in this context, the plurality of values in
increasingly large number of interventions that could our society makes agreement on the concept of futility
possibly prolong the life of virtually any dying patient. difficult if not impossible. Several groups have therefore
For example, extracorporeal membrane oxygenation attempted to arrive at a value-free understanding of the
can replace heart and lung function for up to several concept.2(121 The most promising candidate thus far
weeks. Physicians now use this intervention when they is the notion of "physiologic futility" As the guidelines
expect organ systems eventually to recover or while on the termination of life-sustaining treatment prepared
they await organs for transplantation. However, it could by the Hastings Center state, if a treatment is "clearly futile
prolong the life of almost anyone with cardiorespira- in achieving its physiological objective and so offer[s] no
tory failure, reversible or not. Patients thus kept alive physiological benefit to the patient, the professional has
may remain conscious and capable of communicating. no obligation to provide it."2° For example, the physio-
Caregivers do not now offer this therapy to terminally logic objective of mechanical ventilation is to maintain
ill patients, presumably because it would be futile. This adequate ventilation and oxygenation in the presence of
judgment has gone largely unchallenged, yet it is not respiratory failure, and the physiologic objective of CPR
obvious why a clinician's unilateral decision not to use is to maintain adequate cardiac output and respiration
I .11 1111.111
in the presence of cardiorespiratory failure. The New patient. In addition, the tendency to remember cases
York State Task Force on Life and the Law mistakenly that are unusual or bizarre predisposes physicians to
concludes that CPR is physiologically futile when it will make decisions on the basis of their experiences with
"be unsuccessful in restoring cardiac and respiratory "miraculous" cures or unexpected tragedies.
function or [when] the patient will experience repeated Schneiderman and colleagues recently argued that
arrest in a short time period before death occurs."21 CPR a treatment should be considered futile when 100 con-
is physiologically futile only when it is impossible to per- secutive patients do not respond to it."' But how similar
form effective cardiac massage and ventilation (such as must the patients be? In assessing the efficacy of mech-
in the presence of cardiac rupture or severe outflow ob- anical ventilation to treat pneumonia, for example, is it
struction). Saying that CPR is physiologically futile when sufficient simply to recall the 100 most recent patients
it will be unsuccessful in restoring cardiac function is who received artificial ventilation for pneumonia? Or
like saying that mechanical ventilation is physiologically must this group be stratified according to age, etiologic
futile if it cannot restore respiratory function. The im- organism, or coexisting illness? Clearly, many of these
mediate physiologic effect of the intervention differs from factors will make an important difference.
the broader and more uncertain question of prognosis.
Physiologic futility, understood in narrow terms, Futility and Resource Allocation
comes close to providing a value-free understanding of
futility. Unfortunately, it applies to a very small number Although medical practice has increasingly empha-
of real cases involving CPR. Similarly, since in the case sized patients' autonomy there is growing pressure
of Mrs Wanglie mechanical ventilation could maintain on physicians to slow the increase in health care costs
adequate oxygenation and ventilation, her treatment by foreclosing some options. Thus, we have a tension
could not be considered futile in the physiologic sense. between the value of autonomy exercised in the form
Even the use of extracorporeal membrane oxygenation of consent to use or omit various interventions, and
in terminally ill patients cannot be considered physio- the desirability of a more Spartan approach to the
logically futile, since it can maintain circulation and consumption of medical resources. We promote pa-
ventilation. The concept of physiologic futility there- tients' freedom to request whatever the medical menu
fore falls short of providing guidance in most cases has to offer, but we also require that interventions be
resembling those described above. guided by considerations of cost and the likelihood of
benefit.26 Unfortunately, there is no consensus about
Futility and Statistical Uncertainty what constitutes a just method of balancing the prefer-
ences of individual patients against the diverse needs
In most medical situations, there is no such thing as never. of society.
Futility is almost always a matter of probability. But what To some, the concept of futility provides at least
statistical cutoff point should be chosen as the threshold a partial solution to this dilemma it offers a reason to
for determining futility? The statement from the Coun- limit therapy without the need to define a fair proced-
cil on Ethical and Judicial Affairs of the AMA concludes ure for allocating resources. This approach allows treat-
that physicians have no obligation to provide futile CPR, ments to be denied on the grounds that they are simply
but it fails to specify any level of statistical certainty at not indicated, apart from the matter of cost. Despite its
which the judgment is warranted.8 The AMA statement attractions, there are good reasons why we should not
fails to acknowledge that this is even an issue. Should use this concept to solve problems of allocation.
each physician decide independently what probability of First, arguments based on the futility concept
success should be considered to indicate futility? conceal many statistical and value-laden assumptions,
Even if we could agree on a statistical cutoff point whereas strategies based on resource allocation force
for determining futility, physicians are often highly these assumptions to be stated explicitly. Societies may
unreliable in estimating the likelihood of success of a choose to limit the use of therapies that may be of value
therapeutic intervention. Psychological research22'23 and have a reasonable likelihood of success in some
has shown that estimates of probability are susceptible cases. For example, the much discussed Oregon plan
to "severe and systematic errors."22 Empirical studies for allocating Medicaid funds27 seeks to reflect commun-
have corroborated the limitations of clinical assess- ity values in ranking various health care goals (placing
ment in estimating both prognosis24 and diagnosis.25 preventive care ahead of cosmetic surgery, for example).
Even in theory, statistical inferences about what might Since rationing policies make explicit the values and
happen to groups of patients do not permit accur- probabilities that futility-based arguments leave implicit,
ate predictions of what will happen to the next such it is clearly preferable to develop and adopt them rather
Truog/Brett/Frader: The Problem with Futility 179
than use futility arguments as a cover for limiting the objective and dispassionate standard for determining
availability of scarce and expensive resources. benefits and burdens. Nevertheless, even after providers
Another problem with invoking the idea of futil- make sustained attempts to clarify patients' preferences,
ity in the debate over allocation is that we have no some patients or surrogates will continue to demand
reason to believe that it is applicable in enough cases life-sustaining interventions when the caregivers feel
to make a difference in the scarcity of medical resour- deeply troubled about providing them. In many such
ces. Although it may be true that beds in the inten- cases unrestrained deference to the wishes of the patient
sive care unit (especially those used for extracorporeal or surrogate conflicts with two other values that do not
membrane oxygenation) are relatively scarce, it seems require a unilateral judgment of the futility of treatment:
unlikely that patients similar to Helga Wanglie occupy professional ideals and social consensus.
an important fraction of those beds, let alone account The ideals of medical professionals include respect
for a major proportion of the cost of medical care in for patients' wishes, to be sure, but they also include
the United States. From a macroeconomic perspective other values, such as compassionate action and the
at least, we must remain skeptical that an appeal to the minimization of suffering. Consider, for example, a
idea of futility will get us very far. bedridden victim of multiple strokes who has con-
tractures and bedsores and who "communicates" only
Moving beyond Futility by moaning or grimacing when she is touched. Phys-
icians asked to perform chest compressions, institute
Our rejection of futility as a useful concept does not mechanical ventilation, or use other life-sustaining
imply that we endorse patients' unrestricted demands interventions in such a patient may regard these ac-
for interventions such as those described in our proto- tions as cruel and inhumane.31 Moreover, physicians
typical scenarios. On the contrary when providers and other caregivers have a legitimate interest in seeing
oppose such demands they are usually acting from a that their knowledge and skills are used wisely and
profound sense that further treatment would be fun- effectively. For example, if surgeons were repeatedly
damentally wrong. Our task is to take account of that pressured to perform operations that they believed to
sense of wrongness without resorting to unilateral, be inappropriate, they would certainly suffer a loss of
provider-initiated declarations of futility. dignity and sense of purpose. Although appealing to
In many of the situations in which questions of professional ideals can serve as a convenient means
futility arise, providers believe that the treatment in of protecting the interests of physicians at the expense
question would not be in the patient's interests, even of patients' values, these ideals are legitimate factors to
from the patient's perspective, and that any insistence weigh against other values. To dismiss this perspective
by the patient (or surrogate) on further interventions is as irrelevant in decision-making is to deny an essential
based on faulty reasoning, unrealistic expectations, or part of what it means to practice medicine.
psychological factors, such as denial or guilt. In these Although we believe that health care professionals
circumstances, providers are obligated to make every should not be required to take part in care that violates
effort to clarify precisely what the patient intends to their own morals, the law in this area remains uncertain.
achieve with continued treatment. If the patient's goals On the one hand, courts have upheld a state interest in
appear to reflect unrealistic expectations about the protecting the ethical integrity of the medical profession.
probable course of the underlying illness or the prob- This may provide some basis for protecting doctors who
able effect of medical interventions, providers should at- wish to refrain from cruel or inhumane treatment, despite
tempt to correct those impressions. Because inadequate the wishes of the patient or surrogate.32 On the other
or insensitive communication by providers probably hand, in the two cases that have led to court decisions
accounts for a substantial proportion of unrealistic re- (those of Helga Wanglie3 and of Jane Doe in Atlanta33) the
quests, such discussions will successfully resolve many judges upheld the surrogates' decision-making authority.
conflicts.14'28 Empirical studies of ethics consultations Clearly, this area of the law remains to be defined.
have demonstrated precisely this point.29.30 Finally, social consensus is yet another expression of
Although this appeal to the patient's interests may the values at stake in some medical decisions. In a plur-
seem to contain some of the same ambiguities as argu- alistic society, differences in personal values and interests
ments using the concept of futility, there is a subtle but occasionally run so deep that they cannot be resolved by
important distinction between the two. Judgments about the introduction of additional facts or by further private
what is in the patient's interest are properly grounded debate. At certain critical junctures, the resolution of these
in the patient's perspective, whereas judgments cast in conflicts may require an explicit public process of social
the language of futility falsely assume that there is an decision-making.34 Social consensus has been sought, for
I I 31 111111
example, to address the issue of fair allocation of resour- approach fails, physicians and other caregivers should ask
ces.27 The involvement of society is also essential when themselves whether the care requested is consistent with
the most highly charged questions of morality are at stake, their professional ethics and ideals. When these ideals
as in the increasingly heated debate over euthanasia.35 appear to be violated, either alternative venues for such
In the prototypical scenarios described at the outset care should be found or the conflict should be addressed
of this article, an ongoing attempt to achieve social in a public forum. This broader review could be provided
consensus is perhaps most conspicuous with regard through institutional mechanisms, such as the hospital's
to the prolongation of life for patients in a persistent ethics committee, or by the courts. The public scrutiny
vegetative state. From a legal perspective, the relevant that attends such cases will further the debate over the ap-
decisions began with the case of Karen Quinlan36 and propriate use of medical resources and foster the develop-
have extended through that of Nancy Cruzan.37 These ment of consensus through legislation and public policy.
cases have increased awareness of the ethical issues
raised by the situation of patients in a persistent vege- Conclusion
tative state and have helped to consolidate the view
that it is acceptable to withdraw life-sustaining treat- In outlining the perspectives of the principal
ment from patients in such a state. Controversy does stakeholders—patients and their surrogates, phys-
remain about who has the ultimate authority to make icians, and society—we have avoided the construction
these decisions. Some hold that the choice must remain of a rigid formula for resolving conflicts over interven-
with the patient or surrogate, whereas others believe tions frequently regarded as futile. Because of clinical
that under some circumstances this prerogative may heterogeneity, pluralistic values, and the evolutionary
be overridden. For example, the Hastings Center38 and nature of social consensus, most clinical decision-
the Society of Critical Care Medicine39 have concluded making on behalf of critically ill patients defies reduc-
that providing intensive care to patients in a persistent tion to universally applicable principles.
vegetative state is generally a misuse of resources, and The notion of futility generally fails to provide an
the President's Commission stated that such patients ethically coherent ground for limiting life-sustaining
should be removed from life support if such action is treatment, except in circumstances in which narrowly
necessary to benefit another patient who is not in a per- defined physiologic futility can be plausibly invoked.
sistent vegetative state.4° It is unclear how this debate Futility has been conceptualized as an objective entity
will conclude, but the confluence of medical, legal, and independent of the patient's or surrogate's perspective,
ethical thinking about the persistent vegetative state is but differences in values and the variable probabilities
an example of how social consensus may evolve. of clinical outcomes undermine its basis. Furthermore,
In summary, the Wanglie case demonstrates how assertions of futility may camouflage judgments of com-
the resolution of these conflicts must proceed on many parative worth that are implicit in debates about the allo-
levels. Most such cases will benefit from sustained at- cation of resources. In short, the problem with futility is
tempts to clarify the patient's values and the likelihood that its promise of objectivity can rarely be fulfilled. The
of the various relevant outcomes and to improve com- rapid advance of the language of futility into the jargon of
munication with patients or their surrogates. When this bioethics should be followed by an equally rapid retreat.
References
1. Jacobellis v. State o f Ohio, 84 S Ct 1676 (1964). 6. Appendix A the proposed legislation In: Do not re-
2. Hippocrates. The art. In: Reiser SJ, Dyck AJ, Curran suscitate orders: the proposed legislation and report of
WJ. eds. Ethics in medicine: historical perspectives and the New York State Task Force on Life and the Law. 2nd
contemporary concerns. Cambridge, Mass.: MIT Press, ed. New York: The Task Force, 1986:83.
1977:6-7. 7. 1984 Amendments to the Child Abuse Prevention and
3. Capron AM. In re Helga Wanglie. Hastings Cent Rep Treatment Act. Pub Law 98-457.1984.
1991 ;21 (5): 26-8. 8. Council on Ethical and Judicial Affairs, American
4. Lantos JD, Singer PA, Walker RM, et al. The illusion of Medical Association. Guidelines for the appropriate use
futility in clinical practice. Am J Med 1989;87:81-4. of do-not-resuscitate orders. JAMA 1991;265:1868-71.
5. Standards for cardiopulmonary resuscitation (CPR) 9. Miles SH. Informed demand for "non-beneficial" med-
and emergency cardiac care (ECC). V. Medicole- ical treatment. N Engl J Med 1991;325:512-5.
gal considerations and recommendations. JAMA 10. Brain-damaged woman at center of lawsuit over life-
I 974;227:Supp1:864-6. support dies. New York Times July 5,1991:A8.
Truog/Brett/Frader: The Problem with Futility
11. Paris J J, Crone RK, Reanlon F. Physicians' refusal of re- 25. Poses RM, Cebul RD, Collins M, Fager SS. The accur-
quested treatment: the case of Baby L N Engl J Med acy of experienced physicians' probability estimates
1990;322:1012-15. for patients with sore throats: implications for decision
12. Blackball U. Must we always use CPR? N Engl J Med making. JAMA 1985;254:925-9.
1987;317:1281-5. 26. Aaron H, Schwartz WB. Rationing health care: the
13. Hackler JC, Hiller F C. Family consent t o orders not t o resusci- choice before us. Science 1990;247:418-22.
tate: reconsidering hospital policy. JAMA 1990;264:1281-3. 27. Eddy DM. What's going on in Oregon? JAMA
14. Murphy DJ. Do-not-resuscitate orders: time for 1991;266:417-20.
reappraisal in long-term care institutions. JAMA 28. Youngner SJ. Who defines futility? JAMA
1988;260:2098-101. 1988;260:2094-5.
15. Bedell SE, Delbanco TL, Cook E F, Epstein FH. Survival 29. Brennan TA. Ethics committees and decisions to limit
after cardiopulmonary resuscitation in the hospital. care: the experience at the Massachusetts General Hos-
N Engl J Med 1983;309:569-76. pital. JAMA 1988;260:803-7.
16. Danis M, Gerrity MS, Southerland LI, Patrick DL. 30. La Puma J. Consultations in clinical ethics-issues and
A comparison of patient, family, and physician assess- questions in 27 cases. West J Med 1987;146:633-7.
ments of the value of medical intensive care. Crit Care 31. Braithwaite 5, Thomasma DC. New guidelines on fore-
Med 1988;16:594-600. going life-sustaining treatment in incompetent patients:
17. Danis M, Jarr SL, Southerland LI, Nocella RS, Patrick an anti-cruelty policy. Ann Intern Med 1986;104:711-5.
DL. A comparison of patient, family, and nurse evalua- 32. Meisel A. The right to die. New York John Wiley &
tions of the usefulness of intensive care. Crit Care Med Sons, 1989:104.
1987;15:138-43. 33. In re: Doe, Civil Action No. 093064 (Fulton County,
18. Danis M, Patrick DL, Southerland LI, Green ML. Pa- GA, October 17, 1991).
tients' and families' preferences for medical intensive 34. Callahan D. Medical futility, medical necessity:
care. JAMA 1988;260:797-802. the-problem-without-a-name. Hastings Cent Rep
19. Schneiderman LJ, Jecker NS, Jonsen AR. Medical futil- 1991;21(4):30-5.
ity its meaning and ethical implications. Ann Intern 35. Misbin RI. Physicians' aid in dying. N Engl J Med
Med 1990;112:949-54. 1991;325:1307-11.
20. The Hastings Center. Guidelines on the termination 36. In the Matter of Karen Ann Quinlan, an alleged in-
of life-sustaining treatment and the care of the dying. competent. 355 A.2d 647; or 70 NJ 10. March 31,1976.
Bloomington: Indiana University Press, 1987:32. 37. Annas GJ. Nancy Cruzan and the right to die. N Engl J
21. Appendix C: New York Public Health Law Article Med I 990;323:670-3.
29-B-orders not to resuscitate. In: Do not resuscitate 38. The Hastings Center. Guidelines on the termination
orders: the proposed legislation and report of the New of life-sustaining treatment and the care of the dying.
York State Task Force on Life and the Law. 2nd ed. New Bloomington: Indiana University Press, 1987:112.
York: The Task Force, 1986:96. 39. Task Force on Ethics of the Society of Critical Care Medi-
22. Tversky A, Kahneman D. Judgment under uncertainty: cine. Consensus report on the ethics of foregoing life-
heuristics and biases. Science 1974;185:1124-31. sustaining treatments in the critically ill Crit Care Med
23. Elstein AS. Clinical judgment: psychological research 1990;18:1435-9.
and medical practice. Science 1976;194:696-700. 40. President's Commission for the Study of Ethical Problems
24. Poses RM, Bekes C, Copare FJ, Scott WE. The answer in Medicine and Biomedical and Behavioral Research.
to "What are my chances, doctor?" depends on whom Deciding to forego life-sustaining treatment: ethical,
is asked: prognostic disagreement and inaccuracy for medical, and legal issues in treatment decisions. Wash-
critically ill patients. Crit Care Med 1989;17:827-33. ington, DC: Government Printing Office, 1983:188-9.
Should Physicians Be Gatekeepers offer all technologically feasible and clinically beneficial
medical services to all patients. Individually, though,
of Medical Resources? physicians practise medicine under a basic ethical tenet
Milton C. Weinstein which compels them to do whatever is in their power to
help their patients. Herein lies one of the fundamental
The Medical Commons ethical issues in modern medicine: how can physicians
fulfil their moral obligations as fiduciary agents for in-
Resources available for health care are finite. This dividual patients while being responsible stewards of
means that it is impossible for physicians collectively to the finite pool of resources?
Source Fromfourrial of Medical Ethics 27 (2001), 4,268-74. Reprinted with permission of the BM J Publishing Group.
I I DII 111111
In a 1975 article in the New England Journal of arbitrary barriers to introducing new treatments while
Medicine,' Howard Hiatt likened the situation in med- older treatments of questionable efficacy continue to be
ical care to a parable described by Garrett Hardin in used.
a classic article, entitled "The Tragedy of the Com- This essay begins with a view of this problem
mons."' According to Hiatt's adaptation of Hardin's from the perspective of society as a whole. Adopting a
parable, physicians are like herdsmen who used to feed utilitarian stance, different allocations of resources may
their cattle (patients) on a common pasture. Acting in be judged by how well they achieve some generally
their own interests, the herdsmen could allow their agreed upon measure of social benefit. This representa-
cattle to feed on the land without limit, as long as their tion leads to cost-effectiveness analysis as a guide to
numbers and appetites were small compared to the resource allocation from the societal perspective. The
resources on the land. As the number and appetites next section describes the fundamental principles
of the cattle grew to the point where, collectively, their of cost-effectiveness analysis as a guide to resource
wants exceeded the capacity of the common resource, allocation.
the desires of the herdsmen to extract the maximal The following three sections of the essay contrast
nutrition for their cattle led to overgrazing. At first, the the perspectives of patients, consumers (i.e., patients ex
less aggressive cattle failed to get adequate nutrition ante), and physicians on resource allocation. Evidently
and died. Later, herdsmen were forced out of business. there are differences among these perspectives regard-
And, in the end, the rich pasture turned into an over- ing the degree to which resource cost should enter into
grazed wasteland. decision-making. Concepts that overlap economics and
Although Hardin's essay was written in the context moral philosophy, such as moral hazard and fiduciary
of population growth, Hiatt saw its relevance to health relationships, are introduced along the way. I conclude
care. Physicians, each acting in the best interests of their that neither patients nor physicians can be expected to
own patients, collectively reach the limits of health attend to the rationing of the medical "commons" with-
care resources, with the result that access to care and out some form of collective action but that collective
quality of care are compromised. There is no obvious action alone will not succeed without genuine accept-
ethical solution to the problem of rationing the medical ance of the need to ration health care resources by both
commons, because any solution involves comparisons physicians and patients.
between the value of health services provided to dif- The remainder of the essay concerns the struc-
ferent patients with different conditions. Nonetheless, ture of collective solutions, the purpose of which is to
the remedy must lie in some form of collective action: permit physicians to fulfil their responsibilities as fidu-
physicians, like the herdsmen in Hardin's parable, can ciary agents to patients, while being responsible gate-
save the commons only by adhering to a set of mutually keepers of resources. Forms of collective action may
acceptable covenants which govern and limit their use range from government regulation to decentralised
of the shared resource. budgeting of providers, to non-binding clinical guide-
If health care were "free," there would be no need lines. The promises and limitations of such devices as
to limit its use. Health care is not "free," because the capitated payments, limited access to technology, and
use of resources (physician time, hospital beds, health voluntary guidelines are also considered. An analogy is
care budgets) by some precludes the use of those redrawn with public regulation of pollution. The essay
sources by others. The overall result of failing to adopt concludes with observations regarding the role of cost-
covenants that lead to restrained use of resources by effectiveness analysis at the bedside.
well-meaning physicians is unacceptable. If a society
mandates universal access to health care, and if all phys- The Cost-Effectiveness Paradigm
icians provide their patients with the most beneficial
treatments available, then the cost of health care will be Suppose that a society wants to provide the maximum
unacceptably high to their patients, either as taxpayers, aggregate health improvement in its population, but it
payers of insurance premiums, at the point of care, or in has limited resources to do so. Each potential health
combination. The alternatives are compromises, either intervention delivered to a defined group of persons
with the principle of universal access or with the prin- with a particular condition yields a health improvement
ciple of unlimited care. Failure to acknowledge these and entails a cost. If health improvement is measured in
trade-offs can lead to inefficiencies and inequities that units which reflect the values of the society, and if costs
compromise both principles: hidden barriers to access are measured in units which reflect the extent to which
such as queues and administrative hassles, exclusion of the resource budget is depleted by the intervention,
entire segments of the population, or erecting rigid and then the maximum societal health improvement can
Weinstein: Should Physicians Be Gatekeepers of Medical Resources?
be achieved by applying the following simple decision money," namely, the cost per quality adjusted year of
rule: rank order medical interventions in decreasing life gained.3-5 Thus, interventions having low values
order of their expected health improvement per unit of this ratio go to the top of the rank list, and those
expected cost, and adopt them from the top of the rank with high values may fall below the line for a particular
list to the point on the list that resources are depleted.3 budget.
The ratio of benefit to cost from each intervention rep- Some programmes may actually save more resour-
resents its "value for money." To make this decision rule ces than they consume while still providing a health
operational even at the societal level, one needs meas- improvement. Such "cost-saving" programmes auto-
ures of predicted health improvement or benefit and matically go to the top of the rank list because they
predicted resource cost. (For present purposes, com- do not deplete the resource pool; indeed, they add
plexities arising from uncertainty regarding anticipated to it. Contrary to popular beliefs, however, very few
benefits and costs are set aside, and are summarised in programmes, even in the area of prevention, are truly
terms of the average, or expected, values of each.) cost-saving.' 8
The health benefit can be measured in units that What is the ethical basis for seeking to allo-
reflect the preferences of the community, considering cate society's health care resources with the aim of
their desire for increased longevity but also the value maximising quality adjusted life expectancy? If each
they place on limitation of function, pain, and other individual measured the value of his or her own
dimensions of health-related quality of life.4 One such health in terms of quality adjusted life expectancy,
measure is the quality adjusted life year (QALY).3-5 and if each of them paid for his or her own health care
Quality adjusted life years measure the number of years through individual savings accounts, then economic
of remaining life, each adjusted by a preference weight theory would lead them each to allocate their own
(generally between zero and one, or possibly even less assets according to the cost-per-QALY rule.9 Because
than zero for health states judged worse than dead) of the uncertainties about the need for health care
that reflects the relative value of the health state on a during one's life, however, and for reasons of equity,
scale between perfect health (one) and dead (zero).3-6 health care services are typically covered by pooled
Since the amount of health benefit is uncertain before risk-sharing arrangements in the form of insurance or
the intervention, the measure of benefit for a health national health care. Under these arrangements, the
care intervention can be expressed as an average across measure of aggregate QALYs at the societal level entails
similar persons in the target population, in terms of an interpersonal comparison: the implicit assumption
quality adjusted life expectancy. The QALY gain for an is that a QALY is a QALY, no matter who gets it. Much
intervention should include all health consequences, has been written regarding the ethical justification
both positive and negative, and is therefore a measure for such a social value. Citizens behind a primordial
of net health benefit.5 veil of ignorancel" might elect to pre-specify that
health resource allocations be made so as to maximise
their expected utility as viewed from behind the veil.
Monetary Terms
If their preferences after the veil is removed are to
Costs are usually measured in monetary terms—dollars, maximise their own quality adjusted life expectan-
pounds, euros—but it should always be remembered cies, then a society that maximises aggregate QALYs
that money is only a proxy for the real resources— might appeal to them.
physicians, nurses, hospital facilities, pharmaceutical Rawls, however," comes to a rather different con-
development and production—that are consumed clusion from the same initial premise. He concludes
in providing the service.5 From society's perspective, that individuals behind the veil would attach more
costs are also net of any savings in future health care weight to gains in well-being should they emerge with
resources that might have been consumed but for the disadvantages, such as poverty or poor health. Thus,
intervention—such as the costs of treating strokes that more weight would be applied to health improvements
are averted because of a blood pressure control pro- for those persons in the most disadvantaged positions
gramme. From a societal perspective, costs also include in society. Alternatives to QALYs that reflect societal
resources contributed by patients and family members preferences for differential weighting of health gains to
and other caregivers, including their time, and not just the least healthy have been proposed by Nord."
resources financed through health insurance (public or Whether a society wishes to count all QALYs
private) or billed to the patient. equally or to weight them inversely to the health status
In common practice, the ratio reported from cost- of the beneficiary, the result is a utilitarian measure
effectiveness analyses is the reciprocal of "value for of societal health benefit that could be used to guide
.1 .11 1 1.11
II ill muu
resource allocations in health care through an appro- larger as resource constraints upon patients are their
priately constructed ratio. own time, including travel to the doctor's office, and
The fundamental question in this essay is this: if other out-of-pocket costs such as transportation and
there is general consensus that resources should be al- child care.
located in such a way as to maximise aggregate health In general, these constraints on physicians and
benefit, who is responsible for allocating the resources? patients are relatively weak deterrents to the use of
Should the patient voluntarily deny himself health ser- maximally beneficial medical care. They are not strong
vices out of a sense of communal obligation to conserve enough by themselves to allocate the commons, and
the commons? Should physicians be expected to bal- to the extent that they limit care, they may not do
ance their responsibilities to do the best they can for so efficiently.
their patients with a responsibility to be the gatekeeper
of the commons at each and every encounter with pa-
The Role of Consumers (Patients ex ante)
tients? Or is some form of collective action required
whereby citizens empower their physicians to practise As patients, people have a different view of health care
medicine within a system that imposes limits on the costs than they have as consumers. While people may
resources available to them? My conclusion is that a expect that physicians do everything possible for them
combination of all three is necessary in order to allocate when they are sick, they complain bitterly when their
medical resources efficiently. insurance premiums rise, when the prices of goods and
services go up because labour costs to employers reflect
The Role of Patients
rising health care benefits, or when their taxes go up.
Consumers have to pay for their collective use of
Patients expect physicians, as fiduciary agents, to do health care resources, but they don't want to bear re-
everything in their power to provide them with the best sponsibility for the collective costs at the point of their
possible health care. Americans in particular, perhaps own care. Part of the motivation for individuals is ex-
in contrast with citizens of other industrialised coun- plained by the economic theory of insurance—groups
tries, demand the maximal use of available technolo- of people can become better off by pooling risks and
gies. They are none the less accustomed to the fact avoiding major losses in the event they become sick.
that their physicians are already forced to ration care As a result, citizens demand that health insurance be
because of constraints that are neither financial nor provided by their employers (or made available at
under their control. For example, physicians have lim- nominal cost) or by government. Moreover, subsidised
ited time during the day, which results in barriers to health insurance enables citizens who would not
scheduling office appointments or conducting the most otherwise be able to afford basic health care to obtain
thorough possible examinations. Access to facilities, it; there is a redistributional aspect to the provision of
such as hospital beds and especially diagnostic tech- health insurance.
nologies such as magnetic resonance imaging and com- When patients have medical insurance, they face
puted tomography, may be limited, forcing physicians different incentives in going to the doctor to seek care.
to delay, or even forgo, some diagnostic information. Price becomes less of a factor, because the patient does
In sum, physicians ration care to some extent, or at not bear the full cost of care at the time care is sought.
least set priorities, and patients know that other pa- In effect, the price facing the patient is lower than the
tients may take precedence for their physician's atten- full social cost of health care. The gap between per-
tion, depending on the urgency and severity of their ceived price and resource cost creates an incentive to
problems. Unfortunately, the services that physicians utilise more health care than the patient would other-
are most likely to forgo under these pressures are those wise be willing to buy. This phenomenon is known as
which patients do not actively seek, much less demand. "moral hazard" in the economics literature.14 It tends to
Some of these services, such as periodic screening for promote excessive use of the "medical commons," lead-
colorectal and breast cancer, or blood pressure mon- ing to increases in the cost of health insurance beyond
itoring and treatment, may be far more cost-effective what consumers believe is reasonable.
(by the cost-per-QALY criterion) than the services that Ironically, the word "moral" in "moral hazard"
physicians are compelled, by their sense of obligation suggests that patients are at risk of being "immoral"
to patients, to provide. if they overutilise health care services relative to the
Patients themselves ration their own health care value of the services they receive. This nomenclature
to some degree. In the United States, insured patients stands in sharp contrast to the predominant ethical
often face co-payments or deductibles. Looming even problem faced by physicians in the presence of limited
resources: their ethical obligation to the patient at hand performance may suffer). They are limited by resource
makes them immoral if they underutilise services rela- constraints in hospitals and laboratories, such as inten-
tive to the maximum they could do. In reality, neither sive care beds and magnetic resonance equipment.
the physician placed in the fiduciary role as agent for Sometimes constraints force the physician to
the patient, nor the patient facing artificially low prices make explicit choices between the interests of differ-
at the point of care, can be faulted for providing and ent patients, as in the setting of emergency triage or
expecting, respectively the best health care technology in the selection of organ transplant recipients. In these
can offer. situations, the choices concern patients with names
and faces, all of whom are under the care of a single
The Role of Physicians
physician or provider organisation. The question at
the societal level is whether physicians should be ex-
Physicians and patients engage in what economists pected to allocate resources between their patients and
refer to as principal—agent relationships. This refers other nameless, faceless patients who could, perhaps,
to the fiduciary trust that patients (the principals) obtain more benefit if the resources were conserved to
place in physicians (the agents) to make decisions that benefit them. The question, in other words, is whether
maximise the well-being of their patients and to act as physicians should consider the cost-effectiveness of the
advocates for their patients in the health care system. decisions they make for their patients, recognising that
This fiduciary responsibility has been described by resources are limited. Is it the physician's responsibility
Fried as entailing doing whatever is possible to "[pre- to protect the medical commons?
serve] life capacities for the realization of a reasonable, An entirely different view of the physician's ethical
realistic life plan."15 responsibility would be as an agent for society at large
As ideal agents for patients, physicians would rather than for individual patients. Under this view of
consider not only the health consequences of their de- agency, the physician would be compelled to allocate
cisions but also the economic and psychological con- resources in the most cost-effective manner in order to
sequences for their patients. Out-of-pocket costs, time achieve maximal value for money on behalf of society.
and inconvenience, and reassurance from diagnostic In such a world, physicians would make decisions that
tests are all part of what patients value. A perfect agent are in less than the best interests of their individual
would consider all of these and weigh them against patients, because not all medically beneficial proced-
one another if necessary, as would the patient. A phys- ures would be provided. But, on aggregate, consumers
ician who recommends an expensive, new treatment would be better off because more health benefit would
that is not covered by the patient's insurance but that be achieved.
is equally effective as an insured procedure would not Or would they? Such an ethic would comprom-
be acting as a good agent. But suppose the patient pre- ise the fiduciary relationship between patient and
ferred a far more expensive, insured procedure whose physician. It would place the physician in a position
only advantage was that it avoided a small scar or a of making trade-offs with faceless patients, and then
few hours less waiting time. Where does the physician's explaining to patients why it was not "cost-effective" to
obligation as an agent stop? Clearly, physicians often do extra computed tomography (CT) or to prescribe the
depart from perfect agency even with the most honour- more expensive drug with a slightly better side effect
able and selfless of intentions. profile. Physicians trained to do the best for individual
Principal—agent relationships do not always patients would balk at this social agency role.
achieve the goal of perfect proxy decision-making
and selfless advocacy if the incentives facing the agent
Major Sea Change
lead to deviations from the decision that is best for the
principal. For example, physicians may obtain different The dilemma, then, is that consumers collectively, con-
levels of remuneration, professional stature, or satisfac- cerned about health care costs, have an interest in cost-
tion from their actions, and these incentives compete consciousness in the clinic and at the bedside, while as
with the incentives that are aligned with the interests patients they expect a perfect agency relationship. One
of their patients, such as a sense of obligation and cog- possible class of solutions may entail collective action
nisance of external monitoring of their quality of care. to constrain (not necessarily regulate!) physicians to act
As noted before, physicians are often unable to be in the collective interest while allowing them to strive to
perfect agents for their patients because of constraints maximise the welfare of their individual patients. This
placed upon them. They have limited time during is the type of solution that economists advocate, for
the day (and even if they work extra hours, their example, to induce manufacturers to restrict pollution
or to induce consumers to recycle. But in health care, regulation of medical services. Managed care organisa-
it is not clear that consumers will accept such collect- tions, bound by bottom-line concerns, keep a close watch
ive solutions until they first buy into the premise that on the resource allocations of their affiliated providers.
resources are limited. I believe that a combination of The result has been less this sort of decentralised decision-
incentives, constraints, and a major sea change in cit- making than a system dominated by prior approval and
izen attitudes toward the finitude of medical care will sanctions against providers for overutilisation. Whether
be required to save the medical commons. insurers and managed care organisations actually consider
cost-effectiveness in their regulation of care is uncertain16
The Role of Collective Action If they do, their perspective on both cost and effective-
ness departs from that of society at large by virtue of their
The goal of a collective solution to protect the medical short time horizons, responsiveness to popular demands
commons would be served by placing constraints on over clinical effectiveness, and other factors.17 The result
physicians' choices such that, even as they strive to do has been a large number of dissatisfied consumers and an
the best they can on behalf of their patients, the result even larger proportion of disgruntled physicians.
of their doing so leads to cost-effective resource alloca- While some economists might applaud the decen-
tion. The constraints would, in effect, force physicians tralised, provider-centred solution in theory, it does
to consider the opportunity cost of their decisions, just have a number of practical flaws. For one, physicians
as shopping consumers are bound by the cash in their would have to fend off patients whose demands for
wallet and their credit balance. The consumer seeks health care services were not being met. There would
to maximise her well-being, subject to a budget con- be a tendency to give more attention to the loudest,
straint. The idea is to get the physician to take cogni- most assertive patients, relative to the cost-effectiveness
sance of the collective budget constraint. of their claims on the resource pool, and relatively less
Many examples of incentives to consider oppor- attention to more passive patients. Perhaps a major
tunity costs pervade the existing health care systems barrier to successful implementation of the provider-
of the industrialised world. Capitated (per patient) centred model is the need for buy-in on the part of
payments to physicians encourage them to use their both physicians and consumers. I will return to this
time efficiently. Capitated budgets for hospitals have essential ingredient later.
a similar effect at the institutional level. One problem
with capitation as an incentive at the local or indi- The Role of Physicians under
vidual physician level, however, is that it mitigates Resource Constraints
the advantage of insurance and risk pooling. If all
providers allocate their separate budgets to maximise A responsible physician who cares for a panel of patients,
health improvement, the overall consequence may be but who is either faced with a resource budget or account-
suboptimal if the most cost-effective opportunities to able for the resources he consumes, would be placed
improve health are spread unevenly among providers. in a situation similar to the emergency room physician
Over time, monitoring of caseloads and case-mix could performing triage. The physician would be responsible
be used to adjust the budget levels and thereby equalise for setting or implementing priorities for care and could
the incremental value of resources in different settings. invoke cost-effectiveness data to help guide these prior-
If physicians consider both QALYs (or another so- ities. This would enable physicians to consider the in-
cially desirable measure of health outcome) and cost cremental health benefit they could offer to each patient,
when deciding how to allocate their budgets, the result with perhaps some added moral consideration given to
would produce the societal allocation implied by cost- applying resources to the most desperate cases.n What-
effectiveness analysis. Physicians in that setting would ever metric the physicians use, whether implicitly or ex-
be led to consider the cost per QALY of alternative plicitly, they would be striving to maximise the welfare of
decisions, and the result would be the maximum pos- their patient population. Like parents taking care of their
sible production of QALYs by that physician. If phys- several children—decisions about what clothing to buy
icians consider other values, for example, if they give for each, which ones to send to sports camp, which to
additional priority to the sickest patients as suggested send to college—physicians could approach these inter-
by Nord,' then the result would reproduce the cost- personal comparisons in a caring, compassionate way.
effective result based on those values. Given the complexity of medical decisions, and in
The current situation in the United States, domin- light of the growing recognition of the role of evidence
ated by managed care, has largely eschewed this sort of on effectiveness and cost in medical decisions, phys-
decentralised, provider-centred system in favour of direct icians must work together to develop and interpret the
evidence to support a mutually acceptable framework standards on behalf of their shareholders, but the laws
for cost-effective decision-making. This entails physician are enforced, and they learn to do the best they can for
participation at two levels: helping to set the constraints the shareholders within the constraints.
within which they practise and formulating flexible
guidelines to help them allocate the resources under their Concluding Remarks:The Need for"Buy-In".
stewardship. These, then, are the additional roles phys-
icians must play in order to ensure resource allocations Should physicians do cost-effectiveness analysis at the
that are responsive to patients' needs and preferences. bedside, even if they are not compelled to ration care
If physicians accept this role of stewardship for their by externally imposed constraints? To do so would
portion of the medical commons, they can continue to place them in an untenable position as agents for two,
exercise their roles as agents and advocates for patients. sometimes adversarial principals—the individual pa-
tient in the clinic and the larger community. Patients
would lose the undivided advocacy to which they are
An Apt Analogy: Pollution Control
accustomed, and the pressures on physicians to be re-
It would be unrealistic, and a violation of the trust pa- sponsive would be overwhelming.
tients place in their physicians, to expect physicians to Collective action is required to constrain individual
practise cost-effectively without some form of external providers of health care to practise within their collective
constraint. The ethical forces acting upon physicians means. These constraints could be implemented in and
will pull in the opposite direction if physicians are left to by organisations as diverse as managed care organisations
make trade-offs between their own patients and other, and national health services. Or they could be agreed
unseen, patients. The force compelling them to attend to upon by consensus of provider organisations and citizens.
their own patients is, and should be, stronger, with the Decision-making within these constraints would invite
result that society's commons problem is left unattended. cost-effectiveness analysis by individual providers, but
The situation is analogous to the control of pollu- now the choices would be among members of the phys-
tion. As in health care, self-interest (it does not matter ician's own "family" of patients, just as the parent decides
whether the self-interest reflects economic or health how to tend to the needs and wants of her several chil-
values) competes with the interests of the greater society. dren. The physician would be free to advocate for each
The managers of companies, especially publicly owned patient, to do the best he can to "[preserve] life capacities
ones, have a primary fiduciary obligation to their share- for the realization of a reasonable, realistic life plan."15
holders. If profit maximisation entails the discharge of Such a means of protecting the "medical commons"
pollutants, then it would be a violation of their agency could result in resentment on the part of physicians and
relationship to shareholders to volunteer to regulate patients alike, unless both accept the underlying ration-
pollution more than their shareholders would want. ale for the constraints. The system of utilisation review
But other members of society, who are not sharehold- by managed care organisations in the United States has
ers, have an interest in pollution control. Their interest spawned a movement in support of a return to physician
in the company, while perhaps not zero because the control of decision-making. But a return to unconstrained
costs of pollution control would be passed on to them use of technologies and resources will fail because of
in part through higher product prices, would be greatly the limits at the societal level. The only way out of this
attenuated by comparison with the adverse effects dilemma is for citizens and physicians to accept the con-
of the pollution. Should these managers be expected cept and consequences of resource limits, just as they
to restrict pollution voluntarily below the levels their accept speed limits, zoning laws, and other self-imposed
shareholders would want? Would it be ethical for them constraints in the interest of the greater good.
to do so? Recognising that a valid argument could be
made that it would be unethical for the managers to fail
Attitude of Consumers
to abide by their shareholders' wishes, society adopts
pollution control regulations (or pollution discharge How can this acceptance of resource constraints come
quotas or taxes). Responsible companies, while they about? There is some indication that physicians are
may lobby for less stringent standards, higher quotas, or already willing to accept the idea of limits, as long as
lower taxes, generally accept the rationale for collective they have some degree of control of or participation
action to protect the greater interest of the commun- in the process of setting them. Medical specialty soci-
ity. They then set about their business of maximising eties have begun to incorporate cost-effectiveness into
profits within the constraints. Sometimes they attempt their formulation of clinical guidelines. A recent study
to bypass the constraints or advocate loosening of the showed good concordance between the rankings by a
II MILLI
consensus panel of physicians of appropriateness of a glamorous medical interventions, such as screening for
medical technology in different indications-coronary colorectal cancer and adult vaccinations against influ-
angiography after myocardial infarction-and cost- enza and pneumonia, are underutilised in favour of the
effectiveness ratios for these indications.'8 procedures consumers want and demand. An informed
The biggest obstacle to cost-effective resource al- populace, aware of the finite benefits of health care
location along these lines is the attitude of consumers. services and of the rationale for constrained choice in
People who expect everyone to have access to all pos- the clinic and at the bedside, is an essential ingredient if
sible medical care regardless of cost are bound to be dis- physicians are to be burdened with the responsibility for
appointed. In the United States, the sustenance of this cost-effective decision-making. The evidence suggests
myth has come at a high price. Even in times of economic that the medical profession will accept that responsibil-
prosperity, an increasing proportion of Americans is un- ity if it is given the authority to set the rules by which
insured, and millions of them have limited or no access its members are to play and if their patients are at peace
to medical care. Some of the most cost-effective but least with the principle of living within their means.
References
1. Hiatt HH. Protecting the medical commons: who is respon- 11. Rawls J. A theory of justice. Cambridge, MA: Harvard
sible? New England Journal of Medicine 1975;293:235-41. University Press, 1971.
2. Hardin G. The tragedy o f the commons. Science 12. Nord E. An alternative to QALYs: the saved young life
1968;162:1243-8. equivalent. British Medical Journal 1992;305:875-7.
3. Weinstein MC, Stason WB. Foundations of cost- 13. Aaron H J, Schwartz WB. The painful prescription: ra-
effectiveness analysis for health and medical practices. tioning hospital care. Washington: The Brookings Institu-
New England Journal of Medicine 1977;296:716-21. tion, 1984.
4. Patrick DL, Erickson P. Health status and health policy: 14. Pauly MV. The economics of moral hazard. American
allocating resources to health care. New York: Oxford Economic Review 1968;58:231-7.
University Press, 1993. 15. Fried C. Medical experimentation: personal integrity and
5. Gold MR, Siegel JE, Russell LB, Weinstein MC, eds. Cost- social policy. New York: American Elsevier Publishing
effectiveness in health and medicine. New York: Oxford Co, 1974.
University Press, 1996. 16. Prosser LA, Koplan JP Neumann PJ, Weinstein MC.
6. Torrance GW. Measurement of health state utilities for eco- Barriers to using cost-effectiveness analysis in managed
nomic appraisal. Journal of Health Economics 1986;5:1-30. care decision making. American Journal of Managed Care
7. Russell LB. Is prevention better than cure? Washington: 1999;6:173-9.
The Brookings Institution, 1986. 17. Weinstein MC. Principles of cost-effective resource
8. Weinstein MC. The costs of prevention. Journal of Gen- allocation in health care organizations. International
eral Internal Medicine 1990;5(suppl):89-92s. Journal of Technological Assessment of Health Care
9. Garber AM, Phelps CE. Economic foundations of 1990;6:93-103.
cost-effectiveness analysis. Journal of Health Economics 18. Kuntz KM, Tsevat J, Weinstein MC, Goldman L. Expert
1997;16:1-31. panel vs decision-analysis recommendations for post-
10. Harsanyi JC. Cardinal welfare, individualistic ethics, and discharge coronary angiography after myocardial
interpersonal comparisons of utility. Journal of Political infarction. Journal of the American Medical Association
Economy 1955;63:309-21. 1999;282:2246-51.
Caring in a Crisis: The Ethical to relieve suffering whenever possible. However,

this duty does not exist in a vacuum and depends
Obligations of Physicians and Society on the provision of goods and services referred to
during a Pandemic as reciprocal obligations, which must be provided
by governments, health care institutions, and other
relevant bodies and agencies. The obligations of
government and society to physicians can be seen
Inherent in all health care professional Codes of as comparable to the obligations of physicians to
Ethics is the duty to provide care to patients and their patients.
Source: From CMA POLICY (2008). Reprinted with permission of the Canadian Medical Association.
Canadian Medical Association: Caring in a Crisis
The recent experience of Canadian physicians resources are needed and how these resources would
during the SARS epidemic in Toronto has heightened the best be employed during a pandemic. These resources
sensitivities of the medical profession to several issues might include physician services but could also include
that arose during the course of dealing with that ill- access to vaccines and medications, as well as access
ness. Many of the lessons learned (and the unanswered to equipment such as ventilators or to a bed in the
questions that arose) also apply to the looming threat of intensive care unit. According to paragraph 43 of the
an avian flu (or other) pandemic. Canadian physicians CMA Code of Ethics, physicians have an obligation to
may be in a relatively unique position to consider these "Recognize the responsibility of physicians to promote
issues given their experience and insight. equitable access to health care resources."
The intent of this policy is to highlight the ethical In addition, physicians can reasonably b e expected
issues of greatest concern to practising Canadian phys- to participate in the process of planning for a pandemic
icians which must be considered during a pandemic. or other medical disaster. According to paragraph 42
In order to address these issues before they arise, the of the CMA Code of Ethics, physicians should "Recog-
CMA presents this paper for consideration by individ- nize the profession's responsibility to society in matters
ual physicians, physician organizations, governments, relating to public health, health education, environ-
policy-makers, and interested bodies and stakeholders. mental protection, legislation affecting the health and
Although many of the principles and concepts could well-being of the community and the need for testi-
readily be applied to other health care workers, the mony at judicial proceedings." This responsibility
focus of this paper will be on physicians. could reasonably be seen to apply both to individual
Policies regarding physicians in training, including physicians as well as the bodies and organizations that
medical students and residents, should be clarified in represent them.
advance by the relevant bodies involved in their over- Physicians also have an ethical obligation to recog-
sight and training. Issues of concern would include nize their limitations and the extent of the services they
the responsibilities of trainees to provide care during a are able to provide. During a pandemic, physicians
pandemic and the potential effect of such an outbreak may be asked to assume roles or responsibilities with
on their education and training. which they are not comfortable, nor prepared. Para-
graph 15 of the CMA Code of Ethics reminds physicians
A. Physician Obligations during a Pandemic to "Recognize your limitations and, when indicated,
recommend or seek additional opinions or services."
The professional obligations of physicians are well However, physicians have moral rights as well as
spelled out in the CMA Code of Ethics and other docu- obligations. The concept of personal autonomy allows
ments and publications and are not the main focus of physicians some discretion in determining where, how,
this paper. However, they will be reviewed and dis- and when they will practise medicine. They also have
cussed as follows. an obligation to safeguard their own health. As stated
Several important principles of medical ethics will in paragraph 10 of the CMA Code of Ethics, phys-
be of particular relevance in considering this issue. icians should "Promote and maintain your own health
Physicians have an obligation to be beneficent to their and well-being."
patients and to consider what is in the patient's best The SARS epidemic has served t o reopen the ethical
interest. According to the first paragraph of the CMA debate. Health care practitioners have been forced to
Code of Ethics (2004), "Consider first the well-being reconsider their obligations during a pandemic, includ-
of the patient." ing whether they must provide care to all those in need
Traditionally physicians have also respected the regardless of the level of personal risk. As well, they
principle of altruism, whereby they set aside concern have been re-examining the obligation of governments
for their own health and well-being in order to serve and others to provide reciprocal services to physicians
their patients. While this has often manifested itself and the relationship between these obligations.
primarily as long hours away from home and family
and a benign neglect of personal health issues, at times B. Reciprocal Obligations towards Physicians
more drastic sacrifices have been required. During
previous pandemics, many physicians have served self- While there has been much debate historically (and
lessly in the public interest, often at great risk to their especially more recently) about the ethical obliga-
own well-being. tions of physicians towards their patients and society
The principle of justice requires physicians to con- in general, the consideration of reciprocal obligations
sider what is owed to whom and why, including what towards physicians is a relatively recent phenomenon.
uW I II ;ill III W
During the SARS epidemic, a large number of • Physicians should be made aware of a clear plan
Canadian physicians unselfishly volunteered to assist for resource utilization, including:
their colleagues in trying to bring the epidemic under - how physicians will be relieved of duties after a
control. They did so, in many cases, in spite of signifi- certain time;
cant personal risk and with very little information about - clearly defined roles and expectations, espe-
the nature of the illness, particularly early in the course cially for those practising outside of their area
of the outbreak. Retrospective analysis has cast signifi- of expertise;
cant doubt and concern on the amount of support and - vaccination/treatment plans—will physicians
assistance provided to physicians during the crisis. (and their families) have preferential access
Communication and infrastructure support was poor based on the need to keep caregivers healthy
at best. Equipment was often lacking and not always and on the job;
up to standard when it was available. Psychological - triage plans, including how the triage model
support and counselling was not readily available at might be altered and plans to inform the public
the point of care, nor was financial compensation for of such.
those who missed work due to illness or quarantine. Physicians should have access to the best equip-
Although the Ontario government did provide retro- ment needed and should be able to undergo extra
spective compensation for many physicians whose training in its use if required.
practices were affected by the outbreak, the issue was • Politicians and leaders should provide reassur-
addressed late and not at all in some cases. ances that satisfy physicians that they will not be
It is clear that Canadian physicians have learned "conscripted" by legislation.
greatly from this experience. The likelihood of indi-
viduals again volunteering "blindly" has been reduced During a Pandemic
to the point where it may never happen again. There • Physicians should have access to up-to-date, real-
are expectations that certain conditions and obliga- time information
tions will be met in order to optimize patient care and • Physicians should be kept informed about de-
outcomes and to protect health care workers and their velopments in Canada and globally.
families. • Communication channels should be opened with
Because physicians and other health care providers other countries (e.g., Canada should participate in
will be expected to put themselves directly in harm's way wHo initiatives to identify the threats before they
and to bear a disproportionate burden of the personal arrive on our doorstep).
hardships associated with a pandemic, the argument • Resources should be provided for backup and
has been made that society has a reciprocal obligation relief of physicians and health care workers.
to support and compensate these individuals. • Arrangements should be made for timely provision
According to the University of Toronto Joint Centre of necessary equipment in an ongoing fashion.
for Bioethics report, We Stand on Guard for Thee, "[The • Physicians should be compensated for lost clin-
substantive value of] reciprocity requires that society ical earnings and to cover expenses such as lost
support those who face a disproportionate burden in wages, lost group earnings, overhead, medical
protecting the public good, and take steps to minimize care, medications, rehabilitative therapy, and other
burdens as much as possible. Measures to protect relevant expenses in case of quarantine, clinic can-
the public good are likely to impose a disproportion- cellations, or illness (recognizing that determining
ate burden on health care workers, patients and their exactly when or where an infection was acquired
families." may be difficult).
Therefore, in order to provide adequate care for • Families should receive financial compensation in
patients, the reciprocal obligation to physicians re- the case of a physician family member who dies as
quires providing some or all of the following: a result of providing care during a pandemic.
• In the event that physicians may be called upon
Prior to a Pandemic in a pandemic to practise outside of their area of
• Physicians and the organizations that represent expertise or outside their jurisdiction, they should
them should be more involved in planning and contact their professional liability protection pro-
decision-making at the local, national, and inter- vider for information on their eligibility for protec-
national levels. In turn, physicians and the organ- tion in these circumstances.
izations that represent them have an obligation to • Interprovincial or national licensing programs
participate as well. should be developed to provide physicians with
Canadian Medical Association: Caring in a Crisis 191
backup and relief and ensure experts can move since there is now some time to plan for the next pan-
from place to place in a timely fashion without demic and to ensure that reciprocal obligations can be
undue burden. met prior to its onset. Physicians have always provided
• Psychological and emotional counselling and sup- care in emergency situations without questioning what
port should be provided in a timely fashion for they are owed. According to paragraph 18 of the CMA
physicians, their staff, and family members. Code of Ethics, physicians should "Provide whatever
• Accommodation (i.e., a place to stay) should appropriate assistance you can to any person with
be provided for physicians who have to travel an urgent need for medical care." However, in situa-
to another locale to provide care or who don't tions where obligations can be anticipated and met
want to go home and put their family at risk, in advance, it is reasonable to expect that they will
when this is applicable, i.e., the epidemiology be addressed. Whereas a physician who encounters
of the infectious disease causing the pandemic an emergency situation at the site of a car crash will
indicates substantially greater risk of acquiring act without concern for personal gain or motivation,
infection in the health care setting than in the a physician caring for the same patient in an emer-
community. gency department will rightly expect the availability of
• Billing and compensation arrangements should proper equipment and personnel.
ensure physicians are properly compensated for In order to ensure proper patient care and physician
the services they are providing, including those safety, and to ensure physicians are able to meet their
who may not have an active billing number in the professional obligations and standards, the reciprocal
province where the services are being provided. obligations outlined above should be addressed by the
appropriate body or organization.
After a Pandemic
• Physicians should receive assistance in restarting
Conclusion
their practice (replacing staff, restocking overhead,
communicating with patients, and any other costs If patient and physician well-being is not optimized
related to restarting the practice). by clarifying the obligations of physicians and society
• Physicians should receive ongoing psychological prior to the next pandemic in spite of available time
support and counselling as required. and resources necessary to do so, there are many who
would call into question the ethical duty of phys-
C. How Are Physician Obligations and icians to provide care. However, the CMA believes
Reciprocal Obligations Related? that, in the very best and most honourable traditions
of the medical profession, its members will provide
Beyond a simple statement of the various obligations, care and compassion to those in need. We call on
it is clear that there must be some link between these governments and society to assist us in optimizing
different obligations. This is particularly important this care for all Canadians.
Why Medical Professionals Have experiments and the abuse of prisoners there and then?
It is good that conscientious objection rights exist to
No Moral Claim to Conscientious protect good doctors refusing to participate in such
Objection Accommodation in crimes." There are more sensible versions of this argu-
Liberal Democracies ment, such as one published by an admittedly not "very
courageous" doctor who wrote in the British Medical
Udo Schuklenk and Ricardo Smalling Journal that he would have hoped to object on con-
science grounds to the abuse of prisoners in Stalinist
Introduction Russia and claims that knowing that he was "part of
an independent medical profession with allegiance to
It is not unusual for students in any given bioethics something higher and more enduring than the regime
class to offer something like the following defence of of the day" would have increased the odds of him doing
conscientious objection rights: "Remember the Nazi the right thing.' We know, of course, in bioethics, that
Source: From Journal of Medical Ethics 43 (2016), 4. Reprinted with permission.

MI I I I Al II I La
whenever a Nazi analogy is brought up to defend a par- be instances of conscientious objection in other kinds
ticular normative stance, the odds are that that stance is of societies. There might also be conscientious objec-
weakly supported, if not outright indefensible. Unlike tion in other contexts, for instance, in case of military
contemporary liberal democracies, the Nazis and conscripts. This article will not address those contexts.
Stalin's Soviet Union did not respect conscientious ob- We are concerned only with conscientious objectors
jectors and neither did Pol Pot's henchmen and hench- who decided to join a particular profession (in this case
women. Well-known pacifists such as Hermann Stohr, medicine) voluntarily and who then wish to be exempt
the leader of the German branch of an international from providing services that are typically expected of
pacifist organisation, perished in German concentra- that profession.
tion camps.' Conscientious objection typically does not
flourish in dictatorships. Toleration of conscientious Conscientious Objection in Liberal
objectors in liberal democracies does little to support Democracies: Why Should We Respect It?
conscientious objectors in North Korea. The concept
has evidently little currency there: its utility and legit- Let us begin by trying to understand what conscience
imacy cannot be defended by pointing to Nazi Germany actually is. Daniel Sulmasy describes it as something
or Stalinist Russia. The same cannot be said for liberal that seems to operate both retrospectively as well as
democracies where respect for both individual as well prospectively, it impacts on particular past actions, and
as professional autonomy ranks highly. The medical it impacts on how we evaluate normatively possible
historian John J. Michalczyk got it probably right when future actions.' He conceptualises conscience both as
he noted that "those who invoke the Nazi analogy in a our conviction that we should act in accordance with
broad or general fashion are pressing the limits of valid our individual understanding of what morality de-
analogy simply because the broader the scope of their mands of us and on wilfully and voluntarily acting in
reference, the harder it becomes to understand exactly accordance with what we consider to be morally good
what they think the Holocaust was, and thus why it and right' It is uncontroversial that asking someone
is of moral relevance to the current issue."3 We aim to to violate their conscience-based convictions in mat-
take the ethical debate about conscientious objection in ters that are of great importance to them is also asking
medicine back to where it currently properly belongs, such individuals to accept a potentially fairly high
namely, liberal Western democracies where some med- psychological cost. Typically, when we act contrary to
ical doctors wish to see their private moral or religious our conscience, we will be plagued by guilt and pos-
objections to the provision of certain professional med- sibly worse. Hermann Stohr died rather than recant his
ical procedures accommodated by regulatory regimes. pacifist convictions. There are others like him.
This is invariably at a cost to patients hoping to access Why then should we, prima facie at least, toler-
medical services that they are legally entitled to access. ate conscientious objectors in liberal societies? There
Liberal democracies rightly do not take a stance are various reasons that have been suggested for this.
on the substance of their citizens' moral or religious or One is obvious from the above. Prima facie no soci-
other convictions; what is protected in liberal democ- ety should force conscientious objectors to suffer the
racies' constitutions is the citizens' right to hold such psychological cost that they would incur if they were
beliefs and live by them—within reason. Typically, no forced to act against their conscience. Peter West-Oram
distinction is and should be drawn between religious and Alena Buyx have offered a number of other rea-
and other moral convictions, but that is not always the sons. They write, "the right to exempt oneself from
case.4'5 In the kinds of societies that we are concerned the fulfilment of a generally held duty is typically
about in this paper, the vast majority of litigated cases justified on the grounds that such a right is vital for
are triggered by religious conscientious objectors as the preservation of freedom of conscience. The latter
opposed to secularists or atheists.6 That in its own is itself argued to be a core value of pluralist, liberal-
right is not a reason to disregard such complaints, be- democratic states, and 'a moral right' Further, the
cause the protection of an individual's rights to adopt rights to freedom of conscience and conscientious
significant beliefs and live by them is at the heart of objection are argued to be constitutive of liberty and
what living one's own life in a liberal polity is all about. autonomy and to be necessary for the preservation of
For all practical intent and purposes, we are discuss- individual moral integrity. In promoting these goods,
ing predominantly religiously motivated conscientious the rights are argued to be vital for the adequate tolera-
objectors in the medical profession who ask that their tion of different moral and philosophical perspectives
objections to the delivery of particular professional ser- in a pluralistic society"9 These are powerful reasons.
vices are protected by the secular state. There might On this reading, respect for an individual's conscience
Schuklenk/Smalling: Why Medical Professionals Have No Moral Claim to Conscientious Objection Accommodation
is something of a corollary of the principle of respect for follow but his own reason, yet it is not so with
persons. Mark Wicclair has suggested in an influential him that lives in a Commonwealth, because the
paper that a blanket refusal to tolerate conscientious law is the public conscience by which he hath al-
objection would constitute a significant threat to some ready undertaken to be guided. Otherwise in such
doctors' moral integrity? Similar views are expressed diversity as there is of private consciences, which
by Pellegrino? It is not surprising then that there ap- are but private opinions, the Commonwealth must
pears to be a broad consensus between religious voices needs be distracted, and no man dare to obey the
in bioethics° and liberals° according to whom the fail- sovereign power farther than it shall seem good in
ure to tolerate, protect, and respect some conscientious his own eyes.18
objectors is incompatible with what it means to live
together in a liberal multicultural society. It is also fair Poignantly, Hobbes makes this point in a chapter on
to say that some bioethicists have voiced more or less "those things that weaken or tend to the dissolution of
strong opposition to the view that we should accom- a commonwealth." Of course, Hobbes is not quite our
modate medical doctors' conscientious objections.14-18 archetypical defender of liberal democracies, but the
As mentioned in the "Introduction" section, based point he is making here is valid; it applies to the case
on the review of hundreds of legal cases reported in of conscientious objectors in liberal democracies, too.
Brian Leiter's book Why Tolerate Religion?, respect for Let us begin our argument by making the gen-
conscience in the 21st century translates into a fairly eral case for why conscientious objection in medicine
one-sided affair: it is fought for and demanded by re- should not be accommodated.
ligious healthcare professionals without much regard We recognise that conscientiously objecting
for actual patient care and health outcomes or indeed healthcare professionals can have various rationales to
respect for these patients' moral choices. Considering support their opposition to the participation of doc-
the significant power differential between patients and tors in particular medical procedures. They include
doctors, this is remarkable in its own right Whenever typically a recourse to tradition, the Hippocratic Oath,
there is a conflict, on this account, a doctor's private the Bible, the Quran, and any number of other docu-
ideological convictions generally take precedence in ments that have no legitimate bearing on the practice of
the professional practice of medicine. Various bioethi- 21st century medicine. It is easy to show that even the
cists have tried to suggest limiting criteria to avoid objecting doctors' standard recourse to the Hippocratic
the possibility that arbitrary stances taken by doctors Oath when it comes to abortion and assisted dying is
could pass successfully the conscience muster." Typ- not always credible.19,2° It is evidently the case that
ically, conscientious objections occur in the context of other ideological convictions held by particular doc-
reproductive health (e.g., abortion, in vitro fertilisation, tors motivate their deployment of the argument from
contraceptives), end-of-life care (e.g., assisted dying), tradition, in the guise of the Hippocratic Oath, because
and also affect negatively gays, lesbians, and patients only the lines from the Oath that suit a particular ob-
with gender dysphoria. jecting doctor's interests are typically cited, while those
they disagree with are quietly ignored. But in any case,
Why Conscientious Objection Has No the Oath is not a defensible ethical guide to modern
Place in the Practice of Medicine medical practice." Mercier might be right in this con-
text when she writes that
It is worth noting that scepticism about the importance
of an individual's conscience claims within a commun- religion is all about believing that one's beliefs are
ity has been expressed by influential thinkers for many right, but not about having right beliefs. If first-
years. Thomas Hobbes had this to say in his Leviathan: order religious beliefs had content, their content
could be checked against the truth. It is precisely
another doctrine repugnant to civil society is that because such beliefs lack content that one can
whatsoever a man does against his conscience go on about believing that one believes them
is sin; and it dependeth on the presumption of despite any and every evidence. But the price of
making himself judge of good and evil. For a man's second-order belief in vacant first-order beliefs is
conscience and his judgement is the same thing; self-deception."
and as the judgement, so also the conscience may
be erroneous. Therefore, though he that is sub- On this background, it is fortunate from the con-
ject to no civil law sinneth in all he does against scientious objector's perspective that secular liberal
his conscience, because he has no other rule to democracies do not typically test whether the views
II I I I II 111111
conscientious objectors profess to subscribe to are on the provision of particular procedures, should be
defensible. What matters is that they are deeply held accommodated, given that this toleration subverts the
or, more to the point, that the conscientious objectors very objectives the profession is designed to achieve.
claim that they hold those convictions deeply.23 Even on This does not deny anyone the right to hold any
this count we are incoherent. A female Muslim doctor number of private religious and moral views, as they
refusing to see a male patient would not be granted a see fit and as they choose to hold. What we are deny-
conscientious objection exemption, whereas a pharma- ing is that professionals are entitled to subvert the ob-
cist refusing to sell contraceptives in some countries jectives of the professions they voluntarily joined by
might. Let us assume that both of these conscience prioritising their private beliefs over the professional
decisions were loudly proclaimed to be genuinely held delivery of services to the public, especially when they
by the professionals in question. Why should one be are monopoly purveyors of these services. Legal scholar
respected and the other one not? It could be argued Alta Charo called it right, when she wrote, "claiming an
that in liberal democracies, constitutional documents unfettered right to personal autonomy while holding
and human rights legislation variously protect against monopolistic control over a public good constitutes an
gender-based discrimination but may not guarantee abuse of the public trust—all the worse if it is not in
access to contraception. While this may be true, it fact a personal act of conscience but, rather, an attempt
misses the point of the argument—that the substance at cultural conquest."25
of conscience-based objections that are protected is Various authors have warned that an unqualified
arbitrary. Take as an example the unfolding debates right to conscientious objection in medicine would
around marriage equality. Until quite recently this was a result in harmful consequences, for instance, for
concept alien to liberal constitutions and human rights women wanting to have an abortion, who could be
legislations. Those with conscience-based objections denied timely access to this healthcare procedure not
to same sex marriage were free to assert those beliefs because of legal constraints but because of healthcare
against such couples. However, that is increasingly no professionals refusing to provide such a healthcare
longer the case. Nothing has changed about the same service on conscience grounds. Frequently, a middle-
sex couples or the beliefs of the objectors; society has ground is sought, as legal systems are indeed rare where
simply decided that such objections will no longer be conscientious objection rights cannot be overridden
tolerated. by a competing stronger right (e.g., a patient's right to
Furthermore, just as we cannot test the plausibility access healthcare in a timely manner). Typically, then,
of the ideological dicta that lead to conscientious objec- the argument is not over whether conscience rights are
tions (there is no test for the existence of "God," for defensible but about what kind of compromise is rea-
instance, or truth of the Bible), it is also impossible to sonable for everyone affected.26,27
ascertain whether conscientious objectors actually hold
the views they profess to hold. The US Supreme Court Unfettered versus Fettered Conscientious
writes on this issue, Objection Accommodation
It could be objected that the argument advanced in this
What principle of law or logic can be brought to paper is too simplistic as it appears to target only un-
bear to contradict a believer's assertion that a par- fettered conscientious accommodation stances, when
ticular act is "central" to his personal faith? Judging the focus in today's policy debates is about fettered con-
the centrality of different religious practices is akin scientious objection accommodation. It may be argued
to the unacceptable "business of evaluating the that the question really is: where and how could society
relative merits of differing religious claims" . . . it is draw reasonable lines that take into account societal
not within the judicial ken to question the central- interests as well as those of individual objectors? The
ity of particular beliefs or practices to a faith, or preceding section explains to some extent why we have
the validity of particular litigants' interpretation of taken this stance. The courts in various jurisdictions
those creeds . . . courts must not presume to deter- have already, rightly so, conceded that it is a moot point
mine the place of a particular belief in a religion or trying to establish the truth or even plausibility of the
the plausibility of a religious claim.24 views purportedly held by conscientious objectors.
That is not an insignificant point, because this conces-
If that is the case, it remains unclear why untestable sion opens the door to any number of more or less
conscience claims from privileged professionals who arbitrary and random conscientious objection claims.
voluntarily chose to join a particular profession, and For policy-makers aiming to establish a functioning
who have been endowed by society with a monopoly healthcare system with predictable service delivery and
Schuklenk/Smalling: Why Medical Professionals Have No Moral Claim to Conscientious Objection Accommodation 195
guaranteed service levels to the people who finance have to depend on the goodwill of volunteering doctors
the system, this constitutes an insurmountable prob- who also happen to be the only licensed providers of
lem. It is nigh impossible to predict which healthcare this procedure in the country.
professional, in which part of the system, will demand A second example: At the time of writing, Can-
accommodation for which kinds of purported or real adian legislators grapple with the question of how to
convictions. It is also evidently impossible to verify implement a Supreme Court of Canada judgement
whether objecting healthcare professionals even hold that effectively decriminalises providing assistance
the views they profess to hold. Such claims may merely in dying to competent patients suffering from an ir-
be a convenient way out of the provision of inconven- reversible chronic illness that renders their lives not
ient healthcare services. worth living in their own considered view.2° Dr Cindy
In light of this, it seems to us that whatever at- Ford, the current president of the Canadian Medical
tempt at a principled fettered accommodation is made, Association, mirrors Dr Blackmer's stance on the obli-
it is by necessity resulting in arbitrary outcomes. Re- gations of conscientious objectors. In a statement made
gardless of the accommodations that we make today, to the Canadian parliamentary committee tasked with
these can and likely will be challenged time and again investigating possible legislative options on assisted
by new generations of conscientious objectors. Today dying, Dr Ford reportedly stated that it is unreasonable
it might be abortion and assisted dying, tomorrow it to require conscientious objectors to even refer such
might be the use of the tools of personalised medicine patients to a clinician who they know will be prepared
or something else altogether. Limiting conscientious to render assistance in dying.3° Dr Blackmer agrees
objection accommodation to defensible claims seems with her, while at the same time trying to assure the
impossible to us, unless we overcome the two prob- Canadian parliamentary committee members of reli-
lems mentioned: demonstrate the truth of the founda- able professional service delivery; as he says, "I can
tions of the conscientious objection and demonstrate sit here . . . and guarantee that from simply a number
evidence that objectors actually genuinely hold the perspective, access will not be an issue."31 Dr Black-
views they claim to hold. Failing that, as we will show, mer's argues that given that 30% of Canadian doctors
the inevitably ensuing arbitrary accommodation de- are willing to provide assistance in dying, there should
mands will have harmful real-world consequences as not be a problem for patients wanting to access such
far as healthcare outcomes and patient access to care services. If it was just a numbers game, Dr Blackmer
is concerned. would certainly be correct. Given the relatively small
number of eligible patients likely to ask for assistance
A Few Canadian Examples in dying, there should be a sufficient number of doctors
Lobbyists for doctors usually defend the view that doc- willing to provide assistance to such patients. However,
tors must never be required to provide services that as so often, the devil is in the detail. The problem is—
they object to on conscience grounds. Their focus is we suspect this is the reason why Dr Blackmer does
on ensuring that conscientious objectors do not have not actually offer unconditionally guaranteed access on
to oblige patients asking for such healthcare services behalf of his association—that this likely will not help
or even have to assist them in finding a healthcare seriously sick and vulnerable patients who happen to
provider who will oblige them. A good example of live in remote areas of Canada where only few doctors
this is Dr Jeff Blackmer, the Director of the Canadian happen to practise and where all of them might be op-
Medical Association's Ethics Office. Dr Blackmer writes posed to assisted dying on conscience grounds. It is
on behalf of the Canadian Medical Association that irrelevant then that elsewhere in the country plenty of
medical doctors are neither obliged to provide abortion doctors would be willing to provide assistance in dying
services, nor are they obliged to transfer patients on to on request if a particular patient is unable to travel to
doctors they know will provide abortions to women them. To these patients, Dr Blackmer's "guarantee" is
seeking one.28 Women are legally entitled to access worth very little.
abortions in Canada if they so wish, and typically abor- We should note here that the Canadian Medical
tions are fully funded by provincial public healthcare Associations stance is not unusually radical among its
systems. Canadian doctors are the only professionals peers. The British Medical Association, too, produced
legally entitled to provide such procedures. They are a guideline stipulating that "whenever possible phys-
monopoly providers. Dr Blackmer's considered view on icians who are conscientiously opposed to forgoing
this subject matter is that pregnant women who wish to treatment should be permitted to hand over care of
have access to a medical procedure that they are legally the patient to a colleague," but they are not obliged
entitled to, and that is fully funded by the state, should to do so.32 The entitlement mentality that drives these
II ,d11 1111.1.6/
associations' stances is unsurprisingly supported by the religion; however, it should not do so on the basis that
doctors they represent.33 it is a religious position but on the basis that "in reason
A further example of the potential impact of con- its merits commend themselves."4° Therefore, in order
scientious objection on patients' access to care in Canada to maintain a defensible balance between competing
can be seen in the recent application filed by various beliefs in a liberal and democratic society, such soci-
religious physicians and associations representing them eties have generally considered "the practice of religion
in the Ontario Superior Court. The application challen- and the choices it implies to relate more to individuals'
ges the College of Physicians and Surgeons of Ontario private lives or to voluntary associations."' To this end,
(cPSo) Policy Statement #2-15: Professional Obligations it has been recognised that within the public realm an
and Human Rights, which among other things requires individual's freedom of religion and conscience may be
that physicians "provide a referral to another appropri- legitimately burdened. In another case, the Supreme
ate health-care provider for the elements of care the Court of Canada explicitly acknowledged that legis-
physician is unable to manage directly."34 The appli- lative or administrative actions may increase the cost
cants, including the Christian Medical and Dental Soci- of practising or otherwise manifesting one's religious
ety of Canada and the Canadian Federation of Catholic beliefs.42 Further, such legislative or administrative
Physicians' Societies, described the referral provisions burdens may be justified where they prevent conduct
of the CPSO policy statement as "unconscionable" and that would potentially cause harm to or cause interfer-
a violation of their freedom of conscience and/or free- ence with the rights of others.43 The reasoning of the
dom of religion.35 This example is hardly unique.36 Court supports the view that the conscientious belief of
These kinds of hardline positions disregard the needs a physician may be legitimately halted at the point that
of patients. it interferes with the rights of patients to access services
Statements and actions of this kind strongly sug- they are entitled to receive. That interference, however,
gest that Professor Charo was not attacking strawmen is precisely what happens each time a conscientious
in her article. objector is accommodated and the patients are unable
to access the required service elsewhere.
Respect for Private Conscience Choices Is
Already Limited Respect for Private Conscience Choices Will Result
Regardless of the views espoused by representatives in Avoidable Suboptimal Access to Healthcare
of the Canadian Medical Association and the religious Any society that grants medical professionals a con-
doctors' groups mentioned, there are already limits scientious objection-based opt-out will have to accept
placed on the conscience choices of Canadians in suboptimal health outcomes with regard to proced-
various settings. The Supreme Court of Canada stated ures that are considered at that point in time part
that "freedom of conscience and religion protects the and parcel of good professional healthcare practice
right to entertain beliefs, to declare them openly and in those societies. Why would toleration of conscien-
to manifest them, while at the same time guaranteeing tious objection lead invariably to suboptimal health
that no person can be compelled to adhere directly or outcomes? Take the Canadian assisted dying example.
indirectly to a particular religion or to act in a manner It is apparent that the representatives of Canadian doc-
contrary to his or her beliefs."3' In that same case, the tors are unconcerned about the distress they would
Court went on to say that the "state's duty to protect inflict on—often—dying patients in rural areas who
every person's freedom of conscience and religion cannot access easily a doctor willing to provide as-
means that it may not use its powers in such a way sistance in dying. In many cases, doctors' objection
as to promote the participation of certain believers or could be more than an inconvenience., it could be an
nonbelievers in public life to the detriment of others."38 insurmountable barrier to access. A case in point is the
These principles are seen as cornerstones of main- small Canadian province of Prince Edward Island (also
taining a free and democratic society.39 known as PEI). Women in that province, at the time
This is not to say that physicians and other med- of writing, are required to leave the province if they
ical professionals should be denied the right to hold wish to have an abortion because healthcare providers
religious or conscience beliefs. In liberal democracies, are not offering this service in the province. Empirical
a diversity of beliefs and opinions is welcomed, and in- evidence elsewhere suggests strongly that "travel is a
dividuals are free to advocate for societal acceptance of barrier to accessing legal abortion."44 The situation in
their particular world view. It is also trite to say that the PEI is not unique in the world, far from it. Minerva
law in liberal democratic societies may protect a par- reported recently that some 70% of Italian gynaecolo-
ticular social or moral position espoused by a particular gists conscientiously object to performing abortions,
which is arguably one of the reasons for staggeringly profession. Another aspect of this has to do with the
high backstreet abortion rates in that country.45 Patient unfairness of this situation vis-à-vis the patient. Pa-
interests come clearly last here. As we have seen, the tients, as Ruth Purtilo rightly notes, "can expect to be
representatives of the monopoly providers of these treated fairly. Persons seeking treatment should not
healthcare services in Canada are quite content with be given advantage on the basis of arbitrary favourit-
that outcome, as long as the association's members' ism or be left out on the basis of arbitrary dislike."46
consciences—or, more precisely, their untestable Given the intractability of conscience claims, it is not
claims about the content of their consciences—are not unwarranted to characterise them as essentially arbi-
burdened or otherwise inconvenienced. This attitude trary dislikes. They might not be arbitrary in the eyes
is quite the opposite of what it means to be a profes- of the objector, but we cannot even be certain of that,
sional, where a promise is made to serve the public given our inability to test the objector's conscience
good and to serve patient interests first and foremost. claims. They are arbitrary with regard to what kinds of
Apparently such central features of what it means to be services particular patients can or cannot expect from
a professional can readily be sacrificed when arbitrary, particular professionals belonging to the same profes-
untestable individual conscience claims are mounted sion in the same jurisdiction.
by professionals. The nature of professions is that they provide priv-
ileged services to the public. They are quite unlike polit-
Respect for Private Conscience Choices Will Result ical parties, churches, animal welfare, or environmental
in Avoidable Inequitable Workloads for Doctors organisations. While the latter cater to particular sect-
It seems also uncontroversially true to us that tol- arian audiences, professionals must cater uniformly to
eration of conscientious objectors will have a everyone within the scope of professional practice. The
detrimental impact on healthcare practice in other re- services professionals provide to the public must not
spects. Because patients can expect to be shepherded be subject to their private normative judgements about
around among different healthcare professionals individual needs of members of the public, as long as
choosing "conscientiously" to provide different levels what these members of the public require falls within
of healthcare, there will be an increasing level of up- the scope of professional practice. After all, patients do
heaval visited upon healthcare systems. Worse, if the not seek out doctors for their individual convictions
Canadian Medical Association had its way, patients but to receive professional services that they are trained
would not even be shepherded through the system, to provide.
it would be up to them to find doctors obliging them There are other sounds reasons for refusing to tol-
for particular ailments that might be of concern to erate conscientious objectors in medicine. They have
doctors with any number of idiosyncratic private to do with the voluntary choices made by those who
views of the universe and on what may or may not decided to join the profession.
be ethically acceptable. An important consequence
of such a situation would be that doctors who are Choosing to Join a Profession Is a Voluntary
willing to provide services conscientiously refused by Activity Undertaken by an Autonomous Adult
some or many of their colleagues would have to carry It seems to us that the argument from voluntariness
an inequitable load of such work. We cannot think of is important in this context. Nobody is forced to
a good reason for why they ought to accept such an join a particular profession, medicine included. Pace
inequitable burden and, indeed, why a just society Christopher Cowley's argument that the practice of
should design a health delivery system that burdens medicine for some is a "calling"—one can still choose
such doctors unfairly. not to answer the "call."47 The decision to join the
medical profession is quite distinct to, say, joining the
Respect for Private Conscience Choices Will Result military as a conscript. Medical schools the world all
in Unpredictable and Unfair Service Delivery over reject most applicants, not because the applicants
Because of the unpredictable nature of what it is that are unsuitable to become doctors but because the
conscientious objectors will object to, patients can number of places they have available is always much
never be quite certain about the kinds of services that smaller than the number of people who apply. Anyone
they will be able to receive from a particular doctor, joining a profession knows, or should know, that it is
even their long-time family doctor. That is the oppos- ultimately up to society to determine the scope of pro-
ite of what can reasonably be expected of a profes- fessional practice. Society can do this through profes-
sion and professionals. Uniform service standards are sional bodies, but it can also choose to do so by other
arguably one of the hallmarks of what constitutes a regulatory means. Having a monopoly on the provision
II .41 111111
of services within that scope of practice is not some- To highlight just one example of this, again from
thing that comes cost-neutral to those who choose Canada, the past president of the Canadian Medical
voluntarily to join a particular profession. Indeed, Association, Dr Chris Simpson, on behalf of the as-
given the high-stakes competition to join the medical sociation, referred to assisted dying as a "therapeutic
profession in the first place, it is reasonable to suggest service," a mere 15 years after Wicclair suggested that
that doctors refusing to provide professional services core professional values would make such a descrip-
that are within the scope of practice should be replaced tion impossible.8° It is unlikely that the association
by someone who is willing to undertake the work. If at would refer to assisted dying as a "therapeutic service"
any given time a doctor is unable to continue practising if it thought that the provision of this service would
due to their—ultimately arbitrary—conscience views, violate Wicclair's claimed core values of the profession.
nothing would stop them from leaving the profession If it is a therapeutic service, as we agree it is, Canadian
and taking up a different vocation. This happens across doctors wanting to go the conscientious objector route
industries and professions very frequently. Profession- to avoid providing this service to eligible patients
als can be expected to take responsibility for the volun- could no longer avail themselves of Wicclair's analysis,
tary choices they make. because the "moral integrity of the medical profession"
would not be served by accommodating individual
Does the Nature of a Particular Procedure Matter? doctors who refuse to provide therapeutic services. It
It has been suggested that not all conscientious objec- is important to recognise that medical practice is also
tions fall into the same category. The reason why it a cultural practice that changes over time. No demo-
would not be acceptable for the female Muslim doctor cratic society and no medical profession should leave
in our example to refuse to see a male patient had to do the scope of the provision of professional medical
with that being in violation of the core values in medi- services to the vagaries of its members' personal ideo-
cine. Assisted dying or abortion, on the other hand, logical convictions.
would then be seen if not in violation of traditional
medical values but at least being sufficiently contro- What about Our Societal Ability to Recruit People
versial to justify the accommodation of conscientious to Join the Medical Profession?
objectors. Wicclair argues for one possible threshold An argument could be mounted that if we prevented
that a conscientious objection must meet in order to doctors from excusing themselves from providing
be prima facie worthy of accommodation. He proposes professional services that they happen to feel strongly
that a conscientious objection "has significant moral about, we might find it difficult to find a sufficient
weight only if the core ethical values on which it is number of people willing to become doctors. This
based correspond to one or more core values in medi- seems far-fetched to us, but it is conceivable that some
cine."48 Wicclair is, therefore, supportive of conscien- people might reconsider joining the medical profes-
tious objection to abortion or assisted dying because to sion if the accommodation of their feelings with regard
his mind neither of these medical procedures corres- to particular professional services is not guaranteed.
ponds uncontroversially to these core values. We think We indicated already that the profession and society
that he is mistaken about what he considers to be the would likely be better off if such people chose not to
core values of medicine, which would be applicable join the profession, seeing that they think that their
in these two instances, but we do not wish to debate idiosyncratic views of the universe should be of greater
the merits of this claim here. Rather, his argument also significance than the patients whose interests they
fails because despite his protestations to the contrary promised to further first and foremost as profession-
it is essentially an argument from tradition. What he als. The concern about recruitment problems does not
calls core values could be more aptly described as appear to be based on real-world evidence to begin
traditional values. However, it is in the nature of such with. A recent survey reports that countries such as
values that they are changing. Medicine is currently Sweden, Finland, and Iceland, which do not permit
arguably undergoing a paradigm shift from mainten- their doctors to opt out of the provision of abortion
ance of human life at nearly all cost to a quality-of- service on grounds of conscience, have not experi-
life focused approach.49 The traditional core values enced harmful consequences as far as these countries'
of medical practice are in flux, and that is one reason healthcare systems' ability to train a sufficient number
for why abortion is available in most liberal Western of doctors is concerned."
democracies, and that is the reason why the number of A somewhat related argument suggests that by
jurisdictions that are decriminalising assisted dying is preventing doctors from refusing service based on
steadily increasing. religious or conscience grounds, we would run the
risk of losing the participation of those kinds of in- of practice have been made. However, that is a far
dividuals in the profession. However, that argument cry from suggesting that doctors have a right to such
does not hold. Those who object to particular proced- accommodation.
ures could choose specialities that would not require
that they violate their conscience; for example, they Could Not the State Be the
could opt for dermatology instead of gynaecology if Guarantor of Access?
they are opposed to abortion. A self-imposed limit
on one's choice of speciality is not inherently unfair. Proponents of a fettered accommodation system might
Individuals with certain physical limitations can also argue that we should try to accommodate conscien-
be excluded from certain specialities in the medical tiously objecting healthcare professionals regardless of
profession. That does not preclude them from joining the arguments presented by us so far. They could argue
the medical profession. that, for instance, institutions other than the medical
profession ought to guarantee reliable access in a
Does It Matter When Professionals Joined healthcare system that is proactively being subverted by
a Profession? accommodation-demanding conscientious objectors.
It could be argued that it matters when conscientious Why should not the health ministries in Canada's prov-
objectors joined the profession. The idea here is that inces organise reliable access systems based on doctors
they agreed to join a profession that at that time had who are not conscientiously objecting? We know from
a particular scope of practice that these professionals the mentioned experiences of pregnant women seeking
agreed to work in. In a way, this scope of practice an abortion in the province of PEI that this might not
constitutes a kind of contract between society and be feasible in parts of the country. The odds are fairly
its professionals. If a society wishes to change a pro- high that the more rural an assistance-seeking patient is
fession's scope of practice, it should accommodate located, the more difficult it could become to guarantee
conscientious objectors among those who joined the access. We have yet to see a persuasive reason, given
profession prior to a given controversial change, for the arguments presented so far, why any healthcare
example, by grandmothering them in. Some of the system should burden itself, and ultimately patients,
reasons we mentioned above mitigate against grant- with these sorts of logistical problems when there is
ing objectors a blanket right to such an accommoda- an obvious, more efficient alternative: saying no to the
tion. There is another reason that suggests that this conscience-based accommodation requests of health-
argument is not particularly strong. It is surprising care professionals. A profession that is unable to guar-
that doctors seem to think in large numbers that so- antee reliable access to its services for no other reasons
ciety provided them with a monopoly on the provi- than its desire to accommodate the private ideological
sion of medical services but that society at no time convictions of some or many of its monopoly provider
would add or subtract from the range of services it members is failing in its mission.
considers scope of practice of that profession. This
is puzzling given the nature of medicine as an ever- Conclusion
evolving scientific and cultural pursuit. Most of us
who do not work as medical professionals have en- Medical professionals have no moral claim in liberal
countered employers who wish to change our scope democratic societies to the accommodation of their
of practice. Many academics the world all over are individual conscientious objections. To accommo-
suffering under their employers' desire and pressure date such objections would subvert some of the very
to introduce online learning programmes, despite reasons for why the medical profession was created
our objections on pedagogical and other grounds. in the first place. To accommodate them would also
However, nobody would think that our employers permit such medical professionals to abuse the mon-
are not well within their rights to ask us to contribute opoly privileges that society endowed their profes-
to such programmes. Unlike doctors, we are not even sion with. Medical professionals practising medicine
entrusted with typically well-remunerated service de- in the 21st century would be well advised to accept
livery monopolies. professional service delivery, as defined by the scope
Having said this, we do think that we should of professional practice, as one inevitable corollary of
try at least to accommodate conscientious objectors their voluntary decision to join the profession. For-
who have practised for a long term under a different cing patients to live by the conscientious objectors'
kind of contract than doctors who join the profes- values constitutes an unacceptable infringement on
sion after particular significant changes to the scope the rights of patients.
MU I
References
1. Smith VP. Conscientious objection in medicine: doctors' Amselem: Syndicat Northcrest v. Amselem, [2004] 2 SCR
freedom of conscience. BMJ 2006;332:425. 551, 2004 SCC 47.
2. Braun H. Hermann Stohr (1898-1940). In: Hummel 24. Smith: Employment Division, Department of Human
K-J, Strohm C, eds. Zeugen einer besseren Welt. Christ- Resources of Oregon v. Smith, [1990] 494 US 872.
liche Martyrer des 20. Jahrhunderts. Evangelische Ver- 25. Charo RA. The celestial fire of conscience-refusing to
lagsanstalt. Leipzig: Butzon & Bercker, 2000:87-105. deliver medical care. N Engl J Med 2005;352:2471-3.
3. Michalczykl]. Medicine and ethics in the Third Reich: historical 26. Cantor], Baum K. The limits of conscientious objection:
and contemporary issues. NewYork: Sheed and Ward, 1994:8. may pharmacists refuse to fill prescriptions for emer-
4. Weinstock D. Conscientious refusal and healthcare gency contraception? N Engl J Med 2004;351:2008-12.
professionals: does religion make a difference? Bioethics 27. Lynch HE Conflicts of conscience in health care: an institu-
2014;28:8-15. tional compromise. Cambridge MA: MIT Press, 2008.
5. Leiter B. Why tolerate religion? Princeton: Princeton 28. Blackmer J. Clarification of the CMA's position concern-
University Press, 2013. ing induced abortion. CMAJ 2007;176:1310.
6. Laycock D. Regulatory exemptions of religious behavior 29. Carter: Carter v. Canada (Attorney General), [2015] 1
and the original understanding of the establishment clause. SCR 331, 2015 SCC 5.
Notre Dame Law Review 2006;81:1739-842, at 1839 for 30. Zimonjik P. Doctors with moral objections to assisted
the USA. Strasbourg Consortium for European Court of dying should be able to opt out: committee hears. CBC
Human Rights cases at http://www.strasbourgconsortium. January 27, 2016. http://www.cbc.ca/news/politics/as-
org/. We owe these references to Leiter B 2013, n4. sisted-dying-committee-evidence-hearings-1.3422352
7. Sulmasy DP. What is conscience, and why is respect for (accessed 28 Jan 2016).
it so important? Theo Med Bioeth 2008;29:135-49. 31. Zimonjik P 2016.
8. Sulmasy DP 2008, at 139. 32. As quoted in Wicclair 2000, 218.
9. West-Oram P, Buyx A. Conscientious objection in 33. Lawrence RE, Curlin FA. Physicians' beliefs about
health care provision: a new dimension. Bioethics 2015. conscience in medicine: a national survey. Acad Med
dot 10.1111/bioe.12236. 2009;84:1276-82.
10. Wicclair MR. Conscientious objection in medicine. Bio- 34. College of Physicians and Surgeons of Ontario (CPSO).
ethics 2000;14:205-27. Policy Statement #2-15: Professional Obligations and Human
11. Pellegrino E. The physician's conscience, conscience Rights. March 2015. http://www.cpso.on.ca/CPSO/media/
clauses, and religious belief: a Catholic perspective. documents/Policies/Policy-Items/Human-Rights
Fordham Urban Law J 2002;30:221-44. .pdf?ext=.pdf (accessed 31 Jan 2016).
12. E.g., Pellegrino E 2002. 35. End-of-Life Law and Policy in Canada (EOL). De-
13. E.g., Weinstock D 2014. velopments to watch-Christian Medical and Dental
14. LaFollette E, LaFollette H. Private conscience, public Society of Canada v. College of Physicians and Surgeons
acts. J Med Ethics 2007;33:249-54. of Ontario. 2015. http://eol.law.dal.ca/wp-content/
15. Savulescu J. Conscientious objection in medicine. BMJ uploads/2015/06/2015_03_20_Notice_of Application-3
2006;332:294-7. .pdf (accessed 31 Jan 2016).
16. Schuklenk U. Conscientious objection in medicine: pri- 36. Laycock D. How the Little Sisters of the Poor case puts
vate ideological convictions must not supersede public religious liberty at risk. Washington Post 2015, March 20.
service obligations. Bioethics 2015;29:ii-iii. https://www.washingtonpost.com/opinions/how-the-
17. E.g., Wicclair MR 2000, Sulmasy DP 2008. little-sisters-of-the-poor-put-religious-liberty-at-risk/2016/
18. Hobbes T. Leviathan (chapter 29). London: Andrew 03/20/eaaa6a34-e4b4-11e5-a6f3-21ccdbc5f74e_story.
Crooke, 1651. https://ebooks. adelaide.edu.au/h/hobbes/ html (accessed 21 Mar 2016).
thomas/h681/chapter29.html (accessed 26 Feb 2016). 37. Saguenay: Mouvement laique quebecois v. Saguenay
19. Donahue BS. Infant euthanasia is morally unacceptable. (City), 2015 SCC 16, paragraph 69.
J Thorac Cardiovasc Surg 2015;149:1684-5. 38. Saguenay, paragraph 76.
20. Schuklenk U. The case against assisted euthanasia has not 39. McFarlane: McFarlane v. Relate Avon Ltd [2010] EWCA
been made. J Thorac Cardiovascr Sur 2015;149:1685-6. Civ 880, paragraph 21. Saguenay, paragraph 75.
21. Veatch RM. Hippocratic, religious, and secular medical 40. McFarlane, paragraph 21.
ethics: the points of conflict. Washington DC: Georgetown 41. Lafontaine: Congregation des temoins de Jehovah de St-
University Press, 2012. Jet-time-Lafontaine v. Lafontaine (Village), 2004 SCC 48,
22. Mercier A. Religious belief and self-deception. In: Black- paragraph 67.
ford R, Schuklenk U, eds. 50 Voices of disbelief: why we 42. Edwards Books: R. v. Edwards Books and Art Ltd., [1986]
are atheists. Oxford: Wiley, 2009:41-7, at 47. 2 SCR 713, 1986 CanLII 12 (SCC), paragraph 97.
23. Smith: Employment Division, Department of Human 43. Amselem: Syndicat Northcrest v. Amselem, [2004] 2 SCR
Resources of Oregon v. Smith, [1990] 494 US 872. 551, 2004 SCC 47, paragraph 62.
44. Cameron ST, Ridell J, Brown A, et aL Characteristics 49. Singer P. Rethinking life and death. Text: Melbourne,
of women who present for abortion towards the end 1994.
of the mid-trimester in Scotland: national audit 2013- 50. CBC. Doctor-assisted suicide a "therapeutic service," says
2014. Eur J Contracept Reprod Health Care 2015;1-6. Canadian Medical Association. February 6, 2015. hap://
doi:10.3109/ 13625187.2015.1111326. www. cb c.c a/new s/health/doc t or-a s si s ted- suicide-a -
45. Minerva P. Conscientious objection in Italy. J Med Ethics therapeutic-service-says-canadian-medical-association-
2015;41:170-3. 1. 2947779 (accessed 30 Jan 2016).
46. Purtilo RB. Professional-patient relationship: ethical 51. Fiala C, Gemzell Danielsson K, Heikinheimo 0,
issues. In: Jennings B, ed. Bioethics. 4th edn. Farmington et al. Yes we can! Successful examples of disallow-
Hill, MI: Macmillan, 2014:2547-55, at 2551. ing "conscientious objection" in reproductive health
47. Cowley C . A defense o f conscientious objection in medicare. EurJ Contracept Reprod Health Care 2016;1-6. do
cine: a reply to Schuklenk and Savulescu. Bioethics 2015. i:10.3109/13625187.2016.1138458. Schuklenk U,
doi:10.1111/bioe.12233. Smalling R, 4. Law, ethics and medicine. J Med Ethics
48. Wicclair 2000, at 205. 2016;0: 1-7. doi: 10 .1136/medethics-2016-103560.
The Functions and Limitations than to resolve moral dilemmas, and we have other
(and better) means besides codes of ethics to resolve
of Professional Codes of Ethics moral dilemmas. Resolving moral dilemmas requires
Dale Beyerstein the application of a moral theory which is more general
than the moral theory which informs the code of ethics
Many people become cynical about codes of ethics of a particular profession. So the cynicism implied in
because they suspect that such codes are of no help in the above criticism of codes of ethics is misplaced: we
resolving moral dilemmas. In fact, some cynics take simply should not expect the code of ethics to resolve
delight in pointing out holes in their own professional moral dilemmas.
codes that are large enough to drive a truck through, A second sort of cynicism I wish to address holds
thus making them next to worthless for providing moral that moral dilemmas are irresolvable. The usual argu-
guidance. Other cynics speculate on the reason for this ment for this view rests upon the correct observation
that codes of ethics exist primarily to make professionals that codes of ethics do not resolve moral dilemmas but
look moral, not to promote morality in the profession—in leaps from this premise to the unwarranted conclusion
other words, to allow professionals to protect themselves, that moral dilemmas are not capable of rational resolu-
either individually or collectively. It is not my purpose tion. The reason that this inference is unwarranted is
here to either affirm or deny this claim. As far as I can see, that it ignores the fact that moral theory can be invoked
it is perfectly possible for professions to design codes of in those cases where codes provide no guidance. For-
ethics for this purpose, just as it is possible for individual tunately it is not my job here to actually resolve any
professionals to use the codes of their professions for this moral dilemmas. My aim is much less ambitious: to
purpose. But it is equally possible for whole professions simply direct you to the right place and let you take it
and individuals to use codes for other, more respectable from there. I shall begin by distinguishing professional
ends, and it is these ends which I wish to explore here. codes from moral theories.
The point of this paper, then, is not to examine The Oxford English Dictionary defines "code" in the
how codes of ethics are in fact used but rather to say a sense we are interested in as "a systematic collection or
few things about how they might be used. My hope is digest of the laws of a country, or of those relating to
that this will shed some light on the question whether a particular subject." Two points in this definition are
there is any point in a profession's having a code of important for our present purposes. First is the empha-
ethics and in a professional's ever bothering to consult sis on "systematic": a good code of ethics will present
one. The argument of this paper is twofold: the first its injunctions in a systematic manner. However, note
point is that the intuition behind the cynicism I have the second point: the connection with the laws of a
alluded to is perfectly correct—that is, codes of ethics country. The laws of the land merely assert what is
cannot resolve moral dilemmas. The second is that it to be done or, more usually, what must not be done.
does not matter very much that codes cannot fulfill They do not explain why this is so or provide the rea-
this function. Codes of ethics have other purposes sons behind these pronouncements. Thus, if the legal
Source: FromApplied Ethics: A Reader, ed. E.R Winkler and J.R Coombs (Blackwell, 1993). Reprinted with permission.
II all 111111
system is systematic, a mere inspection of the code is the weight we do on the principles we use to guide our
not sufficient to reveal this. It is only by examining the lives; to explain why we rank them in the order we
code in the light of legal and moral theory that we can do; and to provide reasons for preferring one principle
judge the code to be adequate or inadequate. over another when they conflict. Those of us who act
A code of ethics is essentially a societal code "writ in the world, as professionals do, are primarily inter-
small"—that is, one that governs the conduct of a ested in this function of a moral theory as a method for
subset of the population who engage in a particular set providing justifications for what we think we ought to
of tasks that are not commonly engaged in by the wider do. In moral dilemmas where two morally good actions
public. Since only a small number of people are doing are mutually exclusive, a moral theory resolves the di-
these things, and usually only those people doing them lemma by explaining why one action has a higher value
have the knowledge to tell whether they are done well than the other. How does it do this? Actions are done
or badly, it would be too cumbersome and without for a purpose, an end, and so what we need when we
point to have these rules listed in the whole society's resolve a conflict between actions is a theory of value:
code. Instead, they can be listed in a code made avail- what kinds of things are valuable? Actions? Subjective
able to the profession itself and to those who deal dir- states such as pleasures, contemplations, intentions,
ectly with these professionals. But the main point of a hopes . . . ? How is it that we come to have duties? Is
code is that it is a statement of the rules which govern, it when the opportunity to do some sort of good pre-
or which ought to govern, a profession. sents itself to us or only when we put ourselves under
Such a statement of rules will necessarily be in- an obligation, as when we make a promise? Do the
complete. The point of listing a set of rules is so that duties that we contract when we take on a client give
people can follow them, and if the list is so long that we our clients special claims on us, even when they are
cannot find the rules when we want to refer to them, suffering less, or are generally less morally worthy, than
then the list is of no use to us. This is not to say that a non-client? Who counts from the moral point of view
the rules of the system are or must be incomplete; my anyway? Who has interests that we must take into ac-
point is that the statement of them must be incomplete count when we decide what to do? Do fetuses, gerbils,
if the list is to be of any use. For example, the law set- trees, or the environment as a whole have interests? If
ting the speed limit at 50 kph on city streets is clear and so, how can we weigh them against our own? These
unambiguous: driving at 51 kph counts as a violation. are the kinds of questions that we need to answer in
But if a man is pulled over by a policeman and explains order to solve difficult moral dilemmas. Fortunately we
that he was speeding in order to get his wife to the need not answer them now in order for me to make the
hospital before she delivers the baby, this constitutes central point of this paper that, given the functions
an exception to the law but not one that is recognized that we expect a code of ethics to serve, we find that it
in the law itself as an exception. The reason for this is will in most cases be helpless in answering these kinds
threefold. First, the number of exceptions is so large of questions and therefore helpless in resolving moral
that the statute could not be learned by the public if all dilemmas. To show this I need to outline the functions
the known exceptions were listed. Secondly, the traffic of a code of ethics.
police officer still would have to use his or her judgment A professional code serves four functions: (1) it
to determine whether this case constituted the type of gives guidance to the individual professional; (2) its
situation the exception was framed to capture. Third, principles are statements of the agreed-upon standards
it is arguable that it would be impossible to state in of ethical practice and thus provide guidance for ethics
advance all possible types of situations where an excep- committees struck to rule on professional misconduct
tion ought to be made. So, the procedure we follow is of members; (3) it is a public pronouncement of the
to rely on the individual judgment of the police officer ethical principles agreed upon by the majority of the
involved at the time. In probably 99.9 per cent of cases, profession and thus serves to inform clients about
no exceptions apply, and in only about 0.1 per cent what they have a right to expect from their profes-
of cases is an individual judgment called for. Similarly, sionals; and (4) it informs other professionals with
the professionals' code of ethics will work very well for whom the professional must work about the kind of
the bulk of cases, and it would be foolish to attempt cooperation they have a right to expect from the profes-
to write into the code provisions for moral dilemmas sional and the limits to the cooperation that a profes-
which will be encountered very rarely and which re- sional ought to give. All four of these functions always
quire a great deal of moral deliberation to resolve. involve a consensus of the profession. Their point is
A moral theory, on the other hand, is not itself a set to announce the moral principles that the profession
of principles but is an attempt to explain why we place has already agreed to endorse. Now, this agreement
Beyerstein: The Functions and Limitations of Professional Codes of Ethics
must have been arrived at either by imposition by a in a code tend to be general statements that everyone
group of people whose authority is unquestioned or could interpret in a way he or she finds acceptable. But
by some sort of democratic decision-making process. this generality is precisely what makes these statements
In the former case, there is little to recommend that useless as a guide to solving a moral dilemma
decision. Part of what makes someone a professional is The fourth function of a code is to inform other
that she is thought to have the capacity for independent professionals with whom the professional must work
judgment; therefore, it is contrary to the professional about the kind of cooperation they have a right to
ethos for her to do something simply because she is expect from the professional and the limits of the
told to. This point holds even when the person doing cooperation a professional ought to give. Often, in
the ordering is another member of the same profes- such a case, the individual is pressured by these other
sion.' Therefore, this leaves the second option, that the professionals to do something he or she morally cannot
provision of the code will have been arrived at by the do, and the code serves to buttress the professional's
usual methods of democratically arriving at a consen- resolve to resist this pressure and to provide an author-
sus: negotiation and bargaining. And we know what is ity for the professional's arguments against the other
the usual result of this process: agreement about that professionals: "I cannot do what you request because
which is obvious and prevarication over that which is our code forbids it." It also indicates to the others that
difficult or contentious. there is no point in simply asking some other member
This necessity for compromise severely weakens of the profession to do this job, since the other mem-
the first function of a code, to give guidance to indi- bers are bound by the same code. But, again, if the code
vidual professionals. While the provisions of the code is to serve this purpose, it must express the resolve of
may be adequate to handle the clear and obvious moral the majority of the profession. If it did not, it would be
problems that arise, we shall see in a moment, when disingenuous for a professional to state that provision
we examine the nature of moral dilemmas, why the of the code as evidence that this is the case.
methods of compromise and Trasle-066 are unable to Dilemmas sometimes arise in these cases where
produce a code that is adequate to the task of resolving one professional, who has one set of duties, asks for
moral dilemmas. cooperation from a second professional, who has a
Now let us tum to the second function of the code different, conflicting set of duties. In this sort of case,
we have specified: to state the standards of ethical prac- there will be a conflict between the two profession-
tice that are invoked by an ethics committee to judge a als' codes of ethics. An example of this sort of conflict
professional's conduct. These standards must be prom- between codes would be that of a psychiatrist who is
ulgated in advance, just as a law must, if someone is approached by his client's lawyer asking that the client
to be accused of violating it. We consider it unjust to be kept in hospital two weeks longer, although she was
charge a citizen with violating a law that was promul- ready for discharge, so that the lawyer could have the
gated after the alleged offense was committed, and it trial deferred and would therefore have time to prepare
would be similarly unjust for a professional society's her case for court.
disciplinary body to discipline a member for something In this sort of case, each professional has a code
that is not clearly forbidden by the society's code of of ethics which enjoins him or her to promote certain
ethics in advance of the action which is the subject of ends. So, here, we have a conflict between codes, as
the investigation. If a professional stands in front of an well as a conflict about which end—an honest diag-
ethics committee charged with misconduct, then that nosis or a fair trial—is most important from the moral
person is accused of behavior or judgment which falls point of view: It is understandable in these cases that
very much below the standards of his or her colleagues. each professional, given his or her training and the
Therefore, there must exist an agreed-upon standard values that led that person into the profession, will
which this person has failed to meet. come to value things differently. And it is perfectly
The third function of the code, that is, its provid- natural that, having once decided that one of these is
ing a public pronouncement of the ethical principles the most important, she will struggle to bring it about.
of the profession, again requires the consensus of the But it will not do to say about this sort of case that
majority of the profession. As already mentioned, this one professional ought to follow his code and the other
pronouncement serves to inform clients about what professional hers and both simply ought to try to bring
they have a right to expect from the professionals with about what their respective codes enjoin. In the sort of
whom they deal, and therefore such a pronouncement dilemma we are considering, both professionals have
would be hypocritical if it were not agreed upon by the as their ultimate responsibility to look after the client's
vast majority. It is for this reason that many provisions overall interests, and in so far as both professionals are
II 4111 'ma
trying to achieve this end, they cannot be indifferent It seems to me that there are two approaches to
to what ultimately happens. Thus, both professionals the study of moral dilemmas that we might take. One
must re-examine their codes and look for guidance approach focuses on what it is really like to be in a
from moral theory to determine what ought to happen moral dilemma—on the psychological and phenom-
in this particular case and which code has to be revised enological attributes of the situation: the angst we ex-
in order to be consistent with that result. perience while we are deliberating about what to do
To summarize, then, it appears that three of the and the regret and sense of loss we experience after
four functions of a code of ethics necessarily involve a the decision has been made. This is the approach we
consensus already having been established among the find in European philosophy and among those who
members of the profession. But with respect to the first have been influenced by this tradition, as well as in
function of a moral code, to provide guidance in eth- the continent's literature. The other approach focuses
ical deliberation, an appeal to what the majority would on the method of deliberation itself. According to this
do cannot, by itself, serve as a moral reason for acting. tradition, if it really were the case that we had no reason
The reason for this can be seen if we examine briefly to prefer one option in a moral dilemma to the other,
the nature of moral deliberation. Moral deliberation then there is no role for reason to play. It would not
necessarily involves the moral agent confronting the matter which option we chose, since ex hypothesi there
situation facing him or her, asking the question, "What is nothing to be gained by the choice. So we really have
ought I to do?" This question is not equivalent to the no deliberation problem, even if we do have a human
question, "What is the consensus of my colleagues problem—which the novelist is much better equipped
about what to do?" The question "What ought Ito do?" to deal with than the philosopher.
remains to be asked even after finding out what others My position is the second one, though I cannot
recommend. That is not, of course, to say that seeking justify it in this paper. All I have space for here is to
the advice of your colleagues is worthless. Your col- point out a consequence of this position. It is only
leagues' advice may be exactly right My point is that when we think that there really is a reason to prefer
what makes it right has nothing to do with the numbers one course of action over another, but that we just do
of people who take it to be so. Therefore, the fact that not yet know what it is, that we have a problem. All we
your professional colleagues have agreed to put a rule can do is either try to find empirical facts of the situa-
in your code does not make this the right action. tion which we have not yet seen which might make
Moral dilemmas arise when we are faced with in- a morally relevant difference or refine our theories so
compatible actions, one of which promotes one end, that this difference will be apparent. Paying attention to
or one person's end, and the other of which promotes our code is of no use in either of these two endeavors.
some other end, or some other person's end. Thus, we If I am right, we see something as a moral di-
are faced with a trade-off: at most, we can satisfy one lemma, rather than as a usual garden-variety moral
of these ends. But this conflict between ends is not all problem that we handle every day precisely because we
there is to a moral dilemma. This much is common have not gathered the facts or have not done enough
to almost all moral deliberations. The second feature moral theorizing. Here we have in a nutshell the reason
of a dilemma is that we see both ends as ones that we why codes cannot help us resolve moral dilemmas.
must promote: we recognize a moral loss if we fail to The articulation of a code always occurs after the real
promote either one of these ends, even if the reason moral deliberation has taken place, after we have put
we do not promote it is the impossibility of promot- enough thought into a few of our dilemmas to resolve
ing both ends together. In most moral deliberations we them so that they are now just moral problems. But
encounter, we see the gain from promoting one end as in a profession which is advancing, there will be new
so obviously greater than that to be achieved by any al- situations which arise of sufficient complexity to tax
ternative that the sacrifice of the other end is obviously the individual's moral sense and which have not yet
worth it. But in a moral dilemma, we do not see which been subjected to the moral deliberation of the entire
end to promote: all alternatives seem all to involve ends profession. These will be the moral dilemmas.
which we ought not to sacrifice. I think I have now established the point that we
Now, it is at this point that I shall actually say should not look to codes of ethics for the resolution
something controversial. Up until now, I take it that of moral dilemmas. But given our comparison between
my remarks about moral theories and moral codes have codes of ethics and society's laws, this should come
been merely descriptive and would be accepted by as no surprise. Being a moral person in society as a
philosophers no matter which of the available alterna- whole involves much more than simply obeying soci-
tive moral theories they might defend. ety's laws: it involves paying attention to moral theory.
Beyerstein: The Functions and Limitations of Professional Codes of Ethics
Similarly, being a moral professional who is capable of therefore conclude that moral theory is incapable of
resolving the moral dilemmas that arise in one's profes- resolving moral dilemmas. Such a view is short-sighted.
sional life from time to time involves much more than We should remember that professionals make hundreds
acting in accordance with a profession's code of ethics. of moral judgments every day with confidence, without
It involves, in addition, reasoning about what one is any reason at all to think that these judgments are dif-
doing by paying attention to a wider moral theory. That ficult or controversial. These moral judgments are not
was the "pessimistic" part of this paper. But every paper moral dilemmas precisely because our common-sense
ought to end on an optimistic note. I actually have two. moral theories are adequate to cope with them; we know
First, as I have already mentioned, a code of exactly how to handle the trade-offs between good and
ethics should serve three other functions, and nothing bad that arise with them. In these cases, it does not
I have said entails that it cannot perform these tasks matter whether professional codes of ethics pronounce
adequately. But, more importantly, when we find that on what to do when they arise, for the answer is so ob-
our professional codes are inadequate to resolve moral vious. On the other hand, genuine moral dilemmas arise
dilemmas, we should remember that we have recourse infrequently, but these are the ones we remember—for
to moral theory. The reason that some people despair good reason, of course, because these are the ones that
about the prospect of resolving ethical dilemmas with do require the bulk of our attention. The optimistic
moral theories is that they focus only on the most dif- though undefended conclusion is that moral theory will
ficult moral dilemmas that have not been resolved and help us resolve these difficult cases as well.
Notes
1. If this is a necessary condition for a vocation counting to follow orders during an emergency on the general
as a profession, then it would seem to rule out certain principle(s) that the person giving them is likely to
vocations—nursing and the military, to name two—as know what is best and/or that when a task requires the
professions. In both these vocations, practitioners take cooperation of several people, things will get ruined if one
direction, and even orders, from superiors. However, this person "breaks rank" or wastes precious time questioning.
is the point of the qualification "simply": nurses have So my point is not that it is always inconsistent to follow
come to realize the problem with professionals simply an order. What is inconsistent with professionalism is
following orders and have in their codes addressed this to agree in advance to follow orders from a superior, no
issue. The general consensus among nurses is that a nurse matter what the circumstances. Soldiers come dangerously
ought not to follow an order that violates her professional close to agreeing to this, and for this reason I take the
responsibilities. This does not mean, though, that a nurse term "professional soldier" to function like "professional
must stop to question every order given: it is perfectly athlete," the point of the honorific being to distinguish
consistent with the notion of professional responsibility careerists from amateurs.
4.3 Cases
Case 1 constitute treatment. Therefore, they did not need the

family's consent to withdraw treatment. Nor did they
Hassan Rasouli and Medical Futility
need to refer the matter to the Ontario Consent and
See the description of this case from the introduction Capacity Board, which was established to adjudicate
in Chapter 1 (page 2). Of further note is the physicians' disputes with patients and families about treatment.
justification for wanting to withdraw Mr Rasouli's life The Rasouli case raises acute questions about the
support. Their decision was grounded in a belief that nature and boundaries of professional ethics and profes-
their professional ethics, in particular their right to sional autonomy and how these are to be understood in
professional autonomy, permitted them to override, or relation to patient and family rights to autonomy. The
say "no," to the family's wish that life support be con- physicians' position relied on assumptions about the
tinued. The doctors argued before the courts that usefulness of the notion of futility as a ground for as-
life support for Mr Rasouli was futile and so did not serting professional autonomy (discussed and critiqued
II al .1111i1
in the Truog et al. reading, page 176). Mr Rasouli had with the patient's body, and is closely associated
shown some evidence of being conscious that everyone with the provision of palliative care. Withdrawal
acknowledged. His family was firmly convinced that he of life support is inextricably bound up with care
would find his life to be meaningful even at a low level that serves health-related purposes . . . By remov-
of consciousness or perhaps just because he was alive. If ing medical services that are keeping a patient alive,
so, his life support seems not to have been futile, at least withdrawal of life support impacts patient auton-
according to the values of Mr Rasouli and his family. omy in the most fundamental way. The physicians'
The physicians appeared to place their judgment of attempt to exclude withdrawal of life support from
what would make for a worthwhile life over Mr Rasouli the definition of "treatment" . . . cannot succeed.
and his family's. The Rasouli case may also have been an
instance, anticipated by Truog et al., of using the notion The dissenting minority took the view that the case did
of futility to avoid discussion about another issue— not need to be referred to the Consent and Capacity
namely, fair allocation of scarce health care resources. Board because the relevant legislation did not apply to
In a split 5-2 decision, the Supreme Court of the withdrawal of treatment. The minority added that
Canada ruled to refer the dispute to the Ontario Con- patients cannot insist on treatments that are "medic-
sent and Capacity Board for review.' The majority ally futile, harmful, or contrary to professional medical
decision said nothing directly about the legitimacy of standards of care." It argued that the case should be
decisions not to treat based on futility. However, it did referred to the Ontario Superior Court for a determina-
explicitly state that withdrawal of life support was a tion of such facts. Neither the majority nor the minority
treatment decision under the law in Ontario, stating: took account of discussions challenging the appropri-
ateness of using the concept of futility to refuse treat-
[Withdrawal of life support aims at the health- ment. The case was never referred to the Consent and
related purpose of preventing suffering and indignity Capacity Board or the Ontario Superior Court, and
at the end of life, often entails physical interference Mr Rasouli remains on life support.
Note
1. Cuthbertson v Rasouli, 2013 SCC 53, [2013] 3 S.C.R. 341.
Case 2 the first half of the twentieth century had included

SARS in Toronto: A Duty to Care explicit mention of duties to care for the sick during
epidemic outbreaks, but these had been removed by
In late 2002 and through the first half of 2003, a viral the middle years of that century. During the SARS out-
outbreak of Severe Acute Respiratory Syndrome (SARS) break, a small minority of health care providers refused
threatened to become a global pandemic. In Canada, to go to work in hospitals where SARS patients were
the outbreak focused mainly in Toronto where ap- being treated. Two years after the outbreak, the Joint
proximately 400 people became ill, 44 died, and some Centre for Bioethics (JCB) at the University of Toronto
25,000 people were quarantined before the outbreak produced a report that attempted to provide an ethical
was brought under control.' About 40 per cent of those framework for advance planning for infectious disease
infected were health care workers. Three health care outbreaks. One of its aims was to clarify the rationale
workers died (two nurses and a doctor). The SARS epi- for health care workers' duties to provide care during
demic generally exposed Canada's unpreparedness for an outbreak of a virulently infectious disease. The JCB
responding to a pandemic. In the years immediately stated the following reasons for recognizing a duty of
following, concerted efforts were made by all levels of care in such circumstances:
government and by health care professional bodies to
create such plans. 1. The ability of physicians and health care workers
One of the central issues raised by the SARS epi- to provide care is greater than that of the public,
demic concerned health care providers' professional thus increasing their obligation to provide care.
duty to provide care to infected patients at some risk 2. By freely choosing a profession devoted to care for
to their own health. Physicians' ethics codes from the ill, they assume risks.
Cases 207
3. The profession has a social contract that calls on Association's policy document "Caring in a Crisis: The
members to be available in times of emergency. (In Ethical Obligations of Physicians and Society during
addition, they largely work in publicly supported a Pandemic" (see page 188 in this chapter) is partly a
systems in many countries.)2 response to such requests. It recognizes a qualified
duty to provide care. The Canadian Nurses Associa-
The JCB called upon health care professions tion developed a guidance on this topic at the same
to address the "vacuum" in their ethics codes and time and has incorporated it into its ethics code.3 To
policy documents to provide clear guidance to their what extent these guidance documents live up to the
members about their obligations and the risks they JCB's proposed framework, and to what extent that
would be expected to take. The Canadian Medical framework is justified, remain matters of discussion.
Notes
1. "SARS Outbreak in Canada," Canadian Environmental 3. "Ethical Considerations for Nurses in a Natural or Human-
Health Atlas, Social Sciences and Humanities Made Disaster, Communicable Disease Outbreak or Pandemic"
Research Council, http://www.ehatlas.ca/sars- in Code of Ethics for Registered Nurses, Canadian Nurses
severe-acute-respiratory-syndrome/case-study/ Association (2017), https://www.cna-aiic.ca/html/en/
sars-outbreak-canada. Code-of-Ethics-2017-Edition/files/assets/basic-html/
2. Joint Centre for Bioethics, "Stand on Guard for Thee" page-l.html#.
(2005), http://wwwjcb.utoronto.ca/people/documents/
upshur_stand_guard.pdf.
Case 3 not properly digest her food, and was losing weight.
Is It Ever Right to Violate Rules Surgery was scheduled and cancelled three times. Her
surgeon finally said, "Enough is enough" and told her
of Rationing? to go to emergency and gave her a list of symptoms
to tell the physicians there. These symptoms, he said,
We (two of the editors) had a philosopher friend who would give her a bed, and he would then come and
was quite ill and required surgery but was on a wait- perform the necessary surgery. And that is how our
ing list and deemed not ill enough to be admitted to friend (who taught ethics) got her surgery. Events like
hospital under guidelines that control surgical admis- this happen every day in Canada's health care system.
sions for immediate treatment. The abdominal pains Was the surgeon wrong to violate the hospital's rules
she suffered did, however, significantly interfere with of rationing? Was our friend wrong to do what he told
her life. She had a hard time working or sleeping, could her to do?
Case 4 warning was issued by the College of Physicians

BC Physicians "Cherry-Picking" and Surgeons in its June 2008 bulletin. Dr Morris
VanAndel stated that it is "unacceptable for doctors
Patients for Ease of Care, Refusing to turn away patients just because their care will
Those in Need be complex, time-consuming, and likely to involve
consultations with family members." Dr VanAndel
In April 2008, British Columbia Health Minister clarified his view that it is acceptable to reject new
George Abbott made a public announcement con- patients only when the physician does not offer the
demning a growing trend in which family phys- services the patient requires, such as geriatric ser-
icians were turning away new patients who had or vices for elderly patients or obstetrics services for
who had previously survived cancer. A similar stern pregnant patients.
II iI 111111
Nicole Adams, a spokesperson for the BC Cancer Any trend toward selecting patients on the basis
Agency, confirmed that cancer patients report having a of the ease of providing their care is likely fuelled in
very hard time finding family physicians. Alixe Cormick part by a continuing shortage of family physicians in
provides one example, having called nearly 20 physicians British Columbia and by a fee structure that pays $30
before finding one who would attend to her 64-year- for a usual visit for individuals under 80 years old and
old mother, Beatrice Hanson, during the time she was $45 for those over that age. An estimated 200,000 BC
waiting to see an oncologist at the BC Cancer Agency residents did not have a family physician as of De-
for her pancreatic cancer. Her own family physicianb cember 2015, despite efforts over several years by the
receptionist refused to book an appointment for her provincial government to remedy the situation, includ-
mother, specifically because of the cancer, but did offer ing some measures to pay more for complex care pa-
an appointment to Ms Cormick for a Botox injection, a tients.2 Given the scarcity of family physicians in BC, is
cosmetic service the physician had just begun offering. Dr VanAndel correct that there are no circumstances
Every one of the nearly 20 offices Ms Cormick contacted under which a family physician can ethically refuse to
turned her away when they found out she was calling see a patient requiring particularly time-consuming care
for a cancer patient. "I kept being told that they aren't if the physician offers the services the patient requires?
properly compensated for patients who have to be seen For example, if a family physician's practice is almost
so much."' By the time she found a physician willing, full and already has a substantial number of complex
as a favour to a friend, to see Ms Cormick's mother, Ms care patients, is there a professional duty to accept fur-
Hanson was in Vancouver General Hospital's palliative ther complex care patients? How do we determine what
care unit, where she subsequently died. counts as "cherry-picking" less complicated patients?
Notes
1. Pamela Fayerman, "Health Minister Upset with Recent 2. Ron Seymour, "Doctor Shortage Linked to Government
Trend of Doctors Refusing to Take New Patients with a Incentive Program," Kelowna Daily Courier 9 December
History of Cancer," Vancouver Sun 23 April 2008, Section 2015, http://www. kelownadailycourier. ca/news/article_
A, 1-2. cf dalObe-9f 00-11 e5-87e7-f f ec8edf76c 1 .htm 1.
Case 5 Their refusal to help, they say, is an expression of profes-

Helping Residents to Live at Risk sional autonomy. A response to this position is that, ex-
cepting obvious, imminent, and serious risks, residents
Residents in long-term care facilities often wish to may be in a better position to judge whether risks of
engage in behaviours that put their health at risk. harm are worth it on balance. If so, referring to profes-
These behaviours can include wanting to smoke sional autonomy may be a cover for paternalism after all.
cigarettes, drink alcohol, or eat food that threatens It is becoming increasingly recognized that com-
their health (for example, when there is a choking petent residents in care facilities can choose to live at
risk for residents with swallowing difficulties). Resi- risk. In 2015, the Fraser Health Authority in British
dents are often fully competent to make choices to Columbia adopted a policy entitled "Supporting Resi-
live at risk, and some accommodation is often made dents to Live at Risk in Residential Care and Assisted
for smoking and alcohol. But residents frequently Living," which explicitly recognized residents' "right
have mobility issues that make it difficult for them to to live at risk" under its residential care and assisted
obtain or consume these goods. A question arises over living programs. The policy went further, however,
whether health care providers have any responsibility in stating that residents "will be supported to live
to assist residents in acquiring or consuming these according to their preferences, values and beliefs to
potentially harmful goods or otherwise to facilitate the greatest extent possible—to reflect the level of
risky behaviour. autonomy they would enjoy outside the facility."2
Health care providers are split on this issue. Those These words and the title of the policy apparently
who refuse to help residents are often adamant that impose requirements on health care providers to offer
their motive is not paternalistic.' Rather, they claim assistance to residents when they make choices to
a professional duty of non-maleficence or "to do no live at risk. What this means in practice is unclear,
harm" prevents them from acting on residents' behalf. and the boundaries between patient autonomy and
Cases 209
professional autonomy remain to be worked out. adopted elsewhere and, if so, how they will be re-
Nevertheless, an obligation to help residents live at ceived and applied by health care providers are issues
risk is acknowledged. Whether such policies will be for today and coming years.
Notes
1. Alister Browne, "Helping Residents t o Live at Risk," 2. Fraser Health Authority, "Supporting Residents to Live
Cambridge Health Quarterly of Healthcare Ethics 12 (2003): at Risk in Residential Care and Assisted Living" (2015),
83-90. http://www.fraserheahh.ca/media/201501PolicySupporting
ResidentsLiveAtRisk.pdf.
Case 6 chances of this working were or if it worked whether

the patient would be severely disabled.
When Is Treatment Futile?
The resuscitation procedure required intravenous
heated saline solutions, about 100 units of rare AB
On a cold mid-January morning in 2016, Queen's Uni- positive type blood, and a cardiac bypass to permit
versity physics student Tayyab Jafar attempted to end "extracorporeal rewarming" of Jafar's blood. A spe-
his life by taking an overdose of sedatives and lying cial ventilator was required to prevent lung damage.
down to die in the snow in Kingston, Ontario. Jafar had Jafar was resuscitated, though the cardiac surgeon
been under psychiatric care and had threatened sui- noted, "we may just have saved him into a chronic
cide previously. One of his friends found Jafar's "good vegetative state."
bye" text message and went looking for him. When During Jafar's recovery, his body weight went from
he found Jafar, there were no vital signs, and his core 140 pounds to about 80 pounds. He was in great pain
body temperature had fallen to 20.8 degrees Celsius, during his treatment and rehabilitation and asked to be
16 degrees below normal and about 14 degrees below allowed to die. The hospital staff continued treatment.
where hypothermia occurs. No one knew how long Jafar required extensive rehabilitation, including learn-
Jafar had been there. It had been five hours since the ing to walk again and to use his hands. He recovered and
text message was sent. There was a slight chance of re- returned to study at Queen's in the fall. In response to a
suscitation only if Jafar had become hypothermic after question about whether he was happy to have a second
he had taken the sedatives. No one knew which hap- chance at life, Jafar answered, "I'm happy because I want
pened first. The emergency responders brought Jafar to do many things . . . I'm pretty much happy at this
to Kingston General Hospital where it was decided to point." The surgeon commented, "I've been doing this
attempt resuscitation, though it was unknown what the for 25 years and he's the first survivor I had."'
Note
1. Mary Ormsby, "This Queens Student Froze this-queen%E2%80%99s-student-froze-to-death-on-

to Death on a Kingson Pier. Here Is How He a-kingston-pier-here%E2%80%99s-how-he-
Came Back to Life," Toronto Star 19 February came-back-to-life/ar-AAn6jTr?li=AAadgLE&Tocid=
2017, http://www.msn.com/en-ca/news/canada/ spartandhp.
4.4 Study Questions
1. How is patient autonomy bound u p with patient well-being, and how does this affect
the decisions of health care providers' to refuse requested treatments?
II iull 111E11
2. What difficulties arise when health care providers wish to say "no" to a patient's re-
quest for a treatment that is harmful to the patient, harmful to others, futile, or uses
scarce health care resources?
3. Is a health care provider's "conscientious objection" or "conscientious refusal" to pro-
vide a requested treatment an expression of professional autonomy? Why/why not?
4. Is futility ever a reasonable ground on its own for saying "no" to a patient's request for
treatment?
5. What professional duties, if any, do health care providers assume to provide care at
some risk to themselves—for example, during a pandemic or in dealing with abusive
or violent mentally ill patients?
6. How might the fiduciary duty of health care providers to act in their patients' best
interests lead to a "tragedy of the medical commons"? How might this "tragedy" be
avoided in a way that is consistent with the professional obligations of health care
providers?
7. What is the main value of professional ethics codes? How much useful ethical guid-
ance is contained within them?

Codes of Ethics
Canadian Medical Association. 2004. "CMA Code of Ethics." https://www.cma.ca/
Assets/assets-library/document/en/advocacy/policy-research/CMA_Policy_Code_of_
ethics_of the_Canadian_Medical_Association_Update_2004_PD04-06-e.pdf.
Canadian Nurses Association. 2017. "Code of Ethics for Registered Nurses." 2017. https://
www.cna-aiic.ca/html/en/Code-of-Ethics-2017-Edition/files/assets/basic-html/page-1.
html#.
Professional Ethics
Gampel, E. 2006. "Does Professional Autonomy Protect Medical Futility Judgments?"
Bioethics (20): 92-104.
Hebert, Philip C. 2014. Doing Right: A Practical Guide to Ethics for Medical Trainees and
Physicians. 3rd edn. Oxford University Press.
Pellegrino, E. 2002. "The Physician's Conscience, Conscience Clauses, and Religious Belief:
A Catholic Perspective." Fordham Urban Law Journal (30): 221-44.
Reid, Lynette. 2005. "Diminishing Returns: Risk and the Duty to Care in the SARS Epidemic."
Bioethics (1): 348-61.
Schneiderman, Lawrence J., Nancy S. Jecker, and Albert R. Jonsen. 1996. "Medical Futility:
Response to Critics." Annals of Internal Medicine 125 (8): 669-74.
Wicclair, Daniel. 2000. "Conscientious Objection in Medicine." Bioethics 14 (3): 205-27.
Yeo, Michael, Anne Moorehouse, Pamela Kahn, and Patricia Rodney. 2010. Concepts and
Cases in Nursing Ethic. 3rd edn. Broadview Press.
Reproducidve ahiics
5.1 Introduction
When people think of ethical issues raised by human reproduction, their attention is often
focused on abortion. But human reproduction raises a host of thorny ethical issues both
before and after abortion may be contemplated. It is useful therefore to organize discussion
of reproductive ethics chronologically. To start, there are ethical issues that arise before
reproduction begins. Roughly speaking, these are about controlling the circumstances of
reproduction, including when to reproduce, with whom, who makes those decisions, and
what support, if any, should be provided to assist people to reproduce. Once reproduction
is started, the issues shift to decision-making over whether to stop or continue reproduc-
tion. Stopping reproduction is about abortion and the moral claims, if any, the fetus has
on decisions to abort it. Continuing reproduction is about obligations an expectant mother
and others have to protect and promote the health and well-being of the fetus and future
child. The topics in this chapter are presented in this order. We have tried to cover many
of the important issues around starting and continuing reproduction, though the issues are
more extensive than we have space for (see the suggested readings). The main part of the
chapter is devoted to readings on abortion.
Starting Reproduction
Christine Overall's "Reproductive Freedom, Autonomy, and Reproductive Rights"

(see page 218) addresses the nature and limits of rights to start reproduction. Overall
argues that a moral right to reproduction is grounded in both consequentialist and de-
ontological reasoning. She notes that for many people, having children is a profound
source of value and not having them a source of great sorrow or regret. It is also important
that children be born into families that want them. These are the consequentialist rea-
sons for a right to reproduction. The deontological argument is that reproductive rights
respect women's freedom to make choices to start or refuse to start reproduction. Thus,
they preserve a woman's autonomy and control over her body. Overall argues that the
consequentialist reasons are grounds for providing medical care and assistance to women
who have made decisions to reproduce. Such positive assistance should be available with-
out unjust discrimination—for example, regardless of marital status or sexual orientation,
as is the case for other medical care. The deontological reasons protect women's and
men's choices to control when, where, and with whom to reproduce, including the right
to refuse to reproduce and decisions to donate reproductive gametes to assist others to
reproduce. These are negative deontological claims, or rights not to be interfered with.
M 111
5 Reproductive Ethics
However, Overall is clear that these reproductive rights do not guarantee a baby to
anyone, nor do they by themselves morally justify a decision to reproduce that includes
a broader range of potentially competing moral considerations. A recognition of positive
and negative claims or rights to reproduce and their limits is not controversial in modern
liberal democratic societies, but they deserve to be carefully stated as a backdrop for
discussions of other aspects of starting reproduction.
One issue about starting reproduction not addressed in Overall's article is sex selection
of children, or "preconception gender selection" of offspring (PGS). This practice is frequently
condemned as contributing to sexism toward women where there are cultural preferences for
boys. John A. Robertson argues, however, in his article on PGS (see page 226) that, assum-
ing PGS is safe and effective (as it is today), such concerns about sexism do not apply
when parents make decisions to have a child of a different gender from the child(ren)
they already have. Nor is there anything necessarily problematic about choosing the
sex of a child because the parents prefer to have the experience of raising a boy or a
girl. Sexism is relevant only when PGS is based on a belief in the superiority or greater
value of boys versus girls (or vice versa). Robertson defends a controversial position,
but it is difficult to see an argument against it, given deontological assumptions about
reproductive freedom stated by Overall. Robertson acknowledges, however, the possibil-
ity that such preferences could lead to imbalances in numbers between the sexes that are
socially destabilizing. There is no evidence at this time, however, to hold that preferences
for different gendered child-raising experiences are common enough that they would
lead to imbalances between the sexes in significant numbers in modern industrialized
societies.
Robertson's position is an implicit criticism of the law in Canada, which prohibits
PGS. This law was adopted following the recommendations in 1993 of the report titled
Proceed with Care: The Final Report of the Canadian Royal Commission on New Reproductive
Technologies. This report held that the practice of PGS "violates respect for human life and
dignity, sexual equality, protection of the vulnerable, and appropriate use of resources."'
Whether the royal commission adequately answers Robertson's arguments is an interest-
ing question.
commercial surrogacy A further issue (among many) addressed by the royal commission is commercial
payment to a woman beyond surrogacy. The relevant material is excerpted in the article "Preconception Arrange-
reimbursement of expenses
ments" (see page 233). Again, the royal commission recommended a general prohibition,
for agreeing to carry a
pregnancy for another person
and this was also adopted in law in Canada. Anyone who offers or pays for or assists
or couple who will become commercial surrogacy is criminally liable to fine of up to $500,000 and up to 10 years in
the child's parent(s). prison.2 The royal commission argued that commercial surrogacy should be prohibited
on the grounds that it offended human dignity by commodifying women's reproductive
capacities and by commodifying children. It also held that commercial surrogacy would
have negative psychosocial consequences for the surrogate mother, would cause harms
to the children produced, and would facilitate exploitive power imbalances between sur-
rogates and their contracting partners. However, the royal commission did not defend
a total ban on surrogacy. It recommended permitting women to volunteer as surrogate
mothers as long as they do not accept payment beyond personal expenses related to the
pregnancy. The current law is uncertain as to whether this is permitted, but there are
businesses and lawyers who will assist such "non-commercial surrogacy" arrangements,
including by facilitating offshore international surrogacy contracts, which are not cov-
ered by Canadian law.
Given the difficulty of establishing a boundary between reimbursement for ex-
penses and payment for services, some forms of commercial surrogacy are undoubt-
edly being practised in Canada today. From an ethical perspective, the topic remains
controversial, since the current criminal prohibition of commercial surrogacy imposes
a clear restriction on women's rights to control their bodies and reproductive choices.
Moreover, in the over 20 years since the royal commission's report, it has been difficult
Starting Reproduction
to find evidence confirming negative consequences of the practice from First World
jurisdictions where it occurs and when there are regulations to address issues around
exploitation and the health of the surrogate. A large number of studies from the United
States and the United Kingdom show that surrogate mothers are generally happy with
their experience.3 Arguments about children's rights and welfare have also been chal-
lenged.' As a result, Canadian surrogacy advocates and some researchers have argued
that the law is outdated and should be changed. Commercial surrogacy is permitted in
many states in the United States but prohibited in many European countries, including
the United Kingdom.
Stopping Reproduction—Abortion
Most countries in the world regulate abortion. It is generally prohibited outright in South
America, predominantly Catholic and Muslim countries, and Africa. Most other coun-
tries, including those in Europe, the United States, Australia, and Russia, typically allow
abortions up to a certain time and restrict them after that. In sharp contrast, Canada and
China are the only countries without any legislative or judicial restrictions whatsoever.
In the case of Canada, this was not for want of trying. Canada's abortion law was struck
down as unconstitutional by the Supreme Court of Canada in 1988.5 The House of Com-
mons rewrote and approved the law in 1989, but it did not pass the Senate. The govern-
ment of the day declined to try again, a decision followed by all subsequent governments.
The question we take up in this chapter is whether this is ethically acceptable and if not,
what restrictions should be placed on access to abortions and why. We begin with the
logic of the problem.
The standard view is that the central question concerns the rights status of the fetus. fetus (as used in biology)
Women have an uncontroversial right to control their bodies. Thus, if the fetus has no a pre-natal organism in which
right to life, there is no case for setting that right aside, and abortion must be available all organ systems are in place
and functioning (approx.
on demand. But if the fetus has a right to life either at conception or some later time, the 12-36 weeks old).
woman's right to control her body is limited. How limited depends on when that right
comes into being and how strong it is. We will call this the "status of the fetus" approach,
and it is the view taken by Don Marquis in "Why Abortion Is Immoral" (see page 238),
Mary Anne Warren in "On the Moral and Legal Status of Abortion" (see page 248), and
L.W. Sumner in "A Third Way" (see page 262).
Others, however, think that the status of the fetus is largely or entirely irrelevant to the
debate. In our readings, Judith Jarvis Thomson in "A Defence of Abortion" (see page 253)
argues that even if the fetus has a full right to life, abortion will still be morally permissible
in a wide range of circumstances, and Susan Sherwin in "Abortion through a Feminist
Ethics Lens" (see page 268) presents a feminist position according to which abortion must
be permissible in all circumstances from conception through birth. We will call this the
"priority of the woman" approach.
Fetal Development and Terminology

It will be useful to start by understanding the nature of fetal development and by introdu-
cing some terminology. Human beings come into existence when an egg and sperm unite
to form a zygote with a complete genetic code. Within 24 hours, cellular cleavage starts
to take place and continues as the zygote moves down the Fallopian tube into the uterus.
Upon entering the uterus, the multi-cell zygote—now called a "conceptus"—starts to
burrow into the uterine wall. After two weeks, implantation is complete, and the organism
is called an "embryo." A number of dramatic changes then rapidly take place. A rudiment-
ary heart starts to beat in three to four weeks; at six weeks the organism is recognizably
human; brain activity can be monitored at eight weeks, and spontaneous movement by
II iII MILLI
the organism (motility) occurs at around 10 weeks. At 12 weeks, all organ systems are
in place and functioning, and the organism is called a "fetus." The woman may feel the
fetus moving ("quickening") at 12 to 16 weeks, and it is capable of independent survival
("viable") sometime between 20 and 28 weeks. Birth typically occurs at nine months, and
at that time the organism is called an "infant."
In the philosophic literature on abortion, it is common to refer to an organism at any
fetus (as generally used in stage of fetal development as a "fetus," rather than use that term in the above biological
philosophic discussions sense, and we will follow that practice. "Abortion" means terminating a pregnancy, and
of abortion) a human
this is to be distinguished from killing the fetus, which is called "feticide."
organism at any stage of fetal
development from zygote to
This distinction is important because whereas in many cases pregnancy cannot be
(but not including) infant. terminated without killing the fetus, once the fetus is viable it can be. A woman's right to
control her body will (at most) give rise to a right to abortion. Feticide requires a separate
abortion termination of
argument. We are now ready to consider the morality of abortion.
pregnancy.
Status of the Fetus: Approaches

feticide killing the fetus.
The Conservative Theory

Conservative Theory the The Conservative Theory holds that the fetus has a right to life from the point of con-
fetus has a right to life from ception onwards. There are two versions of the theory. One holds that the fetus acquires
the point of conception a small right to life at the point of conception, and as the organism develops the right
onwards.
becomes stronger, and a woman accordingly requires increasingly weighty reasons to have
an abortion. The problem facing this version is that of giving the view operational signifi-
cance. To have practical application, the theory must tell us when abortions are permis-
sible and when not. But it is not clear what non-arbitrary rationale can be provided for
saying (even roughly) at what point in fetal development a woman may have an abortion
for what reason.
The second version of the Conservative Theory holds that the fetus acquires a full
right to life (i.e., a right indistinguishable in strength from that of a full-fledged human
being) at conception. This entails that that abortions that kill the fetus are justified only
in the circumstances in which it would be justified to kill a full-fledged innocent human
being. Since that is seldom, abortions are seldom if ever justified. This is the view Don
Marquis holds in "Why Abortion Is Immoral" (see page 238). He argues that what makes
it prima facie wrong (i.e., wrong in the absence of special circumstances) to kill a full-
fledged human being is that the killing takes away that beings future. But killing the fetus
likewise takes away its future, and thus it must also be prima facie wrong to kill it. Thus,
the fetus and a full-fledged human being have the same right to life. A standard objection
to the Conservative view is that the same reasoning makes celibacy and contraception
wrong as well, for engaging in those things likewise cancel a potential future. Marquis's
reply is that this objection does not face his version of the Conservative Theory. He argues
that in celibacy and contraception, there is no identifiable single being whose future is
taken away. A possible future is cancelled, to be sure, but no one now exists who has a
right to that future.
If Marquis had argued (as other Conservatives do) that the fetus has a right to life
because it is a potential full-fledged human being, he arguably would be committed to
opposing celibacy and contraception. This is because an egg and sperm are objects that
together constitute a potential full-fledged human being and thus to prevent their union
must be as wrong as permitting it and then killing the resulting organism. He would also
be vulnerable to the objection that potential Xs do not have the rights of actual Xs. For
example, medical students and citizens are potential doctors and prime ministers but do
not have the rights of each. There is no reason, then, to recognize the fetus as having a full
right to life in virtue of its potentiality. These are serious problems for other Conservative
views, but Marquis's account escapes them.
The Liberal Theory

The Liberal Theory holds that the fetus lacks a right to life sufficient to compete with a woman Liberal Theory the fetus
right to control her body at all stages of its development. This view is represented in our read- lacks a right to life at all stages
of fetal development.
ings by Mary Anne Warren in "On the Moral and Legal Status of Abortion" (see page 248).
For Warren, the key question of abortion is "Is the fetus a person?" where a person is char-
acterized by five traits: (1) consciousness, (2) reasoning, (3) self-motivated activity, (4) the
capacity to communicate, and (5) the presence of self-consciousness and self-awareness.
Warren does not claim that these traits are individually necessary or any subset of them
jointly sufficient to be a person but merely that a being that had none of them is not a
person.
Warren further claims that all and only persons have full moral rights and acknow-
ledges that many genetic human beings will thereby lack full moral rights. These include
the permanently unconscious, those without appreciable mental capacity, and fetuses. In
her view, it is plain that the fetus is not a person in the sense described above and that "to
ascribe full moral rights to an entity which is not a person is as absurd as to ascribe moral
obligations and responsibilities to such an entity." Warren anticipates the objection that
while these beings may not have full moral rights by virtue of being persons, they may be
ascribed such rights because they are like persons or are potential persons. In reply, Warren
contends that even if a right to life could be ascribed to the fetus on this basis, "such a right
could not possibly outweigh the right of a woman to obtain an abortion, since the rights of
an actual person invariably outweigh" those of pseudo-persons or potential persons.
A much-discussed alternative Liberal Theory is that developed by Michael Tooley in
"In Defense of Abortion and Infanticide" (see Suggested Further Reading on page 286).
This makes a valuable addition to Warren's account because whereas Warren treats it as
self-evident that only persons have a right to life, Tooley provides a reason for think-
ing this. In Tooley's view, the key element in ascribing a right to life to an organism is
self-consciousness (Warren's fifth characteristic of a person). This is because (he argues)
an entity cannot have a right to life unless it is capable of having an interest in its own
continued existence, and this requires that it has a concept of a continuing self as a subject
of experiences and other mental states. Thus, an entity that has no concept of self cannot
have a right to life, and since it is plain that a fetus does not have such a concept, a fetus
does not have a right to life.
Both Warren's and Tooley's theories have the consequence, common to all Liberal
Theories, that since newborn infants do not have a right to life, infanticide must be mor- infanticide killing an infant.
ally permissible. Both also embrace that view on roughly the same grounds, and Warren
sets these out in the postscript to her article.
The Middle Theory

For those who find the Conservative and Liberal theories unacceptable, the Middle Middle Theory the fetus
Theory beckons. The standard Middle Theory holds that the fetus lacks a right to life up lacks a right to life up to
some point or threshold
to a particular point and after that point has a right to life. Sometimes it is held that the
stage in fetal development
right to life starts out very weak and grows stronger as the fetus matures (just like in the and has a right to life after
gradualist version of the Conservative Theory), and sometimes that a full right to life sud- that time.
denly springs into existence. Middle theories also differ on what that point is. Traditional
candidates include when it is recognizably human, acquires brain activity, makes spontan-
eous movements, and becomes viable. Two problems, however, face selecting the magic
moment. The first is to show that any characteristic selected is significant enough to make
such a momentous moral change. The second is to identify at what point that characteristic
comes into being in a process of development that does not have sharp discontinuities any-
where. These difficulties have made it seem to many that we are forced to choose between
the extreme positions.
L.W. Sumner in "A Third Way" (see page 262), however, argues that a middle ground sentience the ability to
can be found and proposes "sentience" as the criterion of moral standing. Sentience have experiences.
II ill I111.1.1
refers to the ability to have experiences. These include primitive sensations like pleasure
and pain as well as more sophisticated mental states such as wants, aims, tastes, emotions,
and sentiments. The moral relevance of sentience comes (for Sumner) from the fact that
the ability to have any of the above experiences gives an organism interests and morality
has to do with the protection and promotion of interests. Since it is plausible to suppose
the fetus can experience pain at some point in fetal development, Sumner concludes that
the fetus has moral standing at that point. It does not follow that abortion is always wrong
from that point on Exceptions for special considerations such as the health of the mother
or fetus are possible. But on the assumption that abortion will cause the fetus pain, it
does follow that some regulation of the practice is apt, and thus Canada is wrong to allow
abortion on demand.
Sumner acknowledges that the criterion of sentience does not allow us to identify
a precise point at which abortion should be regulated, since sentience gradually phases
into existence. But he argues that it is possible to identify a threshold period before
which we can be morally certain that the fetus cannot experience pain and after which
we can be morally certain it can. On the assumption that the capacity to feel pain tracks
physiological development, this period can be located in the second trimester around the
time of viability. The more precise dating of when abortions are not permissible required
by regulation (Sumner argues) can be arbitrarily identified within this non-arbitrarily
identified period.
Priority of the Woman: Approaches

Judith Jarvis Thomson, in "A Defense of Abortion" (see page 253), argues that ascribing a
right to life to the fetus will not serve those who oppose abortion in the clear simple way
they often think it will. Even if the fetus has a full right to life, she argues, abortion will be
justifiable in a wide range of circumstances. To show this, Thomson invents one of the best-
known thought experiments in philosophy. Suppose that a famous unconscious violinist
has failing kidneys, and to save him the Society of Music Lovers attach him to a sleeping
woman whose kidneys can service both. This will save his life, but the woman will have
to live in this compromised state for nine months. She awakes and wants to detach him.
Can she do so?
Thomson says it is absurd to say "no" and that this immediately shows that it is not
wrong to kill an innocent human being with a full right to life. To be at all plausible, that
claim must be amended to "it is always wrong to kill such a human being unjustly."
The question then becomes when would such a killing be unjust, and Thomson argues
that in the case of the violinist, it would be only if he had a right to the continued use
of the woman's body. Thomson further contends that he would have such a right only
if the woman had explicitly or implicitly granted it to him. Thus (to return to the case
of abortion), abortion in the case of rape is not wrong as the violinist case immediately
shows, nor is it wrong in cases of pregnancy when the woman has taken sufficient pre-
cautions to make it implausible to say she has consented to letting the fetus share her
body. Thomson's view thus does not show that abortion is never wrong but only that its
wrongness cannot be straightaway inferred from an alleged right to life of the fetus. To
conclude that abortion would be wrong in any given case requires showing that in that
case, the woman gave the fetus a right to use her body, and Thomson leaves it open as to
whether that can ever be shown.
In "Abortion through a Feminist Ethics Lens," Susan Sherwin (see page 268) takes
a stronger view. Whereas Thomson holds that there may be occasions on which abortion
is "unjust killing," Sherwin does not. According to Sherwin, the fetus is morally signifi-
cant only through its relational status as part of a pregnancy occurring within a particular
woman's body. As such, the fetus is not a moral person and cannot be ascribed a right to
life. In this Sherwin and the Liberal agree that women must have a right to abortion on
demand. But their views differ on the permissibility of infanticide. The Liberal is commit-
ted to endorsing infanticide because, on the Liberal view, the right to abortion follows from
the claim that the fetus does not have a right to life. Sherwin, by contrast, holds that a
woman right to abortion follows from the right of a woman to control her pregnancy, and
this leaves open the question as to whether the infant has a right to life once the pregnancy
is terminated.
Continuing Reproduction
Women sometimes do things (e.g., smoke, sniff glue) or fail to do things (e.g., follow a diet,
get bed rest) that put their fetus at risk. The possible effects on the fetus and any child that
may eventuate from a fetus carried to term (which we will call a "child-to-be") range from
unlikely, mild, and transitory to highly probable, devastating, and lifelong. What obligation,
if any, does the woman have to avoid these risks? If she does not avoid them voluntarily
should the law be used to force her to? These are the questions we take up in this section with
two selections from the Royal Commission on New Reproductive Technologies. In the first
selection, "Judicial Interventions in Pregnancy and Birth" (see page 277), the commission
gives the official recommendation of the report. The commission argues that women have
an ethical obligation to care for their fetus and child-to-be, but if they do not, judicial
intervention should not be invoked to force them to.
In On Liberty, John Stuart Mill argued that if an act does or threatens harm to the
interests of others, it falls in principle under the scope of the law.6 Mill did not, to be sure,
think that the law should always or even often be used. In his view, it should be used only
if making the conduct illegal would: (1) be effective in curbing the harm; (2) not generate
other and greater harms than it prevents; (3) be the mildest means to curb the evil; and
(4) except as a last resort, be thought justifiable by those subjected to it. For Mill, any
intervention that fails to meet this last condition (to use his purple phrase) "partakes of the
degradation of slavery."' (See the account of Mill's theory of liberty in the Introduction to
this text, page 18.)
The commission argued (in effect; it did not mention Mill) that judicial intervention in
obstetrical cases would fail on every one of Mill's points. Thus it firmly and unequivocally
concluded that judicial intervention is never permissible in either the interest of the fetus
or the child-to-be. This absolutist position on issues that are filled with ethical ambiguity is
striking, and Suzanne Rozell Scorsone, in the second selection from the Royal Commission,
"Dissent on Judicial Intervention in Pregnancy and Birth" (see page 281), disagrees. She
argues that while society should be reluctant to appeal to the law in obstetrical matters, its
use cannot be absolutely ruled out.
Notes
1. Proceed with Care: The Final Report of the Royal Commission on New Reproductive Technologies, vol. 2,
p. 916 (Minister of Government Services Canada, 1993).
2. The Assisted Human Reproduction Act (S.C. 2004 c. 2), section 6. http://laws-lois.justice.gc.ca/eng/
acts/A-13.4/page-1.html.
3. Karen Busby and Delaney Vun, "Revisiting The Handmaid's Tale: Feminist Theory Meets Empirical
Research on Surrogate Mothers," Canadian Journal of Family Law 26 (2010), 13-94; Karen Busby,
"Is It Time to Legalize Commercial Surrogacy in Canada?" Impact Ethics (3 February 2015), https•!/
impactethics.ca/2015/02/03/is-it-time-to-legalize-commercial-surrogacy-in-canada.
4. Jason K.M. Hanna, "Revisiting Child-Based Objections to Commercial Surrogacy," Bioethics 24
(2010), 342-7.
5. R v Morgentaler, 11988] 1 S.C.R. 30
6. J.S. Mill, On Liberty (1858), ch. 1, para. 11.
7. J.S. Mill, Principles of Political Economy (1848), bk. 5, ch. 11, sec. 2.
I I .111 111141
5.2 Starting Reproduction
Reproductive Freedom, Autonomy, the person to do Y. The protective function of repro-

ductive rights must be clarified.
and Reproductive Rights In this chapter, I develop a taxonomy of reproduct-
Christine Overall ive rights—what they are, how they are justified, and
what their limits might be. My focus here is on moral
The ground of the right to become a parent is rights, not legal rights. A moral right is an entitlement
indeed the interests of the potential parents. Be- that we have good reason to accept, an entitlement that
coming a parent is something that lends shape and is an expression of one's humanity and that belongs to
meaning to one's life and often to a life that one an individual by virtue of her or his being a human
shares with another parent; and evidence suggests person. A moral right may or may not be legally recog-
that the interest is one that is very widely shared. nized; that is, a moral entitlement may or may not be
So it is a natural candidate to ground a right. That encoded within the laws of the state in which an indi-
right cannot be absolute, for two reasons: first, vidual lives, but it is no less real and morally significant
the standard feasibility test may in some cases even if it is not recognized by the state (Kates 2004).
be a problem, and it would be unreasonable to For example, the 1994 United Nations International
demand that unlimited medical resources be de- Conference on Population and Development (uNicPD)
voted to solving it; second, there are issues about affirmed "the basic right of all couples and individuals
the suitability of potential parents that worried to decide freely and responsibly the number, spacing
(for example) J.S. Mill, who believed that the re- and timing of their children and to have the informa-
sponsibilities of parenthood were beyond the cap- tion and means to do so, and the right to attain the
acity of some potential parents. So not everyone highest standard of sexual and reproductive health.
can, or should, become a parent. It also includes the right of all to make decisions con-
—Sarah Hannan and Richard Vernon, cerning reproduction free of discrimination, coercion
Parental Rights and violence as expressed in human rights documents"
(quoted in Kates 2004, 57). Even in the most repres-
There are three main reasons for starting this explora- sive regimes where there are few or no reproductive
tion of the ethics of procreation with an examination services or protections, individuals nonetheless still
of rights. First, moral rights are fundamental to much possess moral reproductive rights, despite their not
ethical debate, and reproductive rights have typically being legally recognized.
been emphasized in traditional discussions of procre- I both identify our moral reproductive rights and
ative issues such as abortion and IVF. But as the quota- discuss the kinds of state resources, services, and pro-
tion from Sarah Hannan and Richard Vernon suggests, tections that those rights may indicate. A state that pro-
reproductive rights are more complex than is often vides the resources, services, and protections supported
recognized. It is therefore necessary to delineate the by our moral reproductive rights is in fact according to
scope and limits of reproductive rights; this discussion its citizens legal reproductive rights, even if those legal
provides background for the rest of the book. Second, rights are not formally stated and acknowledged.
some people might assume that deciding to have a Carol A. Kates claims that there is no need to recog-
child is easily justified on the grounds that doing so is nize any special fundamental reproductive rights; we can
simply an expression of reproductive rights.' I examine be content with "a general right to liberty" (2004, 63).
whether this assumption is correct. Third, like many But subsuming reproductive rights under a general
other rights, reproductive rights may also play a pro- right to liberty may not be adequate both because of the
tective role, providing a moral defense against social profound value of reproduction to most individuals and
and legal demands for certain kinds of procreative be- societies as well as because of the far-reaching impact of
havior. If a person has a right to do X, then it is morally procreation, particularly on women. We might therefore
wrong to prevent the person from doing X. If a person say that reproductive rights have both a consequential-
has a right not to do Y, then it is morally wrong to force ist and a deontological foundation.
Source: From Why Have Children? The Ethical Debate, Chapter 2 (The MIT Press, 2012), 19-33. Reprinted with permission.
Overall: Reproductive Freedom, Autonomy, and Reproductive Rights
First, having children is, for many people, deeply because reproductive rights are foundational, they
definitive of their identity and their life's value. For cannot be disregarded or voided.
others, remaining childless is equally essential. Fail- Nonetheless, to say that essential reproductive
ing to have a child when one wants to be a parent can rights must be recognized does not imply that there
be a source of immense sorrow and regret. Becoming is an unlimited right to reproduce. Respect for bodily
a parent against one's wish can be a lifelong burden. freedom and autonomy does not constitute a license to
Hence, the protection of procreative choices by means procreate. For example, as Onora O'Neill points out,
of the recognition of reproductive rights is necessary if, like Jean JacquesRousseau and his mistress, indi-
both to ensure that peoples lives go well and to pre- viduals simply produced infants and then abandoned
vent the misery deprivation, and even oppression that them at a foundling hospital, we would not be inclined
results when people have little or no control over their to say such persons were justifiably exercising their
procreative behavior. In addition, the well-being of right to procreate (1979, 25).
children is dependent on the recognition of reproduct- I propose that the general idea of reproductive
ive rights. Children may suffer if they are unwanted. rights should be analyzed in terms of two distinct
Within families, unwanted children may be handi- prima facie rights—that is, rights that are "conditional
capped by scarce material resources, lack of attention, on not being overridden by other relevant moral
and stressed parenting. Within society at large, as we principles" (Frankena and Granrose 1974, 80). These
know from the past, unwanted children are likely to prima facie rights are the right to reproduce and the
face inadequate provisions for their education, health right not to reproduce. Prima facie rights are strong en-
care, and eventual employment. There are thus strong titlements that make legitimate calls on the behavior of
consequentialist reasons for articulating and protecting others—either individuals or the state (or both)—but
reproductive rights. they are not defeasible. They can sometimes, though
Second, women are particularly vulnerable with re- perhaps only rarely, be superseded by other moral
spect to procreation. Most women are fertile from their requirements—for example, when rights are in conflict
early teens until their late forties. Women conceive, with each other.
gestate, deliver, and breastfeed their babies. Women Let's look first at the right to reproduce. In my early
are still expected to be the primary caregivers for work on reproductive ethics and social policy (Overall
their children. As a feminist, I therefore take women's 1987, 1993), I distinguished between two different
bodily freedom (the absence of physical, legal, or social kinds of right to reproduce: the positive or welfare right
constraints on one's decisions about one's body) and and the negative or liberty right.
autonomy (the capacity to be self-determining, espe-
cially with respect to one's body) to be the sine qua The Right to Reproduce in the
non for women's equality and full citizenship. The Positive or Welfare Sense
deontological basis for reproductive rights is that they
are indispensable to protecting women's personhood. The prima facie right to have children in the positive
Without moral recognition and legal protection of or welfare sense is an entitlement to have a child by
their bodily freedom and autonomy, women are little any means one may choose and to be provided with
more than procreative slaves. It is essential to respect all possible assistance in reproduction. Although there
women's bodily freedom and autonomy because it is arguably is such a right, and it provides important
simply wrong to subject women to forced reproduc- protections, it is nonetheless limited in some import-
tion; it is wrong to use women as a means to others' ant ways.
reproductive goals. Such treatment violates their per- Reproductive services should be seen for what they
sonhood. (A similar claim might also be made about are: a category of health care. As a category of health
men with regard to the use of men's gametes, but the care, reproductive services ought to be available to and
fact that gestation is a female condition makes respect accessible by potential patients on medical grounds.
for bodily freedom and autonomy particularly signifi- Social discrimination in the provision of reproductive
cant for women.) services is just as unfair and unjustified as social dis-
Thus, there are both consequentialist and deonto- crimination in the provision of other forms of health
logical justifications for reproductive rights. Those care, and anyone who opposes social discrimination
rights, properly delineated and understood, provide the for the latter should oppose it for the former. Hence,
foundation for the ethics of procreation. They are a ne- an individual patienth gender identity, race, sexual
cessary (though not sufficient) criterion for evaluating orientation, marital status, and other identity charac-
procreative decisions. And, as I argue in later chapters, teristics are not justified criteria either for providing or
220 5 Reproductive Ethics
for refusing medical treatment because they are not rel- Fourth, there is no empirical evidence that social
evant to health care for the individual's medical condi- identity in itself either compromises or enhances par-
tion. For example, if a married heterosexual woman is enting abilities. It is unjustified to generalize about all
entitled to reproductive treatment X, then so is a single persons with a particular identity. Contrary to social
woman or a lesbian woman. The single woman and the conservatives' protests, single women have been suc-
lesbian woman should not be denied access to donor cessfully rearing children in greater and greater num-
insemination or IVF just because of their marital status bers for the past half-century. The evidence shows that
or sexuality. lesbians' children turn out fine (Gartrell and Bos 2010).
In general, the social identity of potential patients And, in contrast, some married, heterosexual parents
is irrelevant to whether they should be permitted physically and psychologically abuse their children.
reproductive services. For a number of reasons, it is One's age, impairments, and health history are
morally wrong for the medical profession to use social similarly in principle relevant to one's eligibility for
identity criteria to determine who is entitled to become reproductive services only on medical grounds—
a parent, who is competent to become a parent, or, that is, if these characteristics demonstrably make
for that matter, who lacks the characteristics for such services either ineffective or dangerous to the
parenthood. individual seeking them. It is not up to physicians
First, physicians are not trained or qualified to to make judgments about whether a patient's age or
determine who is and is not competent to become a impairment renders her ineligible for motherhood.
parent. A physician's role is to provide medical care, (In chapter 7, however, I discuss the significance of
not to serve as a gatekeeper to parenthood. There is no one's age for the moral justification of one's procre-
reason to believe that physicians have a special ability ative choices, and in chapter 8 I say more about par-
to discern parenting capacities. Moreover, attempting enting decisions by persons with impairments.) Thus,
to be such a gatekeeper would compromise physicians' if a person is simply not healthy enough to undergo
moral responsibilities to serve their patients' best in- a particular medical treatment, if there is little or no
terests because it would make reference to standards prospect that the medical treatment will help the in-
that are independent of and even incompatible with the dividual, or if it in fact will even harm the individual,
patient's health care. then it is legitimate not to provide the treatment. For
Second, individuals who do not seek reproductive example, if the body of an older woman or a woman
services are not subjected to any test or qualification with an impairment does not respond to hormonal
for parenting. If there is no screening system for pro- hyperstimulation, then it is medically unjustified to
spective parents who do not need reproductive servi- persist in attempting to extract ova from her. The
ces, it seems unjust to subject to screening those who cessation of treatment would be based on empirical
have the misfortune of needing medical help. Although evidence of likely ineffectiveness or harm, not on the
some philosophers have argued that there should be a mere fact that the patient is a certain age or has a
licensing system for prospective parents (see, for ex- particular impairment.
ample, Tittle 2004), such a system would be a severe Hence, the right to reproduce in the positive sense
imposition on people's bodily freedom and autonomy, can be construed as protecting individuals against un-
would be grotesquely unwieldy, would consume enor- justified discrimination in access to reproductive servi-
mous resources that can better be used to support good ces, which ought not to be denied to individuals on any
health and good parenting, and would not likely im- grounds but medical ones.
prove the care of children. However, because the positive right to reproduce
Third, there is no clear empirical method for pre- draws upon others' material resources and services,
dicting which individuals will be good parents and it cannot be boundless. To what sorts of reproduct-
which will not. Certain factors no doubt make parent- ive services might a positive right to reproduce allow
ing easier (perhaps training about children's develop- access? My answer here should be considered some-
ment and needs) or harder (serious socioeconomic what tentative and suggestive only, for much depends
deprivation). If so, then instead of requiring physicians on two factors: a society's general level of affluence
to make ad hoc predictions in individual cases, it would and the resulting availability of money and resources
be better for a society that is concerned about the qual- for health care, and the kinds of decisions the society
ity of parenting to provide good education, health care, may make about health-care goals and priorities. For
and socioeconomic support for all its citizens, includ- example, societies with scarce resources are entitled to
ing children, parents, and prospective parents. decide whether they will provide IVF services. In some
societies, other services may legitimately be considered fact that some women have become pregnant and given
more immediately valuable, including preventive health birth, thanks to IVF, at the age of sixty or older, there
services, pregnancy care, infant and maternal care, are good medical reasons, related to the woman's health
medical treatments for various impairments, and so on. risks, her ability to sustain a pregnancy, and perhaps her
I do not attempt here to determine how such priorities offsprings prospective physical health, not to permit
might be evaluated. The issue is complex and difficult; IVF for every postmenopausal woman who may want it.
debating the relative value of different health-care goals I am not saying that no "older" woman should ever have
and priorities would require a book of its own. My dis- access to IVF; I am simply reiterating my earlier point
cussion in the rest of this section is intended simply that health-care professionals are entitled to use med-
to indicate the kinds of considerations that would be ical criteria to determine the patient's suitability for the
relevant to determining the health-care implications of procedure. The situation must be decided on a case-by-
the positive right to reproduce, given adequate resour- case basis according to the woman's medical condition.
ces and a careful assessment of priorities. Third, there is no automatic entitlement to have
I suggest that the right to reproduce requires min- as many individual IVF treatments as the patient may
imally the provision of health-care services for healthy wish for. In some cases, a large number of treatments
pregnancy and delivery, including well-supported may turn out to be medically futile. If so, then there is a
home birth and midwifery when these options are significant question of health-care resource allocation.
chosen. Depending on the society's assets and resour- If a society does determine that it will offer IVF, as I have
ces, the positive right to reproduce may also entitle pa- argued, it must not discriminate on irrelevant grounds
tients to reproductive services that alleviate infertility, in providing access to such services, but it is entitled to
including the surgical repair of damaged fallopian tubes place limits on the number of treatments an individual
in women, the reversal of a vasectomy in men, and, woman may have. Doing so is in part a matter of medical
for women, intrauterine sperm injection or trials of IVF effectiveness but in part also a matter of straightforward
with a willing, competent, informed, and autonomous fairness with respect to a limited and expensive service.
partner.2 All of these procedures increase the likelihood A woman may have no more than x treatments simply
that an individual may become a parent, although of because other women also need IVF treatments, and if
course they do not guarantee anyone a baby, especially one woman has many treatments, then others will have
in the case of IVF, for which success rates are still much fewer. Thus, issues of fairness among individuals with
less than 50 percent .3 similar needs dictate that no woman can just demand
A society may nonetheless be entitled to place ap- a large number of IVF treatments as part of her right to
propriate limits on how some reproductive services are reproduce and expect automatic compliance.
provided. IVF is a good example. Having access to IVF So far, then, I have suggested that there is a posi-
does not automatically entitle a woman to determine tive right to reproduce and that in a society with good
how the procedure will be carried out, for that issue health-care resources, exercising that right might include
is a medical matter. Being provided with IVF does not access to a variety of treatments intended to support
give a woman patient the entitlement to have as many pregnancy and birth and to enhance, repair, or restore
embryos as she wishes inserted into her uterus. Al- fertility. Access to these services should not be denied
though inserting a large number of embryos increases based on social identity characteristics such as marital
the woman's chances of becoming pregnant, it also status or sexual orientation, nor should it be denied on
exacerbates the probability of higher-order multiples the basis of age, health status, or impairment unless the
(HOM)—triplets, quadruplets, quintuplets, sextuplets, latter are medically relevant to the provision of services.
and more. Such pregnancies are dangerous for the But the right to reproduce in the positive or welfare
woman and for the fetuses; Homs require vastly more sense has one additional important and indefeasible
health-care services—during gestation and birth, limit: it does not and cannot include an entitlement to
in the neonatal period, and often beyond—than do consumerlike activities such as buying and selling gam-
singletons. (Think of the case of American "Octomom" etes or purchasing the services of a contract procreator,
Nadya Suleman, who after IVF ended up gestating because no one has a right to the use of others' sperm or
and delivering eight fetuses.) Hence, a hospital is en- ova or a right to have another woman gestate and give
titled to set a policy limiting the number of embryos birth to a baby for them. There can be no entitlement
implanted—perhaps to two or three. to such activities because to satisfy such supposed en-
Second, access to IVF does not include an automatic titlements would require that one or more other people
entitlement to use the procedure at any age. Despite the must provide bodily products or services. There cannot
11 .111 1111.11
be a moral requirement for anyone to provide sperm right to reproduce in that manner.5 Saying this does
or eggs or to serve as a gestator, such a requirement not mean that all of these activities are necessarily
would violate other women's or men's bodily freedom wrong, only that they cannot be justified as the exer-
and autonomy.4 Respect for women's and men's bodily cise of a moral right, for no one has an obligation to
freedom and autonomy obviates the possibility of provide gametes or gestational services.
compelling some women to provide ova or to undergo An additional limit to the prima facie right to re-
pregnancy or of compelling men to provide sperm, produce in the positive sense is that individuals do not
even to facilitate procreation by other women. have a right to the best possible baby or even to the
Some individuals may (for whatever reasons, in- kind of baby they most want. Although women cer-
cluding perhaps the dubious desire for genetic "immor- tainly are entitled to good prenatal and neonatal care,
tality") wish to donate their gametes (sperm or ova). there can be no guarantees about the characteristics of
Such donations can be very helpful to people who the resulting infant. Ben Bova claims, "An individuaN
are struggling with infertility. But the donator cannot desire to produce offspring that are as close to the
expect a guarantee that the gametes will actually be ideal that the parent can envision seems well within
used or even that a health-care facility should have to the rights of any citizen" (1998, 194). He makes this
take the gametes if there are medical reasons not to. claim by contrast to the idea that a government would
If a particular jurisdiction offers sperm or egg be entitled to run a eugenics program, an idea that he
banks, donors to which have given willing, competent, rejects. But individuals don't have an entitlement to
informed, and autonomous consent, then women who engage in their own private eugenics program either.
are candidates for using donated sperm or eggs ought Although there is a strong case for helping people with
not to be discriminated against on irrelevant grounds if infertility to increase their chances of procreating, the
they seek access to them. But no individual has an obli- case for helping people to obtain a "designer baby" is
gation to donate or sell their gametes; hence, no other much weaker.
individual has a right to be given them or buy them. People are, of course, free to choose among con-
If a particular society does not offer gamete donation, senting adults with whom to reproduce—or they
or if it does offer the service and no one chooses to should be. Such decisions in themselves have an effect
supply gametes, then, nonetheless, the rights of those on the kind of child they will have. But whether they
who may want or need others' gametes have not been are entitled to the complex genetic services—including
violated. IVF and embryo screening—that would enable them
There is never a guarantee that an individual will to make more precise choices is a further issue and a
be able to have a baby, and it is a mistake to say that a complicated one. Much depends on the costs of such
woman (or a man) has a "right to a baby" or a "right to services compared to other health-care costs. Societies
a family" This claim is true for several reasons. There have to make policy decisions as to which kinds of
is no guarantee of obtaining a baby biologically even health care should be provided. Consider, for example,
through the use of one's own gametes and pregnancy, the possible funding of hearing aids for the hearing im-
for some women cannot conceive, some pregnan- paired, prostheses for those missing all or part of a limb,
cies end prematurely, and some births fail to yield a and wheelchairs for those whose mobility is reduced. It
live infant. But there are also no guarantees morally is not at all obvious that these items are less important
and politically, for people are not entitled to keep a than funding the capacity to choose the best baby pos-
child, even a child that came out of their body, whom sible. Even in a nation where health care is not publicly
they abuse. There is no guarantee through adoption funded, there will still be questions as to what sorts of
because no one has a right to someone else's baby, al- medical research, education, and training should be
though at times some women are generous enough to supported and encouraged; as I pointed out earlier,
give up their infants for adoption. There is no guaran- choices about health-care priorities have to be made.
tee through the use of someone else's gametes because The potential justification of seeking to produce
no one has a right to someone else's gametes and no the best possible baby also depends on the reason that
one owes an in-need person their gametes. There is screening is needed. It would be far more controversial
no guarantee through contract pregnancy because no to screen embryos for intelligence, beauty, or athletic
one has a right that a woman will provide gestational ability than it would ever be to screen them for the
services and no woman owes another person the child presence of diseases, such as Tay-Sachs, that will cause
of her pregnancy. Insofar as reproduction for some unrelieved suffering. In chapter 7, I discuss whether
people requires the use of and access to other individ- parents have any obligation to enhance their offspring
uals' bodies and bodily products, there is no positive by choosing the best available embryo, and in chapter 8
I return to the topic of prenatal screening when I ex- The Right Not to Reproduce
plore whether there is an obligation in some circum-
stances not to procreate because of the fetus's condition. Women and men also have a clear right not to reproduce.
The right not to reproduce means that human beings are
The Right to Reproduce in the Negative entitled not to be compelled to reproduce against their
or Liberty Sense will. No one should be compelled to give away or sell
their gametes or embryos against their will. No woman
Although there are several actual and potential limits should be compelled to undergo pregnancy or continue
on the right to reproduce in the positive sense, there it against her will. This right is implicit in what I said ear-
are no comparable limits on the prima facie right to lier, in discussing limits on the right to reproduce, about
reproduce in the negative or liberty sense. This right is the absence of entitlement to other people's reproductive
the entitlement not to be interfered with in procreation, resources and services. Indeed, I suggest that the right
and it is primarily this reproductive right that is asserted not to reproduce is morally more basic than the right
in the consensus statement from the UNICPD quoted to reproduce, at least in the sense that it is strongly un-
earlier. The right to reproduce in this sense means ethical to violate one individualb right not to reproduce
the freedom to decide when, where, and with whom in order to serve another individualb right to reproduce.
one will have one's biological children and how many. As I noted in chapter 1, pregnancy sometimes re-
Of course, no man is entitled to have his female partner sults because contraception is simply not used, whether
produce a baby for him, any more than a woman has an because it is not available or not chosen or because the
entitlement to be given her male partnerb sperm. But if individuals are unaware of it or the need for it. In other
two people agree to reproduce together using their own cases, heterosexual intercourse is voluntary, but although
gametes, they are entitled not to be interfered with or children are not wanted, there is no effective contracep-
prevented by third parties. tion or contraceptive methods are inadequate or are in-
What, then, about the right to reproduce in the correctly used. And conception is sometimes unchosen
negative sense with respect to women in a same-sex because it is the result of rape. Therefore, respecting the
relationship? They have the same reproductive rights right not to reproduce requires comprehensive educa-
as persons in heterosexual relationships; the state has tion about sex and procreation, access to safe and effect-
no right to prevent them from having children if they ive contraception and to timely and effective abortion
choose. But their situation is, of course, reproductively services, and protection from compulsory heterosexual
more complex. They need the collaboration of one intercourse, whether by assault or by forced marriage.
more person. As I explained earlier, no one has an Of course, just as there is no guarantee that one will re-
entitlement to another personb reproductive gametes. ceive a baby after pregnancy and birth, so also there is
But if a third person agrees willingly, competently, and no guarantee that contraception will work. That is a fact
autonomously to collaborate by providing his sperm, that women must always take into account. It is also a
then women in a same-sex relationship are entitled not fact that men must accept, as I show in chapter 3.
to be interfered with in their decisions about when, The justification for recognizing the negative right to
where, and with whom to procreate or about how reproduce is simply that if it is not respected, then women
many children to have. especially, but men too, have no choice of or control over
Honoring the right to reproduce in the negative their biological reproductive behavior, and therefore their
sense would also require the provision of at least a bodily freedom and autonomy are violated. All women,
minimally healthy physical environment and working like men, have a right not to reproduce—that is, a right
conditions that do not damage or compromise one's not to be forced to reproduce, whether through sexual
procreative capacities. People are entitled to be pro- slavery, denial of access to contraception, or lack of access
tected from environments and employments that are to abortion.' Because some women have historically been
dangerous to their health, including their reproductive forced against their will into marriage or prostitution,
health. It would also mean no interference in or lim- women's right not to reproduce is important whatever
itation of one's biological procreative behavior. Thus, their sexual orientation may be. Part of what is morally
respecting the negative right to reproduce requires wrong about forced marriage and prostitution is that it
that there can be no forced sterilization, no coercive makes women vulnerable to unwanted procreation; it
contraception,6 no forced marriage or prostitution, no violates their right not to reproduce.
racist marriage or domestic partnership laws, no forced Any violation of the right not to reproduce is ser-
abortions, and no forced caesarians (all of which are, of ious. No individual should be forced to provide her or
course, wrong on other grounds, too). his gametes to another person, whether for procreation
II .111 11141/1
or for research. No woman should ever be forced to itself morally justify the individual's decision to repro-
undergo pregnancy against her will. In the words of duce. At the same time, the right to reproduce in the
D.S. Hutchinson, in unwanted pregnancy "the body is positive sense is limited by a variety of factors and does
being altered for a purpose that does not belong to the not guarantee a baby to anyone.
person. But this is a case where the body is not just In other words, a general appeal to reproductive
used, but transformed. The transformation is different rights cannot be used to vindicate a wholesale entitle-
in kind from the bodily changes involved in ordin- ment to procreate, to be given a baby, or to parent.
ary actions, for it is a transformation of a very special Among the reasons for this claim, the most obvious is
nature. It results in two people where there was previ- that in making the decision to procreate, a new human
ously only one." In unwanted pregnancy, the woman's being will be brought into existence. As S.L. Floyd and
body "is treated as a means to a project that is alien to D. Pomerantz put it in regard to a putatively unlimited
[her;] it involves an alienation of the body from the right to self-determination, "It does not take into account
person." It thereby undermines her integrity because the impossibility of a being's actual or hypothetical con-
"a commitment to one's body is given in the human sent to becoming a child" (2004, 232). Of course, by the
condition" (1982, 71). If the right not to reproduce is very nature of the situation, the prospective child cannot
not respected, then the result is reproductive slavery: consent. And the impossibility of the child's consent
the compulsory and unwilled use of people's bodies for does not make all procreation wrong. At the very least,
procreative purposes—whether they are other individ- however, the prospective well-being of the child-to-be is
uals' or the states purposes. (Reproductive coercion— surely relevant to the ethics of choosing to have children.
the production of babies through the forced, often That issue is one to which I return several times.
violent use of women's bodies—is a standard feature A second reason that the right to reproduce,
of enslaved societies and is part of what makes enslave- whether positive or negative, is not by itself sufficient
ment morally execrable.) People need to be able to to justify procreation has to do, broadly speaking, with
protect their reproductive capacities and not have them resource issues, including the constraints generated by
exploited or coerced to serve purposes that are not the growing human population and the planet's limited
their own. As I argued at the beginning of this chapter, carrying capacity. I explore this latter set of issues in the
genuine reproductive freedom is one of the essential penultimate chapter.
building blocks of a just and flourishing society. A third reason that the right to reproduce is not
by itself sufficient to justify procreation is provided by
Conclusion the need for a procreative partner. It still takes two to
make a baby, if not by sexual intercourse, then by other
What does this discussion o f reproductive rights tell us means of combining egg and sperm. Several times in
about the "why have children?" ethical decision? this chapter, I have added the caveat that one must
The right not to reproduce and the right to repro- have a willing, informed, competent, and autonomous
duce in the negative sense (the entitlement to freedom reproductive partner. No one has a right to another per-
from interference) are grounded in general human in- son's sexual services or procreative capacities, and no
terests; people need protection from compulsory pro- one has a right of access to another person's gametes.
creation as much as (or perhaps even more than) they No one is entitled to obtain them by force or by subter-
need protection from denied procreation. The appeal fuge. Consent to providing reproductive capacities and
to the right not to reproduce also provides important gametes is essential.
protections to women (as well as to men) insofar as The content and structure of reproductive rights
it provides a strong prima facie argument against any leave open the possibility that there might sometimes be
obligation to procreate. If a woman has a right not to an obligation not to reproduce. Acting on one's rights,
reproduce, then she is prima facie not obligated to including the right to reproduce, is not always mor-
reproduce. In chapters 4 and 5, I investigate whether ally justified. One may have the right to do something
there are any factors that can create such an obligation that is nonetheless not morally justified. Reproductive
despite the prima facie right not to reproduce. rights are a necessary but not sufficient condition for
Moreover, the appeal to the right to reproduce in justifying choosing to have children.
the negative sense is inadequate to serve as a complete Therefore, the next topic to be considered is the
ethical justification for choosing to have children be- situation where potential parents disagree—in par-
cause, as subsequent chapters show, many other factors ticular, where they disagree as to whether to continue
are morally relevant to procreative decisions. The fact a pregnancy or not. In such cases, one or the other
that interfering in an individual's procreative behavior partner does not consent to procreate. How might such
violates the individual's right to reproduce does not by disputes be resolved?
Notes
1. For a history of the development of the concept of but no individual, no matter how needy, has a right to
reproductive rights, see Kates 2004. the use of the blood or organs from another person if the
2. How such services should be funded is a separate issue latter person does not want to donate them. If there is
that I cannot explore here. Nonetheless, it seems plausible a system of blood and organ donation in which donors
to me that if these services are part of the health-care participate willingly, knowledgeably, and consensually,
system and especially if prospective patients are not then a needy patient is of course entitled to be served
wealthy, then the state should pay for them. (It is also by that system if he or she meets the criteria of medical
often argued that public financing of IVF, in particular, eligibility. But the system cannot compel donors to
will reduce the incidence of gestation of Homs (triplets, participate, and there is not a right to the use of other
quadruplets, and so on], and hence of all the medical persons' body parts.
problems (including blindness, cerebral palsy, and death] 6. Note, however, that I am not saying that women
and resulting costs borne by the public that are associated (or men, for that matter) are not entitled to set, as
with such infants (Barwin 2009] because if women do not a condition for their participation in heterosexual
have to pay for IVF themselves, then they will be less likely sexual activity, the requirement that their partner use
to expect that, to increase the odds of becoming pregnant, contraception. We are entitled to do so in order to
large numbers of fertilized eggs be implanted at each protect (insofar as the effectiveness of contraception
attempt but will be content with the implantation of only permits) our procreative future. In such cases, the use of
one or two at a time.) contraception is not coerced: because each participating
3. In Canada, for example, the overall live birth rate with partner is entitled to say "no" to the sexual activity if he
IVF is 27 percent per cycle (Bouzayen and Eggertson or she does not want to use contraception, there is no
2009, 243). compulsion.
4. Another moral reason for rejecting commercial contract 7. In this book, I do not discuss the general arguments for
pregnancy is that it is the sale of babies (Overall 1993). and against the justification of abortion. First, there is
However, I set this argument aside. a huge body of literature about the abortion issue, so
5. Philosopher Elisabeth Gedge has suggested that my attempting to summarize it here would take me too far
position may compromise an analogous claim on behalf from my original topic. Second, I think the question
of a positive right to general health care (personal whether to abort an embryo or fetus that has already come
communication, June 19, 2008) and such an implication into existence is different from the question primarily
is morally problematic. However, there are always limits at issue in this book, which is whether to conceive that
that must be placed on the availability of services, even embryo or fetus in the first place. However, in chapter 7 I
if those services are essential to life itself. For example, a make a few comments about the possibility that there may
system of blood and organ donation is highly desirable, be an obligation to have an abortion.
References
Barwin, Norman. 2009. A Private Issue in Need of a Public Hannan, Sarah, and Richard Vernon. 2008. Parental Rights:
Solution. The Globe and Mail May 21, IC2. A Role-Based Approach. Theory and Research in Education
Bouzayan, Renda, and Laura Eggerston. 2009. In Vitro 6 (2):173-89.
Fertilization: A Private Matter Becomes Public. Canadian Hutcheson, D.S. 1982. Utilitarianism and Children. Canadian
Medical Association Journal 181 (5):243. Journal of Philosophy 12 (1):61-73.
Floyd, S.L, and D. Pomerantz. 2004. Is There a Natural Right Kates, Carol A. 2004. Reproductive Liberty and Overpopula-
to Have Children? in Should Parents Be Licensed? Debating tion. Environmental Values 13:51-79.
the Issues, ed. Peg Tittle, 230-32. Amherst, NY: Prome- O'Neill, Onora. 1979. Begetting, Bearing, and Rearing. In
theus Books. Having Children: Philosophical and Legal Reflections on
Frankena, William K., and John T. Granrose, eds. 1974. Parenthood, ed. Onora O'Neill and William Ruddick,
Introductory Readings in Ethics. Englewood Cliffs, NJ: 25-38. New York: Oxford University Press.
Prentice Hall. Overall, Christine. 1987. Ethics and Human Reproduction:
Gartrell, Nanette, and Henry Bos. 2010. US National Longi- A Feminist Analysis. Boston: Allen and Unwin.
tudinal Lesbian Family Study: Psychological Adjustment Overall, Christine. 1993. Human Reproduction: Principles,
of 17 year old Adolescents. Pediatrics 126 (10):28-36. Practices, and Policies. Toronto: Oxford University Press.
u 1 11 II al III1Jj
Preconception Gender Selection' intrauterine insemination (Fugger et al. 1998). A lower

success rate (72%) was reported for male selection.4
John A. Robertson At this early stage of development much more re-
search is needed to establish the high degree of safety
Advances in genetics and reproductive technology and efficacy of flow cytometry methods of PGS that
present prospective parents with an increasing number would justify widespread use. With only one published
of choices about the genetic makeup of their children. study of outcomes to date, it is too soon to say whether
Those choices now involve the use of carrier and pre- the 92% success rate in determining female gender will
natal screening techniques to avoid the birth of chil- hold for other patients, much less that male selection
dren with serious genetic disease, but techniques to will reach that level of efficacy. Animal safety data have
choose nonmedical characteristics will eventually be shown no adverse effect of the dye or laser used in the
available. One nonmedical characteristic that may soon technique on offspring, but that is no substitute for
be within reach is the selection of offspring gender by more extensive human studies (Vidal et al. 1999). In
pre-conception gender selection (PGS). addition, if flow cytometry instruments are to be used
Gender selection through prenatal diagnosis and for sperm separation purposes, they may be classified
abortion has existed since the 1970s. More recently, as medical devices that require US Food and Drug
preimplantation sexing of embryos for transfer has Administration (FDA) approval. Finally, the holder of
been developed (Tann and Handyside 1993; The Ethics the process patent—the USDA—will have to agree to
Committee of the American Society of Reproduct- license the process for human uses.
ive Medicine 1999). Yet prenatal or preimplantation If further research establishes that flow cytometry
methods of gender selection are unattractive because is a safe and effective technique for both male and
they require abortion or a costly, intrusive cycle of in female PGS, and regulatory and licensing barriers are
vitro fertilization OW and embryo discard. Attempts overcome, then a couple wishing to choose the gender
to separate X- and Y-bearing sperm for preconception of their child would need only provide a sperm sample
gender selection by sperm swim-up or swim-through and undergo one or more cycles of intrauterine in-
techniques have not shown consistent X- and Y-sperm semination with separated sperm.5 A clinic or physician
cell separation or success in producing offspring of the that offers assisted reproductive technologies (ART) and
desired gender. invests in the flow cytometry equipment could run the
The use of flow cytometry to separate X- and separation and prepare the X- or Y-enriched sperm for
Y-bearing sperm may turn out to be a much more insemination, or it could have the sperm processed by
reliable method of enriching sperm populations for a clinic or firm that has made that investment. Flow
insemination. Laser beams passed across a flowing cytometry separation would not be as cheap and easy
array of specially dyed sperm can separate most of as determining gender by taking a pill before inter-
the 2.8% heavier X- from Y-bearing sperm, thus pro- course, but it would be within reach of most couples
ducing an X-enriched sperm sample for insemina- who have gender preferences in offspring.6
tion.2 Flow cytometry has been used successfully in
over 400 sex selections in rabbit, swine, ovine, and Demand for Preconception Gender Selection
bovine species, including successive generations in
swine and rabbit (Fugger et al. 1998). A human preg- Unknown at present is the number of people who have
nancy was reported in 1995 (Levinson, Keyvanfar, offspring gender preferences robust enough to incur
and Wu 1995). the costs and inconvenience of PGS. Although polls
The United States Department of Agriculture have often shown a preference for firstborn males, they
(USDA), which holds a patent on the flow cytometry have not shown that a large number of couples would
separation process, has licensed the Genetics and IVF be willing to forego coital conception in order to select
Institute in Fairfax, Virginia, to study the safety and the gender of their children. If PGS proves to be safe
efficacy of the technique for medical and "family balan- and effective, however, it may be sought by two groups
cing" reasons in an institutional review board—approved of persons with gender preferences.
clinical trial? In 1998 researchers at the Institute re- One group would seek PGS in order to have a child
ported a 92.9% success rate for selection of females of a gender different from that of a previous child or
in 27 patients, with most fertilizations occurring after children. A preference for gender variety in offspring
Source: From The American Journal of Bioethics 1 (2001), 1, 2-9. Reprinted with permission.
Robertson: Preconception Gender Selection
would be strongest in families that have already had autonomy. Yet doing so leads to the risk that children
several children of one gender. They may want an addi- will be treated as vehicles of parental satisfaction rather
tional child only if they can be sure that it will be of than as ends in themselves and could accelerate the
the gender opposite to their existing children.' Couples trend toward negative and even positive selection of
who wish to have only two children might use PGS for offspring characteristics. The dilemma of reconciling
the second child to ensure that they have one child of procreative liberty with the welfare of offspring and
each gender. If social preferences for two-child families families will only intensify as genetic technology is fur-
remain strong, some families may use PGS to choose the ther integrated with assisted reproduction and couples
gender of the second child. seek greater control over the genes of offspring.
A second group of PGS users would be those per-
sons who have strong preferences for the gender of the Arguments for Preconception
first child. The most likely candidates here are persons Gender Selection
with strong religious or cultural beliefs about the role
or importance of children with a particular gender. The strongest argument for preconception gender se-
Some Asian cultures have belief systems that strongly lection is that it serves the needs of couples who have
prefer that the firstborn child be a male. In some cases strong preferences about the gender of their offspring
the preference reflects religious beliefs or traditions and would not reproduce unless they could realize
that require a firstborn son to perform funeral rituals to those preferences. Because of the importance of repro-
assure his parents' entrance into heaven (for a discus- duction in an individual's life, the freedom to make
sion of son preferences in India and China, see Macklin reproductive decisions has long been recognized as
1999, 148-51). In others it simply reflects a deeply a fundamental moral and legal right that should not
embedded social preference for males over females. be denied to a person unless exercise of that right
The first-child preference will be all the stronger if a would cause significant harm to others (Robertson
one-child-per-family policy is in effect, as occurred for 1994, 22-42). A corollary of this right, which is now
a while in China (Greenlagh and Li 1995, 627). While reflected in carrier and prenatal screening practices to
the demand for PGS for firstborn children is likely to be prevent the birth of children with genetic disease, is
strongest in those countries, there has been a sizable that prospective parents have the right to obtain pre-
migration of those groups to the United States, Canada, conception or prenatal information about the genetic
and Europe.8 Until they are more fully assimilated, im- characteristics of offspring so that they may decide in a
migrant groups in Western countries may retain the particular case whether or not to reproduce (Robertson
same gender preferences that they would have held in 1996, 424-35).9
their homelands. Although offspring gender is not a genetic disease,
Other persons with strong gender preferences a couple's willingness to reproduce might well depend
for firstborn children would be those who prize the on the gender of expected offspring. Some couples
different rearing or relational experiences they think with one or more children of a particular gender might
they would have with children of a particular gender. refuse to reproduce if they cannot use PGS to provide
They may place special value on having their firstborn gender variety in their offspring or to have additional
be male or female because of personal experiences or children of the same gender (E.E Fugger, personal
beliefs. Numerous scenarios are likely here, from the communication to author). In other cases they might
father who very much wants a son because of a desire have such strong rearing preferences for their firstborn
to provide his child with what he lacked growing up child that they might choose not to reproduce at all
to the woman who wants a girl because of the special if they cannot choose that child's gender. Few persons
closeness that she thinks she will have with a daughter contemplating reproduction may fall into either group,
(Belkin 1999). but for persons who strongly hold those preferences,
the ability to choose gender may determine whether
The Ethical Dilemma of Preconception they reproduce.
Gender Selection In cases where the gender of offspring is essen-
tial to a couple's decision to reproduce, the freedom
The prospect o f preconception gender selection appears to choose offspring gender would arguably be part of
to pose the conflict—long present in other bioethical their procreative liberty (Robertson 1996, 434). Since
issues—between individual desires and the larger respect for a right is not dependent on the number of
common good. Acceding to individual desires about the persons asserting the right, they should be free to use
makeup of children seems to be required by individual a technique essential to their reproductive decision
II 41/1 111611
unless the technique would cause the serious harm that required providers of PGS to select for males and
to others that overcomes the strong presumption that females in equal numbers would prevent such imbal-
exists against government interference in reproductive ances.1° A serious threat of a sex-ratio imbalance would
choice. Until there is a substantial basis for thinking surely constitute the compelling harm necessary to
that a particular use of PGS would cause such harms, justify limits on reproductive choice.
couples should be free to use the technique in con- It may also be difficult to show that children born
stituting their families. The right they claim is a right after PGS were harmed by use of the technique. Par-
against government restriction or prohibition of PGS. It ents who use PGS may indeed have specific gender
is not a claim that society or insurers are obligated to role expectations of their children, but so will parents
fund PGS or that particular physicians must provide it. who have a child of a preferred gender through coitus.
Children born with the desired gender after PGS will
Arguments against Preconception presumably be wanted and loved by the parents who
Gender Selection sought this technique. Parents who choose PGS should
be informed of the risk that the technique will not suc-
There are several arguments against preconception ceed and counseled about what steps they will take if a
gender selection. Although such methods do not harm child of the undesired gender is born.' If they commit
embryos and fetuses or intrude on a woman's body themselves in advance to the well-being of the child,
as prenatal gender selection does, they do raise other whatever its gender, the risk to children should be
important issues. One concern is the potential of such slight. However, it is possible that some couples will
techniques to increase or reinforce sexism, either by abort if the fetus is of the undesired gender. PGS might
allowing more males to be produced as first or later thus inadvertently increase the number of gender-
children or by paying greater attention to gender itself. selection abortions.
A second concern is the welfare of children born as a Finally, technological assistance in reproduction is
result of PGS whose parents may expect them to act in now so prevalent and entrenched that a ban on PGS
certain gender-specific ways when the technique suc- would probably have little effect on the use of genetic
ceeds but who may be disappointed if the technique and reproductive technologies in other situations. With
fails. A third concern is societal. Widely practiced, PGS some form of prenatal screening of fetuses occurring in
could lead to sex-ratio imbalances, as have occurred over 80% of United States pregnancies, genetic selec-
in some parts of India and China due to female in- tion by negative exclusion is already well installed in
fanticide, gender-driven abortions, and a one-child- contemporary reproductive practice. Although there
per-family policy (Sen 1990). Finally, the spread of are valid concerns about whether positive forms of se-
PGS would be another incremental step in the growing lection, including nonmedical genetic alteration of off-
technologization of reproduction and genetic control spring genes, should also occur, drawing the line at all
of offspring. While each step alone may appear to be uses of PGS will not stop the larger social and techno-
justified, together they could constitute a threat to the logical forces that lead parents to use genetic know-
values of care and concern that have traditionally in- ledge to have healthy wanted offspring. If a particular
formed norms of parenting and the rearing of children. technique can be justified on its own terms, it should
not be barred because of speculation of a slippery slope
Evaluation of Ethical and Social Issues toward genetic engineering of offspring traits (for an
analysis of the slippery-slope problem with genetic se-
Concerns about sex-ratio imbalances, welfare of off- lection, see Robertson 1994, 162-5.
spring, and technologizing reproduction may be less
central to debates over PGS than whether such practices Is Gender Selection Inherently Sexist?
would be sexist or contribute to sexism. If the number
of persons choosing PGS is small or the technique is A central ethical concern with PGS is the effect of such
used solely for offspring gender variety, sex-ratio im- practices on women, who in most societies have been
balances should not be a problem. If use patterns did subject to disadvantage and discrimination because
produce drastic changes in sex ratios, self-correcting of their gender. Some ethicists have argued that any
or regulatory mechanisms might come into play. For attention to gender, male or female, is per se sexist and
example, an overabundance of males would mean should be discouraged, regardless of whether one can
fewer females to marry, which would make being male show actual harmful consequences for women (see
less desirable and provide incentives to increase the Grubb and Walsh 1994; and Wertz and Fletcher 1989).
number of female births. Alternatively, laws or policies Others have argued that there are real differences
between male and female children that affect parental contribute to sexism or further disadvantage women.
rearing experiences and thus legitimate nonsexist rea- As Christine Overall, a British feminist bioethicist, has
sons for some couples to prefer to rear a girl rather than put it, "sexual similarity or sexual complementarity are
a boy or vice versa, either as a single child or after they morally acceptable reasons for wanting a child of a cer-
have had a child of the opposite gender. tain sex" (1987, 27; quoted in Mahowald 2000, 117).
To assess whether PGS is sexist, we must first be Psychological research seems to support this
clear about what we mean by sexism. The Compact OED position. It has long been established that there are
(1991, 1727) defines sexism as "the assumption that differences between boys and girls in a variety of do-
one sex is superior to the other and the resultant dis- mains, such as (but not limited to) aggression, activity,
crimination practised against members of the supposed toy preference, psychopathology, and spatial ability
inferior sex, especially by men against women." By this (Maccoby and Jacklin 1974; Gilligan 1980; Kimura
definition, sexism is wrong because it denies the es- and Hampson 1994; Feingold 1994; Collaer and Hines
sential moral, legal, and political equality between men 1995; and Halpern 1997). Whether these differences
and women. Under this definition, if a practice is not are primarily in-born or learned, they are facts that
motivated by judgments or evaluations that one gender might rationally lead people to prefer rearing a child
is superior to the other or does not lead to discrimina- of one gender rather than another, particularly if one
tion against one gender, it is not sexist. has already had one or more children of a particular
Professor Mary Mahowald, an American bioethi- gender. Indeed, Supreme Court Justice Ruth Bader
cist writing from an egalitarian feminist perspective, Ginsburg, a noted activist for women's rights before
makes the same point with a consequentialist twist: her appointment to the Supreme Court, in her opinion
striking down a male-only admissions policy at the Vir-
Selection of either males or females is justifiable ginia Military Institute (United States v. Virginia, 116 S.
on medical grounds and morally defensible in other Ct . 2264 [1996]), noted that:
situations [emphasis added] so long as the inten-
tion and the consequences of the practice are not Physical differences between men and women . . .
sexist. Sexist intentions are those based on the are enduring " [T] he two sexes are not fungible; a
notion that one sex is inferior to the other, sexist community made up exclusively of one [sex] is dif-
consequences are those that disadvantage or ad- ferent from a community composed of both." . . .
vantage one sex vis-à-vis the other (2000, 121). "Inherent differences" between men and women,
we have come to appreciate, remain cause for
In my view, the OED definition, modified by celebration.
Mahowald's attention to consequences, is a persuasive
account of the concept of sexism. If that account is Some persons will strongly disagree with this ac-
correct, then not all attention to the biologic, social, count of sexism and argue that any attention to gender
cultural, or psychological differences between the sexes difference is inherently sexist because perceptions of
would necessarily be sexist or disadvantage females. gender difference are themselves rooted in sexist stereo-
That is, one could recognize that males and females types. They would argue that any offspring gender
have different experiences and identities because of preference is necessarily sexist because it values gender
their gender and have a preference for rearing a child difference and thus reinforces sexism by accepting the
of one gender over another, without disadvantaging the gendered stereotypes that have systematically harmed
dispreferred gender or denying it the equal rights, op- women (Grubb and Walsh 1994; and Wertz and Fletcher
portunities, or value as a person that constitutes sexism. 1989, 2112). According to them, a couple with three boys
If this conjecture is correct, it would follow that who use PGS to have a girl are likely to be acting on the
some uses of PGS would clearly be sexist, while others basis of deeply engrained stereotypes that harm women.
would clearly not be. It would be sexist to use PGS to Similarly, a couple's wish to have only a girl might con-
produce males because of a parental belief that males tribute to unjustified gender discrimination against both
are superior to females. It would be nonsexist to use PGS men and women, even if the couple especially valued fe-
to produce a girl because of a parental recognition that males and would insist that their daughter receive every
the experience of having and rearing a girl will be differ- benefit and opportunity accorded males.
ent than having a boy. In the latter case, PGS would not Resolution of this controversy depends ultimately
rest on a notion of the greater superiority of one gender on one's view of what constitutes sexism and what ac-
over another, nor, if it occurred in countries that legally tions are likely to harm women. Although any recog-
recognize the equal rights of women, would it likely nition of gender difference must be treated cautiously,
U I H 3u 11111I
I submit that recognizing and preferring one type of choice, public policies that bar all nonmedical uses of
childrearing experience over the other can occur with- PGS or that restrict it to choosing gender variety in off-
out disadvantaging women generally or denying them spring alone could be found unconstitutional or illegal.
equal rights and respect. On this view, sexism arises not If there are physical, social, and cultural differences be-
from the recognition or acceptance of difference but tween girls and boys that affect the rearing or relational
from unjustified reactions to it. Given the biological experiences of parents, individuals and couples would
and psychological differences between male and female have the right to implement those preferences as part
children, parents with a child of one gender might of their fundamental procreative liberty. The risk that
without being sexist prefer that their next child be of exercising rights of procreative liberty would hurt off-
the opposite gender. Similarly, some parents might also spring or women—or contribute to sexism generally—
prefer that their firstborn or only child be of a particu- is too speculative and uncertain to justify infringement
lar gender because they desire a specific rearing and of those rights.
companionship experience. The claim of a right to choose offspring gender is
If it is correct that using PGS for offspring diversity clearest in the case of PGS for gender variety. If flow
is sexist, then those who deny that biological gender cytometry or other methods of PGS are found to be safe
differences exist, or who assume that any recognition and effective, there would be no compelling reason to
of them always reinforces sexism or disadvantages ban or restrict their nonmedical use by persons seek-
women, will not have carried the burden of showing ing gender variety in the children they rear. Couples
that a couple's use of PGS for offspring gender variety with one child or several children of a particular gender
or other nonintentionally sexist uses is so harmful to might, without being sexist or disadvantaging a par-
women that it justifies restricting procreative choice. ticular gender, prefer to have an additional child of the
Until a clearer ethical argument emerges, or there is opposite gender. ART clinics should be free to proceed
stronger empirical evidence that most choices to select with PGS for offspring variety in cases where couples
the gender of offspring would be harmful, policies to are aware of the risk of failure and have undergone
prohibit or condemn as unethical all uses of nonmedic- counseling that indicates that they will accept and love
ally indicated PGS would not be justified. children of the dispref erred gender if PGS fails. Clinics
The matter is further complicated by the need to providing PGS should also ask couples to participate in
respect a woman's autonomy in determining whether a research to track and assess the effects of PGS on chil-
practice is sexist. If a woman is freely choosing to engage dren and families.
in gender selection, even gender-selection abortion, she The use of PGS to determine the gender of firstborn
is exercising procreative autonomy. One might argue in children is a more complicated question. The choice
response that the woman choosing PGS or abortion for to have one's first or only child be female has the least
gender selection is not freely choosing if her actions are risk of being sexist, because it is privileging or giving
influenced by strong cultural mores that prefer males over first place to females, who have traditionally been
females. Others, however, would argue that the straighter disfavored.'4 The use of PGS to select firstborn males
path to equal rights is to respect female reproductive au- is more problematic because of the greater risk that
tonomy whenever it is exercised, even if particular exer- this choice reflects sexist notions that males are more
cises of autonomy are strongly influenced by the sexist highly valued. It is also more likely to entrench male
norms of her community (Mahowald 2000, 188). dominance. The danger of sexism is probably highest
in those ethnic communities that place a high premium
Public Policy and Preconception Gender on male offspring, but it could exist independently of
Selection those settings.
Yet restricting PGS to offspring gender variety and
Because of the newness of PGS and uncertainties about firstborn females may be difficult to justify. Given that
its effects, the best societal approach would, of course, individuals could prefer to have a boy rather than a
be to proceed slowly, first requiring extensive studies of girl because of the relational and rearing experiences
safety and efficacy and then at first only permitting PGS he will provide, just as they might prefer a girl for those
for increasing the gender variety of offspring in par- reasons, it might be difficult to show that all prefer-
ticular families.13 Only after the demographic and other ences for firstborn males are sexist. Nor could one
effects of PGS for gender variety have been found ac- easily distinguish firstborn male preferences when the
ceptable should PGS be available for firstborn children. couple demanding them is of a particular ethnic origin.
However, given the close connection between par- Although the risk that firstborn male preferences would
ental gender preferences for offspring and reproductive be sexist is greatest if the PGS occurred in a country
in which those beliefs prevailed, the chance that PGS nonmedical preconception gender selection available
would contribute to societal sexism lessens greatly if for selecting female, if not also for male, offspring.
the child is reared in a country that legally protects the The nonmedical use of PGS raises important ethical,
equal status of women and men. legal, and social issues, including the charge that
If prohibitions on some or all nonmedical uses of any or most uses of PGS would be sexist and should
PGS could not be justified and might even be unconsti- therefore be banned or discouraged. Assessment of
tutional, regulation would have to take different forms. this charge, however, shows that the use of PGS to
One form would be to deny public or private insurance achieve offspring gender variety and (in some cases)
funding of PGS procedures, which would mean that even firstborn gender preference may not be inher-
only those willing to pay out-of-pocket would utilize ently sexist or disadvantaging of women. Although it
them. Another form would be for the physicians who would be desirable to have extensive experience using
control access to PGS techniques to take steps to assure PGS to increase the variety of offspring gender before
that it is used wisely. If they comply with laws banning extending it to firstborn gender preferences, it may not
discrimination, physician organizations or ART clinics be legally possible to restrict the technique in this way.
could set guidelines concerning access to PGS. They However, practitioners offering PGS should restrict
might, for example, limit its use to offspring gender their PGS practice to offspring gender variety until fur-
variety or firstborn female preferences only. As a condi- ther debate and analysis of the issues has occurred. In
tion of providing services, they might also require that any event, physicians offering PGS should screen and
any couple or individual seeking PGS receive counsel- counsel prospective users to assure that persons using
ing about the risks of failure and commit to rear a child PGS are committed to the well-being of their children,
even if its gender is other than that sought through whatever their gender.
PGS.15 Although such guidelines would not have the A policy solution that gives practitioners and pa-
force of statutory law, they could affect the eligibility tients primary control without direct legal or social
of ART clinics to list their ART success rates in national oversight, although not ideal, may be the best way
registries and could help define the standard of care in to deal with new reprogenetic techniques. Society
malpractice cases. should not prohibit or substantially burden repro-
ductive decisions without stronger evidence of harm
Conclusion than PGS now appears to present. Ultimately, the use
of PGS and other reprogenetic procedures will depend
The successful development o f flow cytometry on whether they satisfy ethical norms of care and
separation of X- and Y-bearing sperm would make concern for children while meeting the needs of pro-
safe, effective, and relatively inexpensive means of spective parents.
References
Belkin, L. 1999. Getting the girl. New York Times Magazine, fertilization, or intracytoplasmic sperm injection. Human
25 July, 26. Reproduction 13:2367.
Chen, D.W. 1999. Asian middle class alters a rural enclave. Gilligan, C. 1980. In a different voice. Cambridge, MA:
New York Times, 28 December, Al. Harvard University Press.
Collaer, M.L., and M. Hines. 1995. Human behavioral Glover, J. 1989. Ethics of new reproductive technologies: The
sex differences: A role for gonadal hormones during Glover report to the European Commission. DeKalb, IL:
early development? Psychological Bulletin 118(1): Northern Illinois University Press.
55-107. . 1994. Comments on some ethical issues in sex selec-
Compact OED, The new ed. 1991. New York: Oxford tion 6. Paper presented at International Symposium on
University Press. Ethics in Medicine and Reproductive Biology, July. [Paper
Ethics Committee of the American Society of Reproductive on file with author.]
Medicine, The. 1999. Sex selection and preimplantation Greenhalgh, S., and J. Li 1995. Engendering reproductive
genetic diagnosis. Fertility and Sterility 72(4): 595. policy and practice in peasant China: For a feminist dem-
Feingold, A. 1994. Gender differences in personality: A meta- ography of reproduction. Signs 20:601.
analysis. Psychological Bulletin 116(3): 429-56. Grubb, A., and P Walsh. 1994. Gender-vending II. Dispatches 1
Fugger, E.F., S.H. Black, K. Keyvanfar, and J.D. Schulman. (summer).
1998. Births of normal daughters after microsort sperm Halpern, D.F. 1997. Sex differences in intelligence Implications
separation and intrauterine insemination, in-vitro for education. American Psychologist 52(10): 1091-102.
II A 11111.1
Jones, O.D. 1992. Sex selection: Regulating technology Overall, C 1987. Ethics and human reproduction. Boston:
enabling predetermination of a child's gender. Harvard Allen and Unwin.
Journal of Law and Technology 6:51. Robertson, J.A. 1994. Children of choice: Freedom and the
Kimura, D., and E. Hampson. 1994. Cognitive pattern in men new reproductive technologies. Princeton, NJ: Princeton
and women is influenced by fluctuations in sex hormones. University Press.
Current Directions in Psychological Science 3(2): 57-61. . 1996. Genetic selection of offspring characteristics.
Levinson, G., K. Keyvanfar, and J.C. Wu. 1995. DNA based Boston University Law Review 76:421.
X-enriched sperm separation as an adjunct to preim- Sen, A. 1990. More than 100 million women are missing.
plantation genetic testing for the prevention of Xlinked New York Review of Books 20 December, 61-6.
disease. Human Reproduction 10:979-82. Tarin, J.J., and A.H. Handyside. 1993. Embryo biopsy strategies
Maccoby, E.E., and C.N. Jacklin. 1974. The psychology of sex for preimplantation diagnosis. Fertility and Sterility 59:943.
differences. Palo Alto: Stanford University Press. Vidal, E, J. Blanco, E.F. Fugger, et al. 1999. Preliminary study
Macklin, R. 1999. Against relativism. New York: Oxford Uni- of the incidence of disomy in sperm fractions after micro-
versity Press. sort flow cytometry. Human Reproduction 14:2987.
Mahowald, M.B. 2000. Genes, women, equality. New York: Wertz, D.C., and J.C. Fletcher. 1989. Fatal knowledge: Prenatal
Oxford University Press. diagnosis and sex selection. Hastings Center Report 19:21.
Notes
1. "Preconception gender selection" (PGs) rather than and sought boys and girls in roughly equal numbers
"preconception sex selection" (Pss) is used throughout (Edward Fugger, personal communication to author,
this article to convey the importance of the social and 23 February 2000).
psychological meanings with which biologic sex is 8. See Chen (1999). The article describes immigration of
invested for prospective parents and society generally. a middle class family from the Indian state of Gujarat
Because earlier versions of this article used "sex" rather to Bridgewater, New Jersey, a suburb 40 miles from
than "gender," commentators may not have had the New York City
opportunity to revise their comments in response to 9. It should be emphasized that the right claimed here is a
the change. For the discussion at hand, either "sex" or negative right against government interference, not an
"gender," "Pss" or "PGS" is acceptable. obligation of a particular provider or public or private
2. When combined with the X-chromosomes of oocytes, insurers to provide those services.
X-bearing sperm can produce only XX or female 10. See Glover (1994). Professor Glover has apparently
offspring. Similarly, Y-bearing sperm combined with the changed his position from the more negative one he took
X chromosome of oocytes can produce only XY or male in his earlier Ethics of New Reproductive Technologies: The
offspring. Glover Report to the European Commission (1989, 141-4).
3. See the study's web page, http://www.microsortnet. This See also Jones (1992).
article uses the term "gender variety" rather than "family 11. The risk arises because flow cytometry separation can only
balancing" to avoid the misconception that a family is provide a greatly enriched sample of X- or Y-bearing sperm
"unbalanced" if it has many or only children of one gender. for insemination. It cannot guarantee that every sperm in
(I am grateful to George Annas for this suggestion.) the sample is either X or Y.
4. See the study's web page, http://www.microsortnet. 12. Wertz and Fletcher overlook how one could have gender
Because Y-bearing sperm are smaller and contain less preferences based on perceptions of experiential and
DNA, there is more chance that the sorting machine will rearing differences, rather than on differences in the worth
fail to distinguish X's and Y's and thus provide samples or rights of women, when they assert that any form of
that are insufficiently enriched with Y-sperm to give a high gender selection violates the principle of equality between
chance of having a male child. the genders "because it is premised upon a belief in sexual
5. Presumably, flow cytometry separation of sperm could inequality"
occur with donor as well as couple sperm. It could also be 13. FDA approval of the safety and efficacy of flow cytometry
requested by couples undergoing IVF or intracytoplasmic methods of PGS would also be required before widespread
sperm injection (ics0 who request that the sperm use.
provided be enriched or chosen to effect the gender of 14. Persons taking a more purist approach to sexism would, of
choice. course, differ with this assessment.
6. The current cost of $1500 per insemination cycle should 15. They might also require that consumers agree to
decrease as further progress in the field occurs. participate in research so that policymakers will have
7. Persons requesting PGS for gender balancing in the Fairfax reliable information about the uses of PGS and the
study had an average of 3.4 children of the same gender problems it presents.
Royal Commission on New Reproductive Technologies: Preconception Arrangements
Preconception Arrangements uphold the value of children in and of themselves.

Children are not a commodity, nor are they instru-
Royal Commission on New Reproductive Technologies ments to be used to serve the purposes of others. The
commodification of children entailed by preconception
[. . .] [P]reconception arrangements raise ethical and arrangements ignores these essential values.
legal issues that are neither straightforward nor easy to Moreover, commercial preconception arrange-
deal with. As Commissioners listened to the continu- ments commodify women's reproductive functions and
ing debate about this practice, one conclusion became place women in the situation of alienating aspects of
evident: proponents and opponents are not likely to themselves that should be inherently inalienable. A
change each other's minds about the ethical and social preconception contract obliges the gestational mother
dimensions of preconception arrangements. Views on to sell an intimate aspect of her human functioning to
preconception arrangements are based on fundamen- provide someone else with a genetically related child;
tally different convictions about human nature and the capacity to become pregnant and bear a child is re-
about how the world works or ought to work; there- duced to a marketable service. We do not allow people
fore, assessments of the actual or potential implica- to give up their freedom and become slaves, even if
tions of preconception arrangements for women, for they make a choice to do so, because of our collective
children, for couples, and for our evolution as a society conviction that this would negate the value we attach
also differ. to human dignity and the inalienability of the person.
Similarly, assigning a commercial value to the human
function of reproduction would result eventually in a
Commercial Arrangements
new and, in our view, undesirable social understanding
Using our ethical framework and standards, the Com- of the value and dignity of women, their reproductive
mission finds commercial preconception arrangements capacity, and their bodily integrity.
offensive on several grounds. Commercial preconception contracts by their
First, they offend human dignity by commodify- nature—the exchange of money for a child—contradict
ing women's reproductive capacities and commodify- one of the fundamental tenets of the Commission's
ing children; they contradict the principle that human ethical framework. On these grounds alone, we could
reproduction should not be commercialized in any recommend prohibition of such arrangements, since
way. Second, we see actual and potential harms for we believe that all public policy in this field should
families, for individual women and children, and for be based on the principle of non-commercialization of
specific groups within society. Finally, we believe that reproduction.
public policy that condones or supports the establish- The evidence is clear that in commercial precon-
ment of adversarial relationships is fundamentally ception contracts the principal motivation of both
flawed; public policy should seek instead to support the broker and the gestational woman is money. Far
and encourage humane, non-conflictual family and from being the idyllic situation portrayed by brokers—
social relationships. Any attempt to legitimize or sup- gestational woman as "altruistic angel" giving the gift
port commercial preconception arrangements through of a child to a couple who is happy but infertile—
public policy would represent the antithesis of this goal. commercial preconception contracts are business
transactions. The child is a product being sold by one
Commodification of Children and Reproduction party and bought by the other.
The fundamentally repugnant aspect of preconcep-
tion arrangements is that they instrumentalize human Harms to Individuals
beings through the deliberate act of creating a child for The Commission heard strong arguments that precon-
the express purpose of giving it up, usually in exchange ception arrangements are detrimental to the autonomy
for money. The premise of commercial preconception of gestational mothers. We concur. Far from enhan-
contracts is that a child is a product that can be bought cing the gestational woman's autonomy, as some have
and. sold on the market. The moral point of view re- argued, the practice circumscribes and dictates the
quires that people be treated as ends in themselves, not gestational woman's behaviour by specifying contrac-
as a means to the ends of others. We must therefore tual obligations, including the obligation to be treated
Source: From Proceed with Care: Final Report of the Royal Commission on New Reproductive Technologies (1993), 661-93.
Reprinted with permission.
X NI I 11 I I ill Illlll
by medical personnel selected by the commissioning implications of their ability to bear children. We reject
couple, to have an abortion if the commissioning this argument. Although they are strongly held values,
couple so decides, and to renounce her maternal feel- freedom and autonomy do not include the right to
ings and rights even before conception. Again, we be- engage in activities that will result in harms to others,
lieve there is clear evidence of the potential for coercion particularly, as in this case, to the child that is even-
and exploitation of gestational women because of the tually born; the limits of autonomy become apparent
disparities in power and resources between gestational when its exercise will harm others, as the commodifica-
women and commissioning couples. tion of children most certainly does.
A commissioning couple uses a woman as a vehicle Harm to children born as a result of these arrange-
to serve their own ends. As the New York State Task ments cannot be ignored. Commissioners reject the
Force on Life and the Law observed, they seek the birth argument that these harms could be outweighed by
components of gestation from the gestational woman the opportunity for life, as this argument assumes the
while denying the personal, emotional, and psycho- very factor under deliberation—the child's conception
logical dimensions of her experiences and self. If she and birth. We concur with the assessment of the New
succeeds in denying her emotional responses during York State Task Force on Life and the Law: "The as-
this profound experience, she is dehumanized in the sessment for public policy occurs prior to conception
process. If she fails, her attachment to the child produ- when the . . . arrangements are made. The issue then
ces a conflict that cannot be resolved without anguish is not whether a particular child should be denied life,
for all involved. but whether children should be conceived in circum-
Moreover, informed choice is a necessary compon- stances that would place them at risk." Nor do we see
ent of autonomy. As we have seen, however, brokers any practical way of protecting the interests of the child
arranging preconception arrangements are not neutral that will eventually be born or even of ensuring that
intermediaries; they act in the interests of the commis- they are taken into account in negotiating and conclud-
sioning couple. If the gestational mother has her own ing a preconception arrangement.
lawyer, it is often one recommended by the broker. Commercial preconception arrangements do pro-
Furthermore, since so little is known about the psycho- duce benefits—but the benefits are to brokers and
social effects of these arrangements on the participants commissioning couples at the expense of the interests
and the resulting child, the woman's decision cannot of vulnerable women and of children who had no part
be made in light of all the information that might influ- in the arrangement.
ence it. All these factors undoubtedly undermine her
capacity to exercise informed choice in deciding to Harms to Society
enter a preconception arrangement. We agree with those who argued that commercial
We also conclude that concerns about negative preconception arrangements have potentially nega-
psychosocial consequences for the gestational mother tive consequences not only for individuals but also
are well founded, particularly because it is impossible for women collectively and for other groups in society.
for her to predict, at the time she signs the contract, These arrangements reinforce social attitudes about
how she will feel about fulfilling its terms after the motherhood as the role that defines women's status and
child has been born. value in society. Furthermore, preconception arrange-
Even if fully informed choice were possible, soci- ments could create broad social harms by diminishing
ety has established certain limits on what people are the dignity of reproduction and undermining society's
free to make choices about. Such situations are rare but commitment to the inherent value of children.
central to our definition of the kind of society we want Even if the number of commercial preconception
to live in. Thus, in a caring society, personal autonomy arrangements to date has been relatively small, over
is not a value that trumps all others, and society may time such arrangements would be bound to have a
see fit to place limits on the exercise of free choice detrimental effect on the way society perceives women,
when the choice concerns an activity that society re- children, and reproduction generally.
gards as fundamentally incompatible with values such In short, we reject the notion that public policy
as respect for human dignity and the inalienability of can be based on a description of procreation in terms
the person. of a market production model—which is, essentially,
We heard the view that preconception arrange- that such arrangements should be permissible and
ments enhance women's autonomy by giving substance legally enforceable because commissioning couples
to their right to control their own bodies—by allowing are willing to pay and gestational women are free to
them to decide for themselves the meaning and social sell their labour. Second, we do not accept that the
freedom to procreate automatically assumes a right genuine and legitimate desire of couples who are
to state support—whether in the form of enforceable infertile to have children, but this should not take
contracts or publicly supported medical services—for precedence over the best interests of children, which
the exercise of that right. Finally, we reject the argu- lie in not being the object of a contract, agreement, or
ments of proponents because they are premised on paid transaction.
incomplete or inaccurate depictions of preconception We believe further that preconception arrange-
arrangements, making them an inappropriate basis for ments contradict the ethic of care, as they result—
public policy, which must take into account not only inevitably and intentionally—in the breaking of
the interests of the participants but also the other inter- parental bonds and in strain on family relationships.
ests affected, including those of the resulting child, as They can also result in long, acrimonious conflict, in
well as the potential for individual and/or social harm. court and in the media, between the gestational woman
We also reject the argument of medical necessity. and the commissioning couple; far from preventing
We find it unacceptable that one party—the gestational conflict, preconception arrangements actually make
woman—should be called upon to bear all the medical conflict more likely.
risks of pregnancy and birth, and possibly those of in In reaching our conclusions, we also took into ac-
vitro fertilization and zygote transfer, while all benefits count the Commission's public hearings, consultations,
accrue to the commissioning couple. In no other cir- and surveys, which illustrated the ambivalence of
cumstances does society accept that a healthy person Canadians' attitudes toward commercial arrangements.
be placed at medical risk for the benefit of someone else Canadians have seen what has happened in the United
for money, even when that condition is life-threatening. States, where several cases have come to public atten-
In this case, the commissioning woman's infertility is tion through bitter custody disputes, and do not want
not life-threatening. to see these events repeated here. At the same time,
The commonly understood motivation for com- as we saw in our survey of people across the country,
missioning couples to enter into a preconception ar- Canadians attach great value to having children and
rangement is because the commissioning woman is appear reluctant to deny others the opportunity to have
unable to conceive and/or carry a fetus to term. This a child, even if it means permitting a practice they do
makes preconception arrangements appear as a last not condone or would not engage in themselves. It is
resort when infertility is untreatable or treatment has not clear, however, that public opinion about precon-
been unsuccessful. As we have seen, however, this is ception arrangements is based on a full understanding
not always the case; not everyone who seeks a pre- of their nature and implications because much of the
conception arrangement is involuntarily infertile, a public information about them comes from brokers or
member of a couple, or even of childbearing age. But others with an interest in portraying the practice in a
even in cases where it is true, preconception arrange- positive light.
ments are not an acceptable remedy. In some circumstances a preconception arrange-
ment may seem a reasonable response to a particular
Goals of Public Policy situation. For example, where a woman has a serious
health problem that prevents her from carrying a preg-
As we have argued elsewhere in this report, one goal nancy, she might seek a gestational woman to carry
of public policy in the field of new reproductive tech- a fetus conceived using the commissioning woman's
nologies should be to seek to prevent conflict—or, at eggs and her partner's sperm. Given the broader social
the very least, to avoid knowingly setting up situations harms we have described, however, we do not believe
where conflict is bound to result. Instead, we would that using another woman's reproductive capacity is
seek to foster healthy family and social relationships justifiable even in this situation, as difficult as it might
through such means, for example, as promoting greater be for the individuals involved to accept their inability
social acceptance of family ties based on other than a to have a genetically related child.
genetic component. We do not deny that a public response in the
The Commission recognizes the value of public form of regulation could help to control some of the
policy that supports people's attempts to establish pressures and abuses identified with respect to precon-
families, and we uphold women's right to autonomy. ception arrangements—for example, by requiring the
We believe, nevertheless, that preconception ar- provision of independent legal advice for gestational
rangements can cause damage to children, families, women or by making counselling mandatory for all
and society as a whole and can actually limit, rather parties to an arrangement. We are sceptical, however,
than enhance, women's autonomy. We recognize the that any regulatory scheme could ensure that all parties
uu I II ill 111111
were able to make free and informed choices. More- recognize the vulnerability of couples who are infertile
over, a regulatory approach by its nature would invite and their emotional needs, we believe that making
disputes and conflict. payment for such arrangements should be prohibited.
Even if a regulatory system could be designed We also believe it is essential, in particular, to prohibit
to overcome these obstacles, the deepest and most others from assisting in such arrangements—for ex-
serious harms of preconception arrangements would ample, brokers and physicians—by criminalizing the
remain. No regulatory system could remedy the basic knowing provision of assistance. With these principles
affront to human dignity occasioned by the com- in mind, the Commission reviewed the options avail-
modification of children and the commercialization of able and came to the following recommendations to
reproduction. prohibit commercial preconception arrangements.
Our first goal is to ensure that the status of this
Non-Commercial Arrangements practice is uniform across the country, to discourage
people from travelling to parts of the country where it
The Commissionb conclusions with respect to non- is permitted. Evidence before the Commission shows
commercial preconception arrangements between that arrangements can take place across provincial/
family members or close friends are similar. We do territorial borders. Thus, prohibition only at the prov-
not believe such arrangements should be undertaken, incial level would not be effective with respect to such
sanctioned, or encouraged. The motivation might be arrangements. Hence, we sought a comprehensive,
sincere and generous, but the arrangement still results uniform, and effective approach to preconception ar-
in the commodification of a child and the reproductive rangements across the country. This can be achieved by
process. Even if no money is involved, no one should prohibiting certain activities aimed at facilitating such
have the right to make a "gift" of another human being; arrangements for gain. Accordingly, the Commission
this is offensive to the human dignity of the child. recommends that
Non-commercial arrangements present the poten-
tial for coercion in the form of family pressure to par- 199. The federal government legislate to prohibit
ticipate. They also give rise to the possibility of damage advertising for or acting as an intermediary to
to family relationships before or after the child is born, bring about a preconception arrangement; and
as well as even greater potential for confusion on the to prohibit receiving payment or any financial or
part of the child, because of continuing contact be- commercial benefit for acting as an intermediary,
tween the birth mother and the commissioning couple. under threat of criminal sanction. It should also
Moreover, the arrangement still results in a healthy legislate to prohibit making payment for a pre-
woman being placed at medical risk for the benefit of conception arrangement, under threat of criminal
someone else. sanction.
At the same time, we recognize that the practice
may continue to some degree and so demands a public This proposed criminal prohibition will serve as an
policy response, particularly because of the uncertain- effective deterrent to commercial preconception ar-
ties surrounding the legal status of a child born as a rangements. Given our recommendations with respect
result of such an arrangement and the need to ensure to donor insemination and in vitro fertilization (donor
that the child's best interests are served. We wish to insemination restricted to anonymously donated
make it clear, however, that our recommendations in sperm collected by a licensed facility, in vitro fertil-
this latter regard are not intended to sanction the prac- ization restricted to licensed facilities and offered only
tice but simply to recognize that it is probably going to for diagnosed fallopian tube blockage), physicians
occur and that, in the absence of public policy, signifi- too would be barred from participating in any such
cant harm to children could result. arrangement.
Second, statutory provisions making preconcep-
Recommendations tion contracts unenforceable would operate as a strong
deterrent to the practice, because they would generate
Commissioners are strongly of the view that preconcep- uncertainty for the commissioning couple, whether or
tion arrangements are unacceptable and do not warrant not a broker has been involved. Such provisions would
state support in any form that would signal acceptance ensure that the gestational woman could not be com-
or encouragement of them. We do not advise sanctions pelled by a court to give up custody of a child born as a
with respect to gestational mothers, however, as this result of a preconception agreement. The Commission
would simply compound their vulnerability. While we therefore recommends that
200. Provinces/territories amend their family after a minimum waiting period following the
law legislation to specify that all preconception birth of the child.
agreements, whether or not they involve pay-
ment, are unenforceable against the gestational and that
woman.
205. In any dispute over custody arising from a
Commissioners do not wish to leave the impres- preconception arrangement, the best interests of
sion that we consider non-commercial arrangements the child prevail over the interests of the adults
acceptable or to imply that non-commercial arrange- involved.
ments have no potential to harm individual women,
the status of women generally, children and families, Finally, in support of our international obliga-
or society at large. However, we believe that the most tions, Commissioners believe that Canada should
effective way to deter non-commercial arrangements is demonstrate leadership by supporting policies
to provide for penalties for third parties who facilitate aimed at achieving an international ban on precon-
preconception arrangements. The Commission recom- ception arrangements. Given that Canadians could
mends that go to other countries, particularly the United States,
to seek arrangements not permitted in this country,
201. Self-regulating professional bodies, such as we believe that such a step is needed on the part
provincial colleges of physicians and surgeons of the international community to prevent the ex-
and provincial law societies, adopt strict codes of ploitation of women and the commodification of
conduct, disciplinary measures, and severe penal- children. Adopting a domestic policy would be the
ties, including loss of licence to practise, against first step toward this goal; encouraging other coun-
members involved in brokering or performing as- tries to adopt similar measures would reinforce and
sisted insemination, in vitro fertilization, or zygote/ extend it.
embryo transfer to facilitate a preconception The extent to which Commissioners deplore the
arrangement. practice of preconception arrangements is evident in
our determination to recommend strong measures to
and that discourage these arrangements and to penalize those
who would seek to benefit financially from them.
202. Any facility knowingly providing assisted Our goal is to halt commercial practices entirely
conception procedures in support of a preconcep- and to discourage others from participating in these
tion arrangement lose its licence to provide assisted arrangements. We recognize the value Canadians
conception services. attach to having a genetically related child. In our
view, however, this value cannot be upheld in the
It is important to ensure that the interests of any face of the other values that would have to be sac-
resulting children are protected. In particular, estab- rificed. A caring society has an obligation to ensure
lishing their legal parentage is vital for the children that individual actions—even those intended to
and for the other participants, as it affects the rights benefit others—do not generate greater social harms
and obligations of the parties with respect to custody, and that public policy works to support and foster
access, support, and inheritance. Without clarifica- healthy family and social connections, not to under-
tion of legal parentage, children could be deprived of mine them or set them up to fail.
the support they are owed and become subject to the Preconception arrangements illustrate the ethical
trauma of protracted litigation. For these reasons the dilemmas posed by situations where both benefits and
Commission recommends that harms can result from a practice. In this case, however,
the benefits to a few individuals are far outweighed by
203. All provinces/territories that have not already the harms to others and to society that are likely to
done so amend their family law legislation to result. This is why we have adopted such a strong stand
ensure that a woman who gives birth to a child is against preconception arrangements. Taken together,
considered the legal mother of the child, regardless Commissioners believe, the measures we propose will
of the source of the egg have a strong deterrent effect on preconception ar-
rangements and, in particular, on third-party activities
204. As in the case of adoption, the birth mother in this area, but without compounding the vulnerabil-
be allowed to relinquish her maternal rights only ity of participants. [. . .1
LI I
I I ill 111111
5.3 Stopping Reproduction
Why Abortion Is Immoral seems to be unresolvable. An analysis of the nature

of this standoff suggests a strategy for surmounting it.
Don Marquis Consider the way a typical anti-abortionist argues. She
will argue or assert that life is present from the moment
The view that abortion is, with rare exceptions, ser- of conception or that fetuses look like babies or that
iously immoral has received little support in the recent fetuses possess a characteristic such as a genetic code
philosophical literature. No doubt most philosophers that is both necessary and sufficient for being human.
affiliated with secular institutions of higher education Anti-abortionists seem to believe that (1) the truth of
believe that the anti-abortion position is either a symp- all of these claims is quite obvious, and (2) establishing
tom of irrational religious dogma or a conclusion gen- any of these claims is sufficient to show that abortion is
erated by seriously confused philosophical argument. morally akin to murder.
The purpose of this essay is to undermine this general A standard pro-choice strategy exhibits similarities.
belief. This essay sets out an argument that purports The pro-choicer will argue or assert that fetuses are not
to show, as well as any argument in ethics can show, persons or that fetuses are not rational agents or that
that abortion is, except possibly in rare cases, seriously fetuses are not social beings. Pro-choicers seem to be-
immoral, that it is in the same moral category as killing lieve that (1) the truth of any of these claims is quite
an innocent adult human being. obvious, and (2) establishing any of these claims is suffi-
The argument is based on a major assumption. cient to show that an abortion is not a wrongful killing.
Many of the most insightful and careful writers on In fact, both the pro-choice and the anti-abortion
the ethics of abortion—such as Joel Feinberg, Michael claims do seem to be true, although the "it looks like
Tooley, Mary Anne Warren, H. Tristram Engelhardt, Jr, a baby" claim is more difficult to establish the earlier
L.W. Sumner, John T. Noonan, Jr, and Philip Devine— the pregnancy. We seem to have a standoff. How can
believe that whether or not abortion is morally permis- it be resolved?
sible stands or falls on whether or not a fetus is the sort As everyone who has taken a bit of logic knows, if
of being whose life it is seriously wrong to end. The any of these arguments concerning abortion is a good
argument of this essay will assume, but not argue, that argument, it requires not only some claim characteriz-
they are correct. ing fetuses, but also some general moral principle that
Also, this essay will neglect issues of great im- ties a characteristic of fetuses to having or not having
portance to a complete ethics of abortion. Some anti- the right to life or to some other moral characteristic
abortionists will allow that certain abortions, such as that will generate the obligation or the lack of obliga-
abortion before implantation or abortion when the life tion not to end the life of a fetus. Accordingly, the argu-
of a woman is threatened by a pregnancy or abortion ments of the anti-abortionist and the prochoicer need a
after rape, may be morally permissible. This essay will bit of filling in to be regarded as adequate.
not explore the casuistry of these hard cases. The pur- Note what each partisan will say. The anti-
pose of this essay is to develop a general argument for abortionist will claim that her position is supported
the claim that the overwhelming majority of deliberate by such generally accepted moral principles as "It is
abortions are seriously immoral. always prima facie seriously wrong to take a human
life," or "It is always prima facie seriously wrong to end
the life of a baby." Since these are generally accepted
moral principles, her position is certainly not obviously
A sketch of standard anti-abortion and pro-choice wrong. The prochoicer will claim that her position is
arguments exhibits how those arguments possess cer- supported by such plausible moral principles as "Being
tain symmetries that explain why partisans of those a person is what gives an individual intrinsic moral
positions are so convinced of the correctness of their worth," or "It is only seriously prima facie wrong to
own positions, why they are not successful in convin- take the life of a member of the human community."
cing their opponents, and why, to others, this issue Since these are generally accepted moral principles, the
Source: From Journal of Philosophy 86 (1989) (4), 183-202. Reprinted by permission of the author and the Journal of Philosophy.
Marquis: Why Abortion Is Immoral
pro-choice position is certainly not obviously wrong. of her major principle that is supposed to explain the
Unfortunately, we have again arrived at a standoff. wrongness of taking human life, but which will not also
Now, how might one deal with this standoff? The make abortion immoral. This is no easy task. Appeals
standard approach is to try to show how the moral to social utility will seem satisfactory only to those who
principles of one's opponent lose their plausibility resolve not to think of the enormous difficulties with
under analysis. It is easy to see how this is possible. On a utilitarian account of the wrongness of killing and
the one hand, the anti-abortionist will defend a moral the significant social costs of preserving the lives of the
principle concerning the wrongness of killing which unproductive.' A pro-choice strategy that extends the
tends to be broad in scope in order that even fetuses at definition of "person" to infants or even to young chil-
an early stage of pregnancy will fall under it. The prob- dren seems just as arbitrary as an anti-abortion strategy
lem with broad principles is that they often embrace that extends the definition of "human being" to fetuses.
too much. In this particular instance, the principle "It Again, we find symmetries in the two positions and we
is always prima facie wrong to take a human life" seems arrive at a standoff.
to entail that it is wrong to end the existence of a living There are even further problems that reflect sym-
human cancer-cell culture, on the grounds that the cul- metries in the two positions. In addition to counter-
ture is both living and human. Therefore, it seems that example problems, or the arbitrary application
the anti-abortionist's favoured principle is too broad. problems that can be exchanged for them, the standard
On the other hand, the prochoicer wants to find anti-abortionist principle "It is prima facie seriously
a moral principle concerning the wrongness of kill- wrong to kill a human being," or one of its variants,
ing which tends to be narrow in scope in order that can be objected to on the grounds of ambiguity. If
fetuses will not fall under it. The problem with narrow "human being" is taken to be a biological category, then
principles is that they often do not embrace enough. the anti-abortionist is left with the problem of explain-
Hence, the needed principles such as "It is prima facie ing why a merely biological category should make a
seriously wrong to kill only persons" or "It is prima moral difference. Why, it is asked, is it any more rea-
facie wrong to kill only rational agents' do not explain sonable to base a moral conclusion on the number of
why it is wrong to kill infants or young children or the chromosomes in one's cells than on the colour of one's
severely retarded or even perhaps the severely mentally skin?4 If "human being," on the other hand, is taken
ill. Therefore, we seem again to have a standoff. The to be a moral category, then the claim that a fetus is a
anti-abortionist charges, not unreasonably, that pro- human being cannot be taken to be a premise in the
choice principles concerning killing are too narrow to anti-abortion argument, for it is precisely what needs
be acceptable; the prochoicer charges, not unreason- to be established. Hence, either the anti-abortionist's
ably, that anti-abortionist principles concerning killing main category is a morally irrelevant, merely biological
are too broad to be acceptable. category, or it is of no use to the anti-abortionist in
Attempts by both sides to patch up the difficulties establishing (non-circularly, of course) that abortion is
in their positions run into further difficulties. The anti- wrong. Although this problem with the anti-abortionist
abortionist will try to remove the problem in her pos- position is often noticed, it is less often noticed that the
ition by reformulating her principle concerning killing pro-choice position suffers from an analogous problem.
in terms of human beings. Now we end up with: "It The principle "Only persons have the right to life"
is always prima fade seriously wrong to end the life of also suffers from an ambiguity. The term "person" is typ-
a human being." This principle has the advantage of ically defined in terms of psychological characteristics,
avoiding the problem of the human cancer-cell culture although there will certainly be disagreement concern-
counterexample. But this advantage is purchased at a ing which characteristics are most important. Suppos-
high price. For although it is clear that a fetus is both ing that this matter can be settled, the prochoicer is
human and alive, it is not at all clear that a fetus is left with the problem of explaining why psychological
a human being. There is at least something to be said characteristics should make a moral difference. If the
for the view that something becomes a human being prochoicer should attempt to deal with this problem by
only after a process of development, and that therefore claiming that an explanation is not necessary, that in fact
first-trimester fetuses and perhaps all fetuses are not we do treat such a cluster of psychological properties
yet human beings. Hence, the anti-abortionist, by this as having moral significance, the sharp-witted anti-
move, has merely exchanged one problem for another.' abortionist should have a ready response. We do treat
The prochoicer fares no better. She may attempt to being both living and human as having moral signifi-
find reasons why killing infants, young children, and cance. If it is legitimate for the prochoicer to demand
the severely retarded is wrong which are independent that the anti-abortionist provide an explanation of the
xi .11 I .111
II ill 1111.1.1
connection between the biological character of being Of course, it might not make sense to attribute
a human being and the wrongness of being killed rights to a being that would never in its natural his-
(even though people accept this connection), then it tory have certain psychological traits. This modest
is legitimate for the anti-abortionist to demand that connection between psychological personhood and
the prochoicer provide an explanation of the connec- moral personhood will create a place for Karen Ann
tion between psychological criteria for being a person Quinlan and the temporarily unconscious. But then it
and the wrongness of being killed (even though that makes a place for fetuses also. Hence, it does not serve
connection is accepted).5 Feinberg has attempted to Feinberg's pro-choice purposes. Accordingly, it seems
meet this objection (he calls psychological personhood that the prochoicer will have as much difficulty bridg-
"commonsense personhood"): ing the gap between psychological personhood and
personhood in the moral sense as the anti-abortionist
The characteristics that confer commonsense has bridging the gap between being a biological human
personhood are not arbitrary bases for rights and being and being a human being in the moral sense.
duties, such as race, sex or species membership; Furthermore, the prochoicer cannot any more
rather they are traits that make sense out of rights escape her problem by making "person" a purely moral
and duties and without which those moral attrib- category than the anti-abortionist could escape by the
utes would have no point or function. It is because analogous move. For if "person" is a moral category,
people are conscious; have a sense of their personal then the prochoicer is left without the resources for
identities; have plans, goals, and projects; experi- establishing (non-circularly, of course) the claim that a
ence emotions; are liable to pains, anxieties, and fetus is not a person, which is an essential premise in
frustrations; can reason and bargain, and so on—it her argument. Again, we have both a symmetry and a
is because of these attributes that people have values standoff between pro-choice and anti-abortion views.
and interests, desires and expectations of their Passions in the abortion debate run high. There
own, including a stake in their own futures, and are both plausibilities and difficulties with the stan-
a personal well-being of a sort we cannot ascribe dard positions. Accordingly, it is hardly surprising that
to unconscious or non-rational beings. Because of partisans of either side embrace with fervor the moral
their developed capacities they can assume duties generalizations that support the conclusions they pre-
and responsibilities and can have and make claims analytically favour, and reject with disdain the moral
on one another. Only because of their sense of self, generalizations of their opponents as being subject to
their life plans, their value hierarchies, and their inescapable difficulties. It is easy to believe that the
stakes in their own futures can they be ascribed counterexamples to one's own moral principles are
fundamental rights. There is nothing arbitrary merely temporary difficulties that will dissolve in the
about these linkages. (op. cit., p. 270) wake of further philosophical research, and that the
counterexamples to the principles of one's opponents
The plausible aspects of this attempt should not are as straightforward as the contradiction between A
be taken to obscure its implausible features. There and 0 propositions in traditional logic. This might sug-
is a great deal to be said for the view that being a gest to an impartial observer (if there are any) that the
psychological person under some description is a ne- abortion issue is unresolvable.
cessary condition for having duties. One cannot have There is a way out of this apparent dialectical
a duty unless one is capable of behaving morally, and quandary. The moral generalizations of both sides are
a being's capability of behaving morally will require not quite correct. The generalizations hold for the most
having a certain psychology. It is far from obvious, part, for the usual cases. This suggests that they are all
however, that having rights entails consciousness accidental generalizations that the moral claims made
or rationality, as Feinberg suggests. We speak of the by those on both sides of the dispute do not touch on
rights of the severely retarded or the severely men- the essence of the matter.
tally ill, yet some of these persons are not rational. We This use of the distinction between essence and
speak of the rights of the temporarily unconscious. accident is not meant to invoke obscure metaphysical
The New Jersey Supreme Court based their decision categories. Rather, it is intended to reflect the rather
in the Quinlan case on Karen Ann Quinlan's right to atheoretical nature of the abortion discussion. If the
privacy, and she was known to be permanently un- generalization a partisan in the abortion dispute adopts
conscious at that time. Hence, Feinberg's claim that were derived from the reason why ending the life
having rights entails being conscious is, on its face, of a human being is wrong, then there could not be
obviously false. exceptions to that generalization unless some special
case obtains in which there are even more powerful Therefore, when I die, I am deprived of all of the value
countervailing reasons. Such generalizations would not of my future. Inflicting this loss on me is ultimately
be merely accidental generalizations; they would point what makes killing me wrong. This being the case, it
to, or be based upon, the essence of the wrongness of would seem that what makes killing any adult human
killing, what it is that makes killing wrong. All this sug- being prima facie seriously wrong is the loss of his or
gests that a necessary condition of resolving the abor- her future.°
tion controversy is a more theoretical account of the How should this rudimentary theory of the
wrongness of killing. After all, if we merely believe, but wrongness of killing be evaluated? It cannot be faulted
do not understand, why killing adult human beings for deriving an "ought" from an "is," for it does not.
such as ourselves is wrong, how could we conceivably The analysis assumes that killing me (or you, reader)
show that abortion is either immoral or permissible? is prima fade seriously wrong. The point of the an-
alysis is to establish which natural property ultimately
II explains the wrongness of the killing, given that it is
wrong. A natural property will ultimately explain the
In order t o develop such an account, w e can start from wrongness of killing, only if (1) the explanation fits
the following unproblematic assumption concerning with our intuitions about the matter and (2) there is
our own case: it is wrong to kill us. Why is it wrong? no other natural property that provides the basis for
Some answers can be easily eliminated. It might be a better explanation of the wrongness of killing. This
said that what makes killing us wrong is that a killing analysis rests on the intuition that what makes killing
brutalizes the one who kills. But the brutalization con- a particular human or animal wrong is what it does to
sists of being inured to the performance of an act that that particular human or animal. What makes killing
is hideously immoral; hence, the brutalization does wrong is some natural effect or other of the killing.
not explain the immorality It might be said that what Some would deny this. For instance, a divine com-
makes killing us wrong is the great loss others would mand theorist in ethics would deny it. Surely this
experience due to our absence. Although such hubris is denial is, however, one of those features of divine-
understandable, such an explanation does not account command theory which renders it so implausible.
for the wrongness of killing hermits, or those whose The claim that what makes killing wrong is the
lives are relatively independent and whose friends find loss of the victim's future is directly supported by two
it easy to make new friends. considerations. In the first place, this theory explains
A more obvious answer is better. What primarily why we regard killing as one of the worst of crimes.
makes killing wrong is neither its effect on the mur- Killing is especially wrong, because it deprives the
derer nor its effect on the victim's friends and relatives, victim of more than perhaps any other crime. In the
but its effect on the victim. The loss of one's life is one second place, people with AIDS or cancer who know
of the greatest losses one can suffer. The loss of one's life they are dying believe, of course, that dying is a very
deprives one of all the experiences, activities, projects, bad thing for them. They believe that the loss of a future
and enjoyments that would otherwise have constituted to them that they would otherwise have experienced is
one future. Therefore, killing someone is wrong, pri- what makes their premature death a very bad thing for
marily because the killing inflicts (one of) the greatest them. A better theory of the wrongness of killing would
possible losses on the victim. To describe this as the require a different natural property associated with kill-
loss of life can be misleading, however. The change in ing which better fits with the attitudes of the dying.
my biological state does not by itself make killing me What could it be?
wrong. The effect of the loss of my biological life is the The view that what makes killing wrong is the loss
loss to me of all those activities, projects, experiences, to the victim of the value of the victim's future gains
and enjoyments which would otherwise have consti- additional support when some of its implications are
tuted my future personal life. These activities, projects, examined. In the first place, it is incompatible with
experiences, and enjoyments are either valuable for the view that it is wrong to kill only beings who are
their own sakes or are means to something else that biologically human. It is possible that there exists a
is valuable for its own sake. Some parts of my future different species from another planet whose members
are not valued by me now, but will come to be valued have a future like ours. Since having a future like that
by me as I grow older and as my values and capacities is what makes killing someone wrong, this theory en-
change. When I am killed, I am deprived both of what tails that it would be wrong to kill members of such
I now value which would have been part of my future a species. Hence, this theory is opposed to the claim
personal life, but also what I would come to value. that only life that is biologically human has great moral
3 II 11 31 11311
worth, a claim which many anti-abortionists have of such ad hoc theories seems to be a function of how
seemed to adopt. This opposition, which this theory desperately one wants such theories to work. The claim
has in common with personhood theories, seems to be that the primary wrong-making feature of a killing is
a merit of the theory. the loss to the victim of the value of its future accounts
In the second place, the claim that the loss of for the wrongness of killing young children and infants
one future is the wrong-making feature of one's being directly; it makes the wrongness of such acts as ob-
killed entails the possibility that the futures of some vious as we actually think it is. This is a further merit
actual non-human mammals on our own planet are of this theory. Accordingly, it seems that this value of
sufficiently like ours that it is seriously wrong to kill a future-like-ours theory of the wrongness of killing
them also. Whether some animals do have the same shares strengths of both sanctity-of-life and person-
right to life as human beings depends on adding to the hood accounts while avoiding weaknesses of both. In
account of the wrongness of killing some additional ac- addition, it meshes with a central intuition concerning
count of just what it is about my future or the futures what makes killing wrong.
of other adult human beings which makes it wrong to The claim that the primary wrong-making feature
kill us. No such additional account will be offered in of a killing is the loss to the victim of the value of its
this essay. Undoubtedly, the provision of such an ac- future has obvious consequences for the ethics of abor-
count would be a very difficult matter. Undoubtedly, tion. The future of a standard fetus includes a set of
any such account would be quite controversial. Hence, experiences, projects, activities, and such which are
it surely should not reflect badly on this sketch of an identical with the futures of adult human beings and
elementary theory of the wrongness of killing that it is are identical with the futures of young children. Since
indeterminate with respect to some very difficult issues the reason that is sufficient to explain why it is wrong
regarding animal rights. to kill human beings after the time of birth is a reason
In the third place, the claim that the loss of one's that also applies to fetuses, it follows that abortion is
future is the wrong-making feature of one's being killed prima facie seriously morally wrong.
does not entail, as sanctity of human life theories do, This argument does not rely on the invalid infer-
that active euthanasia is wrong. Persons who are se- ence that, since it is wrong to kill persons, it is wrong
verely and incurably ill, who face a future of pain and to kill potential persons also. The category that is mor-
despair, and who wish to die will not have suffered a ally central to this analysis is the category of having a
loss if they are killed. It is, strictly speaking, the value valuable future like ours; it is not the category of per-
of a humans future which makes killing wrong in this sonhood. The argument to the conclusion that abor-
theory. This being so, killing does not necessarily wrong tion is prima facie seriously morally wrong proceeded
some persons who are sick and dying. Of course, there independently of the notion of person or potential
may be other reasons for a prohibition of active eutha- person or any equivalent. Someone may wish to start
nasia, but that is another matter. Sanctity-of-human-life with this analysis in terms of the value of a human
theories seem to hold that active euthanasia is seriously future, conclude that abortion is, except perhaps in
wrong even in an individual case where there seems to rare circumstances, seriously morally wrong, infer that
be good reason for it independently of public policy fetuses have the right to life, and then call fetuses "per-
considerations. This consequence is most implausible, sons" as a result of their having the right to life. Clearly
and it is a plus for the claim that the loss of a future of in this case, the category of person is being used to state
value is what makes killing wrong that it does not share the conclusion of the analysis rather than to generate
this consequence. the argument of the analysis.
In the fourth place, the account of the wrongness The structure of this anti-abortion argument can
of killing defended in this essay does straightforwardly be both illuminated and defended by comparing it to
entail that it is prima facie seriously wrong to kill chil- what appears to be the best argument for the wrong-
dren and infants, for we do presume that they have ness of the wanton infliction of pain on animals. This
futures of value. Since we do believe that it is wrong latter argument is based on the assumption that it
to kill defenceless little babies, it is important that a is prima facie wrong to inflict pain on me (or you,
theory of the wrongness of killing easily account for reader). What is the natural property associated with
this. Personhood theories of the wrongness of killing, the infliction of pain which makes such infliction
on the other hand, cannot straightforwardly account for wrong? The obvious answer seems to be that the in-
the wrongness of killing infants and young children.' fliction of pain causes suffering and that suffering is a
Hence, such theories must add special ad hoc accounts misfortune. The suffering caused by the infliction of
of the wrongness of killing the young. The plausibility pain is what makes the wanton infliction of pain on me
wrong. The wanton infliction of pain on other adult Hence, Kant's argument for the wrongness of
humans causes suffering. The wanton infliction of pain inflicting pain on animals rests on a claim that, in a
on animals causes suffering. Since causing suffering is world of Kantian moral agents, is demonstrably false.
what makes the wanton infliction of pain wrong and Therefore, the alternative analysis, being more plaus-
since the wanton infliction of pain on animals causes ible anyway, should be accepted. Since this alternative
suffering, it follows that the wanton infliction of pain analysis has the same structure as the anti-abortion
on animals is wrong. argument being defended here, we have further sup-
This argument for the wrongness of the wanton port for the argument for the immorality of abortion
infliction of pain on animals shares a number of struc- being defended in this essay.
tural features with the argument for the serious prima Of course, this value of a future-like-ours argu-
facie wrongness of abortion. Both arguments start with ment, if sound, shows only that abortion is prima
an obvious assumption concerning what it is wrong facie wrong, not that it is wrong in any and all circum-
to do to me (or you, reader). Both then look for the stances. Since the loss of the future to a standard fetus,
characteristic or the consequence of the wrong action if killed, is, however, at least as great a loss as the loss
which makes the action wrong. Both recognize that of the future to a standard adult human being who is
the wrong-making feature of these immoral actions is killed, abortion, like ordinary killing, could be justified
a property of actions sometimes directed at individuals only by the most compelling reasons. The loss of one's
other than postnatal human beings. If the structure of life is almost the greatest misfortune that can happen
the argument for the wrongness of the wanton inflic- to one. Presumably abortion could be justified in some
tion of pain on animals is sound, then the structure circumstances, only if the loss consequent on failing
of the argument for the prima facie serious wrongness to abort would be at least as great. Accordingly, mor-
of abortion is also sound, for the structure of the two ally permissible abortions will be rare indeed unless,
arguments is the same. The structure common to both perhaps, they occur so early in pregnancy that a fetus is
is the key to the explanation of how the wrongness not yet definitely an individual. Hence, this argument
of abortion can be demonstrated without recourse to should be taken as showing that abortion is presump-
the category of person. In neither argument is that cat- tively very seriously wrong, where the presumption is
egory crucial. very strong as strong as the presumption that killing
This defence of an argument for the wrongness another adult human being is wrong.
of abortion in terms of a structurally similar argument
for the wrongness of the wanton infliction of pain on III
animals succeeds only if the account regarding animals
is the correct account. Is it? In the first place, it seems How complete an account o f the wrongness o f killing
plausible. In the second place, its major competition does the value of a future-like-ours account have to
is Kant's account. Kant believed that we do not have be in order that the wrongness of abortion is a conse-
direct duties to animals at all, because they are not per- quence? This account does not have to be an account
sons. Hence, Kant had to explain and justify the wrong- of the necessary conditions for the wrongness of kill-
ness of inflicting pain on animals on the grounds that ing. Some persons in nursing homes may lack valu-
"he who is hard in his dealings with animals becomes able human futures, yet it may be wrong to kill them
hard also in his dealing with men."8 The problem with for other reasons. Furthermore, this account does not
Kant's account is that there seems to be no reason for obviously have to be the sole reason killing is wrong
accepting this latter claim unless Kant's account is re- where the victim did have a valuable future. This an-
jected. If the alternative to Kant's account is accepted, alysis claims only that, for any killing where the victim
then it is easy to understand why someone who is indif- did have a valuable future like ours, having that future
ferent to inflicting pain on animals is also indifferent to by itself is sufficient to create the strong presumption
inflicting pain on humans, for one is indifferent to what that the killing is seriously wrong.
makes inflicting pain wrong in both cases. But, if Kant's One way to overturn the value of a future-like-ours
account is accepted, there is no intelligible reason why argument would be to find some account of the wrong-
one who is hard in his dealings with animals (or crab- ness of killing which is at least as intelligible and which
grass or stones) should also be hard in his dealings with has different implications for the ethics of abortion. Two
men. After all, men are persons: animals are no more rival accounts possess at least some degree of plaus-
persons than crabgrass or stones. Persons are Kant's ibility. One account is based on the obvious fact that
crucial moral category. Why, in short, should a Kantian people value the experience of living and wish for that
accept the basic claim in Kant's argument? valuable experience to continue. Therefore, it might be
II ill 111111
said, what makes killing wrong is the discontinuation This suggests that one can retain what is intuitively
of that experience for the victim. Let us call this the plausible about the desire account without a challenge
discontinuation account.9 Another rival account is based to the basic argument of this paper.
upon the obvious fact that people strongly desire to It is also worth noting that, if future desires
continue to live. This suggests that what makes killing have moral force in a modified desire account of the
us so wrong is that it interferes with the fulfillment of a wrongness of killing, one can find support for an
strong and fundamental desire, the fulfillment of which anti-abortion ethic even in the absence of a value of
is necessary for the fulfillment of any other desires we a future-like-ours account. If one decides that a mor-
might have. Let us call this the desire account.'° ally relevant property, the possession of which is suf-
Consider first the desire account as a rival ac- ficient to make it wrong to kill some individual, is the
count of the ethics of killing which would provide the desire at some future time to live one might decide to
basis for rejecting the anti-abortion position. Such an justify one's refusal to kill suicidal teenagers on these
account will have to be stronger than the value of a grounds, for example then, since typical fetuses will
future-like-ours account of the wrongness of abortion have the desire in the future to live, it is wrong to kill
if it is to do the job expected of it. typical fetuses. Accordingly, it does not seem that a
To entail the wrongness of abortion, the value of desire account of the wrongness of killing can provide
a future-like-ours account has only to provide a suf- a justification of a pro-choice ethic of abortion which
ficient, but not a necessary, condition for the wrong- is nearly as adequate as the value of a human-future
ness of killing. The desire account, on the other hand, justification of an anti-abortion ethic.
must provide us also with a necessary condition for The discontinuation account looks more promis-
the wrongness of killing in order to generate a pro- ing as an account of the wrongness of killing. It seems
choice conclusion on abortion. The reason for this is just as intelligible as the value of a future-like-ours ac-
that presumably the argument from the desire account count, but it does not justify an anti-abortion position.
moves from the claim that what makes killing wrong Obviously, if it is the continuation of one's activities,
is interference with a very strong desire to the claim experiences, and projects, the loss of which makes kill-
that abortion is not wrong because the fetus lacks a ing wrong, then it is not wrong to kill fetuses for that
strong desire to live. Obviously, this inference fails if reason, for fetuses do not have experiences, activities,
someone's having the desire to live is not a necessary and projects to be continued or discontinued. Accord-
condition of its being wrong to kill that individual. ingly, the discontinuation account does not have the
One problem with the desire account is that we anti-abortion consequences that the value of a future-
do regard it as seriously wrong to kill persons who like-ours account has. Yet, it seems as intelligible as the
have little desire to live or who have no desire to live value of a future-like-ours account, for when we think
or, indeed, have a desire not to live. We believe it is of what would be wrong with our being killed, it does
seriously wrong to kill the unconscious, the sleeping, seem as if it is the discontinuation of what makes our
those who are tired of life, and those who are suicidal. lives worthwhile which makes killing us wrong.
The value-of-a-human-future account renders standard Is the discontinuation account just as good an ac-
morality intelligible in these cases; these cases appear count as the value of a future-like-ours account? The
to be incompatible with the desire account. discontinuation account will not be adequate at all, if it
The desire account is subject to a deeper difficulty. does not refer to the value of the experience that may
We desire life, because we value the goods of this life. be discontinued. One does not want the discontinua-
The goodness of life is not secondly to our desire for tion account to make it wrong to kill a patient who
it. If this were not so, the pain of one's own premature begs for death and who is in severe pain that cannot be
death could be done away with merely by an appropri- relieved short of killing. (I leave open the question of
ate alteration in the configuration of one's desires. This whether it is wrong for other reasons.) Accordingly, the
is absurd. Hence, it would seem that it is the loss of discontinuation account must be more than a bare dis-
the goods of one's future, not the interference with the continuation account. It must make some reference to
fulfillment of a strong desire to live, which accounts ul- the positive value of the patient's experiences. But, by
timately for the wrongness of killing. It is worth noting the same token, the value of a future-like-ours account
that, if the desire account is modified so that it does not cannot be a bare future account either. Just having a
provide a necessary, but only a sufficient, condition for future surely does not itself rule out killing the above
the wrongness of killing, the desire account is compat- patient. This account must make some reference to
ible with the value of a future-like-ours account. The the value of the patient's future experiences and pro-
combined accounts will yield an anti-abortion ethic. jects also. Hence, both accounts involve the value of
Marquis: Why Abortion Is Immoral 245
experiences, projects, and activities. So far we still have whatsoever, but I may be wrong because others rightly
symmetry between the accounts. see value even great value in it. Furthermore, my future
The symmetry fades, however, when we focus on can be valuable to me even if I do not value it. This is
the time period of the value of the experiences, etc., the case when a young person attempts suicide, but is
which has moral consequences. Although both ac- rescued and goes on to significant human achievements.
counts leave open the possibility that the patient in our Such young people's futures are ultimately valuable to
example may be killed, this possibility is left open only them, even though such futures do not seem to be valu-
in virtue of the utterly bleak future for the patient. It able to them at the moment of attempted suicide. A fetus's
makes no difference whether the patient's immediate future can be valuable to it in the same way. Accordingly,
past contains intolerable pain, or consists in being in a this attempt to limit the anti-abortion argument fails.
coma (which we can imagine is a situation of indiffer- Another similar attempt to reject the anti-abortion
ence), or consists in a life of value. If the patient's future position is based on Tooley's claim that an entity cannot
is a future of value, we want our account to make it possess the right to life unless it has the capacity to desire
wrong to kill the patient. If the patient's future is in- its continued existence. It follows that, since fetuses lack
tolerable, whatever his or her immediate past, we want the conceptual capacity to desire to continue to live, they
our account to allow killing the patient. Obviously, lack the right to life. Accordingly, Tooley concludes that
then, it is the value of that patient's future which is abortion cannot be seriously prima facie wrong (op. cit.,
doing the work in rendering the morality of killing the pp. 46-7). What could b e the evidence for Tooley's basic
patient intelligible. claim? Tooley once argued that individuals have a prima
This being the case, it seems clear that whether one facie right to what they desire and that the lack of the
has immediate past experiences or not does no work in capacity to desire something undercuts the basis of one's
the explanation of what makes killing wrong. The addi- right to it (op. cit., pp. 44-5). This argument plainly
tion the discontinuation account makes to the value of will not succeed in the context of the analysis of this
a human future account is otiose. Its addition to the essay, however, since the point here is to establish the
value-of-a-future account plays no role at all in render- fetus's right to life on other grounds. Tooley's argument
ing intelligible the wrongness of killing. Therefore, it assumes that the right to life cannot be established in
can be discarded with the discontinuation account of general on some basis other than the desire for life. This
which it is a part. position was considered and rejected in the preceding
section of this paper.
IV One might attempt to defend Tooley's basic claim
on the grounds that, because a fetus cannot apprehend
The analysis of the previous section suggests that al- continued life as a benefit, its continued life cannot be
ternative general accounts of the wrongness of killing a benefit or cannot be something it has a right to or
are either inadequate or unsuccessful in getting around cannot be something that is in its interest. This might
the anti-abortion consequences of the value of a future- be defended in terms of the general proposition that,
like-ours argument. A different strategy for avoiding if an individual is literally incapable of caring about or
these anti-abortion consequences involves limiting the taking an interest in some x, then one does not have a
scope of the value of a future argument. More precisely, right to x or x is not a benefit or xis not something that
the strategy involves arguing that fetuses lack a prop- is in one's interest.°
erty that is essential for the value-of-a-future argument Each member o f this family o f claims seems to be
(or for any anti-abortion argument) to apply to them. open to objections. As John C. Stevens12 has pointed
One move of this sort is based upon the claim that a out, one may have a right to be treated with a certain
necessary condition of one's future being valuable is medical procedure (because of a health insurance policy
that one values it. Value implies a valuer. Given this one has purchased), even though one cannot conceive
one might argue that, since fetuses cannot value their of the nature of the procedure. And, as Tooley himself
futures, their futures are not valuable to them. Hence, has pointed out, persons who have been indoctrinated,
it does not seriously wrong them deliberately to end or drugged, or rendered temporarily unconscious may
their lives. be literally incapable of caring about or taking an inter-
This move fails, however, because of some ambigu- est in something that is in their interest or is something
ities. Let us assume that something cannot be of value to which they have a right, or is something that benefits
unless it is valued by someone. This does not entail that them. Hence, the Tooley claim that would restrict the
my life is of no value unless it is valued by me. I may scope of the value of a future-like-ours argument is
think, in a period of despair, that my future is of no worth undermined by counterexamples.°
I I ill 111111
Finally, Paul Bassenm has argued that, even though and his low self-esteem, regarded his work as unworthy
the prospects of an embryo might seem to be a basis of survival, even though it possessed genuine literary
for the wrongness of abortion, an embryo cannot be a merit. Destruction of such work would surely victim-
victim and therefore cannot be wronged. An embryo ize its author. In such a case, empathy with the victim
cannot be a victim, he says, because it lacks sentience. concerning the loss is clearly impossible.
His central argument for this seems to be that, even Of course, Bassen does not make the possibility of
though plants and the permanently unconscious are empathy a necessary condition of victimizability; he re-
alive, they clearly cannot be victims. What is the ex- quires only mentation. Hence, on Bassen's actual view,
planation of this? Bassen claims that the explanation is this author, as I have described him, can be a victim.
that their lives consist of mere metabolism and mere The problem is that the basic intuition that renders
metabolism is not enough to ground victimizability. Bassen's view plausible is missing in the author'S case.
Mentation is required. In order to attempt to avoid counterexamples, Bassen
The problem with this attempt to establish the has made his thesis too weak to be supported by the
absence of victimizability is that both plants and the intuitions that suggested it.
permanently unconscious clearly lack what Bassen Even so, the mentation requirement on victimiz-
calls "prospects" or what I have called "a future life like ability is still subject to counterexamples. Suppose a
ours." Hence, it is surely open to one to argue that the severe accident renders me totally unconscious for a
real reason we believe plants and the permanently un- month, after which I recover. Surely killing me while
conscious cannot be victims is that killing them cannot I am unconscious victimizes me, even though I am in-
deprive them of a future life like ours; the real reason is capable of mentation during that time. It follows that
not their absence of present mentation. Bassen's thesis fails. Apparently, attempts to restrict the
Bassen recognizes that his view is subject to this dif- value of a future-like-ours argument so that fetuses do
ficulty, and he recognizes that the case of children seems not fall within its scope do not succeed.
to support this difficulty, for "much of what we do for
children is based on prospects." He argues, however, V
that, in the case of children and in other such cases,
"potentiality comes into play only where victimizability In this essay, it has been argued that the correct ethic
has been secured on other grounds" (ibid., p. 333). of the wrongness of killing can be extended to fetal life
Bassen's defence of his view is patently question- and used to show that there is a strong presumption
begging, since what is adequate to secure victimiz- that any abortion is morally impermissible. If the ethic
ability is exactly what is at issue. His examples do not of killing adopted here entails, however, that contracep-
support his own view against the thesis of this essay. tion is also seriously immoral, then there would appear
Of course, embryos can be victims: when their lives to be a difficulty with the analysis of this essay. But this
are deliberately terminated, they are deprived of their analysis does not entail that contraception is wrong.
futures of value, their prospects. This makes them vic- Of course, contraception prevents the actualization of
tims, for it directly wrongs them. a possible future of value. Hence, it follows from the
The seeming plausibility of Bassen's view stems claim that futures of value should be maximized that
from the fact that paradigmatic cases of imagining contraception is prima facie immoral. This obligation to
someone as a victim involve empathy, and empathy maximize does not exist, however, furthermore, noth-
requires mentation of the victim. The victims of flood, ing in the ethics of killing in this paper entails that it
famine, rape, or child abuse are all persons with whom does. The ethics of killing in this essay would entail that
we can empathize. That empathy seems to be part of contraception is wrong only if something were denied
seeing them as victims.15 a human future of value by contraception. Nothing at
In spite of the strength of these examples, the all is denied such a future by contraception, however.
attractive intuition that a situation in which there is Candidates for a subject of harm by contraception
victimization requires the possibility of empathy is sub- fall into four categories: (1) some sperm or other, (2) some
ject to counterexamples. Consider a case that Bassen ovum or other, (3) a sperm and an ovum separately, and
himself offers: "Posthumous obliteration of an author's (4) a sperm and an ovum together. Assigning the harm to
work constitutes a misfortune for him only if he had some sperm is utterly arbitrary, for no reason can be given
wished his work to endure" (op cit., p. 318). The con- for making a sperm the subject of harm rather than an
ditions Bassen wishes to impose upon the possibility of ovum. Assigning the harm to some ovum is utterly arbi-
being victimized here seem far too strong. Perhaps this trary, for no reason can be given for making an ovum the
author, due to his unrealistic standards of excellence subject of harm rather than a sperm. One might attempt
to avoid these problems by insisting that contraception fetus. Since a fetus possesses a property, the possession
deprives both the sperm and the ovum separately of a of which in adult human beings is sufficient to make
valuable future like ours. On this alternative, too many killing an adult human being wrong, abortion is wrong.
futures are lost. Contraception was supposed to be wrong, This way of dealing with the problem of abortion seems
because it deprived us of one future of value, not two. superior to other approaches to the ethics of abortion,
One might attempt to avoid this problem by holding that because it rests on an ethics of killing which is close to
contraception deprives the combination of sperm and self-evident, because the crucial morally relevant prop-
ovum of a valuable future like ours. But here the definite erty clearly applies to fetuses, and because the argument'
article misleads. At the time of contraception, there are avoids the usual equivocations on "human life," "human
hundreds of millions of sperm, one (released) ovum and being," or "person." The argument rests neither on reli-
millions of possible combinations of all of these. There gious claims nor on Papal dogma. It is not subject to the
is no actual combination at all. Is the subject of the loss objection of "speciesism." Its soundness is compatible
to be a merely possible combination? Which one? This with the moral permissibility of euthanasia and contra-
alternative does not yield an actual subject of harm either. ception. It deals with our intuitions concerning young
Accordingly, the immorality of contraception is not en- children. Finally, this analysis can be viewed as resolv-
tailed by the loss of a future-like-ours argument simply ing a standard problem indeed, the standard problem
because there is no non-arbitrarily identifiable subject of concerning the ethics of abortion. Clearly, it is wrong to
the loss in the case of contraception. kill adult human beings. Clearly, it is not wrong to end
the life of some arbitrarily chosen single human cell.
VI Fetuses seem to be like arbitrarily chosen human cells in
some respects and like adult humans in other respects.
The purpose of this essay has been to set out an argu- The problem of the ethics of abortion is the problem
ment for the serious presumptive wrongness of abortion of determining the fetal property that settles this moral
subject to the assumption that the moral permissibil- controversy. The thesis of this essay is that the problem
ity of abortion stands or falls on the moral status of the of the ethics of abortion, so understood, is solvable.
Notes
1. Feinberg, (1986), "Abortion," i n Matters of Life and Death: 5. This seems to be the fatal flaw in Warren's treatment of
New Introductory Essays in Moral Philosophy, ed. Tom Regan this issue.
(Random House: New York), 256-93; Tooley, (1972), 6. I have been most influenced on this matter by Jonathan
"Abortion and Infanticide," in Philosophy and Public Affairs Glover, (1977), Causing Death and Saving Lives (Penguin:
11(1): 37-65; Tooley, (1984), Abortion and Infanticide New York), ch. 3; and Robert Young, (1979), "What Is
(Oxford: New York); Warren, (1973), "On the Moral and So Wrong with Killing People?," in Philosophy, 225 (210):
Legal Status of Abortion," in The Monist 1 (11): 43-61; 515-28.
Engelhardt, (1974), "The Ontology of Abortion," in Ethics, 7. Feinberg, Tooley, Warren, and Engelhardt have all dealt
34 (3): 217-34; Sumner, (1981), Abortion and Moral Theory with this problem.
(Princeton University Press: Princeton); Noonan, (1970), 8. "Duties to Animals and Spirits," (1963), in Lectures on
"An Almost Absolute Value in History," in The Morality Ethics, trans. Louis Infeld (Harper: New York), 239.
of Abortion: Legal and Historical Perspectives, Noonan, 9. I am indebted to Jack Bricke for raising this objection.
ed. (Harvard University Press: Cambridge); and Devine, 10. Presumably a preference utilitarian would press such an
(1978), The Ethics of Homicide (Cornell: Ithaca, NY). objection. Tooley once suggested that his account has
2. For interesting discussions of this issue, see Warren such a theoretical underpinning. See his "Abortion and
Quinn, (1984), "Abortion: Identity and Loss," in Philosophy Infanticide," 44-5.
and Public Affairs 13 (1): 24-54; and Lawrence C Becker, 11. Donald VanDeVeer seems to think this is self-evident. See
(1975), "Human Being: The Boundaries of the Concept," his "Whither Baby Doe?" in Matters of Life and Death, 233.
in Philosophy and Public Affairs 4 (4): 334-59. 12. "Must the Bearer of a Right Have the Concept of That to
3. For example, see my "Ethics and The Elderly: Some Which He Has a Right?," (1984), in Ethics 95 (1): 68-74.
Problems," (1978), in Aging and the Elderly: Humanistic 13. See Tooley again in "Abortion and Infanticide," 47-9.
Perspectives in Gerontology, ed. Stuart Spicker, Kathleen 14. "Present Sakes and Future Prospects: The Status of Early
Woodward, and David Van Tassel (Humanities: Atlantic Abortion," (1982), in Philosophy and Public Affairs 11 (4):
Highlands, NJ), 341-55. 322-6.
4. See Warren, op. cit., and Tooley, "Abortion and Infanticide." 15. Note carefully the reasons he gives on the bottom of p. 316.
111
II ill 111111
On the Moral and Legal Status other species could be. If (1) is acceptable only if the
moral sense is intended, (2) is non-question-begging
of Abortion only if what is intended is the genetic sense.
Mary Anne Warren In "Deciding Who is Human," Noonan argues
for the classification of fetuses with human beings by
pointing to the presence of the full genetic code, and
The question which we must answer in order to pro- the potential capacity for rational thought.2 It is clear
duce a satisfactory solution to the problem of the moral that what he needs to show, for his version of the trad-
status of abortion is this: How are we to define the itional argument to be valid, is that fetuses are human
moral community, the set of beings with full and equal in the moral sense, the sense in which it is analytically
moral rights, such that we can decide whether a human true that all human beings have full moral rights. But,
fetus is a member of this community or not? What sort in the absence of any argument showing that whatever
of entity, exactly, has the inalienable rights to life, lib- is genetically human is also morally human, and he
erty, and the pursuit of happiness? Jefferson attributed gives none, nothing more than genetic humanity can
these rights to all men, and it may or may not be fair to be demonstrated by the presence of the human gen-
suggest that he intended to attribute them only to men. etic code. And, as we will see, the potential capacity for
Perhaps he ought to have attributed them to all human rational thought can at most show that an entity has the
beings. If so, then we arrive, first, at [John] Noonan's potential for becoming human in the moral sense.
problem of defining what makes a being human, and,
second, at the equally vital question which Noonan
2. Defining the Moral Community
does not consider, namely, what reason is there for
identifying the moral community with the set of all Can it be established that genetic humanity is suffi-
human beings, in whatever way we have chosen to cient for moral humanity? I think that there are very
define that term? good reasons for not defining the moral community in
this way. I would like to suggest an alternative way of
1. On the Definition of"Human" defining the moral community, which I will argue for
only to the extent of explaining why it is, or should be,
One reason why this vital second question is so fre- self-evident. The suggestion is simply that the moral
quently overlooked in the debate over the moral status community consists of all and only people, rather than
of abortion is that the term "human" has two distinct, all and only human beings,3 and probably the best way
but not often distinguished, senses. This fact results in of demonstrating its self-evidence is by considering the
a slide of meaning, which serves to conceal the falla- concept of personhood, to see what sorts of entity are
ciousness of the traditional argument that since (1) it is and are not persons, and what the decision that a being
wrong to kill innocent human beings, and (2) fetuses is or is not a person implies about its moral rights.
are innocent human beings, then (3) it is wrong to kill What characteristics entitle an entity to be con-
fetuses. For if "human" is used in the same sense in sidered a person? This is obviously not the place to
both (1) and (2) then, whichever of the two senses is attempt a complete analysis of the concept of person-
meant, one of these premises is question-begging. And hood, but we do not need such a fully adequate analy-
if it is used in two different senses then of course the sis just to determine whether and why a fetus is or isn't
conclusion doesn't follow a person. All we need is a rough and approximate list of
Thus, (1) is a self-evident moral truth,' and avoids the most basic criteria of personhood, and some idea of
begging the question about abortion, only if "human which, or how many, of these an entity must satisfy in
being" is used to mean something like "a full-fledged order to properly be considered a person.
member of the moral community" (It may or may not In searching for such criteria, it is useful to look
also be meant to refer exclusively to members of the beyond the set of people with whom we are acquainted,
species Homo sapiens.) We may call this the moral sense and ask how we would decide whether a totally alien
of "human." It is not to be conf used with what we call being was a person or not. (For we have no right to
the genetic sense, i.e., the sense in which any member assume that genetic humanity is necessary for per-
of the species is a human being, and no member of any sonhood.) Imagine a space traveller who lands on an
Source From The Monist: An International Quarterly Journal of General Philosophical Inquiry 57 (1973), 43-61. Copyright © 1973 The
MONIST: An International Quarterly Journal of General Philosophical Inquiry Peru, Illinois, USA 61354. Reprinted with permission.
Warren: On the Moral and Legal Status of Abortion
unknown planet and encounters a race of beings utterly sufficient. Neither do we need to insist that any one
unlike any he has ever seen or heard of. If he wants to of these criteria is necessary for personhood, although
be sure of behaving morally toward these beings, he has once again (1) and (2) look like fairly good candidates
to somehow decide whether they are people, and hence for necessary conditions, as does (3), if "activity" is
have full moral rights, or whether they are the sort of construed so as to include the activity of reasoning.
thing which he need not feel guilty about treating as, for All we need to claim, to demonstrate that a fetus
example, a source of food. is not a person, is that any being which satisfies none
How should he go about making this decision? of (1)—(5) is certainly not a person. I consider this
If he has some anthropological background, he might claim to be so obvious that I think anyone who denied
look for such things as religion, art, and the manufac- it, and claimed that a being which satisfied none of
turing of tools, weapons, or shelters, since these fac- (1)—(5) was a person all the same, would thereby dem-
tors have been used to distinguish our human from onstrate that he had no notion at all of what a person
our prehuman ancestors, in what seems to be closer is—perhaps because he had confused the concept of
to the moral than the genetic sense of "human." And a person with that of genetic humanity. If the oppon-
no doubt he would be right to consider the presence ents of abortion were to deny the appropriateness of
of such factors as good evidence that the alien beings these five criteria, I do not know what further argu-
were people, and morally human. It would, however, ments would convince them. We would probably have
be overly anthropocentric of him to take the absence to admit that our conceptual schemes were indeed
of these things as adequate evidence that they were irreconcilably different, and that our dispute could not
not, since we can imagine people who have progressed be settled objectively.
beyond, or evolved without ever developing, these cul- I do not expect this to happen, however, since
tural characteristics. I think that the concept of a person is one which is
I suggest that the traits which are most central to very nearly universal (to people), and that it is common
the concept of personhood, or humanity in the moral to both pro-abortionists and anti-abortionists, even
sense, are, very roughly the following: though neither group has fully realized the relevance
of this concept to the resolution of their dispute.
1. consciousness (of objects and events external and/ Furthermore, I think that on reflection even the anti-
or internal to the being), and in particular the cap- abortionists ought to agree not only that (1)—(5) are
acity to feel pain; central to the concept of personhood, but also that it is
2. reasoning (the developed capacity to solve new and a part of this concept that all and only people have full
relatively complex problems); moral rights. The concept of a person is in part a moral
3. self-motivated activity (activity which is relatively concept; once we have admitted that x is a person we
independent of either genetic or direct external have recognized, even if we have not agreed to respect,
control); x's right to be treated as a member of the moral com-
4. the capacity to communicate, by whatever means, munity. It is true that the claim that x is a human being
messages of an indefinite variety of types, that is, is more commonly voiced as part of an appeal to treat x
not just with an indefinite number of possible con- decently than is the claim that x is a person, but this is
tents, but on indefinitely many possible topics; either because "human being" is here used in the sense
5. the presence of self-concepts, and self-awareness, which implies personhood, or because the genetic and
either individual or racial, or both. moral sense of "human" have been confused.
Now if (1)—(5) are indeed the primary criteria of
Admittedly, there are apt to be a great many prob- personhood, then it is clear that genetic humanity is
lems involved in formulating precise definitions of neither necessary nor sufficient for establishing that an
these criteria, let alone in developing universally valid entity is a person. Some human beings are not people,
behavioural criteria for deciding when they apply. But and there may well be people who are not human
I will assume that both we and our explorer know ap- beings. A man or woman whose consciousness has
proximately what (1)—(5) mean, and that he is also able been permanently obliterated but who remains alive is
to determine whether or not they apply. How, then, a human being which is no longer a person; defective
should he use his findings to decide whether or not the human beings, with no appreciable mental capacity,
alien beings are people? We needn't suppose that an are not and presumably never will be people; and a
entity must have all of these attributes to be properly fetus is a human being which is not yet a person, and
considered a person; (1) and (2) alone may well be suf- which therefore cannot coherently be said to have full
ficient for personhood, and quite probably (1)—(3) are moral rights. Citizens of the next century should be
II 2 111111
prepared to recognize highly advanced, self-aware is not fully conscious, in the way that an infant of a few
robots or computers, should such be developed, and months is, and that it cannot reason, or communicate
intelligent inhabitants of other worlds, should such messages of indefinitely many sorts, does not engage in
be found, as people in the fullest sense, and to respect self-motivated activity and has no self-awareness. Thus,
their moral rights. But to ascribe full moral rights to an in the relevant respects, a fetus, even a fully developed
entity which is not a person is as absurd as to ascribe one, is considerably less personlike than is the average
moral obligations and responsibilities to such an entity. mature mammal, indeed the average fish. And I think
that a rational person must conclude that if the right to
3. Fetal Development and the Right to Life life of a fetus is to be based upon its resemblance to a
person, then it cannot be said to have any more right
Two problems arise in the application of these sugges- to life than, let us say, a newborn guppy (which also
tions for the definition of the moral community to the seems to be capable of feeling pain), and that a right of
determination of the precise moral status of a human that magnitude could never override a woman's right to
fetus. Given that the paradigm example of a person is obtain an abortion, at any stage of her pregnancy.
a normal adult human being, then (1) how like this There may, of course, be other arguments in favour
paradigm, in particular how far advanced since concep- of placing legal limits upon the stage of pregnancy in
tion, does a human being need to be before it begins to which an abortion may be performed. Given the rela-
have a right to life by virtue, not of being fully a person tive safety of the new techniques of artificially inducing
as of yet, but of being like a person?, and (2) to what labour during the third trimester, the danger to the
extent, if any, does the fact that a fetus has the potential woman's life or health is no longer such an argument.
for becoming a person endow it with some of the same Neither is the fact that people tend to respond to the
rights? Each of these questions requires some comment. thought of abortion in the later stages of pregnancy with
In answering the first question, we need not attempt emotional repulsion, since mere emotional responses
a detailed consideration of the moral rights of organisms cannot take the place of moral reasoning in determin-
which are not developed enough, aware enough, intel- ing what ought to be permitted. Nor, finally is the
ligent enough, etc., to be considered people, but which frequently heard argument that legalizing abortion, es-
resemble people in some respects. It does seem reason- pecially late in the pregnancy, may erode the level of re-
able to suggest that the more like a person, in the relevant spect for human life, leading, perhaps, to an increase in
respects, a being is, the stronger is the case for regarding unjustified euthanasia and other crimes. For this threat,
it as having a right to life, and indeed the stronger its if it is a threat, can be better met by educating people
right to life is. Thus we ought to take seriously the sug- to the kinds of moral distinctions which we are making
gestion that, insofar as "the human individual develops here than by limiting access to abortion (which limita-
biologically in a continuous fashion . . . the rights of a tion may in its disregard for the rights of women, be just
human person might develop in the same way."4 But we as damaging to the level of respect for human rights).
must keep in mind that the attributes which are relevant Thus, since the fact that even a fully developed
in determining whether or not an entity is enough like a fetus is not personlike enough to have any significant
person to be regarded as having some of the same moral right to life on the basis of its personlikeness shows
rights are no different from those which are relevant to that no legal restrictions upon the stage of pregnancy
determining whether or not it is fully a person—i.e., in which an abortion may be performed can be justified
are no different from (1)—(5)—and that being genetic- on the grounds that we should protect the rights of the
ally human, or having recognizable human facial and older fetus; and since there is no other apparent justifi-
other physical features, or detectable brain activity, or cation for such restrictions, we may conclude that they
the capacity to survive outside the uterus, are simply not are entirely unjustified. Whether or not it would be
among these relevant attributes. indecent (whatever that means) for a woman in her sev-
Thus it is clear that even though a seven- or eight- enth month to obtain an abortion just to avoid having
month fetus has features which make it apt to arouse to postpone a trip to Europe, it would not, in itself, be
in us almost the same powerful protective instinct as is immoral, and therefore it ought to be permitted.
commonly aroused by a small infant, nevertheless it is
not significantly more personlike than is a very small 4. Potential Personhood and the Right to Life
embryo. It is somewhat more personlike; it can appar-
ently feel and respond to pain, and it may even have a We have seen that a fetus does not resemble a person in
rudimentary form of consciousness, insofar as its brain any way which can support the claim that it has even
is quite active. Nevertheless, it seems safe to say that it some of the same rights. But what about its potential,
the fact that if nurtured and allowed to develop natur- carelessness, or even if he had done so deliberately,
ally it will very probably become a person? Doesn't that knowing the consequences. Regardless of how he got
alone give it at least some right to life? It is hard to deny captured, he is not morally obligated to remain in cap-
that the fact that an entity is a potential person is a tivity for any period of time for the sake of permitting
strong prima facie reason for not destroying it; but we any number of potential people to come into actual-
need not conclude from this that a potential person has ity, so great is the margin by which one actual person's
a right to life, by virtue of that potential. It may be that right to liberty outweighs whatever right to life even a
our feeling that it is better, other things being equal, not hundred thousand potential people have. And it seems
to destroy a potential person is better explained by the reasonable to conclude that the rights of a woman will
fact that potential people are still (felt to be) an invalu- outweigh by a similar margin whatever right to life a
able resource, not to be lightly squandered. Surely, if fetus may have by virtue of its potential personhood.
every speck of dust were a potential person, we would Thus, neither a fetus resemblance to a person, nor
be much less apt to conclude that every potential its potential for becoming a person provides any basis
person has a right to become actual. whatever for the claim that it has any significant right
Still, we do not need to insist that a potential to life. Consequently, a woman right to protect her
person has no right to life whatever. There may well health, happiness, freedom, and even her life,' by ter-
be something immoral, and not just imprudent, about minating an unwanted pregnancy, will always override
wantonly destroying potential people, when doing so whatever right to life it may be appropriate to ascribe
isn't necessary to protect anyone's rights. But even if to a fetus, even a fully developed one. And thus, in the
a potential person does have some prima facie right absence of any overwhelming social need for every pos-
to life, such a right could not possibly outweigh the sible child, the laws which restrict the right to obtain
right of a woman to obtain an abortion, since the rights an abortion, or limit the period of pregnancy during
of any actual person invariably outweigh those of any which an abortion may be performed, are a wholly un-
potential person, whenever the two conflict. Since this justified violation of a woman's most basic moral and
may not be immediately obvious in the case of a human constitutional rights.6
fetus, let us look at another case.
Suppose that our space explorer falls into the Postscript on Infanticide, 26 February 1982
hands of an alien culture, whose scientists decide to
create a few hundred thousand or more human beings, One of the most troubling objections to the argument
by breaking his body into its component cells, and presented in this article is that it may appear to justify not
using these to create fully developed human beings, only abortion but infanticide as well. A newborn infant
with, of course, his genetic code. We may imagine that is not a great deal more personlike than a nine-month
each of these newly created men will have all of the fetus, and thus it might seem that if late-term abortion is
original man's abilities, skills, knowledge, and so on, sometimes justified, then infanticide must also be some-
and also have an individual self-concept—in short, that times justified. Yet most people consider that infanticide
each of them will be a bona fide (though hardly unique) is a form of murder, and thus never justified.
person. Imagine that the whole project will take only While it is important to appreciate the emotional
seconds, and that its chances of success are extremely force of this objection, its logical force is far less than
high, and that our explorer knows all of this, and also it may seem at first glance. There are many reasons
knows that these people will be treated fairly. I main- why infanticide is much more difficult to justify than
tain that in such a situation he would have every right abortion, even though if my argument is correct neither
to escape if he could, and thus to deprive all of these constitutes the killing of a person. In this country, and
potential people of their potential lives; for his right to in this period of history, the deliberate killing of viable
life outweighs all of theirs together, in spite of the fact newborns is virtually never justified. This is in part be-
that they are all genetically human, all innocent, and all cause neonates are so very close to being persons that to
have a very high probability of becoming people very kill them requires a very strong moral justification—as
soon, if only he refrains from acting. does the killing of dolphins, whales, chimpanzees, and
Indeed, I think he would have a right to escape other highly personlike creatures. It is certainly wrong
even if it were not his life which the alien scientists to kill such beings just for the sake of convenience, or
planned to take, but only a year of his freedom, or, financial profit, or "sport."
indeed, only a day. Nor would he be obligated to Another reason why infanticide is usually wrong,
stay if he had gotten captured (thus bringing all these in our society, is that if the newborn's parents do not
people-potentials into existence) because of his own want it, or are unable to care for it, there are (in most
.11 II .111 111111
cases) people who are able and eager to adopt it and when the threat cannot be removed without killing the
to provide a good home for it. Many people wait years animal, the person's right to self-protection outweighs
for the opportunity to adopt a child, and some are the desires of those who would prefer that the animal not
unable to do so even though there is every reason to be harmed. Thus, while the moment of birth may not
believe that they would be good parents. The need- mark any sharp discontinuity in the degree to which an
less destruction of a viable infant inevitably deprives infant possesses a right to life, it does mark the end of the
some person or persons of a source of great pleasure mother's absolute right to determine its fate. Indeed, if
and satisfaction, perhaps severely impoverishing their and when a late-term abortion could be safely performed
lives. Furthermore, even if an infant is considered to without killing the fetus, she would have no absolute
be unadoptable (e.g., because of some extremely severe right to insist on its death (e.g., if others wish to adopt it
mental or physical handicap) it is still wrong in most or pay for its care), for the same reason that she does not
cases to kill it. For most of us value the lives of infants, have a right to insist that a viable infant be killed.
and would prefer to pay taxes to support orphanages It remains true that according to my argument neither
and state institutions for the handicapped rather than abortion nor the killing of neonates is properly considered
to allow unwanted infants to be killed. So long as most a form of murder. Perhaps it is understandable that the law
people feel this way, and so long as our society can should classify infanticide as murder or homicide, since
afford to provide care for infants which are unwanted there is no other existing legal category which adequately
or which have special needs that preclude home care, or conveniently expresses the force of our society's disap-
it is wrong to destroy any infant which has a chance of proval of this action. But the moral distinction remains,
living a reasonably satisfactory life. and it has several important consequences.
If these arguments show that infanticide is wrong, In the first place, it implies that when an infant is
at least in this society, then why don't they also show that born into a society which—unlike ours—is so impover-
late-term abortion is wrong? After all, third-trimester ished that it simply cannot care for it adequately without
fetuses are also highly personlike, and many people endangering the survival of existing persons, killing it
value them and would much prefer that they be pre- or allowing it to die is not necessarily wrong—provided
served, even at some cost to themselves. As a potential that there is no other society which is willing and able
source of pleasure to some family, a viable fetus is just as to provide such care. Most human societies, from those
valuable as a viable infant. But there is an obvious and at the hunting and gathering stage of economic de-
crucial difference between the two cases once the infant velopment to the highly civilized Greeks and Romans,
is born, its continued life cannot (except, perhaps, in have permitted the practice of infanticide under such
very exceptional cases) pose any serious threat to the unfortunate circumstances, and I would argue that it
woman's life or health, since she is free to put it up for shows a serious lack of understanding to condemn
adoption, or, where this is impossible, to place it in a them as morally backward for this reason alone.
state-supported institution. While she might prefer that In the second place, the argument implies that when
it die, rather than being raised by others, it is not clear an infant is born with such severe physical anomalies
that such a preference would constitute a right on her that its life would predictably be a very short and/or very
part. True, she may suffer greatly from the knowledge miserable one, even with the most heroic of medical treat-
that her child will be thrown into the lottery of the ment, and where its parents do not choose to bear the
adoption system, and that she will be unable to ensure often crushing emotional, financial, and other burdens
its well-being, or even to know whether it is healthy, attendant upon the artificial prolongation of such a tragic
happy, doing well in school, etc.: for the law generally life, it is not morally wrong to cease or withhold treat-
does not permit natural parents to remain in contact ment, thus allowing the infant a painless death. It is wrong
with their children, once they are adopted by another (and sometimes a form of murder) to practice involuntary
family. But there are surely better ways of dealing with euthanasia on persons, since they have the right to decide
these problems than by permitting infanticide in such for themselves whether or not they wish to continue to
cases. (It might help, for instance, if the natural parents live. But terminally ill neonates cannot make this decision
of adopted children could at least receive some infor- for themselves, and thus it is incumbent upon responsible
mation about their progress, without necessarily being persons to make the decision for them, as best they can.
informed of the identity of the adopting family.) The mistaken belief that infanticide is always tantamount
In contrast, a pregnant woman's right to protect to murder is responsible for a great deal of unnecessary
her own life and health clearly outweighs other people's suffering, not just on the part of infants which are made
desire that the fetus be preserved—just as, when a per- to endure needlessly prolonged and painful deaths, but
song life or limb is threatened by some wild animal, and also on the part of parents, nurses, and other involved
persons, who must watch infants suffering needlessly, other moral issues as well, issues not just about infanti-
helpless to end that suffering in the most humane way. cide and euthanasia, but also about the moral rights of
I am well await that these conclusions, however women and of non-human animals. It is a philosopher's
modest and reasonable they may seem to some people, task to criticize mistaken beliefs which stand in the way
strike other people as morally monstrous, and that of moral understanding, even when—perhaps espe-
some people might even prefer to abandon their previ- cially when—those beliefs are popular and widespread.
ous support for women's right to abortion rather than The belief that moral strictures against killing should
accept a theory which leads to such conclusions about apply equally to all genetically human entities, and only
infanticide. But all that these facts show is that abor- to genetically human entities, is such an error. The over-
tion is not an isolated moral issue; to fully understand coming of this error will undoubtedly require long and
the moral status of abortion we may have to reconsider often painful struggle; but it must be done.
Notes
1. 0 f course, the principle that it is (always) wrong to 4. Thomas L. Hayes, (1967), "A Biological View," in
kill innocent human beings is in need of many other Commonweal 85 (17 March): 677-78; quoted by
modifications, e.g., that it may be permissible to do so to Daniel Callahan, in Abortion: Law, Choice and Morality
save a greater number of other innocent human beings, (1970; Macmillan & Co.: London).
but we may safely ignore these complications here. 5. That is, insofar as the death rate, for the woman, is
2. John Noonan, (1968), "Deciding Who Is Human," in higher for childbirth than for early abortion.
Natural Law Forum, 13: 135. 6. My thanks to the following people, who were kind
3. From here on, we will use "human" to mean genetically enough to read and criticize an earlier version of this
human, since the moral sense seems closely connected to, and paper: Herbert Gold, Gene Glass, Anne Lauterbach,
perhaps derived from, the assumption that genetic humanity Judith Thomson, Mary Mothersill, and Timothy
is sufficient for membership in the moral community. Binkley.
A Defence of Abortion' they are. Arguments of this form are sometimes called
"slippery slope arguments"—the phrase is perhaps
Judith Jarvis Thomson self-explanatory—and it is dismaying that opponents
of abortion rely on them so heavily and uncritically.
Most opposition to abortion relies on the premise that I am inclined to agree, however, that the pros-
the fetus is a human being, a person, from the moment pects for "drawing a line" in the development of the
of conception. The premise is argued for, but, as I fetus look dim. I am inclined to think also that we
think, not well. Take, for example, the most common shall probably have to agree that the fetus has already
argument. We are asked to notice that the development become a human person well before birth. Indeed, it
of a human being from conception through birth into comes as a surprise when one first learns how early in
childhood is continuous; then it is said that to draw its life it begins to acquire human characteristics. By the
a line, to choose a point in this development and say tenth week, for example, it already has a face, arms and
"before this point the thing is not a person, after this legs, fingers and toes; it has internal organs, and brain
point it is a person" is to make an arbitrary choice, a activity is detectable.2 On the other hand, I think that
choice for which in the nature of things no good reason the premise is false, that the fetus is not a person from
can be given. It is concluded that the fetus is, or anyway the moment of conception. A newly fertilized ovum, a
that we had better say it is, a person from the moment newly implanted clump of cells, is no more a person
of conception. But this conclusion does not follow. than an acorn is an oak tree. But I shall not discuss
Similar things might be said about the development any of this. For it seems to me to be of great interest
of an acorn into an oak tree, and it does not follow to ask what happens if, for the sake of argument, we
that acorns are oak trees, or that we had better say allow the premise. How, precisely, are we supposed to
Source: From Philosophy and Public Affairs 1 (1971, (1), 47-66. Reproduced with permission of Blackwell Publishing Ltd.
I II II ,21 111111
get from there to the conclusion that abortion is mor- to life, and violinists are persons. Granted you have a
ally impermissible? Opponents of abortion commonly right to decide what happens in and to your body, but
spend most of their time establishing that the fetus a persons right to life outweighs your right to decide
is a person, and hardly any time explaining the step what happens in and to your body. So you cannot ever
from there to the impermissibility of abortion. Perhaps be unplugged from him." I imagine you would regard
they think the step too simple and obvious to require this as outrageous, which suggests that something
much comment. Or perhaps instead they are simply really is wrong with that plausible-sounding argument
being economical in argument. Many of those who I mentioned a moment ago.
defend abortion rely on the premise that the fetus is In this case, of course, you were kidnapped; you
not a person, but only a bit of tissue that will become a didn't volunteer for the operation that plugged the
person at birth; and why pay out more arguments than violinist into your kidneys. Can those who oppose
you have to? Whatever the explanation, I suggest that abortion on the ground I mentioned make an excep-
the step they take is neither easy nor obvious, that it tion for a pregnancy due to rape? Certainly. They can
calls for closer examination than it is commonly given, say that persons have a right to life only if they didn't
and that when we do give it this closer examination we come into existence because of rape; or they can say
shall feel inclined to reject it. that all persons have a right to life, but that some
I propose, then, that we grant that the fetus is a have less of a right to life than others, in particular,
person from the moment of conception. How does the that those who came into existence because of rape
argument go from here? Something like this, I take have less. But these statements have a rather un-
it. Every person has a right to life. So the fetus has a pleasant sound. Surely the question of whether you
right to life. No doubt the mother has a right to decide have a right to life at all, or how much of it you have,
what shall happen in and to her body; everyone would shouldn't turn on the question of whether or not you
grant that. But surely a person's right to life is stronger are the product of a rape. And in fact the people who
and more stringent than the mother's right to decide oppose abortion on the ground I mentioned do not
what happens in and to her body, and so outweighs it. make this distinction, and hence do not make an ex-
So the fetus may not be killed; an abortion may not ception in case of rape.
be performed. Nor do they make an exception for a case in
It sounds plausible. But now let me ask you to which the mother has to spend the nine months of her
imagine this. You wake up in the morning and find pregnancy in bed. They would agree that would be a
yourself back to back in bed with an unconscious great pity, and hard on the mother, but all the same, all
violinist. A famous unconscious violinist. He has been persons have a right to life, the fetus is a person, and
found to have a fatal kidney ailment, and the Society so on. I suspect, in fact, that they would not make an
of Music Lovers has canvassed all the available medical exception for a case in which, miraculously enough,
records and found that you alone have the right blood the pregnancy went on for nine years, or even the rest
type to help. They have therefore kidnapped you, and of the mother's life.
last night the violinist's circulatory system was plugged Some won't even make an exception for a case
into yours, so that your kidneys can be used to extract in which continuation of the pregnancy is likely to
poisons from his blood as well as your own. The direc- shorten the mother's life; they regard abortion as im-
tor of the hospital now tells you, "Look, we're sorry permissible even to save the mother's life. Such cases
the Society of Music Lovers did this to you—we would are nowadays very rare, and many opponents of abor-
never have permitted it if we had known. But still, they tion do not accept this extreme view. All the same, it is
did it, and the violinist now is plugged into you. To a good place to begin: a number of points of interest
unplug you would be to kill him. But never mind, it's come out in respect to it.
only for nine months. By then he will have recovered
from his ailment, and can safely be unplugged from 1
you." Is it morally incumbent on you to accede to this
situation? No doubt it would be very nice of you if you Let us call the view that abortion is impermissible even
did, a great kindness. But do you have to accede to it? to save the mother's life "the extreme view" I want to
What if it were not nine months, but nine years? Or suggest first that it does not issue from the argument I
longer still? What if the director of the hospital says, mentioned earlier without the addition of some fairly
'Tough luck, I agree, but you've now got to stay in bed, powerful premises. Suppose a woman has become preg-
with the violinist plugged into you, for the rest of your nant, and now learns that she has a cardiac condition
life. Because remember this. All persons have a right such that she will die if she carries the baby to term.
Thomson: A Defence of Abortion
What may be done for her? The fetus, being a person, If anything in the world is true, it is that you do not
has a right to life, but as the mother is a person too, so commit murder, you do not do what is impermissible,
has she a right to life. Presumably they have an equal if you reach around to your back and unplug yourself
right to life. How is it supposed to come out that an from that violinist to save your life.
abortion may not be performed? If mother and child The main focus of attention in writings on abor-
have an equal right to life, shouldn't we perhaps flip a tion has been on what a third party may or may not do
coin? Or should we add to the motherb right to life her in answer to a request from a woman for an abortion.
right to decide what happens in and to her body, which This is in a way understandable. Things being as they
everybody seems to be ready to grant—the sum of her are, there isn't much a woman can safely do to abort
rights now outweighing the fetusb right to life? The most herself. So the question asked is what a third party may
familiar argument here is the following. We are told that do, and what the mother may do, if it is mentioned at
performing the abortion would be directly killing' the all, is deduced, almost as an afterthought, from what
child, whereas doing nothing would not be killing the it is concluded that third parties may do. But it seems
mother, but only letting her die. Moreover, in killing to me that to treat the matter in this way is to refuse to
the child, one would be killing an innocent person, for grant to the mother that very status of person which
the child has committed no crime, and is not aiming at is so firmly insisted on for the fetus. For we cannot
his motherb death. And then there are a variety of ways simply read off what a person may do from what a third
in which this might be continued. (1) But as directly party may do. Suppose you find yourself trapped in a
killing an innocent person is always and absolutely tiny house with a growing child. I mean a very tiny
impermissible, an abortion may not be performed. Or house, and a rapidly growing child—you are already
(2) as directly killing an innocent person is murder, up against the wall of the house and in a few minutes
and murder is always and absolutely impermissible, an you'll be crushed to death. The child on the other hand
abortion may not be performed.' Or (3) as ones duty to won't be crushed to death; if nothing is done to stop
refrain from directly killing an innocent person is more him from growing he'll be hurt, but in the end he'll
stringent than ones duty to keep a person from dying, simply burst open the house and walk out a free man.
an abortion may not be performed. Or (4) if ones only Now I could well understand it if a bystander were to
options are directly killing an innocent person or letting say, "There's nothing we can do for you. We cannot
a person die, one must prefer letting the person die, and choose between your life and his, we cannot be the
thus an abortion may not be performed.' ones to decide who is to live, we cannot intervene."
Some people seem to have thought that these are But it cannot be concluded that you too can do noth-
not further premises which must be added if the con- ing, that you cannot attack it to save your life. However
clusion is to be reached, but that they follow from the innocent the child may be, you do not have to wait pas-
very fact that an innocent person has a right to life.' sively while it crushes you to death. Perhaps a pregnant
But this seems to me to be a mistake, and perhaps the woman is vaguely felt to have the status of house, to
simplest way to show this is to bring out that while which we don't allow the right of self-defence. But if
we must certainly grant that innocent persons have a the woman houses the child, it should be remembered
right to life, the theses in (1) through (4) are all false. that she is a person who houses it.
Take (2), for example. If directly killing an innocent I should perhaps stop to say explicitly that I am
person is murder, and thus is impermissible, then the not claiming that people have a right to do anything
motherb directly killing the innocent person inside whatever to save their lives. I think, rather, that there
her is murder, and thus is impermissible. But it cannot are drastic limits to the right of self-defence. If someone
seriously be thought to be murder if the mother per- threatens you with death unless you torture someone
forms an abortion on herself to save her life. It cannot else to death, I think you have not the right, even to
seriously be said that she must refrain, that she must sit save your life, to do so. But the case under considera-
passively by and wait for her death. Let us look again at tion here is very different. In our case there are only
the case of you and the violinist. There you are, in bed two people involved, one whose life is threatened, and
with the violinist, and the director of the hospital says one who threatens it. Both are innocent: the one who
to you, "It's all most distressing, and I deeply sympa- is threatened is not threatened because of any fault, the
thize, but you see this is putting an additional strain on one who threatens does not threaten because of any
your kidneys, and you'll be dead within the month. But fault. For this reason we may feel that we bystanders
you have to stay where you are all the same. Because cannot intervene. But the person threatened can. In
unplugging you would be directly killing an innocent sum, a woman surely can defend her life against the
violinist, and that's murder, and that's impermissible." threat to it posed by the unborn child, even if doing so
II iY 1111111
involves its death. And this shows not merely that the request that he perform an abortion to save her life, but
theses in (1) through (4) are false; it shows also that the only that he may.
extreme view of abortion is false, and so we need not I suppose that in some views of human life the
canvass any other possible ways of arriving at it from mother's body is only on loan to her, the loan not being
the argument I mentioned at the outset. one which gives her any prior claim to it. One who
held this view might well think it impartiality to say "I
2 cannot choose." But I shall simply ignore this possibil-
ity. My own view is that if a human being has any just,
The extreme view could of course be weakened to say prior claim to anything at all, he has a just, prior claim
that while abortion is permissible to save the mother's to his own body. And perhaps this needn't be argued
life, it may not be performed by a third party, but only for here anyway, since, as I mentioned, the arguments
by the mother herself. But this cannot be right either. against abortion we are looking at do grant that the
For what we have to keep in mind is that the mother woman has a right to decide what happens in and to
and the unborn child are not like two tenants in a small her body.
house which has, by an unfortunate mistake, been But although they do grant it, I have tried to show
rented to both: the mother owns the house. The fact that they do not take seriously what is done in granting
that she does adds to the offensiveness of deducing that it. I suggest the same thing will reappear even more
the mother can do nothing from the supposition that clearly when we turn away from cases in which the
third parties can do nothing. But it does more than this: mother's life is at stake, and attend, as I propose we
it casts a bright light on the supposition that third par- now do, to the vastly more common cases in which a
ties can do nothing. Certainly it lets us see that a third woman wants an abortion for some less weighty reason
party who says "I cannot choose between you" is fool- than preserving her own life.
ing himself if he thinks this is impartiality. If Jones has
found and fastened on a certain coat, which he needs 3
to keep him from freezing, but which Smith also needs
to keep him from freezing, then it is not impartiality Where the mother's life is not at stake, the argument I
that says "I cannot choose between you" when Smith mentioned at the outset seems to have a much stronger
owns the coat. Women have said again and again "This pull. "Everyone has a right to life, so the unborn person
body is my body!" and they have reason to feel angry, has a right to life." And isn't the child right to life
reason to feel that it has been like shouting into the weightier than anything other than the mother's own
wind. Smith, after all, is hardly likely to bless us if we right to life, which she might put forward as ground
say to him, "Of course it's your coat, anybody would for an abortion?
grant that it is. But no one may choose between you This argument treats the right to life as if it were
and Jones who is to have it." unproblematic. It is not, and this seems to me to be
We should really ask what it is that says "no one precisely the source of the mistake.
may choose" in the face of the fact that the body that For we should now, at long last, ask what it comes
houses the child is the mother's body. It may be simply to, to have a right to life. In some views having a right
a failure to appreciate this fact. But it may be something to life includes having a right to be given at least the
more interesting, namely the sense that one has a right bare minimum one needs for continued life. But sup-
to ref use to lay hands on people, even where it would pose that what in fact is the bare minimum a man needs
be just and fair to do so, even where justice seems to re- for continued life is something he has no right at all to
quire that somebody do so. Thus justice might call for be given? If I am sick unto death, and the only thing
somebody to get Smith's coat back from Jones, and yet that will save my life is the touch of Henry Fonda's
you have a right to ref use to be the one to lay hands on cool hand on my fevered brow, then all the same,
Jones, a right to refuse to do physical violence to him. I have no right to be given the touch of Henry Fonda's
This, I think, must be granted. But then what should cool hand on my fevered brow. It would be frightfully
be said is not "no one may choose," but only "I cannot nice of him to fly in from the West Coast to provide
choose," and indeed not even this, but "I will not act," it. It would be less nice, though no doubt well meant,
leaving it open that somebody else can or should, and if my friends flew out to the West Coast and carried
in particular that anyone in a position of authority, Henry Fonda back with them. But I have no right at
with the job of securing people's rights, both can and all against anybody that he should do this for me. Or
should. So this is no difficulty. I have not been arguing again, to return to the story I told earlier, the fact that
that any given third party must accede to the mother's for continued life that violinist needs the continued
use of your kidneys does not establish that he has a does not guarantee having either a right to be given the
right to be given the continued use of your kidneys. He use of or a right to be allowed continued use of another
certainly has no right against you that you should give person's body—even if one needs it for life itself. So the
him continued use of your kidneys. For nobody has right to life will not serve the opponents of abortion in
any right to use your kidneys unless you give him such the very simple and clear way in which they seem to
a right; and nobody has the right against you that you have thought it would.
shall give him this right—if you do allow him to go on
using your kidneys, this is a kindness on your part, and 4
not something he can claim from you as his due. Nor
has he any right against anybody else that they should There is another way to bring out the difficulty. In the
give him continued use of your kidneys. Certainly he most ordinary sort of case, to deprive someone of what
had no right against the Society of Music Lovers that he has a right to is to treat him unjustly. Suppose a
they should plug him into you in the first place. And if boy and his small brother are jointly given a box of
you now start to unplug yourself, having learned that chocolates for Christmas. If the older boy takes the box
you will otherwise have to spend nine years in bed with and refuses to give his brother any of the chocolates, he
him, there is nobody in the world who must try to pre- is unjust to him, for the brother has been given a right
vent you, in order to see to it that he is given something to half of them. But suppose that, having learned that
he has a right to be given. otherwise it means nine years in bed with that violinist,
Some people are rather stricter about the right you unplug yourself from him. You surely are not being
to life. In their view, it does not include the right to unjust to him, for you gave him no right to use your
be given anything, but amounts to, and only to, the kidneys, and no one else can have given him any such
right not to be killed by anybody. But here a related right. But we have to notice that in unplugging your-
difficulty arises. If everybody is to refrain from lulling self, you are killing him; and violinists, like everybody
that violinist, then everybody must refrain from doing else, have a right to life, and thus in the view we were
a great many different sorts of things. Everybody must considering just now, the right not to be killed. So here
refrain from slitting his throat, everybody must refrain you do what he supposedly has a right you shall not
from shooting him—and everybody must refrain from do, but you do not act unjustly to him in doing it.
unplugging you from him. But does he have a right The emendation which may be made at this point
against everybody that they shall refrain from unplug- is this the right to life consists not in the right not to be
ging you from him? To refrain from doing this is to killed, but rather in the right not to be killed unjustly.
allow him to continue to use your kidneys. It could This runs a risk of circularity, but never mind: it would
be argued that he has a right against us that we should enable us to square the fact that the violinist has a right
allow him to continue to use your kidneys. That is, to life with the fact that you do not act unjustly toward
while he had no right against us that we should give him in unplugging yourself, thereby killing him. For if
him the use of your kidneys, it might be argued that you do not kill him unjustly, you do not violate his right
he anyway has a right against us that we shall not now to life, and so it is no wonder you do him no injustice.
intervene and deprive him of the use of your kidneys. But if this emendation is accepted, the gap in the
I shall come back to third-party interventions later. But argument against abortion stares us plainly in the face:
certainly the violinist has no right against you that you it is by no means enough to show that the fetus is a
shall allow him to continue to use your kidneys. As I person, and to remind us that all persons have a right
said, if you do allow him to use them, it is a kindness to life—we need to be shown also that killing the fetus
on your part, and not something you owe him. violates its right to life, i.e., that abortion is unjust kill-
The difficulty I point to here is not peculiar to ing. And is it?
the right to life. It reappears in connection with all I suppose we may take it as a datum that in a case
the other natural rights, and it is something which an of pregnancy due to rape the mother has not given the
adequate account of rights must deal with. For present unborn person a right to the use of her body for food
purposes it is enough just to draw attention to it. But I and shelter. Indeed, in what pregnancy could it be sup-
would stress that I am not arguing that people do not posed that the mother has given the unborn person
have a right to life—quite to the contrary, it seems to such a right? It is not as if there were unborn persons
me that the primary control we must place on the ac- drifting about the world, to whom a woman who wants
ceptability of an account of rights is that it should turn a child says "I invite you in."
out in that account to be a truth that all persons have a But it might be argued that there are other ways
right to life. I am arguing only that having a right to life one can have acquired a right to the use of another
E II II .- 311 IIIIJl
person's body than by having been invited to use it by burglar got in only because of a defect in the bars. It re-
that person. Suppose a woman voluntarily indulges mains equally absurd if we imagine it is not a burglar
in intercourse, knowing of the chance it will issue in who climbs in, but an innocent person who blunders
pregnancy, and then she does become pregnant; is she or falls in. Again, suppose it were like this: people-
not in part responsible for the presence, in fact the very seeds drift about in the air like pollen, and if you open
existence, of the unborn person inside her? No doubt your windows, one may drift in and take root in your
she did not invite it in. But doesn't her partial respon- carpets or upholstery. You don't want children, so you
sibility for its being there itself give it a right to the use fix up your windows with fine mesh screens, the very
of her body?' If so, then her aborting it would be more best you can buy. As can happen, however, and on very,
like the boy's taking away the chocolates, and less like very rare occasions does happen, one of the screens is
your unplugging yourself from the violinist—doing so defective; and a seed drifts in and takes root. Does the
would be depriving it of what it does have a right to, person-plant who now develops have a right to the use
and thus would be doing it an injustice. of your house? Surely not—despite the fact that you
And then, too, it might be asked whether or not voluntarily opened your windows, you knowingly kept
she can kill it even to save her own life: if she voluntar- carpets and upholstered furniture, and you knew that
ily called it into existence, how can she now kill it, even screens were sometimes defective. Someone may argue
in self-defence? that you are responsible for its rooting, that it does have
The first thing to be said about this is that it is a right to your house, because after all you could have
something new. Opponents of abortion have been so lived out your life with bare floors and furniture, or
concerned to make out the independence of the fetus, with sealed windows and doors. But this wont do—for
in order to establish that it has a right to life, just as by the same token anyone can avoid a pregnancy due
its mother does, that they have tended to overlook the to rape by having a hysterectomy, or anyway by never
possible support they might gain from making out that leaving home without a (reliable!) army.
the fetus is dependent on the mother, in order to estab- It seems to me that the argument we are looking at
lish that she has a special kind of responsibility for it, a can establish at most that there are some cases in which
responsibility that gives it rights against her which are the unborn person has a right to the use of its mother's
not possessed by any independent person—such as an body, and therefore some cases in which abortion is
ailing violinist who is a stranger to her. unjust killing. There is room for much discussion and
On the other hand, this argument would give the argument as to precisely which, if any. But I think we
unborn person a right to its mother's body only if her should side-step this issue and leave it open, for at any
pregnancy resulted from a voluntary act, undertaken in rate the argument certainly does not establish that all
full knowledge of the chance a pregnancy might result abortion is unjust killing.
from it. It would leave out entirely the unborn person
whose existence is due to rape. Pending the availability
5
of some further argument, then, we would be left with
the conclusion that unborn persons whose existence is There is room for yet another argument here, however.
due to rape have no right to the use of their mothers' We surely must all grant that there may be cases in
bodies, and thus that aborting them is not depriving which it would be morally indecent to detach a person
them of anything they have a right to and hence is not from your body at the cost of his life. Suppose you
unjust killing. learn that what the violinist needs is not nine years of
And we should also notice that it is not at all your life, but only one hour all you need do to save his
plain that this argument really does go even as far as life is to spend one hour in that bed with him. Suppose
it purports to. For there are cases and cases, and the also that letting him use your kidneys for that one hour
details make a difference. If the room is stuffy, and I would not affect your health in the slightest. Admit-
therefore open a window to air it, and a burglar climbs tedly you were kidnapped. Admittedly you did not give
in, it would be absurd to say, "Ah, now he can stay, anyone permission to plug him into you. Nevertheless
she's given him a right to the use of her house—for she it seems to me plain you ought to allow him to use your
is partially responsible for his presence there, having kidneys for that hour—it would be indecent to refuse.
voluntarily done what enabled him to get in, in full Again, suppose pregnancy lasted only an hour,
knowledge that there are such things as burglars, and and constituted no threat to life or health. And sup-
that burglars burgle." It would be still more absurd to pose that a woman becomes pregnant as a result of
say this if I had had bars installed outside my windows, rape. Admittedly she did not voluntarily do anything
precisely to prevent burglars from getting in, and a to bring about the existence of a child. Admittedly she
did nothing at all which would give the unborn person that in this case she has a right to the touch of his hand
a right to the use of her body. All the same it might well on her brow, and so it would be an injustice in him
be said, as in the newly emended violinist story, that to refuse"? So that I have a right to it when it is easy
she ought to allow it to remain for that hour—that it for him to provide it, though no right when it's hard?
would be indecent in her to refuse. It's rather a shocking idea that anyone's rights should
Now some people are inclined to use the term fade away and disappear as it gets harder and harder to
"right" in such a way that it follows from the fact that accord them to him.
you ought to allow a person to use your body for the So my own view is that even though you ought
hour he needs, that he has a right to use your body for to let the violinist use your kidneys for the one hour
the hour he needs, even though he has not been given he needs, we should not conclude that he has a right
that right by any person or act. They may say that it to do so—we should say that if you refuse, you are,
follows also that if you refuse, you act unjustly toward like the boy who owns all the chocolates and will
him. This use of the term is perhaps so common that it give none away, self-centred and callous, indecent in
cannot be called wrong; nevertheless it seems to me to fact, but not unjust. And similarly, that even suppos-
be an unfortunate loosening of what we would do better ing a case in which a woman pregnant due to rape
to keep a tight rein on. Suppose that box of chocolates ought to allow the unborn person to use her body
I mentioned earlier had not been given to both boys for the hour he needs, we should not conclude that
jointly, but was given only to the older boy. There he he has a right to do so; we should conclude that she
sits, stolidly eating his way through the box, his small is self-centred, callous, indecent, but not unjust, if
brother watching enviously. Here we are likely to say, she refuses. The complaints are no less grave; they
"You ought not to be so mean. You ought to give your are just different. However, there is no need to insist
brother some of those chocolates." on this point. If anyone does wish to deduce "he has
My own view is that it just does not follow from a right" from "you ought," then all the same he must
the truth of this that the brother has any right to any of surely grant that there are cases in which it is not
the chocolates. If the boy refuses to give his brother any, morally required of you that you allow that violin-
he is greedy, stingy, callous—but not unjust. I suppose ist to use your kidneys, and in which he does not
that the people I have in mind will say it does follow have a right to use them, and in which you do not
that the brother has a right to some of the chocolates, do him an injustice if you refuse. And so also for
and thus that the boy does act unjustly if he refuses mother and unborn child. Except in such cases as
to give his brother any. But the effect of saying this is the unborn person has a right to demand it—and
to obscure what we should keep distinct, namely the we were leaving open the possibility that there may
difference between the boy's refusal in this case and be such cases—nobody is morally required to make
the boy's refusal in the earlier case, in which the box large sacrifices, of health, of all other interests and
was given to both boys jointly, and in which the small concerns, of all other duties and commitments, for
brother thus had what was from any point of view clear nine years, or even for nine months, in order to keep
title to half. another person alive.
A further objection to so using the term "right"
that from the fact that A ought to do a thing for B, it
6
follows that B has a right against A that A do it for him,
is that it is going to make the question of whether or We have in fact to distinguish between two kinds of
not a man has a right to a thing turn on how easy it Samaritan: the Good Samaritan and what we might call
is to provide him with it; and this seems not merely the Minimally Decent Samaritan. The story of the Good
unfortunate, but morally unacceptable. Take the case Samaritan, you will remember, goes like this:
of Henry Fonda again. I said earlier that I had no right
to the touch of his cool hand on my fevered brow, even A certain man went down from Jerusalem to
though I needed it to save my life. I said it would be Jericho, and fell among thieves, which stripped
frightfully nice of him to fly in from the West Coast him of his raiment, and wounded him, and de-
to provide me with it, but that I had no right against parted, leaving him half dead.
him that he should do so. But suppose he isn't on the And by chance there came down a certain
West Coast. Suppose he has only to walk across the priest that way; and when he saw him, he passed
room, place a hand briefly on my brow—and lo, my by on the other side.
life is saved. Then surely he ought to do it, it would be And likewise a Levite, when he was at the
indecent to refuse. Is it to be said, "Ah, well, it follows place, came and looked on him, and passed by
M III II 11 II 311 111111
on the other side. But a certain Samaritan, as he respect to abortion, but it is worth drawing attention
journeyed, came where he was; and when he saw to the fact that in no state in this country is any man
him he had compassion on him. compelled by law to be even a Minimally Decent
And went to him, and bound up his wounds, Samaritan to any person; there is no law under
pouring in oil and wine, and set him on his own which charges could be brought against the 38 who
beast, and brought him to an inn, and took care stood by while Kitty Genovese died. By contrast, in
of him. most states in this country women are compelled
And on the morrow, when he departed, he by law to be not merely Minimally Decent Samar-
took out two pence, and gave them to the host, itans but Good Samaritans to unborn persons inside
and said unto him, "Take care of him; and whatso- them. This doesn't by itself settle anything one way
ever thou spendest more, when I come again, I will or the other, because it may well be argued that
repay thee." (Luke 10: 30-35) there should be laws in this country—as there are
in many European countries—compelling at least
The Good Samaritan went out of his way, at some Minimally Decent Samaritanism.8 But it does show
cost to himself, to help one in need of it. We are not that there is a gross injustice in the existing state of
told what the options were, that is, whether or not the the law. And it shows also that the groups currently
priest and the Levite could have helped by doing less working against liberalization of abortion laws, in
than the Good Samaritan did, but assuming they could fact working toward having it declared unconstitu-
have, then the fact they did nothing at all shows they tional for a state to permit abortion, had better start
were not even Minimally Decent Samaritans, not be- working for the adoption of Good Samaritan laws
cause they were not Samaritans, but because they were generally, or earn the charge that they are acting in
not even minimally decent. bad faith.
These things are a matter of degree, of course, but I should think, myself, that Minimally Decent
there is a difference, and it comes out perhaps most Samaritan laws would be one thing, Good Samaritan
clearly in the story of Kitty Genovese, who, as you will laws quite another, and in fact highly improper. But
remember, was murdered while 38 people watched or we are not here concerned with the law. What we
listened, and did nothing at all to help her. A Good should ask is not whether anybody should be com-
Samaritan would have rushed out to give direct assist- pelled by law to be a Good Samaritan, but whether
ance against the murderer. Or perhaps we had better we must accede to a situation in which somebody
allow that it would have been a Splendid Samaritan is being compelled—by nature, perhaps—to be a
who did this, on the ground that it would have involved Good Samaritan. We have, in other words, to look
a risk of death for himself. But the 38 not only did not now at third-party interventions. I have been argu-
do this, they did not even trouble to pick up a phone to ing that no person is morally required to make
call the police. Minimally Decent Samaritanism would large sacrifices to sustain the life of another who
call for doing at least that, and their not having done it has no right to demand them, and this even where
was monstrous. the sacrifices do not include life itself; we are not
After telling the story of the Good Samaritan, Jesus morally required to be Good Samaritans or anyway
said, "Go, and do thou likewise." Perhaps he meant that Very Good Samaritans to one another. But what if a
we are morally required to act as the Good Samaritan man cannot extricate himself from such a situation?
did. Perhaps he was urging people to do more than is What if he appeals to us to extricate him? It seems
morally required of them. At all events it seems plain to me plain that there are cases in which we can,
that it was not morally required of any of the 38 that he cases in which a Good Samaritan would extricate
rush out to give direct assistance at the risk of his own him. There you are, you were kidnapped, and nine
life, and that it is not morally required of anyone that years in bed with that violinist lie ahead of you.
he give long stretches of his life—nine years or nine You have your own life to lead. You are sorry, but
months—to sustaining the life of a person who has no you simply cannot see giving up so much of your
special right (we were leaving open the possibility of life to the sustaining of his. You cannot extricate
this) to demand it. yourself, and ask us to do so. I should have thought
Indeed, with one rather striking class of excep- that—in light of his having no right to the use of
tions, no one in any country in the world is legally your body—it was obvious that we do not have to
required to do anywhere near as much as this for accede to your being forced to give up so much. We
anyone else. The class of exceptions is obvious. can do what you ask. There is no injustice to the
My main concern here is not the state of the law in violinist in our doing so.
Thomson: A Defence of Abortion 261
7 the child to term requires only Minimally Decent Sam-

aritanism of the mother, and this is a standard we must
Following the lead of the opponents of abortion, I not fall below. I am inclined to think it a merit of my
have throughout been speaking of the fetus merely as account precisely that it does not give a general yes or
a person, and what I have been asking is whether or a general no. It allows for and supports our sense that,
not the argument we began with, which proceeds only for example, a sick and desperately frightened 14-year-
from the fetus's being a person, really does establish its old schoolgirl, pregnant due to rape, may of course
conclusion. I have argued that it does not. But of course choose abortion, and that any law which rules this out
there are arguments and arguments, and it may be said is an insane law: And it also allows for and supports
that I have simply fastened on the wrong one. It may be our sense that in other cases resort to abortion is even
said that what is important is not merely the fact that positively indecent. It would be indecent in the woman
the fetus is a person, but that it is a person for whom the to request an abortion, and indecent in a doctor to per-
woman has a special kind of responsibility issuing from form it, if she is in her seventh month, and wants the
the fact that she is its mother. And it might be argued abortion just to avoid the nuisance of postponing a trip
that all my analogies are therefore irrelevant—for you abroad. The very fact that the arguments I have been
do not have that special kind of responsibility for that drawing attention to treat all cases of abortion, or even
violinist, Henry Fonda does not have that special kind all cases of abortion in which the mother's life is not at
of responsibility for me. And our attention might be stake, as morally on a par ought to have made them
drawn to the fact that men and women both are com- suspect at the outset.
pelled by law to provide support for their children. Secondly, while I am arguing for the permissibil-
I have in effect dealt (briefly) with this argument ity of abortion in some cases, I am not arguing for
in section 4 above; but a (still briefer) recapitulation the right to secure the death of the unborn child. It
now may be in order. Surely we do not have any such is easy to confuse these two things in that up to a
"special responsibility" for a person unless we have ascertain point in the life of the fetus it is not able to
sumed it, explicitly or implicitly. If a set of parents do survive outside the mother's body; hence removing
not try to prevent pregnancy, do not obtain an abor- it from her body guarantees its death. But they are
tion, and then at the time of birth of the child do not importantly different. I have argued that you are not
put it out for adoption, but rather take it home with morally required to spend nine months in bed, sus-
them, then they have assumed responsibility for it, they taining the life of that violinist; but to say this is by no
have given it rights, and they cannot now withdraw means to say that if, when you unplug yourself, there
support from it at the cost of its life because they now is a miracle and he survives, you then have a right to
find it difficult to go on providing for it. But if they turn round and slit his throat. You may detach your-
have taken all reasonable precautions against having a self even if this costs him his life; you have no right
child, they do not simply by virtue of their biological to be guaranteed his death, by some other means,
relationship to the child who comes into existence have if unplugging yourself does not kill him. There are
a special responsibility for it. They may wish to assume some people who will feel dissatisfied by this feature
responsibility for it, or they may not wish to. And I am of my argument. A woman may be utterly devastated
suggesting that if assuming responsibility for it would by the thought of a child, a bit of herself, put out
require large sacrifices, then they may refuse. A Good for adoption and never seen or heard of again. She
Samaritan would not refuse—or anyway a Splendid may therefore want not merely that the child be de-
Samaritan, if the sacrifices that had to be made were tached from her, but more, that it die. Some oppon-
enormous. But then so would a Good Samaritan ents of abortion are inclined to regard this as beneath
assume responsibility for that violinist; so would Henry contempt—thereby showing insensitivity to what is
Fonda, if he is a Good Samaritan, fly in from the West surely a powerful source of despair. All the same, I
Coast and assume responsibility for me. agree that the desire for the child's death is not one
which anybody may gratify, should it turn out to be
possible to detach the child alive.
8
At this place, however, it should be remembered
My argument will be found unsatisfactory on two counts that we have only been pretending throughout that the
by many of those who want to regard abortion as mor- fetus is a human being from the moment of concep-
ally permissible. First, while I do argue that abortion tion. A very early abortion is surely not the killing of
is not impermissible, I do not argue that it is always a person, and so is not dealt with by anything I have
permissible. There may well be cases in which carrying said here.
■■ 1 11
Notes
1. I am very much indebted to James Thomson for 5. The thesis in (4) is in an interesting way weaker than those
discussion, criticism, and many helpful suggestions. in (1), (2), and (3): they rule out abortion even in cases in
2. Daniel Callahan, (1970), Abortion: Law, Choice and Morality which both mother and child will die if the abortion is not
(New York), 373. This book gives a fascinating survey of performed. By contrast, one who held the view expressed
the available information on abortion. The Jewish tradition in (4) could consistently say that one needn't prefer letting
is surveyed in David M. Feldman, (1968), Birth Control in two persons die to killing one.
Jewish Law (New York), Part 5; the Catholic tradition, in 6. Cf. the following passage from Pius xii, Address to the
John T. Noonan, Jr, (1970), "An Almost Absolute Value in Italian Catholic Society of Midwives: "The baby in the
History," in The Morality of Abortion, ed. John T. Noonan, Jr maternal breast has the right to life immediately from God.
(Cambridge, MA). Hence there is no man, no human authority, no science,
3. The term "direct" in the arguments I refer to is a technical no medical, eugenic, social, economic or moral 'indication'
one. Roughly, what is meant by "direct killing" is either which can establish or grant a valid juridical ground for
killing as an end in itself, or killing as a means to some a direct deliberate disposition of an innocent human life,
end, for example, the end of saving someone else's life. See that is a disposition which looks to its destruction either
note 6, below, for an example of its use. as an end or as a means to another end perhaps in itself
4. Cf. Encyclical Letter of Pope Pius XI on Christian Marriage, not illicit. The baby, still not born, is a man in the same
St. Paul Editions (Boston, n.d.), 32: "however much we degree and for the same reason as the mother" (quoted in
may pity the mother whose health and even life is gravely Noonan, The Morality of Abortion, 45).
imperiled in the performance of the duty allotted to her by 7. The need for a discussion of this argument was brought
nature, nevertheless what could ever be a sufficient reason home to me by members of the Society for Ethical and Legal
for excusing in any way the direct murder of the innocent? Philosophy, to whom this paper was originally presented.
This is precisely what we are dealing with here." Noonan 8. The need for a discussion of this argument was brought
(The Morality of Abortion, p. 43) reads this as follows: home to me by members of the Society for Ethical and
"What cause can ever avail to excuse in any way the direct Legal Philosophy, to whom this paper was originally
killing of the innocent? For it is a question of that." presented.
A Third Way allegiance of more than a small minority of the public.

For the rest of us who are unwilling to embrace either
L.W. Sumner of the extreme options, the problem has been the lack
of a well-defined middle ground between them. In
[. . .] In most countries in the West, public discus- contrast to the power of the established views more
sion of abortion has been distorted by the domin- moderate alternatives may appear both indistinct
ance of two . . . views. The liberal view, espoused by and indecisive.
"pro-choice" groups, holds that (voluntary) abortion is Public distrust of the established views is well
always morally innocuous and (therefore) that the only grounded: neither stands up under critical scrutiny'
acceptable abortion policy is one which treats abor- If their demise is not to leave us without any credible
tion as another variety of minor elective surgery. The view of abortion, three tasks must be successfully com-
conservative view, espoused by "pro-life" groups, holds pleted. The first is to define a third way with abortion
that abortion is always morally serious and (therefore) and to distinguish it from both of the views which it
that the only acceptable abortion policy is one which will supersede. The second is to give it an intuitive de-
treats abortion as another variety of homicide. fense by showing that it coheres better than either of
Because they define the extremities of the con- its predecessors with our considered moral judgments
tinuum of possible positions, and because each is both on abortion itself and on closely related issues.
sufficiently simple and forceful to be advocated by Then, finally, the third way must be grounded in a
a powerful movement, these established views con- moral theory. The first two of these tasks will be under-
stitute the familiar reference points in our abortion taken here; the more daunting theoretical challenge is
landscape. Yet neither has managed to command the confronted elsewhere.'
Source: From Abortion and Moral Theory (Princeton NJ: Princeton University Press, 1981), 124-60. Reprinted with permission.
Sumner: A Third Way 263
1. Specifications standing with the empirical properties of such beings,

and it must be morally relevant. Its moral relevance is
Despite their opposition, the two established views partly testable by appeal to intuition, for arbitrary or
suffer from similar defects. Collating their failures will shallow criteria will be vulnerable to counterexamples.
provide us with some positive guidelines to follow in But the final test of moral relevance is grounding in a
building a more satisfactory alternative. The central moral theory.
issue in the morality of abortion is the moral status of An adequate view of the fetus promises a morally
the fetus. Let us say that a creature has moral standing significant division between early abortions (before the
if, for the purpose of moral decision-making, it must threshold stage) and late abortions (after the threshold
be counted for something in its own right. To count stage). It also promises borderline cases (during the
for nothing is to have no moral standing to count for threshold stage). Wherever that stage is located, abor-
as much as possible (as much, that is, as any creature tions that precede it will be private matters, since the
does) is to have full moral standing. We may, for the fetus will at that stage lack moral standing. Thus the
purpose of the present discussion, make this rather provisions of the liberal view will apply to early abor-
vague notion more precise by adopting the rights tions: they will be morally innocent (as long as the usual
vocabulary favored by both of the established views. conditions of maternal consent, etc., are satisfied) and
We will suppose that having (some) moral standing is ought to be legally unregulated (except for rules equally
equivalent to having (some) right to life. The central applicable to all other medical procedures). Early abor-
issue in the morality of abortion is then whether fetuses tion will have the same moral status as contraception.
have moral standing in this sense.3 Abortions that follow the threshold stage will be
The conservative view, and also the more naive interpersonal matters, since the fetus will at that stage
versions of the liberal view, select a precise point possess moral standing. Thus the provisions of the con-
(conception, birth, etc.) as the threshold of moral servative view will apply to late abortions: they must
standing, implying that the transition from no stand- be assessed on a case-by-case basis, and they ought to
ing to full standing occurs abruptly. In doing so they be legally permitted only on appropriate grounds. Late
rest more weight on these sudden events than they are abortions will have the same moral status as infanti-
capable of bearing. A view that avoids this defect will cide, except for the difference made by the physical
allow full moral standing to be acquired gradually. It connection between fetus and mother.
will therefore attempt to locate not a threshold point A third way with abortion is thus a moderate and
but a threshold period or stage. differential view, combining elements of the liberal
Both of the established views attribute a uniform view for early abortions with elements of (a weakened
moral status to all fetuses, regardless of their dissimi- version of) the conservative view for late abortions. The
larities. Each, for example, counts a newly conceived policy that a moderate view will support is a moderate
zygote for precisely as much (or as little) as a full-term policy permissive in the early stages of pregnancy and
fetus, despite the enormous differences between them. more restrictive (though not as restrictive as conserva-
A view that avoids this defect will assign moral status tives think appropriate) in the later stages. So far as the
differentially so that the threshold stage occurs some- personal question of the moral evaluation of particu-
time during pregnancy. lar abortions is concerned, there is no pressing need
A consequence of the uniform approach adopted to resolve the borderline cases around the threshold
by both of the established views is that neither can stage. But a workable abortion policy cannot tolerate
attach any significance to the development of the fetus this vagueness and will need to establish a definite
during gestation. Yet this development is the most time limit beyond which the stipulated grounds will
obvious feature of gestation. A view that avoids this come into play. Although the precise location of the
defect will base the (differential) moral standing of the time limit will unavoidably be somewhat arbitrary it
fetus at least in part on its level of development. It will will be defensible as long as it falls somewhere within
thus assign undeveloped fetuses a moral status akin the threshold stage. Abortion on request up to the time
to that of ova and spermatozoa, whereas it will assign limit and only for cause thereafter: these are the ele-
developed fetuses a moral status akin to that of infants. ments of a satisfactory abortion policy. E. . .1
So far, then, an adequate view of the fetus must
be gradual, differential, and developmental. It must
2. A Criterion of Moral Standing
also be derived from a satisfactory criterion of moral
standing. Such a criterion must be general (applicable We are assuming that for a creature to have moral
to beings other than fetuses), it must connect moral standing is for it to have a right to life. Any such right
imposes duties on moral agents; these duties may be The first candidate for a criterion takes a direction
either negative (not to deprive the creature of life) or rather different from that of the remaining three. It is
positive (to support the creature's life). Possession of a commonplace in moral philosophy to attribute to
a right to life implies at least some immunity against (normal adult) human beings a special worth or value
attack by others and possibly also some entitlement to or dignity in virtue of which they possess (among other
the aid of others. As the duties may vary in strength, rights) a full right to life. This position implies that
so may the corresponding rights. To have some moral (some degree of) moral standing extends just as far as
standing is to have some right to life, whether or not (some degree of) this intrinsic value, a higher degree of
it may be overridden by the rights of others. To have the latter entailing a higher degree of the former. E. . .1
full moral standing is to have the strongest right to life Such a criterion is undeniably attractive in certain
possessed by anyone, the right to life of the paradigm respects[.] [. . .1 But it is clear that it cannot serve our
person. Depending on one's moral theory, this right present purpose. A criterion must connect moral stand-
may or may not be inviolable and indefeasible and thus ing with some property of things whose presence or
may or may not impose absolute duties on others. absence can be confirmed by a settled, objective, and
To which creatures should we distribute (some public method of investigation. The property of being
degree of) moral standing? On which criterion should intrinsically valuable is not subject to such verification.
we base this distribution? It may be easier to answer A criterion based on intrinsic value cannot be applied
these questions if we begin with the clear case and work without a theory of intrinsic value. Such a theory will
outward to the unclear ones. If we can determine why supply a criterion of intrinsic value by specifying the
we ascribe full standing to the paradigm case, we may natural properties of things in virtue of which they
learn what to look for in other creatures when deciding possess such value. But if things have moral standing
whether or not to include them in the moral sphere. in virtue of having intrinsic value, and if they have in-
The paradigm bearer of moral standing is an adult trinsic value in virtue of having some natural property,
human being with normal capacities of intellect, emo- then it is that natural property which is serving as the
tion, perception, sensation, decision, action, and the real criterion of moral standing, and the middle term
like. If we think of such a person as a complex bundle of intrinsic value is eliminable without loss. A theory
of natural properties, then in principle we could employ of intrinsic value may thus entail a criterion of moral
as a criterion any of the properties common to all standing, but intrinsic value cannot itself serve as that
normal and mature members of our species. Selecting criterion. E. . .1
a particular property or set of properties will define a A criterion of life (or teleology) is too weak, admit-
class of creatures with moral standing, namely, all (and ting classes of beings (animate and inanimate) who are
only) those who share that property. The extension of not suitable loci for moral rights; being alive is neces-
that class will depend on how widely the property in sary for having standing, but it is not sufficient. A cri-
question is distributed. Some putative criteria will be terion of rationality (or moral agency) is too strong,
obviously frivolous and will immediately fail the tests of excluding classes of beings (human and nonhuman)
generality or moral relevance. But even after excluding who are suitable loci for rights; being rational is suf-
the silly candidates, we are left with a number of serious ficient for having standing, but it is not necessary A cri-
ones. There are four that appear to be the most serious: terion of sentience (or consciousness) is a promising
we might attribute full moral standing to the paradigm middle path between these extremes. Sentience is the
person on the ground that he/she is (a) intrinsically capacity for feeling or affect. In its most primitive form
valuable, (b) alive, (c) sentient, or (d) rational. An intui- it is the ability to experience sensations of pleasure and
tive test of the adequacy of any of these candidates will pain and thus the ability to enjoy and suffer. Its more
involve first enumerating the class of beings to whom developed forms include wants, aims, and desires (and
it will distribute moral standing and then determining thus the ability to be satisfied and frustrated); attitudes,
whether that class either excludes creatures that upon tastes, and values; and moods, emotions, sentiments,
careful reflection we believe ought to be included or in- and passions. Consciousness is a necessary condition
cludes creatures that we believe ought to be excluded. of sentience, for feelings are states of mind of which
In the former case the criterion draws the boundary their owner is aware. But it is not sufficient; it is at
of the moral sphere too narrowly and fails as a neces- least possible in principle for beings to be conscious
sary condition of moral standing. In the latter case the (percipient, for instance, or even rational) while utterly
criterion draws the boundary too broadly and fails as a lacking feelings. If rationality embraces a set of cogni-
sufficient condition. (A given criterion may, of course, tive capacities, then sentience is rooted in a beings af-
be defective in both respects.) E. . .1 fective and conative life. It is in virtue of being sentient
Sumner: A Third Way
that creatures have interests, which are compounded ourselves. The criterion also implies that if we encoun-
either out of their desires or out of the experiences they ter creatures who are rational but nonsentient—who
find agreeable (or both). If morality has to do with the utterly lack affect and desire—nothing we can do will
protection and promotion of interests, it is a plausible adversely affect such creatures (in morally relevant
conjecture that we owe moral duties to all those beings ways). We would be entitled, for instance, to treat them
capable of having interests. But this will include all as a species of organic computer. The same obviously
sentient creatures. holds for forms of artificial intelligence; in deciding
Like rationality, and unlike life, it makes sense whether to extend moral standing to sophisticated ma-
to think of sentience as admitting of degrees. Within chines, the question (as Bentham put it) is not whether
any given mode, such as the perception of pain, one they can reason but whether they can suffer.
creature may be more or less sensitive than another. A criterion of sentience also requires gentle usage
But there is a further sense in which more developed of the severely abnormal. Cognitive disabilities and
(more rational) creatures possess a higher degree of disorders may impair a person's range of sensibility, but
sentience. The expansion of consciousness and of intel- they do not generally reduce that person to the level of
ligence opens up new ways of experiencing the world a nonsentient being. Even the grossly retarded or de-
and therefore new ways of being affected by the world. ranged will still be capable of some forms of enjoyment
More rational beings are capable of finding either ful- and suffering and thus will still possess (some) moral
fillment or frustration in activities and states of affairs standing in their own right. This standing diminishes
to which less developed creatures are, both cognitively to the vanishing point only when sentience is entirely
and affectively, blind. It is in this sense of a broader and lost or never gained in the first place. If all affect and
deeper sensibility that a higher being is capable of a responsivity are absent, and if they cannot be engen-
richer, fuller, and more varied existence. The fact that dered, then (but only then) are we no longer dealing
sentience admits of degrees (whether of sensitivity or with a sentient creature. This verdict accords well with
sensibility) enables us to employ it both as an inclusion the contemporary trend toward defining death in terms
criterion and as a comparison criterion of moral stand- of the permanent loss of cerebral functioning. Although
ing. The animal kingdom presents us with a hierarchy such patients are in one obvious sense still alive (their
of sentience. Nonsentient beings have no moral stand- blood circulates and is oxygenated), in the morally
ing; among sentient beings the more developed have relevant sense they are now beyond our reach, for we
greater standing than the less developed, the upper can cause them neither good nor ill. A criterion of life
limit being occupied by the paradigm of a normal would require us to continue treating them as beings
adult human being. Although sentience is the criter- with (full?) moral standing, whereas a criterion of
ion of moral standing, it is also possible to explain the rationality would withdraw that standing when reason
relevance of rationality. The evolutionary order is one was lost even though sensibility should remain. Again
of ascending intelligence. Since rationality expands a a criterion of sentience enables us to find a middle way.
creature's interests, it is a reliable indicator of the degree Fastening upon sentience as the criterion for pos-
of moral standing which that creature possesses. Crea- session of a right to life thus opens up the possibility
tures less rational than human beings do not altogether of a reasonable and moderate treatment of moral prob-
lack standing, but they do lack full standing [. . .] lems other than abortion, problems pertaining to the
A criterion of sentience can thus allow for the grad- treatment of nonhuman animals, extraterrestrial life,
ual emergence of moral standing in the order of nature. artificial intelligence, "defective" human beings, and
It can explain why no moral issues arise (directly) in persons at the end of life. We need now to trace out its
our dealings with inanimate objects, plants, and the implications for the fetus.
simpler forms of animal life. It can also function as a
moral guideline in our encounters with novel life forms 3. The Morality of Abortion
on other planets. If the creatures we meet have interests
and are capable of enjoyment and suffering, we must The adoption of sentience as a criterion determines the
grant them some moral standing. We thereby constrain location of a threshold of moral standing. Since sen-
ourselves not to exploit them ruthlessly for our own tience admits of degrees, we can in principle construct
advantage. The kind of standing that they deserve may a continuum ranging from fully sentient creatures at
be determined by the range and depth of their sens- one extreme to completely nonsentient creatures at
ibility, and in ordinary circumstances this will vary the other. The threshold of moral standing is that area
with their intelligence. We should therefore recognize of the continuum through which sentience fades into
as equals beings who are as rational and sensitive as nonsentience. In phylogenesis the continuum extends
II '..111 1111:13
from homo sapiens to the simple animals and plants, one performed sometime late in the second trimester
and the threshold area is the boundary between verte- or during the third. An early abortion belongs in the
brates and invertebrates. In pathology the continuum same moral category as contraception: it prevents the
extends from the fully normal to the totally incapaci- emergence of a new being with moral standing. A late
tated, and the threshold area is the transition from con- abortion belongs in the same moral category as infanti-
sciousness to unconsciousness. Human ontogenesis cide: it terminates the life of a new being with moral
also presents us with a continuum from adult to zygote. standing. The threshold of sentience thus extends the
The threshold area will be the stage at which sentience morality of contraception forward to cover early abor-
first emerges, but where is that to be located? [. . .1 tion and extends the morality of infanticide backward
The information we now possess does not enable to cover late abortion. [. . .1
us to date with accuracy the emergence of fetal sen- Settling on sentience as a criterion of moral stand-
tience. Of some judgments, however, we can be reasoning thus leads us to a view of the moral status of the
ably confident. First-trimester fetuses are clearly not yet fetus, and of the morality of abortion, which satisfies
sentient. Third-trimester fetuses probably possess some the constraints set out in Section 1. It is gradual, since it
degree of sentience, however minimal. The threshold locates a threshold stage rather than a point and allows
of sentience thus appears to fall in the second trimester. moral standing to be acquired incrementally. It is dif-
More ancient and primitive than cognition, the ability ferential, since it locates the threshold stage during ges-
to discriminate simple sensations of pleasure and pain tation and thus distinguishes the moral status of newly
is probably the first form of consciousness to appear in conceived and full-term fetuses. It is developmental,
the ontogenetic order. Further, when sentience emerges since it grounds the acquisition of moral standing in
it does not do so suddenly. The best we can hope for one aspect of the normal development of the fetus. And
is to locate a threshold stage or period in the second it is moderate, since it distinguishes the moral status
trimester. It is at present unclear just how far into that of early and late abortions and applies each of the es-
trimester this stage occurs. tablished views to that range of cases for which it is
The phylogenetic and pathological continua yield appropriate.
us clear cases at the extremes and unclear cases in the
middle. The ontogenetic continuum does the same. An Abortion Policy
Because there is no quantum leap into consciousness
during fetal development, there is no dean and sharp A differential view of the morality of abortion leads to a
boundary between sentient and nonsentient fetuses. differential abortion policy—one that draws a legal dis-
There is therefore no precise point at which a fetus ac- tinction between early and late abortions. If we work
quires moral standing. More and better information may within the framework of a liberal social theory, then it
enable us to locate the threshold stage ever more accur- is understood that the state has no right to interfere in
ately, but it will never collapse that stage into a point. the private activities of individuals. An early abortion is
We are therefore inevitably confronted with a class of a private act—or, rather, a private transaction between
fetuses around the threshold stage whose sentience, and a woman and her physician. [. . .1
therefore whose moral status, is indeterminate. [. . .1 Grounds for abortion become pertinent only when
Our common moral consciousness locates con- we turn to late abortions. Since virtually all such abor-
traception and infanticide in quite different moral tions will result in the death of a being that has a right
categories. This fact suggests implicit recognition of a to life (though not all will violate that right), the state
basic asymmetry between choosing not to create a new has a legitimate role to play in governing trade in abor-
life in the first place and choosing to destroy a new life tion at this stage. Legal grounds for late abortion are
once it has been created. The boundary between the a special case of conditions for justifiable homicide.
two kinds of act is the threshold at which that life gains As much as possible (allowing for the unique relation
moral protection. Since gametes lack moral standing, between mother and fetus) these grounds should au-
contraception (however it is carried out) merely pre- thorize abortion when killing would also be justified
vents the creation of a new person. Since an infant has in relevantly similar cases not involving fetuses. Two
moral standing, infanticide (however it is carried out) general conditions for justifiable homicide will be ap-
destroys a new person. A second-trimester threshold plicable to abortions: self-defense and euthanasia.
of moral standing introduces this asymmetry into the The usual legal grounds for abortion provided by
moral assessment of abortion. We may define an early moderate policies may be divided into four categories:
abortion as one performed sometime during the first (a) therapeutic (threat to maternal life or health); (b)
trimester or early in the second and a late abortion as eugenic (risk of fetal abnormality); (c) humanitarian
Sumner: AThird Way
(pregnancy due to the commission of a crime, such There is no formula that can be applied mechanically
as rape or incest); (d) socioeconomic (poverty, family to decide the hard cases. We should look to a statute
size, etc.). If a moderate treatment of late abortion is for only the most perfunctory statement of justify-
coupled (as it should be) with a permissive treatment ing grounds for abortion. Particular decisions (the
of early ones, only the first two categories are neces- analogue of case law) are best undertaken by persons
sary. Therapeutic grounds for abortion follow from a with the relevant medical expertise. This might be a
woman's right of self-defense. The threat, however, hospital or clinic committee established especially to
must be serious in two different respects: the injury in monitor late abortions or an "ethics committee" with
prospect must be more than trivial, and the probability broader responsibilities. In either case, establishing the
of its occurrence must be greater than normal. The risks right sort of screening mechanism is the best means of
generally associated with pregnancy will not here suf- ensuring that the justifying grounds are given a reason-
fice. Further, there must be good medical reason not to able application.
delay until the fetus has a better chance of survival, and There is no need for any special notice of humanitar-
every effort must be made to save the fetus' life if this ian grounds. It is doubtful indeed whether incest ought
is possible. Thus late abortion for therapeutic reasons to be a crime, except in those cases in which someone
ought to be reserved for genuine medical emergencies is being exploited. In any case, any woman who has
in which no other course of action would qualify as become pregnant due to incestuous intercourse will
proper care of the mother. In many putatively moderate have ready access to an early abortion. If she declines
policies, therapeutic grounds for abortion (especially this opportunity and if there is no evidence of genetic
mental health clauses) are interpreted so liberally as to abnormality, she may not simply change her mind later.
cover large numbers of cases that are not by any stretch The same obviously applies to pregnancy due to rape,
of the imagination medical emergencies. This is the including statutory rape. The practical problems should
standard device whereby a policy moderate in principle be approached by providing suitable counseling.
becomes permissive in practice. Since the policy here A permissive policy for early abortions will also
advanced is permissive in principle (for early abor- render socioeconomic grounds redundant. Since social
tions), a strict interpretation of the therapeutic grounds constraints do not normally create an emergency for
for late abortions will be mandatory. which abortion is the only solution, and since women
The same strictures will apply to eugenic grounds. will be able to terminate pregnancies at will in the early
Where there is a substantial risk of some severe anom- stages, there is no need for separate recognition of
aly (rubella, spina bifida, Tay-Sachs disease, etc.), social or economic justifications for abortion.
abortion may be the best course of action for the fetus. An adequate abortion policy is thus a conjunction
This is not obviously the case for less severe defects of a permissive policy for early abortions and a mod-
(Down's syndrome, dwarfism, etc.). Again, there will be erate policy for late abortions. The obvious remaining
no justification for an interpretation of eugenic grounds question is where to draw the boundary between the
so elastic that it permits abortion whenever the child is two classes of cases. When we are dealing with the
unwanted (because, say, it is the "wrong" sex). A rough morality of abortion, borderline fuzziness is both in-
rule of thumb is that late abortion for reasons of fetal evitable and tolerable. Many moral problems turn on
abnormality is permissible only in those cases in which factors that are matters of degree. Where such factors
euthanasia for defective newborns would also be per- are present, we cannot avoid borderline cases whose
missible. Probability will play a different role in the two status is unclear or indeterminate. It is a defect in a
kinds of case, since prenatal diagnosis of these condi- moral theory to draw sharp lines where there are none
tions is often less certain than postnatal. But against or to treat hard cases as though they were easy. But
this reason for delay we must balance the anguish of a what makes for good morals may also make for bad
woman carrying a fetus who may turn out at birth to law. An abortion policy must be enforceable and so
be grossly deformed. Since diagnostic techniques such must divide cases as clearly as possible. A threshold
as ultrasound and amniocentesis cannot be employed stage separating early from late abortions must here
until the second trimester, a permissive treatment of give way to a cutoff point.
early abortions will not eliminate the need for late abor- Since there is no threshold point in fetal develop-
tions on eugenic grounds. ment, any precise upper limit on the application of a per-
Both therapeutic and eugenic grounds can be al- missive policy will be to some extent arbitrary. Clearly it
leged for a wide range of abortions. Some of these cases must be located within the threshold period, thus some-
will be clearly justified, others will be just as clearly time in the second trimester. Beyond this constraint the
unjustified, and the remainder will just be hard cases. choice of a time limit may be made on pragmatic grounds.
I II II Al 111111.1
If a permissive policy for early abortions is to promote defended here as the outcome of a bargaining proced-
their autonomy, women must have time to discover that ure. Instead it has been advanced as the only policy
they are pregnant and to decide on a course of action. congruent with an adequate criterion of moral stand-
This factor will tend to push the cutoff point toward the ing and proper recognition of both a woman's right to
end of the second trimester. On the other hand, earlier autonomy and a fetus's right to life. A differential policy
abortions are substantially safer and more economical does not mediate between alternatives both of which
of scarce medical resources than later ones. This factor are rationally defensible; instead it supersedes alterna-
will tend to pull the cutoff point toward the beginning tives both of which have been discredited.
of the second trimester. Balancing these considerations There is, therefore, a third way with the abortion
would indine one toward a time limit located sometime issue. Its superiority over the established views lies
around the midpoint of pregnancy. But it should not be largely in its sensitivity to a factor which both of them
pretended that there is a unique solution to this policy are committed to ignoring the manifest differences
problem. Differential policies may legitimately vary between a fetus at the beginning and at the end of its
(within constraints) in their choice of a boundary be- prenatal existence. Views which deny the relevance of
tween permissiveness and moderation. E. . .1 this factor deserve to command no more than minor-
The abortion policy here proposed is not novel: ity support. Those who, for this reason, can embrace
a differential policy with a time limit in the second neither of the established views need feel no diffidence
trimester is already in operation in a number of coun- about seeking a middle ground between them. A mod-
tries.4 But these policies seem usually to have been erate and differential view of abortion is capable of
settled on as compromises between the opposed de- drawing the common-sense distinction between early
mands of liberals and conservatives rather than as mat- and late abortions and of showing that such a distinc-
ters of principle. Such compromises are attractive to tion is neither shallow nor arbitrary. The view from the
politicians, who do not seek any deeper justification for middle lacks of course the simplicity which has made
the policies they devise. But there is a deeper justifica- it so easy to market its more extreme counterparts. But
tion for this policy. Although it does define a middle then why should we think that the moral problems
ground between the established views, it has not been raised by abortion are simple?
Notes
1. I will not b e defending this assessment i n the present to showing how that rhetoric is equally available to a
paper. For the arguments, see Abortion and Moral Theory, moderate view. For a generalized notion of moral standing
chs. 2 and 3. freed from all connection with rights, see Abortion and
2. Abortion and Moral Theory, chs. 5 and 6. Moral Theory, Section 23.
3. The adoption of this working definition of moral standing 4. Notably the United States, Great Britain, France, Italy,
should not be construed as a concession that rights are Sweden, the Soviet Union, China, India, Japan, and most
the appropriate category for dealing with the moral issues of the countries of Eastern Europe. The cutoff points in
posed by abortion. But since both of the established these jurisdictions vary from the beginning to the end of
views employ the rhetoric of rights, there is some point the second trimester.
Abortion through a Feminist discussions. Authors and readers commonly presume a

Ethics Lens familiarity with the feminist position and equate it with
liberal defences of women's right to choose abortion,
Susan Sherwin but, in fact, feminist ethics yields a different analysis
of the moral questions surrounding abortion than that
Abortion has long been a central issue in the arena of usually offered by the more familiar liberal defenders
applied ethics, but the distinctive analysis of femin- of abortion rights. Most feminists can agree with some
ist ethics is generally overlooked in most philosophic of the conclusions that arise from certain non-feminist
Source: From Dialogue: Canadian Philosophical Review 30 (1987) (3), 265-84. Reprinted by permission of the author.
Sherwin: Abortion through a Feminist Ethics Lens 269
arguments on abortion, but they often disagree about in women's lives. Clearly, the need for abortion can be
the way the issues are formulated and the sorts of rea- very intense; women have pursued abortions under
sons that are invoked in the mainstream literature. appalling and dangerous conditions, across widely
Among the many differences found between fem- diverse cultures and historical periods. No one denies
inist and non-feminist arguments about abortion, is that if abortion is not made legal, safe, and accessible,
the fact that most non-feminist discussions of abortion women will seek out illegal and life-threatening abor-
consider the questions of the moral or legal permis- tions to terminate pregnancies they cannot accept.
sibility of abortion in isolation from other questions, Anti-abortion activists seem willing to accept this price,
ignoring (and thereby obscuring) relevant connections but feminists judge the inevitable loss of women's lives
to other social practices that oppress women. They are associated with restrictive abortion policies to be a
generally grounded in masculinist conceptions of free- matter of fundamental concern.
dom (e.g., privacy, individual choice, individuals' prop- Although anti-abortion campaigners imagine that
erty rights in their own bodies) that do not meet the women often make frivolous and irresponsible deci-
needs, interests, and intuitions of many of the women sions about abortion, feminists recognize that women
concerned. In contrast, feminists seek to couch their have abortions for a wide variety of reasons. Some
arguments in moral concepts that support their general women, for instance, find themselves seriously ill and
campaign of overcoming injustice in all its dimensions, incapacitated throughout pregnancy; they cannot con-
including those inherent in moral theory itself.' There tinue in their jobs and may face enormous difficulties
is even disagreement about how best to understand in fulfilling their responsibilities at home. Many em-
the moral question at issue: non-feminist arguments ployers and schools will not tolerate pregnancy in their
focus exclusively on the morality and/or legality of per- employees or students, and not every woman is able
forming abortions, whereas feminists insist that other to put her job, career, or studies on hold. Women of
questions, including ones about accessibility and deliv- limited means may be unable to take adequate care of
ery of abortion services must also be addressed. children they have already borne and they may know
Although feminists welcome the support of non- that another mouth to feed will reduce their ability to
feminists in pursuing policies that will grant women provide for their existing children. Women who suffer
control over abortion decisions, they generally envision from chronic disease, or who feel too young, or too old,
very different sorts of policies for this purpose than or who are unable to maintain lasting relationships may
those considered by non-feminist sympathizers. For recognize that they will not be able to care properly
example, Kathleen McDonnell (1984) urges feminists for a child at this time. Some who are homeless, or ad-
to develop an explicitly "'feminist morality' of abortion. dicted to drugs, or who are diagnosed as carrying the
. . . At its root it would be characterized by the deep AIDS virus may be unwilling to allow a child to enter
appreciations of the complexities of life, the refusal to the world under such circumstances. If the pregnancy
polarize and adopt simplistic formulas" (52). Here, I is a result of rape or incest, the psychological pain of
propose one conception of the shape such an analysis carrying it to term may be unbearable, and the woman
should take. may recognize that her attitude to the child after birth
will always be tinged with bitterness. Some women
Women and Abortion have learned that the fetuses they carry have serious
chromosomal anomalies and consider it best to prevent
The most obvious difference between feminist and them from being born with a condition bound to cause
non-feminist approaches to abortion can be seen in suffering. Others, knowing the fathers to be brutal and
the relative attention each gives to the interests and violent, may be unwilling to subject a child to the beat-
experiences of women in its analysis. Feminists con- ings or incestuous attacks they anticipate; some may
sider it self-evident that the pregnant woman is a have no other realistic way to remove the child (or
subject of principal concern in abortion decisions. In themselves) from the relationship.
most non-feminist accounts, however, not only is she Or a woman may simply believe that bearing a
not perceived as central, she is rendered virtually invis- child is incompatible with her life plans at this time,
ible. Non-feminist theorists, whether they support or since continuing a pregnancy is likely to have profound
oppose women's right to choose abortion, focus almost repercussions throughout a woman's entire life. If the
all their attention on the moral status of the developing woman is young, a pregnancy will very likely reduce
embryo or the fetus. her chances of education and hence limit her career
In pursuing a distinctively feminist ethics, it is ap- and life opportunities "The earlier a woman has a baby,
propriate to begin with a look at the role of abortion it seems, the more likely she is to drop out of school;
uu I II ,.M 1111141
the less education she gets, the more likely she is to society it is especially important to ensure that women
remain poorly paid, peripheral to the labour market, have the authority to control their own reproduction.3
or unemployed, and the more children she will have— Despite the diversity of opinion among feminists on
between one and three more than her working childless most other matters, virtually all feminists seem to
counterpart" (Petchesky 1984, 150). In many circum- agree that women must gain full control over their own
stances, having a child will exacerbate the social and reproductive lives if they are to free themselves from
economic forces already stacked against her by virtue male dominance.4 Many perceive the commitment of
of her sex (and her race, class, age, sexual orientation, the political right wing to opposing abortion as part of
or the effects of some disability, etc.). Access to abor- a general strategy to reassert patriarchal control over
tion is a necessary option for many women if they are women in the face of significant feminist influence
to escape the oppressive conditions of poverty. (Petchesky 1980, 112).
Whatever the reason, most feminists believe that Women's freedom to choose abortion is also linked
a pregnant woman is in the best position to judge with their ability to control their own sexuality. Women's
whether abortion is the appropriate response to subordinate status often prevents them from refusing
her circumstances. Since she is usually the only one men sexual access to their bodies. If women cannot end
able to weigh all the relevant factors, most femin- the unwanted pregnancies that result from male sexual
ists reject attempts to offer any general abstract rules dominance, their sexual vulnerability to particular
for determining when abortion is morally justified. men can increase, because caring for an(other) infant
Women's personal deliberations about abortion in- involves greater financial needs and reduced economic
clude contextually defined considerations reflecting opportunities for women.' As a result, pregnancy often
her commitment to the needs and interests of everyone forces women to become dependent on men. Since a
concerned—including herself, the fetus she carries, woman's dependence on a man is assumed to entail
other members of her household, etc. Because there that she will remain sexually loyal to him, restriction
is no single formula available for balancing these com- of abortion serves to channel women's sexuality and
plex factors through all possible cases, it is vital that further perpetuates the cycle of oppression.
feminists insist on protecting each woman's right to In contrast to most non-feminist accounts, femin-
come to her own conclusions. Abortion decisions are, ist analyses of abortion direct attention to the question
by their very nature, dependent on specific features of of how women get pregnant. Those who reject abor-
each woman' experience; theoretically dispassionate tion seem to believe that women can avoid unwanted
philosophers and other moralists should not expect to pregnancies by avoiding sexual intercourse. Such views
set the agenda for these considerations in any univer- show little appreciation for the power of sexual politics
sal way. Women must be acknowledged as full moral in a culture that oppresses women. Existing patterns of
agents with the responsibility for making moral deci- sexual dominance mean that women often have little
sions about their own pregnancies.2 Although I think control over their sexual lives. They may be subject to
that it is possible for a woman to make a mistake in her rape by strangers, or by their husbands, boyfriends,
moral judgment on this matter (i.e., it is possible that a colleagues, employers, customers, fathers, brothers,
woman may come to believe that she was wrong about uncles, and dates. Often, the sexual coercion is not
her decision to continue or terminate a pregnancy), even recognized as such by the participants, but is the
the intimate nature of this sort of decision makes it price of continued "good we—popularity, economic
unlikely that anyone else is in a position to arrive at survival, peace, or simple acceptance. Few women
a more reliable conclusion; it is, therefore, improper have not found themselves in circumstances where
to grant others the authority to interfere in women's they do not feel free to refuse a man's demands for
decisions to seek abortions. intercourse, either because he is holding a gun to her
Feminist analysis regards the effects of unwanted head or because he threatens to be emotionally hurt
pregnancies on the lives of women individually and if she refuses (or both). Women are socialized to be
collectively as a central element in the moral evaluation compliant and accommodating, sensitive to the feel-
of abortion. Even without patriarchy bearing a child ings of others, and frightened of physical power; men
would be a very important event in a woman's life. It are socialized to take advantage of every opportunity to
involves significant physical, emotional, social, and engage in sexual intercourse and to use sex to express
(usually) economic changes for her. The ability to exert dominance and power. Under such circumstances, it is
control over the incidence, timing, and frequency of difficult to argue that women could simply "choose" to
childbearing is often tied to her ability to control most avoid heterosexual activity if they wish to avoid preg-
other things she values. Since we live in a patriarchal nancy. Catherine MacKinnon neatly sums it up: "the
Sherwin: Abortion through a Feminist Ethics Lens
logic by which women are supposed to consent to sex the fetus is lacking some criterion of full personhood.6
[is]: preclude the alternatives, then call the remaining Opponents of abortion have structured the debate so
option 'her choice- (MacKinnon 1989, 192). that it is necessary to define the status of the fetus as
Nor can women rely on birth control alone to either valued the same as other humans (and hence en-
avoid pregnancy. There simply is no form of reversible titled not to be killed) or as lacking in all value. Rather
contraception available that is fully safe and reliable. than challenging the logic of this formulation, many
The pill and the IUD are the most effective means of- defenders of abortion have concentrated on show-
fered, but both involve significant health hazards to ing that the fetus is indeed without significant value
women and are quite dangerous for some. No woman (Tooley 1972, Warren 1973); others, such as Wayne
should spend the 30 to 40 years of her reproductive life Sumner (1981), offer a more subtle account that re-
on either form of birth control. Further, both have been flects the gradual development of fetuses whereby there
associated with subsequent problems of involuntary in- is some specific criterion that determines the degree of
fertility, so they are far from optimum for women who protection to be afforded them which is lacking in the
seek to control the timing of their pregnancies. early stages of pregnancy but present in the later stages.
The safest form of birth control involves the use of Thus, the debate often rages between abortion oppon-
barrier methods (condoms or diaphragms) in combina- ents who describe the fetus as an "innocent," vulner-
tion with spermicidal foams or jelly. But these methods able, morally important, separate being whose life is
also pose difficulties for women. They may be socially threatened and who must be protected at all costs, and
awkward to use: young women are discouraged from abortion supporters who try to establish some sort of
preparing for sexual activity that might never happen deficiency inherent to fetuses which removes them
and are offered instead romantic models of spontan- from the scope of the moral community.
eous passion. (Few films or novels interrupt scenes The woman on whom the fetus depends for sur-
of seduction for the fetching of contraceptives.) Many vival is considered as secondary (if she is considered
women find their male partners unwilling to use bar- at all) in these debates. The actual experiences and
rier methods of contraception and they do not have responsibilities of real women are not perceived as
the power to insist. Further, cost is a limiting factor for morally relevant (unless they, too, can be proved inno-
many women. Condoms and spermicides are expen- cent by establishing that their pregnancies are a result
sive and are not covered under most health care plans. of rape or incest). It is a common assumption of both
There is only one contraceptive option which offers defenders and opponents of women's right to choose
women safe and fully effective birth control: barrier abortion that many women will be irresponsible in their
methods with the back-up option of abortion.6 choices. The important question, though, is whether
From a feminist perspective, a central moral fetuses have the sort of status that justifies interfering in
feature of pregnancy is that it takes place in women's women's choices at all. In some contexts, women's role
bodies and has profound effects on women's lives. in gestation is literally reduced to that of "fetal contain-
Gender-neutral accounts of pregnancy are not avail- ers"; the individual women disappear or are perceived
able; pregnancy is explicitly a condition associated simply as mechanical life-support systems.9
with the female body.' Because the need for abortion The current rhetoric against abortion stresses the
is experienced only by women, policies about abor- fact that the genetic make-up of the fetus is determined
tion affect women uniquely. Thus, it is important to at conception and the genetic code is incontestably
consider how proposed policies on abortion fit into human. Lest there be any doubt about the humanity of
general patterns of oppression for women. Unlike the fetus, we are assailed with photographs of fetuses at
non-feminist accounts, feminist ethics demands that various stages of development demonstrating the early
the effects on the oppression of women be a principal appearance of recognizably human characteristics, e.g.,
consideration when evaluating abortion policies. eyes, fingers, and toes. The fact that the fetus in its early
stages is microscopic, virtually indistinguishable from
The Fetus other primate fetuses to the untrained eye, and lacking
in the capacities that make human life meaningful and
In contrast, most non-feminist analysts believe that the valuable is not deemed relevant by the self-appointed
moral acceptability of abortion turns on the question of defenders of fetuses. The anti-abortion campaign is
the moral status of the fetus. Even those who support directed at evoking sympathetic attitudes toward this
women's right to choose abortion tend to accept the tiny, helpless being whose life is threatened by its
central premise of the anti-abortion proponents that own mother; it urges us to see the fetus as entangled
abortion can only be tolerated if it can be proved that in an adversarial relationship with the (presumably
II j11 MILLI
irresponsible) woman who carries it. We are encour- to ensure that women comply with medical "advice."
aged to identify with the "unborn child" and not with Courts are called upon to enforce the doctors' orders
the (selfish) woman whose life is also at issue. when moral pressure alone proves inadequate, and
Within the non-feminist literature, both defenders women are being coerced into undergoing unwanted
and opponents of women's right to choose abortion Caesarean deliveries and technologically monitored
agree that the difference between a late-term fetus and hospital births. Some states have begun to imprison
a newborn infant is "merely geographical" and cannot women for endangering their fetuses through drug
be considered morally significant. But a fetus inhabits a abuse and other socially unacceptable behaviours. An
woman body and is wholly dependent on her unique Australian state recently introduced a bill that makes
contribution to its maintenance while a newborn is women liable to criminal prosecution "if they are found
physically separate though still in need of a lot of care. to have smoked during pregnancy, eaten unhealthful
One can only view the distinction between being in foods, or taken any other action which can be shown
or out of a woman's womb as morally irrelevant if one to have adversely affected the development of the fetus"
discounts the perspective of the pregnant woman; fem- (Warren 1989, 60).
inists seem to be alone in recognizing her perspective In other words, physicians have joined with anti-
as morally important.10 abortionist activists in fostering a cultural acceptance
Within anti-abortion arguments, fetuses are iden- of the view that fetuses are distinct individuals, who
tified as individuals; in our culture which views the are physically, ontologically, and socially separate from
(abstract) individual as sacred, fetuses qua individuals the women whose bodies they inhabit, and who have
should be honoured and preserved. Extraordinary their own distinct interests. In this picture, pregnant
claims are made to try to establish the individuality and women are either ignored altogether or are viewed as
moral agency of fetuses. At the same time, the women deficient in some crucial respect and hence subject to
who carry these fetal individuals are viewed as pas- coercion for the sake of their fetuses. In the former case,
sive hosts whose only significant role is to refrain from the interests of the women concerned are assumed to
aborting or harming their fetuses. Since it is widely be identical with those of the fetus; in the latter, the
believed that the woman does not actually have to do women's interests are irrelevant because they are per-
anything to protect the life of the fetus, pregnancy is ceived as immoral, unimportant, or unnatural. Focus
often considered (abstractly) to be a tolerable burden to on the fetus as an independent entity has led to pre-
protect the life of an individual so like us.n sumptions which deny pregnant women their roles as
Medicine has played its part in supporting these active, independent, moral agents with a primary inter-
sorts of attitudes. Fetal medicine is a rapidly expand- est in what becomes of the fetuses they carry. Emphasis
ing specialty, and it is commonplace in professional on the fetus's status has led to an assumed licence to
medical journals to find references to pregnant women interfere with women's reproductive freedom.
as "fetal environments." Fetal surgeons now have at
their disposal a repertory of sophisticated technology A Feminist View of the Fetus
that can save the lives of dangerously ill fetuses; in light
of such heroic successes, it is perhaps understandable Because the public debate has been set up as a com-
that women have disappeared from their view. These petition between the rights of women and those of
specialists see fetuses as their patients, not the women fetuses, feminists have often felt pushed to reject claims
who nurture them. Doctors perceive themselves as the of fetal value in order to protect women's claims. Yet,
active agents in saving fetal lives and, hence, believe as Addelson (1987) has argued, viewing abortion in
that they are the ones in direct relationship with the this way "tears [it] out of the context of women's lives"
fetuses they treat. (107). There are other accounts of fetal value that are
Perhaps even more distressing than the tendency more plausible and less oppressive to women.
to ignore the woman's agency altogether and view her On a feminist account, fetal development is
as a purely passive participant in the medically con- examined in the context in which it occurs, within
trolled events of pregnancy and childbirth is the grow- women's bodies rather than in the imagined isola-
ing practice of viewing women as genuine threats to the tion implicit in many theoretical accounts. Fetuses
well-being of the fetus. Increasingly women are viewed develop in specific pregnancies which occur in the
as irresponsible or hostile toward their fetuses, and the lives of particular women. They are not individuals
relationship between them is characterized as adver- housed in generic female wombs, nor are they full
sarial (Overall 1987, 60). Concern for the well-being persons at risk only because they are small and sub-
of the fetus is taken as licence for doctors to intervene ject to the whims of women. Their very existence is
relational, developing as they do within particular is a relational concept that must be defined in terms of
women's bodies, and their principal relationship is to interactions and relationships with others.
the women who carry them. A fetus is a unique sort of being in that it cannot
On this view, fetuses are morally significant, but form relationships freely with others, nor can others
their status is relational rather than absolute. Unlike readily form relationships with it. A fetus has a primary
other human beings, fetuses do not have any in- and particularly intimate relationship with the woman
dependent existence; their existence is uniquely tied in whose womb it develops; any other relationship it
to the support of a specific other. Most non-feminist may have is indirect, and must be mediated through
commentators have ignored the relational dimension the pregnant woman. The relationship that exists be-
of fetal development and have presumed that the moral tween a woman and her fetus is clearly asymmetrical,
status of fetuses could be resolved solely in terms of since she is the only party to the relationship who is
abstract metaphysical criteria of personhood. They capable of making a decision about whether the inter-
imagine that there is some set of properties (such as action should continue and since the fetus is wholly
genetic heritage, moral agency, self-consciousness, lan- dependent on the woman who sustains it while she is
guage use, or self-determination) which will entitle all quite capable of surviving without it.
who possess them to be granted the moral status of However much some might prefer it to be other-
persons (Warren 1973, Tooley 1972). They seek some wise, no one else can do anything to support or harm
particular feature by which we can neatly divide the a fetus without doing something to the woman who
world into the dichotomy of moral persons (who are nurtures it. Because of this inexorable biological re-
to be valued and protected) and others (who are not ality, she bears a unique responsibility and privilege
entitled to the same group privileges); it follows that it in determining her fetus's place in the social scheme
is a merely empirical question whether or not fetuses of things. Clearly, many pregnancies occur to women
possess the relevant properties. who place very high value on the lives of the particu-
But this vision misinterprets what is involved in lar fetuses they carry, and choose to see their pregnan-
personhood and what it is that is especially valued cies through to term despite the possible risks and
about persons. Personhood is a social category, not an costs involved; hence, it would be wrong of anyone
isolated state. Persons are members of a community, to force such a woman to terminate her pregnancy
they develop as concrete, discrete, and specific individ- under these circumstances. Other women, or some
uals. To be a morally significant category, personhood of these same women at other times, value other
must involve personality as well as biological integrity.'2 things more highly (e.g., their freedom, their health,
It is not sufficient to consider persons simply as Kantian or previous responsibilities which conflict with those
atoms of rationality; persons are all embodied, con- generated by the pregnancies), and choose not to
scious beings with particular social histories. Annette continue their pregnancies. The value that women
Baier (1985) has developed a concept of persons as ascribe to individual fetuses varies dramatically from
"second persons" which helps explain the sort of social case to case, and may well change over the course of
dimension that seems fundamental to any moral notion any particular pregnancy. There is no absolute value
of personhood: that attaches to fetuses apart from their relational
status determined in the context of their particular
A person, perhaps, is best seen as one who was development.
long enough dependent upon other persons to Since human beings are fundamentally relational
acquire the essential arts of personhood. Persons beings, it is important to remember that fetuses are
essentially are second persons, who grow up with characteristically limited in the relationships in which
other persons. . . . The fact that a person has a they can participate; within those relationships, they
life history, and that a people collectively have a can make only the most restricted "contributions.""
history depends upon the humbler fact that each After birth, human beings are capable of a much wider
person has a childhood in which a cultural herit- range of roles in relationships with an infinite variety
age is transmitted, ready for adolescent rejection of partners; it is that very diversity of possibility and
and adult discriminating selection and contribu- experience that leads us to focus on the abstraction of
tion. Persons come after and before other persons. the individual as a constant through all her/his relation-
(84-5; her emphasis.) ships. But until birth, no such variety is possible, and
the fetus is defined as an entity within a woman who
Persons, in other words, are members of a social com- will almost certainly be principally responsible for it for
munity which shapes and values them, and personhood many years to come.
II II II AI
No human, and especially no fetus, can exist apart effective counselling is offered for a wide range of
from relationships; feminist views of what is valuable reproductive decisions.15 Providers need to recognize
about persons must reflect the social nature of their that abortion is a legitimate option so that services will
existence. Fetal lives can neither be sustained nor de- be delivered with respect and concern for the physical,
stroyed without affecting the women who support them. psychological, and emotional effects on a patient. All
Because of a fetus's unique physical status—within and too frequently, hospital-based abortions are provided
dependent on a particular woman—the responsibility by practitioners who are uneasy about their role and
and privilege of determining its specific social status treat the women involved with hostility and resent-
and value must rest with the woman carrying it. Fetuses ment. Increasingly, many anti-abortion activists have
are not persons because they have not developed suf- personalized their attacks and focused their attention
ficiently in social relationships to be persons in any on harassing the women who enter and leave abortion
morally significant sense (i.e., they are not yet second clinics. Surely requiring a woman to pass a gauntlet of
persons). Newborns, although just beginning their hostile protesters on her way to and from an abortion is
development into persons, are immediately subject to not conducive to effective health care. Ethical explora-
social relationships, for they are capable of communication of abortion raises questions about how women
tion and response in interaction with a variety of other are treated when they seek abortions;16 achieving legal
persons. Thus, feminist accounts of abortion stress the permission for women to dispose of their fetuses if they
importance of protecting women's right to continue as are determined enough to manage the struggle should
well as to terminate pregnancies as each sees fit. not be accepted as the sole moral consideration.
Nonetheless, feminists must formulate their dis-
Feminist Politics and Abortion tinctive response to legislative initiatives on abortion.
The tendency of Canadian politicians confronted by
Feminist ethics directs us to look at abortion in the vocal activists on both sides of the abortion issue has
context of other issues of power and not to limit dis- been to seek "compromises" that seem to give something
cussion to the standard questions about its moral and to each (and, thereby, also deprives each of important
legal acceptability. Because coerced pregnancy has re- features sought in policy formation). Thus, the House of
percussions for women's oppressed status generally, it Commons recently passed a law (Bill C-43) that allows
is important to ensure that abortion not only be made a woman to have an abortion only if a doctor certifies
legal but that adequate services be made accessible to all that her physical, mental, or emotional health will be
women who seek them. This means that within Canada, otherwise threatened. Many non-feminist supporters
where medically approved abortion is technically recog- of women's right to choose consider this a victory and
nized as legal (at least for the moment), we must pro- urge feminists to be satisfied with it, but feminists have
test the fact that it is not made available to many of the good reason to object. Besides their obvious objection
women who have the greatest need for abortions: vast to having abortion returned to the Criminal Code,
geographical areas offer no abortion services at all, but feminists also object that this policy considers doctors
unless the women of those regions can afford to travel and not women the best judges of a woman's need for
to urban clinics, they have no meaningful right to abor- abortion; feminists have little reason to trust doctors to
tion. Because women depend on access to abortion in appreciate the political dimension of abortion or to re-
their pursuit of social equality, it is a matter of moral spond adequately to women's needs. Abortion must be a
as well as political responsibility that provincial health woman's decision, and not one controlled by her doctor.
plans should cover the cost of transport and service Further, experience shows that doctors are already re-
in the abortion facilities women choose. Ethical study luctant to provide abortions to women; the opportunity
of abortion involves understanding and critiquing the this law presents for criminal persecution of doctors by
economic, age, and social barriers that currently restrict anti-abortion campaigners is a sufficient worry to inhibit
access to medically acceptable abortion services.14 their participation.'' Feminists want women's decision-
Moreover, it is also important that abortion servi- making to be recognized as legitimate, and cannot be
ces be provided in an atmosphere that fosters women's satisfied with a law that makes abortion a medical choice.
health and well-being; hence, the care offered should Feminists support abortion on demand because
be in a context that is supportive of the choices women they know that women must have control over their
make. Abortions should be seen as part of women's reproduction. For the same reason, they actively oppose
overall reproductive health and could be included forced abortion and coerced sterilization, practices that
within centres that deal with all matters of reproduct- are sometimes inflicted on the most powerless women,
ive health in an open, patient-centred manner where especially those in the Third World. Feminist ethics
demands that access to voluntary, safe, effective birth who carry fetuses is not only a more legitimate policy
control be part of any abortion discussion, so that women than is regulating them; it is probably also more effective
have access to other means of avoiding pregnancy.18 at ensuring the health and well-being of more fetuses.
Feminist analysis addresses the context as well as Feminist ethics also explores how abortion policies
the practice of abortion decisions. Thus, feminists also fit within the politics of sexual domination. Most fem-
object to the conditions which lead women to abort inists are sensitive to the fact that many men support
wanted fetuses because there are not adequate financial womens right to abortion out of the belief that women
and social supports available to care for a child. Because will be more willing sexual partners if they believe that
feminist accounts value fetuses that are wanted by the they can readily terminate an unwanted pregnancy.
women who carry them, they oppose practices which Some men coerce their partners into obtaining abor-
force women to abort because of poverty or intimida- tions the women may not wanC9 Feminists understand
tion. Yet, the sorts of social changes necessary if we are that many women oppose abortion for this very reason,
to free women from having abortions out of economic being unwilling to support a practice that increases
necessity are vast; they include changes not only in legal women's sexual vulnerability (Luker 1984, 209-15).
and health care policy, but also in housing, child care, Thus, it is important that feminists develop a coherent
employment, etc. (Petchesky 1980, 112). Nonetheless, analysis of reproductive freedom that includes sexual
feminist ethics defines reproductive freedom as the con- freedom (as women choose to define it). That requires
dition under which women are able to make truly vol- an analysis of sexual freedom that includes women's
untary choices about their reproductive lives, and these right to refuse sex; such a right can only be assured if
many dimensions are implicit in the ideal. women have equal power to men and are not subject to
Clearly, feminists are not "pro-abortion," for they domination by virtue of their sex.
are concerned to ensure the safety of each pregnancy to In sum, then, feminist ethics demands that moral
the greatest degree possible; wanted fetuses should not discussions of abortion be more broadly defined than
be harmed or lost. Therefore, adequate pre- and post- they have been in most philosophic discussions. Only by
natal care and nutrition are also important elements of reflecting on the meaning of ethical pronouncements on
any feminist position on reproductive freedom. Where actual women's lives and the connections between judg-
anti-abortionists direct their energies to trying to prevent ments on abortion and the conditions of domination
women from obtaining abortions, feminists seek to pro- and subordination can we come to an adequate under-
tect the health of wanted fetuses. They recognize that far standing of the moral status of abortion in our society.
more could be done to protect and care for fetuses if the As Rosalind Petchesky (1980) argues, feminist discus-
state directed its resources at supporting women who sion of abortion "must be moved beyond the framework
continue their pregnancies, rather than draining away of a "woman's right to choose" and connected to a much
resources in order to police women who find that they broader revolutionary movement that addresses all of
must interrupt their pregnancies. Caring for the women the conditions of women's liberation" (113).
Notes
1. For some idea of the ways in which traditional moral distinguish between legitimate and illegitimate personal
theory oppresses women, see Morgan (1987) and reasons for choosing abortion chiefly rest, reflect an
Hoagland (1988). offensive conception of women as irresponsible; they
2. Critics continue to want to structure the debate around ought not to be perpetuated. Women, seeking moral
the possibility of women making frivolous abortion guidance in their own deliberations about choosing
decisions and hence want feminists to agree to setting abortion, do not find such hypothetical discussions
boundaries on acceptable grounds for choosing abortion. of much use.
Feminists ought to resist this injunction, though. There 3. In her monumental historical analysis of the early roots
is no practical way of drawing a line fairly in the abstract; of Western patriarchy, Gerda Lerner (1986) determined
cases that may appear "frivolous" at a distance often that patriarchy began in the period from 3100 to 600 BC
turn out to be substantive when the details are revealed, when men appropriated women's sexual and reproductive
i.e., frivolity is in the eyes of the beholder. There is capacity; the earliest states entrenched patriarchy by
no evidence to suggest that women actually make the institutionalizing the sexual and procreative subordination
sorts of choices worried critics hypothesize about: e.g., of women to men.
a woman eight months pregnant who chooses to abort 4. There are some women who claim to be feminists against
because she wants to take a trip or gets in "a tiff' with her choice in abortion. See, for instance, Callahan (1987),
partner. These sorts of fantasies, on which demands to though few spell out their full feminist program. For
Ai II 111611
reasons I develop in this paper, I do not think this is a the relationship are defined by the projections and
consistent position. interpretations of the pregnant woman in the latter stages
5. There is a lot the state could do to ameliorate this of pregnancy if she chooses to perceive fetal movements
condition. If it provided women with adequate financial in purposeful ways (e.g., "it likes classical music,
support, removed the inequities in the labour market, and wine, exercise").
provided affordable and reliable childcare, pregnancy need 14. Some feminists suggest we seek recognition of the
not so often lead to a woman's dependence on a particular legitimacy of non-medical abortion services. This would
man. The fact that it does not do so is evidence of the reduce costs and increase access dramatically, with no
state§ complicity in maintaining women's subordinate apparent increase in risk, provided that services were
position with respect to men. offered by trained, responsible practitioners concerned
6. See Petchesky (1984), especially Chapter 5, "Considering with the well-being of their clients. It would also allow the
the Alternatives: The Problems of Contraception," where possibility of increasing women's control over abortion.
she documents the risks and discomforts associated with See, for example McDonnell (1984) 8.
pill use and nips and the increasing rate at which women 15. For a useful model of such a centre, see Van Wagner and
are choosing the option of diaphragm or condom with the Lee (1989).
option of early legal abortions as backup. 16. See CARA1/Halifax (1990) for women's stories about their
7. See Zillah Eisenstein (1988) for a comprehensive theory of experiences with hospitals and freestanding abortion
the role of the pregnant body as the central element in the clinics.
cultural subordination of women. 17. The Canadian Medical Association has confirmed those
8. Thomson (1971) is a notable exception to this trend. fears. In testimony before the House of Commons
9. This seems reminiscent of Aristotle view of women as committee reviewing the bill, the CMA reported that over
"flower pots" where men implant the seed with all the half the doctors surveyed who now perform abortions
important genetic information and the movement necessary expect to stop offering them if the legislation goes
for development and women§ job is that of passive gestation, through. Since the Commons passed the bill, the threats
like the flower pot. For exploration of the flower pot picture of withdrawal of service have increased. Many doctors
of pregnancy, see Whitbeck (1973) and Lange (1983). plan to abandon their abortion service once the law is
10. Contrast Warren (1989) with Tooley (1972). introduced, because they are unwilling to accept the
11. The definition of pregnancy as a purely passive activity harassment they anticipate from anti-abortion zealots.
reaches its ghoulish conclusion in the increasing Even those who believe that they will eventually win any
acceptability of sustaining brain-dead women on life court case that arises fear the expense and anxiety involved
support systems to continue their functions as incubators as the case plays itself out.
until the fetus can be safely delivered. For a discussion of 18. Therefore, the Soviet model, where women have access
this new trend, see Murphy (1989). to multiple abortions but where there is no other birth
12. This apt phrasing is taken from Petchesky (1986), 342. control available, must also be opposed.
13. Fetuses are almost wholly individuated by the 19. See CARAL/Halifax (1990), 20-1, for examples of this sort
women who bear them. The fetal "contributions" to of abuse.
References
Callahan, Sidney. 1987. "A Pro-life Feminist Makes Her McDonnell, Kathleen. 1984. Not an Easy Choice: A Feminist
Case," in Utne Reader (March/April): 104-14. Re-examines Abortion. The Women's Press: Toronto.
CARAL/Halifax. 1990. Telling Our Stories: Abortion Stories from MacKinnon, Catherine. 1989. Toward a Feminism Theory
Nova Scotia. CARAL/Halifax (Canadian Abortion Rights of the State. Harvard University Press: Cambridge, MA.
Action League): Halifax. Morgan, Kathryn Pauly. 1987. "Women and Moral Madness,"
Eisenstein, Zillah R. 1988. The Female Body and the Law. Uni- in Science, Morality and Feminist Theory, ed. Marsha
versity of California Press: Berkeley. Hanen and Kai Nielsen, eds. Canadian Journal of Philoso-
Hoagland, Sara Lucia. 1988. Lesbian Ethics: Toward New Value. phy, Supplementary Volume 13: 201-26.
Institute of Lesbian Studies: Palo Alto, CA. Petchesky, Rosalind Pollack. 1980. "Reproductive Freedom:
Lange, Lynda. 1983. "Woman is Not a Rational Animal: On Beyond 'A Woman's Right to Choose'," in Women: Sex and
Aristotle's Biology of Reproduction," in Discovering Real- Sexuality, ed. Catharine R Stimpson and Ethel Spector
ity: Feminist Perspectives on Epistemology, Metaphysics, Person. University of Chicago Press: Chicago.
Methodology, and Philosophy of Science, ed. Sandra Harding . 1984. Abortion and Woman's Choice: The State, Sexu-
and Merrill B. Hintikka. D. Reidel: Dordrecht, Holland. ality, and Reproductive Freedom. Northeastern University
Lerner, Gerda. 1986. The Creation of Patriarchy. Oxford: New York. Press: Boston.
Luker, Kristin. 1984. Abortion and the Politics of Motherhood. Thomson, Judith Jarvis. 1971. "A Defense of Abortion," in
University of California Press: Berkeley. Philosophy and Public Affairs 1: 47-66.
Tooley, Michael. 1972. "Abortion and Infanticide," in Philoso- Warren, Mary Amine. 1973. "On the Moral and Legal Status
phy and Public Affairs 2 (1; Fall): 37-65. of Abortion," in The Monist 57: 43-61.
Van Wagner, Vicki, and Lee, Bob. 1989. "Principles into Prac- . 1989. "The Moral Significance of Birth," in Hypatia 4
tice: An Activist Vision of Feminist Reproductive Health (2; Summer): 46-65.
Care," in The Future of Human Reproduction, ed. Christine Whitbeck, Carolyn. 1973. "Theories of Sex Difference," in The
Overall. The Women's Press: Toronto. Philosophic Forum, 5 (1-2; Fall/Winter 1973-4): 54-80.
5.4 Completing Reproduction
Judicial Intervention in resist intervention. Whether overt discrimination is at

work or whether the life circumstances of these women
Pregnancy and Birth are such that their behaviour during pregnancy is more
Royal Commission on New Reproductive Technologies likely to come under scrutiny is difficult to disentangle.
Judicial intervention is an issue for all women in
Canada, however, regardless of socioeconomic status,
[...] Judicial Intervention Defined
because its implications go beyond the consequences for
Our mandate directed us to examine "judicial interven- an individual woman; it is an issue for women more gen-
tions during gestation and birth." This examination in- erally if becoming pregnant means that they waive the
volved considering how legislation and court decisions constitutional protections afforded other citizens. [. . .1
are or may be used to control a pregnant woman's be-
haviour. Such judicial intervention usually occurs when The Ethical View
a woman is believed to be endangering the fetus she is
carrying by refusing medical treatment believed neces- The Commission's position on judicial intervention
sary for fetal health, by abusing drugs or alcohol, or by in pregnancy and birth [. . .1 relies primarily on our
engaging in behaviour such as prostitution. The matter is ethical stance and guiding principles, which we have
typically brought before the courts by a children's aid so- applied throughout our report and in our recom-
ciety, a health care facility, or, in some cases, a physician. mendations. Although many legal and constitutional
Judicial interventions during gestation and birth arguments can be made, our conclusions with respect
can take several forms. Canadian courts, for example, to judicial intervention rest largely on our ethical rea-
have ordered pregnant women to refrain from specified soning. In our view, it is ethically (as well as legally)
behaviours and to undergo certain medical proced- wrong to suggest that pregnant women's rights to
ures considered necessary for the health of the fetus. make decisions about their medical care and treat-
[. . .1 US courts have issued similar directives, ordering ment should be changed or lessened because they are
women to engage or not to engage in certain behav- pregnant. [. . .1
iours during pregnancy, to undergo Caesarian section Consistent with the ethic of care—which is con-
and other medical treatment, and to be incarcerated cerned with preventing conflicts instead of trying to
until they had given birth. resolve them after they arise—we begin by asking
Few cases have reached the courts in Canada, questions about how to ensure the best possible pre-
because the women most likely to encounter this situa- natal health and the maximum degree of well-being for
tion are often in no position to resist and therefore they both the pregnant woman and the fetus. Regardless of
comply with the wishes of a physician or child welfare whether a fetus is a "person" with "rights," it is clear that
authority. An examination of the cases that have been the interests of the fetus are worthy of protection: what
reported shows that the women most likely to be sub- transpires before birth—the behaviour of the woman
jected to judicial intervention are disproportionately during pregnancy, the provision of medical treatment to
poor, Aboriginal, or members of a racial or ethnic her and to the fetus—can seriously affect the health and
minority—all factors that influence their capacity to well-being of the child that is eventually born. Society
Source From Proceed with Care: Final Report of the Royal Commission on New Reproductive Technologies (1993), 951,953,
956-65. Reprinted with permission.
II AI IIILLI
therefore has an interest in promoting the prenatal health deemed essential by the courts later turned out not to have
and well-being of the fetus and of the woman carrying it. been necessary. For example, a woman who went into
From the woman's perspective, however, consid- hiding in defiance of a court order to undergo a Caesarian
ering the interests of her fetus separately from her own section later gave birth vaginally to a healthy child. Thus,
has the potential to create adversary situations with medical and judicial judgements—even those made with
negative consequences for her autonomy and bodily the best of intentions can be mistaken. In addition, the
integrity, for her relationship with her partner, and for very limited time frame within which most such deci-
her relationship with her physician. Judicial interven- sions must be made and acted upon makes the process of
tion is bound to precipitate crisis and conflict, instead judicial intervention unlikely to lead to fully considered,
of preventing them through support and care. It also principled, or constitutional conclusions.
ignores the basic components of women's fundamental Finally, judicial intervention both emerges from
human rights—the right to bodily integrity and the and reinforces a social perception of the role of women
right to equality, privacy, and dignity. Importantly, as in reproduction that instrumentalizes them and deval-
we will see, such measures are also unlikely to be ef- ues their humanity and individuality. At the core of the
fective and may not in fact protect the fetus. impulse toward judicial intervention in pregnancy and
If we impose a legal obligation upon a woman to birth is the view that pregnant women are the means
care for her fetus—even if it were possible to legislate to an end—the birth of healthy children. To the extent
a caring and nurturing relationship—the potential for that judicial intervention reinforces the notion that a
curtailing women's choices and behaviour becomes stag- pregnant woman's role is only to carry and deliver a
gering. The kinds of substances and activities that could healthy child, it denies her existence as an autonomous
pose a danger to the fetus are many, varied, and increas- individual with legal and constitutional rights and is
ing: cigarettes, alcohol, drugs (both legal and illegal), en- dangerous to the rights and autonomy of all women.
vironmental pollutants, strenuous exercise, saunas, and In summary, judicial intervention offers no satis-
inadequate nutrition. As scientific knowledge develops, factory answer to ensuring the well-being of the fetus:
the list is becoming longer. Many women's management it precipitates crisis and conflict, it ignores women's
of pregnancy could be subject to challenge and scrutiny, fundamental constitutional and human rights, it con-
and pregnancy could become the source of potential tributes to an instrumentalized view of their role in
liability suits against women who failed to comply with reproduction—with adverse consequences for women
certain standards of behaviour. In some cases, fearing individually and as a group—and, most important, it
a less-than-perfect outcome, a pregnant woman might is not effective in achieving its goal of protecting fetal
feel compelled to seek abortion instead of care. well-being. Society cannot care for a fetus, in the ab-
Moreover, the threat of judicial intervention could sence of the pregnant woman's cooperation, without
have significant negative effects on fetal and maternal taking control of the woman herself. The physical re-
health. If women knew that they could be confined lationship between the fetus and the pregnant woman
against their will, forced to submit to medical treat- and the dependency of the fetus on the pregnant
ment, or charged with criminal offences, they might woman for sustenance make this impossible. By for-
well avoid seeking medical care. Unfortunately, those cing medical intervention, society would be requiring
who might avoid seeking care would likely be those pregnant women to do something that is asked of no
who need it most—for example, women who are de- other individual: to undergo medical treatment for the
pendent on drugs or alcohol. As a result, health prob- benefit of another. Even a living child has no right to
lems would escape detection and treatment—precisely force a parent to undergo medical procedures for the
the opposite effect sought by those who would use child's benefit, however morally compelling the case
judicial means to intervene. E. . .1 might be. This infringement of bodily autonomy and
The resort to judicial intervention also has serious physical integrity is not justified on any grounds.
implications for the relationship between a pregnant This imperative will not change even as research
woman and her physician. If the physician is perceived to pushes the boundaries of what can be done to treat a
be potentially coercive instead of a caregiver, the woman fetus in utero. For example, if and when surgery on a
might begin to withhold information or stop seeking pre- fetus moves beyond the research stage for a range of
natal care, with detrimental consequences for her health conditions, there may be increasing pressure on preg-
and that of the fetus. These dangers are recognized by nant women to consent to such procedures. Whether
many professional associations of physicians. Moreover, these techniques remain experimental or move into the
experience with judicial intervention has shown the un- realm of accepted practice, they must be offered only
certainties inherent in diagnosis and treatment; in sev- in the context of the ethical and legal considerations
eral cases of judicial intervention, the medical treatment set out in this chapter—that is, in the context of the
Royal Commission on New ReproductiveTechnologies: Judicial Intervention in Pregnancy and Birth
pregnant woman's autonomy and with her informed careful deliberation and consideration of the issues from
consent, based on full knowledge of the nature and all sides. We made this decision not because harm to
risks of the proposed treatment. a fetus is acceptable or even tolerable, but because the
dangers posed by judicial intervention far outweigh any
benefits that a given individual intervention might yield.
An Approach Based on Support and Care
In line with the ethic of care, we believe that the
If we reject judicial intervention in pregnancy and best approach is to seek ways to ensure that the needs of
birth on moral, practical, and legal grounds, we must both the woman and the fetus are met—in other words,
return to the question of how to ensure the health and to prevent a situation developing in which child welfare,
well-being of the fetus and the pregnant woman. How medical, or other authorities might consider judicial
should society respond to a situation where a woman intervention appropriate or necessary. The ethic of care
is not caring for her fetus or engaging in behaviour that offers a means of avoiding the conflicts inherent in judi-
may harm it? In the Commission's view, the answer lies cial intervention by promoting two fundamental values:
in examining the reasons for that behaviour and seek- respect for the rights and autonomy of the pregnant
ing solutions that address them. woman and concern for the health and well-being of the
Some of the situations that give rise to attempts at fetus. The best way to accomplish this is not by compel-
judicial intervention are among the most difficult and ling pregnant women to behave in certain ways, but by
tragic imaginable. The potential for harm is evident; providing a supportive and caring environment in which
the dangers to a fetus of alcohol abuse, drug addiction, they can make informed decisions and choose from
or sexually transmitted diseases are real and potentially among realistic options before and during pregnancy.
devastating. These situations are all the more distress- The situations that lead to judicial intervention are
ing because the caring and nurturing assumed to be inherently distressing because of the commitment we, as
inherent in the relationship between woman and fetus a society and as individuals, have to respect human life
appear to be absent. and dignity—the life and dignity of the pregnant woman,
Although many cases involving refusal to follow expressed through her autonomy, and that of the fetus,
medical advice or to accept surgical or other medical as a potential person. Judicial intervention sacrifices the
treatment have involved drug or alcohol abuse, a human dignity and rights of one for the potential well-
woman's reasons for choosing a particular course of being of the other. Taking the alternative route of care
action may include her socioeconomic circumstances, and assistance means that the human life and dignity of
her educational level, her religious convictions, her both woman and fetus are respected—and it may even
cultural beliefs, her fears, or other deeply held values accomplish what legislation or court decisions cannot:
or personal beliefs. Of relevance in this regard is the establish a caring and nurturing relationship.
fact that most of the women who have been subject Clearly, the vast majority of women will act in a
to judicial intervention to date have been Aboriginal way they believe to be the best interests of their fetus.
women and women of colour. The best way to promote prenatal health is therefore
Whatever the circumstances, judicial intervention to provide the information and support necessary to
does not provide a solution, because it does nothing to enable pregnant women to make healthy choices for
address the circumstances that bring about attempts to the well-being of themselves and their fetuses and
intervene or to create the social conditions and support informing them—in non-coercive, non-judgemental
that help to ensure a successful pregnancy and healthy ways—about the implications of their decisions. This
outcome for both the woman and the child. includes providing safe and accessible contraception
In reaching this conclusion, Commissioners are and abortion services; offering accessible and cultur-
acutely aware of the tragic nature of some of the situa- ally appropriate prenatal care and social services to
tions that give rise to efforts to intervene in a pregnancy. pregnant women; counselling pregnant women about
As members of the helping professions, physicians and healthy lifestyles and ensuring they have the means
child welfare workers face situations that call on their to make these choices, including financial assistance
basic human instinct to help where possible—an im- where necessary, and providing information, outreach,
pulse that is rightly very difficult to resist because it is so and supports in the forms pregnant women need to
fundamental to who they are and the job they do. The make informed choices and realistic decisions about
decision to respect a woman's autonomy and physical care and treatment, particularly for addictions.
integrity and not to intervene must surely be one of the The Canadian Bar Association pointed out, in
most difficult decisions any human being would ever be its brief to the Commission, that recourse to judicial
called upon to make. That is why the Commissioners' intervention should be seen as a failure—a failure to
decision in this matter was reached through long and provide policies and programs that sustain a woman's
right to manage her pregnancy and to support her conclusion that judicial intervention is neither an
decisions with appropriate services and resources in acceptable nor an effective method of achieving that
the community. By itself, prohibiting judicial inter- goal. Because the woman's consent and cooperation
vention does not fulfil our responsibility as a society are needed to ensure a positive outcome for the fetus,
to promote the health and well-being of pregnant it follows that the most effective way of caring for
women. Meeting this responsibility also requires the fetus is through appropriate support and caring
appropriate programs, services, and outreach de- for the pregnant woman. The Commission therefore
signed specifically to support pregnant women who recommends that
are in the difficult circumstances we have outlined.
This is not the case at present. In fact, overall, the 273. Judicial intervention in pregnancy and birth
behaviour that attracts judicial intervention may be not be permissible. Specifically, the Commission
less threatening to fetal and neonatal health than the recommends that
well-documented effects of poverty on a much larger
number of pregnancies. F. . .1 a) medical treatment never be imposed upon a
[A] variety of appropriately designed supportive pregnant woman against her wishes;
programs for pregnant women can at the same time b) the criminal law, or any other law, never
help to ensure the well-being of the fetus. In particu- be used to confine or imprison a pregnant
lar, with respect to pregnant women who endanger woman in the interests of her fetus;
the health of their fetuses by using alcohol or drugs, c) the conduct of a pregnant woman in relation
the Board of Trustees of the American Medical Asso- to her fetus not be criminalized;
ciation has recommended that "[p]regnant substance d) child welfare or other legislation never be
abusers should be provided with rehabilitative treat- used to control a woman's behaviour during
ment appropriate to their specific physiological and pregnancy or birth; and
psychological needs." Similar conclusions have been e) civil liability never be imposed upon a woman
reached by this Commission and by others studying for harm done to her fetus during pregnancy.
the problem of drug use during pregnancy. What is
required is ready access to facilities and services that 274. Unwanted medical treatment and other
provide outreach, counselling, and treatment designed interferences, or threatened interferences, with
specifically for pregnant women that are appropriate the physical autonomy of pregnant women be
to their needs. recognized explicitly under the Criminal Code as
criminal assault.
Conclusion and Recommendations
and that
In summary, trying to use the law and the courts to
protect fetal health can only be counterproductive. 275. All provinces/territories ensure that they have
Such laws may on the surface, have appeal, because in place
we all support the goal of the well-being of the fetus,
and enacting them may appear to be a logical extension a) information and education programs directed
of society's interest in the health of the fetus. But there to pregnant women so that they do not in-
is nothing in our experience to demonstrate that such advertently put a fetus at risk;
laws work in practice. Indeed, there is strong evidence b) outreach and culturally appropriate support
to the contrary, particularly because the instruments services for pregnant women and young
available to the courts—forcing action under penalty of women in potentially vulnerable groups; and
fines or incarceration—are brutally blunt and patently c) counselling, rehabilitation, outreach, and sup-
unsuited to the goal of promoting anyone's health or port services designed specifically to meet the
well-being. Clearly, if protecting the fetus is the goal, needs of pregnant women with drug/alcohol
other methods are needed. addictions.
A societal interest in pregnancy and birth—to
maximize the chances for the birth of a healthy In conclusion, it is the Commission's view that
child—is a goal Commissioners strongly endorse; almost all pregnant women will take steps to maxi-
it is an important and worthy goal. But our exam- mize their chances of a healthy birth if they have
ination of the legal, ethical, and social implications ready access to the information, prenatal care, social
of judicial intervention leads to the inescapable services, and income support necessary to do so. In
Royal Commission on New Reproductive Technologies: Judicial Intervention in Pregnancy and Birth 281
the Commission's view, extending care to the fetus General Sources

by giving the pregnant woman the support she needs
provides the best hope for enhancing the health Rodgers, S. "Juridical Interference with Gestation and Birth."
and well-being of both the fetus and the woman In Research Volumes of the Royal Commission on New
carrying it. Reproductive Technologies, 1993.
Dissent on Judicial Intervention individuals and o f those whom they affect. Their cases
should not be predetermined in service to the inter-
in Pregnancy and Birth (Royal ests of some other or larger aggregate group, such as
Commission on New Reproductive women, whose cause (or rather one available sociopo-
Technologies) litical interpretation of whose cause) that individual
has not explicitly embraced, since no one may be used
Suzanne Rowell Scorsone as a means to an end, however worthy that end.
There may be instances in which judicial interven-
[. .] I do not concur with the recommendation that tion would enable and defend a woman's best interests,
judicial intervention in pregnancy not be permissible, her actual consent and autonomy against the coercion
nor do I concur with the associated legislative measures. arising from some particular factor in her situation. One
Words like "never" are, in my view, far too absolute. such example would be the case of a severely addicted
Intervention is generally inadvisable, but should not be woman who states clearly and explicitly that she wants
entirely precluded. The existing possibility of recourse her child to be healthy but whose withdrawal symp-
to the courts, a disinterested forum with accepted toms would demonstrably drive her to seek the drug
legitimacy for mediation and resolution of conflict in she abuses were she not in mandatory treatment. Since
matters of human welfare, remains necessary in an area only judicial review and possible intervention would
so fraught with ambivalence on the part of all parties in allow the nature of her most fundamental consent and
very specific and particular personal difficulties. the actual expression of her choice and autonomy to
Nor would I support a departure from the normal be ascertained and enabled, even autonomy would in
protections of all individuals from medical or other inter- some instances require the continued existence of the
vention, whatever their sex. Application of the severe possibility of judicial intervention in pregnancy. The
sanctions of the criminal law uniquely to interventions Commission report itself, in what appears to me to be a
directed at pregnant women appears to me to be unjusti- contradiction of its own position, refers to the appoint-
fied. Intervention in pregnancy is not fundamentally dif- ment of a legal guardian for a person found mentally
ferent from other forms of medical or social intervention, incompetent. Such a finding and appointment requires
and women are not so different from men in their es- court examination of the case and would therefore in
sence or before the law that the protections and sanctions fact constitute judicial intervention. It is precisely the
governing them should be of different orders. ineradicable need for the availability of objective as-
Such an absolutization could, moreover, have sessment such as this in grave cases that is the point of
negative and discriminatory implications, calling into this dissent.
question the equality of men and women before the law Questions of the existence or non-existence of in-
There are many issues in which attention to the dependent legal or constitutional rights of the fetus are
collective status of women and the autonomy of women irrelevant to the issue. The state has been declared by the
as women would be of proportionately overriding im- Supreme Court of Canada to have an interest in the fetus,
portance; this is not one of them. The consequences for which means that this interest must have some possibil-
individual vulnerable human beings, both woman and ity and venue of exercise. The principle of the require-
child, are too severe and personal and the variability of ment of consent to treatment, including treatment in the
circumstances is too great to be resolved by an abso- interests of another of any age or relationship, is accepted
lutized application of a general principle without the both in ethics and in law, which means that a woman is
possibility of review of individual cases. protected in general from non-consensual intervention.
The resolution of the situations of individuals Positing that the fetus "has no legal existence" and that
should be determined in the best interests of those no third party can volunteer to defend the rights of such
Source: From Proceed with Care: Final Report of the Royal Commission on New Reproductive Technologies (1993), 1063-5 Re-
printed with permission.
um I 11 311 II 1 Li
a being is therefore neither strictly accurate in law nor maximize education, service, and care extended to
necessary to the ordinary protection of women. all women, especially to those who are vulnerable or
I do fully concur with my fellow Commission- addicted so that risk to both woman and fetus can
ers, however, in recommendations which would be avoided.
5.5 Cases
Case 1 sex selection favouring boys in some cultural subgroups,

in particular, among immigrants from India. It appears
Sex Selection in Canada
that the proportion of boys born to that group in Canada
increases with the number of children born to a mother
The law in Canada prohibits sex selection of children at so that by the third birth 138 boys are born for every
preconception, but there are loopholes that permit par- 100 girls. However, it is not clear how troubling this is
ents to determine the sex of their children. Since Canada if the decisions are motivated to have a child of a differ-
has no laws regulating abortion, a fetus can be termin- ent gender from the one(s) already born and there is no
ated if parents discover its sex and do not wish to have overall effect on the normal biological ratio of boys to
a child of that gender. Information about gender can be girls. Australia is currently considering national guide-
obtained from a variety of prenatal tests. Although sex lines for in vitro fertilization that would allow for sex
disclosure is often discouraged, there is no law against selection for a parent's third child.2
it. Parents can also obtain information about the sex There is also some evidence from those involved in
of a fetus by visiting US clinics that will provide this sex selection clinics in the United States that girls are pre-
information and assist them with choosing the sex of ferred over boys and that preferences are changing toward
their child. Recent studies show that the overall ratio of girls in other countries.3 What changes, if any, there
boys to girls born in Canada is about 105 to 100, which should be to public policy about sex selection in Canada
is consistent with the standard biological odds for each depends crucially on difficult-to-obtain accurate informa-
sex, so it does not appear as if there is a significant prob- tion about current sex selection practices. What policy
lem.' But those studies also provide some evidence of measures can, or should be, taken in these circumstances?
Notes
1. Wency Leung, "Some Couples in Canada Practicing Pre- Law Changes, News.com.au 25 July 2016, http://
natal Sex Selection in Favour of Male Fetuses, Studies www.news.com.au/lifestyle/parenting/babies/
Suggest," The Globe and Mail 11 April 2016, httpf/ australian-parents-may-be-able-to-pick-the-sex-of-
www.theglobeandmail.comilife/health-and-fitness/ their-baby-under-proposed-ivf-law-changes/news-story/
health/some-couples-in-canada-practising-prenatal- e84aa850333379c6ce26deef80fdc815.
sex-selection-in-favour-of-male-fetuses-studies-show/ 3. Douglas Todd, "Economics Drives Revolution in Sex Roles:
artic1e29583670. And We're Not Ready," The Vancouver Sun 12 July 2010,
2. Rebecca Sullivan, "Australian Parents May Be Able httplivancouversun.com/news/staff-blogs/economics-
to Pick the Sex of Their Baby under Proposed 1VF drives-revolution-in-sex-roles-and-were-not-ready.
Case 2 eight weeks gestation. Since her advanced age increases

Prenatal Diagnosis and Abortion the risk of having a fetus with Down syndrome, Mrs K
is offered an amniocentesis to be done at 15 to 20 weeks
or Infanticide through Declining gestation. The procedure involves removing and repli-
Treatment cating fetal body cells from the amniotic fluid surround-
ing the fetus in the uterus so that it may be subjected
Berthe K is a 35-year-old woman who is pregnant for to genetic testing. The test will detect or rule out Down
the first time after years of primary infertility. She is at syndrome and other significant genetic anomalies in the
Cases 283
fetus. Mrs K asks what would happen if the test were likelihood of mental disability in the future. The parents
positive. She is told that the test is intended to provide are informed that there is no way of knowing how pro-
her and her husband with information only, and that found the mental disability will be but that with surgery
the choice of what to do with that information would be the infant will have a long life ahead of him. After much
theirs to make. Their options would be to continue with soul-searching and many tears, the Ks decide to refuse the
the pregnancy or to terminate it. Mrs K is very uncom- surgery. The physician makes the situation clear: 'With-
fortable with the prospect of abortion. After discussing out the surgery, your son will die. With it, he can have a
the matter with family and friends, the Ks make the long, full life!" Mr K replies: "You would have allowed my
decision not to have the amniocentesis. wife to have genetic testing and then an abortion halfway
When the baby is born, Mr and Mrs K discover that through the pregnancy. Whath the difference between that
he does have Down syndrome. He has a heart murmur and allowing our son to die now? The condition threaten-
and a life-threatening intestinal blockage and faces the ing him is the same, and it's our decision to make."
Case 3 she and her husband desperately want a boy. Theirs is a

Abortion of Suspected Female Fetus low-income household, and they are unwilling to take
on the costs of caring for another child unless that child
is male. If the physician can use prenatal testing to de-
Claudia Gonzalez is a 30-year-old married woman with termine that the fetus is male, then she will continue
three daughters under the age of 10. Ms Gonzalez is with the pregnancy. If, however, the tests determine
pregnant for the fourth time, currently at 10 weeks that the fetus is female or if she is denied testing that
gestation, and has come to an appointment with her will confirm the sex of the fetus, then she will have an
physician to discuss prenatal testing. She explains that abortion. Should the physician do the test?
Case 4 However, after a court-ordered psychiatric report found

Ms G and Refusal of Treatment Ms G to be competent, she appealed the enforced drug
treatment.2 The Manitoba Court of Appeal unani-
while Pregnant mously reversed the lower court's decision, ruling that
courts cannot order a mentally competent person to
In August 1996, 22-year-old Winnipeg resident Ms G, undergo treatment against her will.3 Winnipeg's Child
already a mother of three, learned she was pregnant and Family Services appealed to the Supreme Court of
again. Two of her children experienced brain damage Canada in an attempt to determine whose rights should
caused by Ms Gs abuse of solvents—paint thinner, prevail: those of the mother or those of the fetus.4 By
glue, nail polish remover—during the two pregnancies. June 1997, when the case reached the Supreme Court
Concerned about the welfare of both the unborn fetus of Canada, Ms G had given birth to her child and was
and its mother, health care providers brought Ms G's pregnant again.5 Additional arguments, on both sides of
case to the attention of Winnipeg's Child and Family the issue, were heard from 12 intervenors representing
Services, who sought custody of Ms G so they could a range of interests—women's rights, Aboriginal rights,
enforce treatment on her during her pregnancy for the human rights, religious rights. In late October 1997,
well-being of the fetus. Mr Justice Perry Schulman of the Supreme Court issued a 7-2 decision stating that a
the Manitoba Court of Queen's Bench committed Ms G fetus does not have legal rights and that courts cannot,
to drug treatment after ruling that she was not mentally therefore, force a pregnant woman to undergo treat-
competent to refuse it on account of her addiction.' ment in order to prevent harm to the fetus.°
Notes
1. Winnipeg Child and Family Services v G (DF) 2. "Judge's Order for Care 'Ethical,'" The Globe and Mail 8
1996. August 1996, Al.
11111 1 11
3. "Pregnant Women's Rights Affirmed," The Globe and Mail 5. "Woman in Glue-Sniffing Case is Pregnant Again," Ottawa
13 September 1996, Al. Citizen 18 June 1997, Al.
4. "Top Court to Consider Fetal Rights," Toronto Star 19 6. "Lawmakers Must Decide Rights of Unborn, Top Court
October 1996, Al2. Says," Ottawa Citizen 1 November 1997, Al.
Case 5 from her last menstrual cycle; however, a best-guess es-

Endangering Behaviour in timate suggests that she is about five months pregnant.
Christina shows no interest in hearing about proper
a Pregnant Woman prenatal care or in taking care of her medical issues:
her principal concern is with being discharged so that
Christina C. is a 24-year-old woman who lives on the she can get back to work. The HCP5, conferring among
streets of Vancouver and works as a prostitute. She has themselves, believe that she will be using drugs again
been brought to the hospital as the result of a heroin within hours, putting herself and her fetus at risk. They
overdose. Christina is seriously underweight and mal- explain to her that drug use during pregnancy will ser-
nourished, and the needle marks on her body show that iously harm the fetus, perhaps causing it to be born
she has been addicted for a long time. When Christina with mental disabilities. Christina is indifferent: "Social
revives enough for the HCP5 to take a full medical his- services took my boy, they'll take this one, too. Are you
tory she discloses a previous diagnosis of hepatitis C going to let me go now?" As they watch her wander off
and HIV as well as the possibility that she is pregnant. A into the night, the attending physician shakes his head
test confirms her pregnancy. sadly "There's another child the system is going to have
Because Christina is so badly malnourished, it is to take care of. If only we could get her into treatment,
difficult to determine the gestational age of the fetus just until the baby is born, they'd both be better off."
Case 6 linked to the condition.5 Down syndrome individuals

Abortion of a Fetus Due to have specific appearance characteristics, like a smaller-
than-normal and abnormally-shaped head, excess neck
Diagnosis of Down Syndrome skin, flattened nose, small ears and mouth, upward-
slanting eyes, wide short hands with short fingers, and
As revealed by Dr Ken Seethram a t the Canadian Society shortened stature. They can expect delayed social and
of Fertility and Andrology conference, sometime in mental development, learning disabilities, and even
2010 a BC couple discovered that their gestational car- dementia. Impulsive behavior, poor judgment, short
rier [GC] or "surrogate mother" was carrying a fetus with attention span, and slow learning are common prob-
Down syndrome.' The GC had been implanted with an lems, and children with Down may exhibit frustration
embryo created with the commissioning parents' egg and anger as they grow and become aware of their lim-
and sperm. A first-trimester ultrasound indicated that itations.4 There is no way of knowing or estimating the
Down syndrome was likely, which was later confirmed full degree of disability at the fetal or even infant stage.
in a diagnostic test. This may be why data shows that up to 85 per cent
Down syndrome is a genetic condition in which a of pregnant women given a Down syndrome diagnosis
fetus inherits three copies of chromosome 21 instead opt for abortion.°
of the typical two. It occurs in one in 700-900 births. In this case, the commissioning parents urged the
The resultant cellular changes cause health problems, GC to abort the affected fetus. She initially refused,
developmental delays, and learning disabilities.2 More being determined to take the pregnancy to term. How-
than 40 per cent of affected fetuses will have a life- ever, under the agreement the trio had signed, choosing
threatening congenital heart malformation.3 Eye and to continue with the pregnancy would mean absolving
hearing problems are common, as are gastrointestinal the commissioning parents of any responsibility for
blockage, underactive thyroid, chronic constipation, raising the child. This raised thorny legal questions, as
sleep apnea, and a tendency to hip dislocation.4 In- Canada's Assisted Human Reproduction Act does not
creased risk of developing leukemia has also been address parentage or parental rights, leaving that up to
Cases 1 285
provincial statutes, which have yet to be written. Before substandard product because of a genetic condition."
this could become an issue, however, the GC chose to Juliet Guichon from the University of Calgary ques-
abort the fetus, in part because of her obligations to her tioned whether the rules of commerce should apply
family which included two children of her own.' to the creation of children: "It's kind of like stopping
The case drew criticism from various Canadian the production line: 'Oh, oh, there's a flaw-1 Disability
bioethicists. Francoise Baylis from Dalhousie Univer- rights' activists were likewise appalled by the sugges-
sity said that the case demonstrates the dangers of tion that a fetus is considered inferior or defective be-
treating life like a commodity: "The child is seen by the cause it faces developmental challenges, not worthy of
commissioning parents as a product, and in this case a being allowed to live.
Notes
1. Tom Blackwell, "Couple Urged Surrogate t o Abort Fetus 4. Down Syndrome Trisomy 21 (2010) National Institutes
Due to Defect," National Post 6 October 2010, http:// of Health, http://wwwncbi.nlm.nih.gov/pubmedhealth/
life.nationalpost.com/2010/10/06/couple-urged-surrogate- PMH0001992 (accessed 26 June 2011).
mother-to-abort-fetus-because-of-defect (accessed 26 June 5. National Cancer Institute, n.d., "Children with Down
2011). Syndrome," http://www.cancer.govicancertopics/pdq/
2. Down Syndrome Research Foundation, http://www.dsrf treatment/childAML/HealthProfessional/page8 (accessed
.org/news_%26_information/information_on_down_ June 26, 2011).
syndrome (accessed 26 June 2011). 6. James L. Natoli, "Prenatal Diagnosis of Down Syndrome:
3. Canadian Down Syndrome Society, http://wwwcdss.ca/ A Systematic Review of Termination Rates," Prenatal
information.html. Diagnosis 32 (2012) no. 2, 142-53.
5.6 Study Questions
1. What are Christine Overall's consequentialist and deontological arguments for the
right to reproductive freedom? What are the limitations to this right?
2. How do you think John A. Robertson would respond to the Royal Commission on
New Reproductive Technology's arguments against sex selection of children? Who has
the stronger evidence for their position?
3. What is the law on commercial surrogacy i n Canada? What are the arguments for and
against it?
4. What is the Conservative Theory? What do you think is the best argument for it? I s a
Conservative committed to condemning celibacy and contraception?
5. What is the Liberal Theory? What do you think is the best argument for it? Must a
Liberal embrace infanticide? Is infanticide justifiable?
6. What is the Middle Theory? What is the standard objection to it? How does Wayne
Sumner try to overcome that problem, and does he succeed?
7. How does Judith Jarvis Thomson try to show that it is not always wrong to kill a
human being with a full right to life? Under what conditions (if any) would Thomson
say that it is wrong to kill a human fetus with a full right to life?
8. What reason does Susan Sherwin give for saying that a woman has full control over her
pregnancy? Do you agree?
9. If a woman behaves in a way that puts her fetus or child-to-be at risk, should judicial
intervention ever be invoked? What does the Royal Commission on New Reproductive
Technologies say? What does Suzanne Rozell Scorsone say? Who do you think is right,
and why?
II III I u I 111111
Busby, Karen, and Delaney Vun. 2010. "Revisiting The Handmaid's Tale: Feminist Theory
Meets Empirical Research on Surrogate Mothers." Canadian Journal of Family Law 26:
13-94.
Dickens, B.M., "Can Sex Selection Be Tolerated?" 2002. Journal of Medical Ethics 28: 335-6.
Feinberg, Joel, ed. 1984. The Problem of Abortion. Belmont, CA: Wadsworth.
Flagler, Elizabeth, Francoise Baylis, and Sanda Rodgers. 1997. "Ethical Dilemmas That
Arise in the Care of Pregnant Women: Rethinking 'Maternal-Fetal Conflicts."' Canadian
Medical Association Journal 157 (12): 1729-32.
Hanna, Jason K.M. 2010. "Revisiting Child-Based Objections to Commercial Surrogacy."
Bioethics 24: 342-7.
Kluge, E.-H., "There Ought to Be a Law" and Zimmerman, David, "No Easy Answer." 1987.
BC Medical Journal 29 (1).
Murray, Thomas H. 1996. "Moral Obligations to the Not-Yet Born The Fetus as Patient."
In Thomas H. Murray, The Worth of a Child, ch. 5,96-114. University of California Press.
Nelson, Lawrence", and Nancy Milliken. 1990. "Compelled Medical Treatment of Pregnant
Women: Life, Liberty, and Law in Conflict." In Richard T. Hull, ed., Ethical Issues in the
New Reproductive Technologies, 224-40. Belmont, CA: Wadsworth.
Purdy, Laura M. 1990. "Are Women Fetal Containers?" Bioethics 4 (4): 372-91.
Rhoden, Nancy K. 1989. "Cesareans and Samaritans." In John Arras and Nancy Rhoden,
eds., Ethical Issues in Modern Medicine, 3rd ed., 321-8. Mountain View, CA: Mayfield.
Robertson, John A., and Joseph D. Schulman. 1987. "Pregnancy and Prenatal Harm to
Offspring: The Case of Mothers with PKU." Hastings Center Report 17 (4): 23-32.
Savalescu, Julian. 2001. "Procreative Beneficence: Why We Should Select the Best Children,"
Bioethics 15: 413-26.
Steinbock, Bonnie. 2009. "Mother—Fetus Conflict." In Helga Kuhse and Peter Singer, eds,
A Companion to Bioethics, 2nd edn., 149-60. Blackwell Publishing.
Sumner, L.W. 1981. Abortion and Moral Theory. Princeton, NJ: Princeton University Press.
Tooley, Michael. 1983. Abortion and Infanticide. Oxford: Clarendon Press.
—. 1972. "In Defense of Abortion and Infanticide." Philosophy & Public Affairs 2
(1): 37-65.
End-of-Life Dacision-Making
6.1 Introduction
This chapter discusses three related but distinct issues about life and death. The first con-
cerns withholding or withdrawing life-sustaining treatment (wLST). The second is about
what is commonly called "assisted suicide and active euthanasia" but is now in Canada
called "physician-assisted death" (PAD) or "medical assistance in dying" (MAID). PAD became
legal in Canada in 2015, and legislation regulating it was passed in 2016. There is much
controversy over both PAD and the ensuing legislation. We will first examine the question
of whether PAD should be legal at all and then turn to the question of how the law should
regulate PAD. The third issue concerns death itself, and the chapter ends with a short essay
on defining death. It will emerge that the problem of deciding when to end life and the
problem of defining death have much in common.
Withholding and Withdrawing Life-Sustaining Treatment
Canadian law and biomedical ethics have now firmly established that it is permissible to
WLST from a wide range of capable and incapable patients. This has not always been so.
Section 197 of the Criminal Code of Canada states that no treatment necessary for life
can be withheld and section 199 that no treatment necessary for life can be withdrawn.
This was read literally and understood to be the law of the land until the late 1970s. The
current law slowly evolved from this position through a series of court decisions in Canada
and the United States.' These decisions began with the right of capable patients in a ter-
minal condition to refuse life-saving treatment and expanded that right. Treatment could be
withdrawn as well as withheld. The patient did not have to be terminal. Foods and fluids as
well as ventilation could be withheld or withdrawn. The patient's substitute decision-maker
A Guide to Acronyms Related .W End-of-Life

Decision-Making
PAD = physician-assisted death
MAID = medical assistance in dying
WLST = withholding and withdrawing life-sustaining treatment
SCC = Supreme Court of Canada
DDE = doctrine of double effect
288 6 End-of-Life Decision-Making
or physician could make the decision. Through these developments, Canada arrived at
the current law, which allows capable patients and their substitute decision-makers or
physicians to WLST if it is not wanted or beneficial. The reading from the Health Law In-
stitute, Dalhousie University, "Withholding and Withdrawal of Potentially Life-Sustaining
Treatment" (see page 295) summarizes the current law on WLST. We see a similar story of
gradualism repeating when we look at the debate on implementing PAD.
Physician-Assisted Death: Definitions, Ethics, and Law
The permissibility of physicians to WLST raises the question of whether they should also be
allowed to practise PAD/MAID. However, any systematic discussion of this topic must begin
by working through a tangle of definitions of the terms from the box above and others
besides. We begin with "euthanasia" and "assisted suicide." Although Canada has aban-
doned these terms in favour of PAD/MAID, they are in common use in most of the world.
Understanding them will enable us to follow the world literature and will also help us to
understand the terminology of PAD and MAID. The Scc used the term PAD, whereas the
Canadian federal legislation that emerged uses the term MAID. We shall use "PAD" except for
cases where the discussion of the Canadian legislation requires using "MAID."
Terminology
euthanasia bringing about Euthanasia consists of bringing about the death of a person in the person's interest.
the death of a person in that The italicized clause is what distinguishes euthanasia from other reasons for bringing about
person's interest. the death of a person. Usually death is not regarded as a benefit for the patient, but in
euthanasia it is, and this makes it possible to make out a moral case for the practice.
passive euthanasia We can also distinguish between two kinds of euthanasia: "active" and "passive." Passive
withholding or withdrawing euthanasia is WLST. Active euthanasia is an intervention such as injecting a patient with
Gfe-sustaining treatment (wLST). a lethal dose. In most jurisdictions, assisted suicide differs from active euthanasia in that
it is the patient versus someone else who causes the death—e.g., by drinking something or
active euthanasia pushing a button. Under Canadian law, however, assisted suicide does not require that a
administering a lethal agent to death occur at all. Assisted suicide is performed when one provides another with the means
another person in order to kill to death, whether or not the person uses those means.
the person. In the terminology of PAD, the distinction between active and passive euthanasia dis-
appears. Passive euthanasia (i.e., wLST) is not euthanasia at all but something to be con-
assisted suicide providing trasted with PAD. Likewise, the distinction between assisted suicide and active euthanasia
a person with the means of disappears and are both encompassed by the term PAD. The qualifying term "P" (Physician)
death. indicates that these acts are permissible only by physicians. The alternative term MAID
("medical assistance in dying") was proposed to allow assisted suicide or active euthanasia
PAD physician-assisted to be performed by a physician or a nurse practitioner, as per amendments to the Criminal
suicide or active euthanasia Code of Canada in 2016.
performed by a physician. Cutting across the distinction between PAD and WLST is another one. This concerns
whether the aid in dying is "voluntary," "non-voluntary" or "involuntary" PAD and WLST
MAID assisted suicide are voluntary if they are done with the consent of the patient or substitute decision-maker,
or active euthanasia non-voluntary if done without the consent of those parties, and involuntary if done (as
performed by a physician or proposed in the Rasouli case in Chapter 1; see page 2) contrary to their request. In this
a nurse practitioner, as per
chapter we will be concerned primarily with voluntary PAD.
amendments to the Criminal
Code of Canada in 2016.
Legal Status of PAD

PAD islegally permitted in Belgium, the Netherlands, Luxembourg (the so-called "Benelux"
countries), and Colombia. It is also legal in four US states (Oregon, Washington, Vermont,
and Montana) and was ruled legal in Canada in 2015. The conditions under which PAD is
Physician-Assisted Death: Definitions, Ethics, and Law 289
permitted in these jurisdictions vary. Some allow assisted suicide but not active euthanasia;
some restrict PAD to capable individuals, whereas others allow substitute decision-makers
to decide. There are also variations in whether advance directives give valid authorization
and eligibility requirements such as age and mental capability. We will consider the ques-
tion of how the legislation is best written later, but for now we look at whether PAD should
be legal at all. Two kinds of arguments can be urged in favour of legalizing PAD, which we
distinguish as "Philosophical" and "Constitutional," and we will look at them in turn.
Philosophical Argument for Legalizing PAD

The philosophical argument for PAD is set out by Dan Brock in "Voluntary Active Eutha-
nasia" (see page 305). According to Brock, "the central argument" for PAD (to put it in
Canadian terms) rests on an appeal to two important values: "individual self-determination"
and "individual well-being." It is hard to deny the importance of these things, and they
combine to make a strong case for legalizing PAD. If we have a moral right to determine
our own destiny, and if we have a moral right to minimize our suffering, we must have
a moral right to request and receive PAD from any physician who is willing to provide it.
Thus, we get a presumptive case for legalizing PAD grounded in values that are recognized
as fundamental to biomedical ethics—namely, autonomy and beneficence.
Constitutional Argument for Legalizing PAD

The constitutional argument for PAD consists of the claim that prohibiting PAD violates
some Charter right or rights. In 1982, a Charter of Rights and Freedoms was added to
Canada's Constitution. This gave those who wanted to change the law on PAD access to
courts, and it was a constitutional argument that made PAD legal in Canada. The case
that decided the issue was the so-called "Carter Case"2 and involved two women, Gloria
Taylor and Kay Carter. Gloria Taylor suffered from ALS and did "not want to die slowly,
piece by piece" or "wracked with pain." Kay Carter suffered from spinal stenosis, which
results in the progressive compression of the spinal cord and ends in immobility and
chronic pain. The argument made in the Scc was that the law prohibiting PAD violated
section 7 of the Canadian Charter of Rights. That section guarantees Canadians a right to
life, liberty, and security of the person. The Scc unanimously found that section 7 rights
were infringed by the prohibition of PAD. Life was infringed because the prohibition of
PAD has the effect of forcing individuals whose natural death would be dreadful to take
their own lives prematurely—that is, before they become so incapacitated by their illness
that they cannot commit suicide. Liberty was infringed because without access to PAD,
individuals who are unable to commit suicide cannot lawfully do what others can and
thus are discriminated against. And security was infringed because without the possibility
of PAD, individuals who do not have the ability to kill themselves will sometimes have to
endure a dreadful death.
Objections to Legalizing PAD

The philosophical and constitutional arguments presented above do not settle the question
in favour of legalizing PAD. They only make a presumptive (or, as it is sometimes called,
a "prima facie") case for legalizing the practice—that is, a case that settles the question in
the absence of countervailing considerations, and there may be some such countervailing
considerations present. If the case is made on moral grounds, there may be more weighty
moral considerations (e.g., the risk to the vulnerable) on the other side. As well, for a
Charter challenge to succeed, it must not only be shown that some article of the Charter
is infringed, it must also be shown that the infringement "cannot be justified in a free and
democratic society."
II i.111 111141
6 End-of-Life Decision-Making
All prior attempts to legalize PAD in Canada failed for one or the other of these reasons.
We must now look at what the arguments against legalizing PAD are and ask whether they
are sufficient to cancel the case for legalizing it. Brock identifies two main arguments against
making PAD legal. The first claims that the practice is intrinsically wrong. The second, that
legalizing it will lead to harmful consequences. We will take these in turn.
PAD Is Intrinsically Wrong

The case for saying that PAD is intrinsically wrong rests on a view that has roots in the Hip-
pocratic tradition according to which physicians are healers. Their proper business is to
protect life and promote health, and hence (so the argument goes) PAD cannot be counten-
anced by physicians. It is not that PAD is wrong because of bad consequences to patients
or society but wrong in itself—intrinsically wrong—because it contradicts the proper aim
of medicine. Two arguments can be presented to show that PAD is intrinsically wrong. The
first is that PAD is intrinsically wrong because it consists of killing the patient. The second
is that PAD is intrinsically wrong because it consists of intentionally killing the patient.
Doctors Must Not Kill

The problem with resting the intrinsic wrongness of PAD on the view that doctors must not
kill is that physicians regularly WLST. And since (so the argument goes) WLST causes death
and causing death is killing, it is not true that doctors must never kill. They do so whenever
they WLST, and WLST is an essential part of medical practice. Brock's article (see page 305)
and James Rachels's "Active and Passive Euthanasia" (see page 301) accept this argument
and conclude that there is no morally relevant difference between WLST and PAD. They thus
argue that if WLST is permissible, so PAD must be.
Daniel Callahan in "When Self-Determination Runs Amok" (see page 317) takes a
different view He argues that the argument endorsed by Brock and by Rachels rests on a
mistake about causation. In Callahan's view, in WLST the doctor does not cause the death
of the patient. Rather, the doctor returns the patient to the state of having an untreated
disease condition, and it is that condition that causes the death. This view is widely ac-
cepted in medicine and the law (see Case 1 on Nancy B., page 341), but it is problematic.
direct or proximate cause We may agree that the untreated disease condition is the direct or proximate cause of
the cause that is closest to the the patient's death. However, if being in an untreated disease condition caused the patient's
event to be explained.
death, and the doctor caused the patient to be in that untreated condition, the doctor
caused the death.
indirect orremote cause a Callahan may say that withholding or withdrawing treatment is merely an indirect
cause that causes the or remote cause. But an indirect cause is still a cause, and it is arbitrary to pick out the
proximate cause.
direct cause as the cause. The fact is that any event has many causes—conditions "but
for" which the event would not have happened—and we typically do not cite them all in
explaining an event. Which cause or causes we pick out to explain an event is determined
by what we think are most unusual or significant in the circumstances. For example, it may
be appropriate to cite the fact that the doctor withdrew treatment as the cause of death if
the speaker thought the hearer did not know that treatment was withdrawn. On the other
hand, it may be more appropriate to cite the disease condition if the speaker thought the
hearer understood that treatment was withdrawn but did not understand why that would
cause death. It is also worth noting that citing an indirect cause is often more useful than
the direct cause. There is an old joke that says pathology is easy. everyone dies of heart
failure. It is true that the direct cause of all death is heart failure, but the task of the patholo-
gist is to say why the heart failed—that is, to identify an indirect cause.
Philosophic controversy over the killing/letting die distinction continues to be lively,
and there is a large literature on it (see the Suggested Further Reading at the end of this
chapter). But many (perhaps most) philosophers dismiss its significance for the PAD debate,
and the SCC briskly set it aside with the comment: "the preponderance of evidence from
Physician-Assisted Death: Definitions, Ethics, and Law
ethicists is that there is no ethical distinction between physician-assisted death and other
end-of-life practices whose outcome is highly likely to be death."3
Doctors Must Not Intentionally Kill

Even if WI_ST does not constitute killing, there is another obstacle to the view that doctors
must never kill their patients. It comes from the fact that it is part of a physician's job to
keep patients free of pain as much as possible. It is also understood both in law and as
good medical practice that physicians can provide patients with sufficient sedation to quell
distress. If they can relieve suffering with non-lethal amounts of drugs, they cannot provide
a lethal dose. But if the distress continues to the point at which distress cannot be quelled
without the sedation killing the patient, the physician can provide a lethal dose of sedation.
When this occurs in the case of dying patients, it is called "terminal sedation." Since it is terminal sedation the
uncontroversial that physicians can sometimes provide terminal sedation, the case for the use of high doses of pain
medication to treat a patient's
intrinsic wrongness of PAD must shift from the claim that physicians must never kill their pa-
suffering at the end of life,
tients to the claim that they must never intentionally kill them. The question now is whether with the awareness that such
the prohibition of intentional killing can be reconciled with the practice of terminal sedation. high dosages will hasten the
The most common way of trying to reconcile providing terminal sedation with phys- patient's death.
icians' putative obligations not to intentionally kill their patients is to appeal to the doctrine of
double effect (DDE). Joseph M. Boyle explains and defends that doctrine in "Medical Ethics
doctrine of double effect
and Double Effect: The Case of Terminal Sedation" (see page 297). The DDE comes into play (DDE) the principle that a
when an agent is faced with a situation in which a single action has a good effect and a bad proposed action that will have
effect and the good effect cannot be brought about without bringing about the bad. The DDE benefits but will also cause
tells us that the action can be performed if the good effect is intended and the evil one merely some harm is permissible if
the action itself is morally
foreseen. This allows physicians to provide terminal sedation if they intend to quell suffering
appropriate and the foreseen
and foresee, but do not intend to cause, the death of the patient. Thus (so the argument goes), harm is not intended.
physicians can always keep their patients free of distress and yet never intentionally kill them.
The DDE is accepted in law in many jurisdictions and by the medical profession, but as
Boyle makes clear, the DDE is philosophically open to challenge, and we raise three ques-
tions about it. First, to judge the permissibility of actions by intentions requires that the
agent's intentions can be identified. But this is not a straightforward task. A doctor provid-
ing terminal sedation, for example, may have a number of things in mind such as to reduce
suffering and kill the patient and honour a promise and ease a burden on the family. It is
unrealistic to suppose that she (or anyone else) will always or even often be able to identify
what she intends, what she merely foresees, and the relative importance of her intentions
in the way the DDE requires.
Second, it is not clear that the DDE enables physicians to provide terminal sedation
and not have an intention to cause death. The utilitarian philosopher Jeremy Bentham
distinguished two kinds of intentions: "direct" and "oblique." Direct intentions are what
the agent aims at. Oblique intentions are what the agent foresees as consequences of her
actions.4 If we define "intentions" as "direct intentions," physicians need never intention-
ally cause death when they provide terminal sedation. On the other hand, if we define "in-
tentions" as (or as including) "oblique intentions," physicians always intend to cause death
as long as they foresee death as a consequence. The reconciliation of terminal sedation
with the alleged duty not to intentionally kill thus relies on picking out the direct intention
as the morally relevant one. But it is not clear what justifies that selection. We ordinarily
count foreseen consequences as intended consequences and hold agents responsible for
them. What, then, can justify excluding them in cases of terminal sedation?
Finally, there is a question of why physicians should be interested in the DDE.
That view is only important to physicians who hold that they must never intentionally
kill patients in the course of their professional practice. But why should physicians want
to hold that view? Intentionally killing another human being is not generally considered
always wrong. It is normally thought permissible (even for physicians) to kill in self-
defence, to protect the defenceless, and in war, for example. Why then should physicians
hold that they cannot intentionally kill as a part of their practice? Without a compelling
answer to this question, the case for legalizing PAD cannot be rejected by appeal to the
claim that doctors must never intentionally kill. There also is disagreement over whether
a compelling answer can be given. Brock (see page 305, particularly the section entitled
"The Role of Physicians") and Callahan (see page 317, particularly the section entitled
"Euthanasia and Medical Practice") take up this question in their articles. Brock argues
that physicians should accept intentional killing as part of their job description; Callahan
defends the view that they should not.
PAD Will Have Bad Consequences

We now turn to the argument that PAD should not be legalized because of potential bad
consequences. Brock provides a list of six such consequences (see page 305, especially
pages 311-313), and similar lists dot the literature. There are two main concerns about
legalizing PAD. The first has to do with risk to individuals requesting it. Here the worry
is whether we can ever have sufficient assurance that those requesting PAD are competent
and their consent is informed, voluntary, not ambivalent, and not compromised by social vulner-
slippery slope a situation abilities such as age or disability. The second is whether legalizing PAD will lead to a slippery
that is likely to lead to an slope. The worry is that legalizing voluntary PAD today will lead to non-voluntary PAD
undesirable consequence,
tomorrow and involuntary PAD the day after. What starts out as policy designed to promote
whether or not one tries to
prevent that.
the dignity and minimize the suffering of individuals thus may end up as one that aims at
eliminating those who are burdens on society. Given this (the argument runs), PAD should
not be legalized.
The possible risks to individuals requesting PAD and undesirable social consequences
slippery slope
have been the main deterrents to legalizing PAD. Until recently, they have been largely based
argument an argument on speculation about what might happen, but now that many jurisdictions permit PAD, the
against allowing a situation evaluation of risk can be put on an empirical basis. The Supreme Court of BC (from which
to occur because it is likely the Carter case came on appeal) undertook just such an evaluation. After what was arguably
to lead to an undesirable to that date the most comprehensive examination of jurisdictions that allow PAD, the court
situation.
concluded that there is no risk to the vulnerable from misuse or abuse, or of a slippery slope,
that would require an absolute prohibition of PAD. At a certain level this was not very surpris-
ing, since the same risks equally apply to WLST and PAD. As there is no evidence of systematic
abuse of WLST, and no evidence of a slippery slope in this context, many have argued that
professional and legal standards that regulate WLST can be expected to apply to PAD without
risk The reading from the BC Supreme Court, "Controlling the Risks of PAD" (see page 321)
provides the concerns Canada had about such risks and the conclusions of the court.
The SCC accepted the BC Supreme Court's evaluation of risks of PAD, and with that,
the last obstacle to legalizing PAD—the fear of bad consequences—was overcome. There
was thus nothing to counterbalance the infringements of the life, liberty, and security of
the person guarantees of section 7 of the Charter, and the SCC ruled, 9-0, that the sections
of the Criminal Code prohibiting PAD are unconstitutional. Specifically, the SCC found that
legislation cannot prohibit PAD for competent adults who (1) clearly consent to the ter-
mination of life and (2) have a grievous medical condition (including an illness, a disease,
or a disability) that is irremediable (meaning that it cannot be alleviated by means accept-
able to the individual) and causes enduring suffering that is intolerable to the individual.
Implementing PAD: Access, Safeguards,

The SCC suspended the effect of its declaration for 12 months (later extended by four
months) to give the federal and provincial governments time to put legislation in place.
To draft the legislation, the federal government of Canada had to decide three issues.
Implementing PAD: Access, Safeguards, Conscientious Objection
These concerned who has access to PAD, what safeguards should be put in place against
abuse and misuse of PAD, and what accommodation should be made for those individuals
or institutions that have conscientious objection to PAD. Current debates on PAD centre conscientious objection/
on these issues, and we look at each in turn. In each case we provide a discussion of the refusal when health
issue, report what legislation the government decided on, and identify issues that still care providers refuse on
moral grounds to provide a
have to be decided.
treatment that is permitted
The reading from the Department of Justice entitled "Legislative Background: Medical by the standards of their
Assistance in Dying (Bill C-14)" (see page 323) explains how the government arrived at profession.
its legislation. The reading "Mature Minors, Mental Illness, Advance Directives, and Con-
scientious Objection" comprises excerpts from recommendations made by the Provincial-
Territorial Expert Advisory Group on Physician-Assisted Dying as well as from the Special
Joint Committee on Physician-Assisted Dying (see page 328). These reports were prepared
to assist the government in writing the legislation and are included to help in evaluating
the legislation and thinking about the issues yet to be settled.
Access
The Scc in the Carter case ruled that any capable adult who has a grievous condition,
experiences suffering that is irremediable in any way acceptable to the person, and who
makes a voluntary and informed choice is entitled to PAD. In Bill C-14, which is the legisla-
tion putting PAD into effect, the federal government of Canada translated the judgement in
Carter into law by saying that individuals are eligible to request and receive MAID only if:
(a) they have a serious and incurable illness, disease or disability,

(b) they are in an advanced state of irreversible decline in capability;
(c) that illness, disease or disability or that state of decline causes them enduring physical
or psychological suffering that is intolerable to them and that cannot be relieved under
conditions that they consider acceptable; and
(d) their natural death has become reasonably foreseeable, taking into account all of their
medical circumstances, without a prognosis necessarily having been made as to the
specific length of time that they have remaining.'
The Department of Justice "Legislative Background" document mentioned above (see page 323)
explains that the above conditions of access are a work in progress and identifies for further
study the possibility of providing access to MAID by "mature minors, advance requests, and
requests where mental illness is the sole underlying medical condition."
Further questions can b e raised about the legislation. Should substitute decision-makers
who can make decisions to WLST from incapable adults, infants, children, and adolescents
be able to request MAID on their behalf as well? One may also ask why candidates for MAID
must have a terminal diagnosis or otherwise have their deaths foreseen. This condition is
insisted on by Quebec in its end of life legislation (Act Respecting End of Life Care, CQ
LR, c. S-32.0001) and all American states that allow PAD but not by any of the Benelux
countries. An argument for the condition is that it limits the risks of MAID by limiting
access to it, but in so doing it also excludes persons with disabilities from MAID. One irony
of including the foreseeable death condition is that Ms Taylor and Ms Carter would not
benefit from the legislation because they were disabled and suffering but not expected to
die soon.
Special Safeguards
Every jurisdiction in the world that allows PAD surrounds i t with special safeguards such as
requiring a second opinion, a mandatory waiting period, witnessing procedures, and so on.
The Scc gave encouragement to attaching special safeguards when it cited the conclusion
H ..111 IIILU
of the trial judge of the BC Supreme Court, saying that the risks inherent in PAD can be
substantially minimized through a "carefully designed system imposing stringent limits
that are scrupulously monitored and enforced."6
The federal government followed this view in Bill C-14. The preamble to the act states
that "robust safeguards are essential to prevent errors and abuse in the provision of MAID,"
and the act goes on to require that requests for MAID be:
(a) made in writing after the person was informed that his or her natural death was rea-
sonably foreseeable;
(b) signed and dated in the presence of two independent witnesses;
(c) accompanied by a written opinion by a second independent medical practitioner or
nurse practitioner confirming the person's eligibility,
(d) made at least 15 days before provision of MAID (unless both practitioners agree that
death or loss of capacity to consent is imminent);
(e) confirmed immediately before providing MAID.?
Not everyone, however, thinks special safeguards are required or desirable. An alternative
to attaching special safeguards is to leave the decision as a private matter between patients,
families, and physicians, just as decisions to WLST are currently left. The Scc also gave
encouragement to this approach when it wrote:
Logically speaking, there is no reason to think that the injured, ill and disabled
who have the option to refuse or to request withdrawal of lifesaving or life-
sustaining treatment, or who seek palliative sedation are less vulnerable or less
susceptible to biased decision-making than whose who might seek more active
assistance in dying.8
Indeed, if there are no greater risks in PAD than in WLST, it would be a logical consequence
that either there should be no special safeguards for either or there should be special
safeguards for both. Brock acknowledges this (see page 305, especially page 315) and sug-
gests that serious consideration be given to attaching such safeguards to WLST. Defences
of going in the opposite direction and rejecting special safeguards for both WLST and PAD
are provided by Glanville Williams in The Sanctity of life and the Criminal Law (1957;
pp. 339-46 of that publication) and James Rachels in The End of life (1986; Chapter 10
of that publication). These works are cited in the Suggested Further Reading at the end
of this chapter.
The Scc acknowledged that physicians have a right to refuse to provide PAD if doing so
conflicts with their moral values. The question is whether they have to do anything to help
those who make the request. Do they have to refer the patient to a physician who will
provide PAD, as they would in the case of abortion? If they do not have to refer, is it enough
to inform the patient that other physicians will provide PAD, or do they have to also tell him
or her how to contact those physicians? Or should there be someone whom an objecting
physician must inform and who has an obligation to ensure a referral? Similar questions
arise with respect to the obligations of publicly funded institutions that have objections to
PAD. Must they allow it on their premises? Or can they transfer patients to another hospital
for the procedure?
Bill C-14 recommends that the Criminal Code be amended (as it now has been) to
specify that no one can be compelled to provide MAID [241.2(9)] but leaves it to physicians'
colleges, the federal Parliament, and the provincial legislatures to grapple with the issues of
conscientious objection. The question remains unsettled.
Defining Death
Defining Death
This chapter has been concerned with the question of when it is appropriate to bring about
death. It is thus natural to want to know when death occurs and how it is to be defined. Alister
Browne takes up these questions in "Defining Death" (see page 339) and argues that 'when death
occurs" is not a time that is discovered (like when the Great Pyramid was built) but rather (like
the age of maturity) a time that is decided. On this view, the line between life and death is not laid
down by God or nature but decided by human beings for human purposes. Thus, lines that are
drawn between life and death at one time can be redrawn in another way at another time. Or, if
it seems best (as it does to some), these lines can be erased altogether in favour of descriptions of
what can be done to whom and when without using the word "death" at all If so, the questions
of PAD and defining death have much in common. Browne's article provides a brief introduction
to this topic and an annotated bibliography for those who want to pursue it further.
Notes
1. See Bernard M. Dickens, "Terminal Care and Related Decisions: A Review o f Legal Developments,"
MODERN MEDICINE of Canada 45 (1990) (4).
2. Carter v Canada (Attorney General), 2015 SCC5.
3. Carter v Canada (Attorney General), 2015 scc5, para. 23. Quoting from the trial judge in Carter v
Canada (Attorney General), 2012, para. 335.
4. Jeremy Bentham, The Principles of Morals and Legislation (1823), ch. VIII, sec. IV.
5. Bill C-14, Statutes of Canada 2016, Ch. 3, An Act to Amend the Criminal Code and to Make Related
Amendments to Other Acts (Medical Assistance in Dying), Sec. 241.2 (2).
6. Carter v Canada (Attorney General), 2015 SCC5, para. 105.
7. Bill C-14, Statutes of Canada 2016, Ch. 3, An Act to Amend the Criminal Code and to Make Related
Amendments to Other Acts (Medical Assistance in Dying), Sec. 241.2 (3), Summary of Key Points.
8. Carter v Canada (Attorney General), 2015 SCC5, para. 115.
6.2 Withholding or Withdrawing Life-Sustaining Treatment and

Physician-Assisted Death
Withholding and Withdrawal of What Does "Withdrawal of Potentially

Life-Sustaining Treatment" Mean?
Potentially Life-Sustaining Treatment
Health Law Institute, Dalhousie University Withdrawal of potentially life-sustaining treatment
means stopping treatment that has the potential to sus-
tain the life of a patient—for example, stopping ventila-
What Does "Withholding of Potentially tory support for a patient in a persistent vegetative state.
Life-Sustaining Treatment" Mean?
Is the Withholding and Withdrawal of
Withholding potentially life-sustaining treatment Potentially Life-Sustaining Treatment
means not starting treatment that has the potential Legal in Canada?
to sustain the life of a patient—for example, not at-
tempting cardiopulmonary resuscitation (CPR) when a Yes, both canb e legal. In the 1993 Supreme Court of Canada
patient's heart suddenly stops beating. case, Rodriguez v. British Columbia (Attorney General), Justice
Source: From http://asOLucis.dal.ca/dhli/cmp_documents/documents/WW_Brochure.pdf (2017). Reprinted by permission of the

author.
II :111 111111
Sopinka, writing for a majority of the Supreme Court of The Patient Is Incompetent but Has a
Canada, made three statements to the effect that there Valid Advance Directive
is a common law right to refuse even potentially life- If the patient has expressed a treatment refusal within
sustaining treatment: a valid advance directive, healthcare providers must re-
spect this refusal. If the patient has appointed a proxy
. . . That there is a right to choose how one's body through a proxy directive, health care providers must
will be dealt with, even in the context of benefi- respect a free and informed refusal by the proxy (unless
cial medical treatment, has long been recognized the healthcare provider believes the proxy is not acting
by the common law. To impose medical treatment in accordance with their powers under the proxy dir-
on one who refuses constitutes battery, and our ective). Provincial legislation dictates the requirements
common law has recognized the right to demand for valid advance directives.
that medical treatment which would extend life be
withheld or withdrawn. The Patient Is a Mature Minor
[. ..] If the patient is a minor who understands and appreci-
ates the consequences of the decision to be made and
Canadian courts have recognized a common law has sufficient independence, under common law their
right of patients to refuse consent to medical decision must be respected. This common law rule
treatment, or to demand that treatment, once can be superseded by statute (as in British Columbia,
commenced, be withdrawn or discontinued (Ciar- Alberta, Saskatchewan, Manitoba and Newfoundland
lariello v. Schacter, [1993] 2 S.C.R. 119). This and Labrador through their child protection statutes).
right has been specifically recognized to exist even However, in accordance with the Supreme Court of
if the withdrawal from or refusal of treatment may Canada's ruling in A.C. v Manitoba (Director of Child and
result in death (Nancy B. v. Hotel-Dieu de Quebec Family Services), these statutes should be interpreted
(1992), 86 D.L.R. (4th) 385 (Que. S.C.); Malette such that acting in the minor's "best interests" under the
v. Shulman (1990), 72 O.R. (2d) 417 (C.A.)). . . . legislation is understood to include respect for the min-
]. • • or's decision-making if the minor is sufficiently mature
and independent. In the context of the withholding and
Whether or not one agrees that the active vs. pas- withdrawal of potentially life-sustaining treatment, the
sive distinction is maintainable, however, the fact scrutiny that the court will apply to the minors matur-
remains that under our common law, the physician ity and independence will be high given the seriousness
has no choice but to accept the patient's instruc- of the decision's potential impact on the minor's life or
tions to discontinue treatment. To continue to health. In provinces and territories without such stat-
treat the patient when the patient has withdrawn utes, the common law mature minor rule will prevaiL
consent to that treatment constitutes battery (Ciar-
lariello and Nancy B., supra). . . . The Patient Is an Immature Minor
If the patient is an immature minor, then the healthcare
More recently, Justice Smith characterized withholding provider must respect a substitute decision-maker's refusal
and withdrawal of potentially life-sustaining treatment to consent to treatment unless the health care provider be-
as "conventionally ethical end-of-life practice[s]" (at 186) lieves that the substitute decision-maker is not acting in
(Carter v Canada (Attorney General), 2012 BCSC 886). the patient's best interests. In such a case, the healthcare
provider should seek an order to treat from the courts.
Must Healthcare Providers Respect Refusals
Made by Patients or Their Substitute Do Healthcare Providers Have the Authority
Decision-Makers? to Withhold or Withdraw Potentially Life-
Sustaining Treatment against the Wishes or
The answer to this question depends upon several Without the Knowledge of a Patient or Their
variables. Substitute Decision-Maker?
The Patient Is Competent The law in Canada on the unilateral withholding and
If the patient is competent and making a free and in- withdrawal of potentially life-sustaining treatment by a
formed refusal, health care providers must respect this healthcare provider is not yet clear. Some courts have
refusal. held that healthcare providers do not have the authority
Brock: Voluntary Active Euthanasia
to unilaterally withhold or withdraw potentially life- Ontario's health care consent legislation requires con-
sustaining treatment, others have said that they do, and sent from a patient's substitute decision-maker prior to
still others have said that the law is unsettled or they the withdrawal of potentially life-sustaining treatment.
have remained silent on the issue. Unfortunately, this decision provides little guidance
On October 18, 2013 some clarity was brought to for provinces and territories other than Ontario as it
the issue when the Supreme Court of Canada released was based on an interpretation of the Ontario legisla-
its decision in Brian Cuthbertson, et at v Hassan Rasouli tion (which others do not share). More litigation, or
by his Litigation Guardian and Substitute Decision Maher, preferably action by provincial legislatures, is needed
Parichehr Salasel. Here, a majority of the Scc found that to resolve the ongoing confusion and controversy.
Voluntary Active Euthanasia proposal to legalize both voluntary active euthanasia

and physician-assisted suicide. Finally, some cases of
Dan W. Brock this kind, such as "It's Over, Debbie," described in the
Journal of the American Medical Association, the "suicide
Since the case of Karen Quinlan first seized public machine" of Dr Jack Kevorkian, and the cancer patient
attention 15 years ago, no issue in biomedical ethics "Diane" of Dr Timothy Quill, have captured wide
has been more prominent than the debate about for- public and professional attention.3 Unfortunately, the
going life-sustaining treatment. Controversy continues first two of these cases were sufficiently problematic
regarding some aspects of that debate, such as forgoing that even most supporters of euthanasia or assisted
life-sustaining nutrition and hydration, and relevant law suicide did not defend the physicians' actions in them.
varies some from state to state. Nevertheless, I believe As a result, the subsequent debate they spawned has
it is possible to identify an emerging consensus that often shed more heat than light. My aim is to increase
competent patients, or the surrogates of incompetent the light, and perhaps as well to reduce the heat, on
patients, should be permitted to weigh the benefits and this important subject by formulating and evaluating
burdens of alternative treatments, including the alterna- the central ethical arguments for and against volun-
tive of no treatment, according to the patient's values, tary active euthanasia and physician-assisted suicide.
and either to refuse any treatment or to select from My evaluation of the arguments leads me, with res-
among available alternative treatments. This consensus ervations to be noted, to support permitting both
is reflected in bioethics scholarship, in reports of pres- practices. My primary aim, however, is not to argue
tigious bodies such as the President's Commission for for euthanasia, but to identify confusions in some
the Study of Ethical Problems in Medicine, The Hast- common arguments, and problematic assumptions
ings Center, and the American Medical Association, in and claims that need more defence or data in others.
a large body of judicial decisions in courts around the The issues are considerably more complex than either
country, and finally in the beliefs and practices of health supporters or opponents often make out; my hope is
care professionals who care for dying patients.' to advance the debate by focusing attention on what I
More recently, significant public and professional believe the real issues under discussion should be.
attention has shifted from life-sustaining treatment In the recent bioethics literature some have en-
to euthanasia—more specifically, voluntary active dorsed physician-assisted suicide but not euthanasia.4
euthanasia—and to physician-assisted suicide. Several Are they sufficiently different that the moral arguments
factors have contributed to the increased interest in for one often do not apply to the other? A paradigm
euthanasia. case of physician-assisted suicide is a patient's ending
In the Netherlands, it has been openly practised his or her life with a lethal dose of a medication re-
by physicians for several years with the acceptance quested of and provided by a physician for that pur-
of the country's highest court.' In 1988 there was an pose. A paradigm case of voluntary active euthanasia
unsuccessful attempt to get the question of whether is a physician's administering the lethal dose, often
it should be made legally permissible on the ballot in because the patient is unable to do so. The only differ-
California. In November 1991 voters in the state of ence that need exist between the two is the person who
Washington defeated a widely publicized referendum actually administers the lethal dose—the physician or
Source: From The Hastings Center Report 22 (1992) (2), 10-22. Reprinted by permission of The Hastings Center and the author.
II .31 111111
the patient. In each, the physician plays an active and from their religious views. The policy issue is only
necessary causal role. whether euthanasia should be permissible, and no one
In physician-assisted suicide the patient acts last who has religious objections to it should be required
(for example, Janet Adkins herself pushed the button to take any part in it, though of course this would not
after Dr Kevorkian hooked her up to his suicide ma- fully satisfy some opponents.
chine), whereas in euthanasia the physician acts last
by performing the physical equivalent of pushing the The Central Ethical Argument for Voluntary
button. In both cases, however, the choice rests fully Active Euthanasia
with the patient. In both the patient acts last in the sense
of retaining the right to change his or her mind until The central ethical argument for euthanasia is famil-
the point at which the lethal process becomes irrevers- iar. It is that the very same two fundamental ethical
ible. How could there be a substantial moral difference values supporting the consensus on patient's rights to
between the two based only on this small difference in decide about life-sustaining treatment also support the
the part played by the physician in the causal process ethical permissibility of euthanasia. These values are
resulting in death? Of course, it might be held that the individual self-determination or autonomy and indi-
moral difference is clear and important—in euthanasia vidual well-being. By self-determination as it bears on
the physician kills the patient whereas in physician- euthanasia, I mean people's interest in making import-
assisted suicide the patient kills him- or herself. But this ant decisions about their lives for themselves accord-
is misleading at best. In assisted suicide the physician ing to their own values or conceptions of a good life,
and patient together kill the patient. To see this, suppose and in being left free to act on those decisions. Self-
a physician supplied a lethal dose to a patient with the determination is valuable because it permits people to
knowledge and intent that the patient will wrongfully form and live in accordance with their own conception
administer it to another. We would have no difficulty of a good life, at least within the bounds of justice and
in morality or the law recognizing this as a case of joint consistent with others doing so as well. In exercising
action to kill for which both are responsible. self-determination people take responsibility for their
If there is no significant, intrinsic moral difference lives and for the kinds of persons they become. A cen-
between the two, it is also difficult to see why public or tral aspect of human dignity lies in people's capacity
legal policy should permit one but not the other, wor- to direct their lives in this way. The value of exercis-
ries about abuse or about giving anyone dominion over ing self-determination presupposes some minimum
the lives of others apply equally to either. As a result, of decision-making capacities or competence, which
I will take the arguments evaluated below to apply to thus limits the scope of euthanasia supported by self-
both and will focus on euthanasia. determination; it cannot justifiably be administered,
My concern here will be with voluntary euthanasia for example, in cases of serious dementia or treatable
only—that is, with the case in which a clearly competent clinical depression.
patient makes a fully voluntary and persistent request Does the value of individual self-determination
for aid in dying. Involuntary euthanasia, in which a extend to the time and manner of one's death? Most
competent patient explicitly refuses or opposes receiv- people are very concerned about the nature of the last
ing euthanasia, and nonvoluntary euthanasia, in which a stage of their lives. This reflects not just a fear of ex-
patient is incompetent and unable to express his or her periencing substantial suffering when dying, but also
wishes about euthanasia, will be considered here only as a desire to retain dignity and control during this last
potential unwanted side effects of permitting voluntary period of life. Death is today increasingly preceded
euthanasia. I emphasize as well that I am concerned with by a long period of significant physical and mental
active euthanasia, not withholding or withdrawing life- decline, due in part to the technological interventions
sustaining treatment, which some commentators charac- of modern medicine. Many people adjust to these dis-
terize as "passive euthanasia." Finally, I will be concerned abilities and find meaning and value in new activities
with euthanasia where the motive of those who perform and ways. Others find the impairments and burdens
it is to respect the wishes of the patient and to provide in the last stage of their lives at some point sufficiently
the patient with a "good death," though one important great to make life no longer worth living. For many
issue is whether a change in legal policy could restrict patients near death, maintaining the quality of one's
the performance of euthanasia to only those cases. life, avoiding great suffering, maintaining one's dignity,
A last introductory point is that I will be examin- and, insuring that others remember us as we wish them
ing only secular arguments about euthanasia, though to become of paramount importance and outweigh
of course many people's attitudes to it are inextricable merely extending one's life. But there is no single,
Brock: Voluntary Active Euthanasia 299
objectively correct answer for everyone as to when, if their view outweigh or override this support. The first
at all, one's life becomes all things considered a burden kind of argument is that in any individual case where
and unwanted. If self-determination is a fundamental considerations of the patient's self-determination and
value, then the great variability among people on this well-being do support euthanasia, it is nevertheless
question makes it especially important that individuals always ethically wrong or impermissible. The second
control the manner, circumstances, and timing of their kind of argument grants that in some individual cases
dying and death. euthanasia may not be ethically wrong, but maintains
The other main value that supports euthanasia is nonetheless that public and legal policy should never
individual well-being. It might seem that individual permit it. The first kind of argument focuses on fea-
well-being conflicts with a person's self-determination tures of any individual case of euthanasia, while the
when the person requests euthanasia. Life itself is com- second kind focuses on social or legal policy. In the
monly taken to be a central good for persons, often next section I consider the first kind of argument.
valued for its own sake, as well as necessary for pursuit
of all other goods within a life. But when a competent Euthanasia Is the Deliberate Killing of an
patient decides to forgo all further life-sustaining treat- Innocent Person
ment then the patient, either explicitly or implicitly,
commonly decides that the best life possible for him or The claim that any individual instance of euthanasia
her with treatment is of sufficiently poor quality that it is a case of deliberate killing of an innocent person is,
is worse than no further life at all. Life is no longer con- with only minor qualifications, correct. Unlike forgo-
sidered a benefit by the patient, but has now become ing life-sustaining treatment, commonly understood as
a burden. The same judgment underlies a request for allowing to die, euthanasia is clearly killing, defined as
euthanasia: continued life is seen by the patient as no depriving of life or causing the death of a living being.
longer a benefit, but now a burden. Especially in the While providing morphine for pain relief at doses
often severely compromised and debilitated states of where the risk of respiratory depression and an earlier
many critically ill or dying patients, there is no object- death may be a foreseen but unintended side effect of
ive standard, but only the competent patient's judg- treating the patient's pain, in a case of euthanasia the
ment of whether continued life is no longer a benefit. patient's death is deliberate or intended even if in both
Of course, sometimes there are conditions, such as the physician's ultimate end may be respecting the pa-
clinical depression, that call into question whether the tient's wishes. If the deliberate killing of an innocent
patient has made a competent choice, either to forgo person is wrong, euthanasia would be nearly always
life-sustaining treatment or to seek euthanasia, and then impermissible.
the patient's choice need not be evidence that continued In the context of medicine, the ethical prohibition
life is no longer a benefit for him or her. Just as with against deliberately killing the innocent derives some
decisions about treatment, a determination of incompe- of its plausibility from the belief that nothing in the
tence can warrant not honouring the patient's choice; currently accepted practice of medicine is deliberate
in the case of treatment, we then transfer decisional killing. Thus, in commenting on the "It's Over, Debbie"
authority to a surrogate, though in the case of voluntary case, four prominent physicians and bioethicists could
active euthanasia a determination that the patient is in- entitle their paper "Doctors Must Not Kill."5 The belief
competent means that choice is not possible. that doctors do not in fact kill requires the corollary
The value or right of self-determination does not belief that forgoing life-sustaining treatment, whether
entitle patients to compel physicians to act contrary by not starting or by stopping treatment, is allowing to
to their own moral or professional values. Physicians die, not killing. Common though this view is, I shall
are moral and professional agents whose own self- argue that it is confused and mistaken.
determination or integrity should be respected as well. Why is the common view mistaken? Consider the
If performing euthanasia became legally permissible, case of a patient terminally ill with ALS disease. She is
but conflicted with a particular physician's reason- completely respirator-dependent with no hope of ever
able understanding of his or her moral or professional being weaned. She is unquestionably competent, but
responsibilities, the care of a patient who requested finds her condition intolerable and persistently re-
euthanasia should be transferred to another. quests to be removed from the respirator and allowed
Most opponents do not deny that there are some to die. Most people and physicians would agree that
cases in which the values of patient self-determination the patient's physician should respect the patient's
and well-being support euthanasia Instead, they com- wishes and remove her from the respirator, though this
monly offer two kinds of arguments against it that in will certainly cause the patient's death. The common
II ill 111111i1
understanding is that the physician thereby allows the son does is to allow to die, not kill.' Here, I want to
patient to die. But is that correct? note two other reasons why the conclusion that stop-
Suppose the patient has a greedy and hostile son ping life support is killing is resisted.
who mistakenly believes that his mother will never The first reason is that killing is often understood,
decide to stop her life-sustaining treatment and that even especially within medicine, as unjustified causing of
if she did her physician would not remove her from the death; in medicine it is thought to be done only acci-
respirator. Afraid that his inheritance will be dissipated dentally or negligently. It is also increasingly widely ac-
by a long and expensive hospitalization, he enters his cepted that a physician is ethically justified in stopping
mother's room while she is sedated, extubates her, and life support in a case like that of the ALS patient. But if
she dies. Shortly thereafter the medical staff discovers these two beliefs are correct, then what the physician
what he has done and confronts the son. He replies, "I does cannot be killing, and so must be allowing to die.
didn't kill her, I merely allowed her to die. It was her Killing patients is not, to put it flippantly, understood
ALS disease that caused her death." I think this would to be part of physicians' job description. What is mis-
rightly be dismissed as transparent sophistry—the son taken in this line of reasoning is the assumption that all
went into his mother's room and deliberately killed her. killings are unjustified causings of death. Instead, some
But, of course, the son performed just the same phys- killings are ethically justified, including many instances
ical actions, did just the same thing, that the physician of stopping life support.
would have done. If that is so, then doesn't the physician Another reason for resisting the conclusion that
also kill the patient when he extubates her? stopping life support is often killing is that it is psych-
I underline immediately that there are important ologically uncomfortable. Suppose the physician had
ethical differences between what the physician and the stopped the ALS patient's respirator and had made the
greedy son do. First, the physician acts with the pa- son's claim, "I didn't kill her, I merely allowed her to
tient's consent whereas the son does not. Second, the die. It was her ALS disease that caused her death." The
physician acts with a good motive—to respect the pa- clue to the psychological role here is how naturally the
tient's wishes and self-determination—whereas the son "merely" modifies "allowed her to die." The characteriza-
acts with a bad motive—to protect his own inheritance. tion as allowing to die is meant to shift felt responsibility
Third, the physician acts in a social role through which away from the agent—the physician—and to the lethal
he is legally authorized to carry out the patient's wishes disease process. Other language common in death and
regarding treatment whereas the son has no such au- dying contexts plays a similar role; "letting nature take
thorization. These and perhaps other ethically import- its course" or "stopping prolonging the dying process"
ant differences show that what the physician did was both seem to shift responsibility from the physician who
morally justified whereas what the son did was morally stops life support to the fatal disease process. However
wrong. What they do not show, however, is that the psychologically helpful these conceptualizations may
son killed while the physician allowed to die. One can be in making the difficult responsibility of a physician's
either kill or allow to die with or without consent, with role in the patient's death bearable, they nevertheless are
a good or bad motive, within or outside of a social role confusions. Both physicians and family members can
that authorizes one to do so. instead be helped to understand that it is the patient's
The difference between killing and allowing to die decision and consent to stopping treatment that limits
that I have been implicitly appealing to here is roughly their responsibility for the patient's death and that shifts
that between acts and omissions resulting in death.° that responsibility to the patient.
Both the physician and the greedy son act in a manner Many who accept the difference between killing
intended to cause death, do cause death, and so both and allowing to die as the distinction between acts and
kill. One reason this conclusion is resisted is that on omissions resulting in death have gone on to argue
a different understanding of the distinction between that killing is not in itself morally different from al-
killing and allowing to die, what the physician does is lowing to die.° In this account, very roughly, one kills
allow to die. In this account, the mother's ALS is a lethal when one performs an action that causes the death of
disease whose normal progression is being held back a person (we are in a boat, you cannot swim, I push
or blocked by the life-sustaining respirator treatment. you overboard, and you drown), and one allows to die
Removing this artificial intervention is then viewed as when one has the ability and opportunity to prevent
standing aside and allowing the patient to die of her the death of another, knows this, and omits doing so,
underlying disease. I have argued elsewhere that this with the result that the person dies (we are in a boat,
alternative account is deeply problematic, in part be- you cannot swim, you fall overboard, I don't throw you
cause it commits us to accepting that what the greedy an available life ring, and you drown). Those who see
no moral difference between killing and allowing to a valued future, and of all the person wanted and
die typically employ the strategy of comparing cases planned to do in that future.
that differ in these and no other potentially morally A natural expression of this account of the
important respects. This will allow people to consider wrongness of killing is that people have a moral right
whether the mere difference that one is a case of kill- not to be killed.9 But in this account of the wrongness
ing and the other of allowing to die matters morally, of killing, the right not to be killed, like other rights,
or whether instead it is other features that make most should be waivable when the person makes a compe-
cases of killing worse than most instances of allowing tent decision that continued life is no longer wanted
to die. Here is such a pair of cases: or a good, but is instead worse than no further life at
all. In this view, euthanasia is properly understood as
Case 1. A very gravely ill patient is brought to a a case of a person having waived his or her right not
hospital emergency room and sent up to the icu. to be killed.
The patient begins to develop respiratory failure This rights view of the wrongness of killing is not,
that is likely to require intubation very soon. At of course, universally shared. Many people's moral views
that point the patient's family members and long- about killing have their origins in religious views that
standing physician arrive at the icu and inform the human life comes from God and cannot be justifiably
Ku staff that there had been extensive discussion destroyed or taken away, either by the person whose life
about future care with the patient when he was it is or by another. But in a pluralistic society like our
unquestionably competent. Given his grave and own, with a strong commitment to freedom of religion,
terminal illness, as well as his state of debilitation, public policy should not be grounded in religious beliefs
the patient had firmly rejected being placed on a which many in that society reject. I turn now to the gen-
respirator under any circumstances, and the family eral evaluation of public policy on euthanasia.
and physician produce the patient's advance direc-
tive to that effect. The icu staff do not intubate the Would the Bad Consequences of Euthanasia
patient, who dies of respiratory failure. Outweigh the Good?
Case 2. The same as Case 1 except that the family The argument against euthanasia at the policy level is
and physician are slightly delayed in traffic and stronger than at the level of individual cases, though
arrive shortly after the patient has been intubated even here I believe the case is ultimately unpersuasive,
and placed on the respirator. The icu staff extubate or at best indecisive. The policy level is the place where
the patient, who dies of respiratory failure. the main issues lie, however, and where moral consider-
ations that might override arguments in favour of eutha-
In Case 1 the patient is allowed to die, in Case 2 he is nasia will be found, if they are found anywhere. It is
killed, but it is hard to see why what is done in Case 2 important to note two kinds of disagreement about the
is significantly different morally than what is done in consequences for public policy of permitting euthanasia.
Case 1. It must be other factors that make most killings First, there is empirical or factual disagreement about
worse than most allowings to die, and if so, euthanasia what the consequences would be. This disagreement is
cannot be wrong simply because it is killing instead of greatly exacerbated by the lack of firm data on the issue.
allowing to die. Second, since on any reasonable assessment there would
Suppose both my arguments are mistaken. Sup- be both good and bad consequences, there are moral
pose that killing is worse than allowing to die and disagreements about the relative importance of different
that withdrawing life support is not killing, although effects. In addition to these two sources of disagreement,
euthanasia is. Euthanasia still need not for that reason there is also no single, well-specified policy proposal for
be morally wrong. To see this, we need to determine legalizing euthanasia on which policy assessments can
the basic principle for the moral evaluation of kill- focus. But without such specification, and especially
ing persons. What is it that makes paradigm cases of without explicit procedures for protecting against well-
wrongful killing wrongful? One very plausible answer intentioned misuse and ill-intentioned abuse, the conse-
is that killing denies the victim something that he or quences for policy are largely speculative. Despite these
she values greatly—continued life or a future. More- difficulties, a preliminary account of the main likely
over, since continued life is necessary for pursuing good and bad consequences is possible. This should
any of a person's plans and purposes, killing brings help clarify where better data or more moral analysis
the frustration of all of these plans and desires as well. and argument are needed, as well as where policy safe-
In a nutshell, wrongful killing deprives a person of guards must be developed.
II j11 IIILI
Potential Good Consequences of A third good consequence of the legalization of

Permitting Euthanasia euthanasia concerns patients whose dying is filled
What are the likely good consequences? First, if eutha- with severe and unrelievable pain or suffering. When
nasia were permitted it would be possible to respect the there is a life-sustaining treatment that, if forgone,
self-determination of competent patients who want it, will lead relatively quickly to death, then doing so
but now cannot get it because of its illegality. We simply can bring an end to these patients' suffering without
do not know how many such patients and people there recourse to euthanasia. For patients receiving no such
are. In the Netherlands, with a population of about 14.5 treatment, however, euthanasia may be the only re-
million (in 1987), estimates in a recent study were that lease from their otherwise prolonged suffering and
about 1,900 cases of voluntary active euthanasia or agony. This argument from mercy has always been the
physician-assisted suicide occur annually. No straight- strongest argument for euthanasia in those cases to
forward extrapolation to the United States is possible which it applies."
for many reasons, among them, that we do not know The importance of relieving pain and suffering is
how many people here who want euthanasia now get it, less controversial than is the frequency with which pa-
despite its illegality. Even with better data on the number tients are forced to undergo untreatable agony that only
of persons who want euthanasia but cannot get it, sig- euthanasia could relieve. If we focus first on suffering
nificant moral disagreement would remain about how caused by physical pain, it is crucial to distinguish pain
much weight should be given to any instance of failure that could be adequately relieved with modern methods
to respect a person's self-determination in this way. of pain control, though it in fact is not, from pain that is
One important factor substantially affecting the relievable only by death." For a variety of reasons, in-
number of persons who would seek euthanasia is the cluding some physicians' fear of hastening the patient's
extent to which an alternative is available. The widespread death, as well as the lack of a publicly accessible means
acceptance in the law, social policy, and medical practice for assessing the amount of the patient's pain, many
of the right of a competent patient to forgo life-sustaining patients suffer pain that could be, but is not, relieved.
treatment suggests that the number of competent persons Specialists in pain control, as for example the pain
in the United States who would want euthanasia if it were of terminally ill cancer patients, argue that there are
permitted is probably relatively small. very few patients whose pain could not be adequately
A second good consequence of making euthanasia controlled, though sometimes at the cost of so sedating
legally permissible benefits a much larger group. Polls them that they are effectively unable to interact with
have shown that a majority of the American public be- other people or their environment. Thus, the argument
lieves that people should have a right to obtain eutha- from mercy in cases of physical pain can probably be
nasia if they want it.m No doubt the vast majority of met in a large majority of cases by providing adequate
those who support this right to euthanasia will never in measures of pain relief. This should be a high priority,
fact come to want euthanasia for themselves. Neverthe- whatever our legal policy on euthanasia—the relief of
less, making it legally permissible would reassure many pain and suffering has long been, quite properly, one
people that if they ever do want euthanasia they would of the central goals of medicine. Those cases in which
be able to obtain it. This reassurance would supple- pain could be effectively relieved, but in fact is not,
ment the broader control over the process of dying should only count significantly in favour of legalizing
given by the right to decide about life-sustaining treat- euthanasia if all reasonable efforts to change pain man-
ment. Having fire insurance on one's house benefits all agement techniques have been tried and have failed.
who have it, not just those whose houses actually burn Dying patients often undergo substantial psycho-
down, by reassuring them that in the unlikely event of logical suffering that is not fully or even principally the
their house burning down, they will receive the money result of physical pain." The knowledge about how to
needed to rebuild it. Likewise, the legalization of eutha- relieve this suffering is much more limited than in the
nasia can be thought of as a kind of insurance policy case of relieving pain, and efforts to do so are probably
against being forced to endure a protracted dying more often unsuccessful. If the argument from mercy is
process that one has come to find burdensome and extended to patients experiencing great and unreliev-
unwanted, especially when there is no life-sustaining able psychological suffering, the numbers of patients to
treatment to forgo. The strong concern about losing which it applies are much greater.
control of their care expressed by many people who One last good consequence of legalizing euthanasia
face serious illness likely to end in death suggests that is that once death has been accepted, it is often more
they give substantial importance to the legalization of humane to end life quickly and peacefully, when that
euthanasia as a means of maintaining this control. is what the patient wants. Such a death will often be
seen as better than a more prolonged one. People who killing of patients and that we shall take disciplin-
suffer a sudden and unexpected death, for example by ary action against doctors who kill. And we must
dying quickly or in their sleep from a heart attack or say to the broader community that if it insists on
stroke, are often considered lucky to have died in this tolerating or legalizing active euthanasia, it will
way. We care about how we die in part because we care have to find nonphysicians to do its killing.14
about how others remember us, and we hope they will
remember us as we were in "good times" with them and If permitting physicians to kill would undermine
not as we might be when disease has robbed us of our the very "moral centre" of medicine, then almost cer-
dignity as human beings. As with much in the treat- tainly physicians should not be permitted to perform
ment and care of the dying, people's concerns differ in euthanasia. But how persuasive is this claim? Patients
this respect, but for at least some people, euthanasia should not fear, as a consequence of permitting vol-
will be a more humane death than what they have often untary active euthanasia, that their physicians will
experienced with other loved ones and might other- substitute a lethal injection for what patients want
wise expect for themselves. and believe is part of their care. If active euthanasia is
Some opponents of euthanasia challenge how restricted to cases in which it is truly voluntary, then
much importance should be given to any of these no patient should fear getting it unless she or he has
good consequences of permitting it, or even whether voluntarily requested it. (The fear that we might in time
some would be good consequences at all. But more fre- also come to accept nonvoluntary, or even involuntary
quently, opponents cite a number of bad consequences active euthanasia is a slippery slope worry I address
that permitting euthanasia would or could produce, below.) Patients' trust of their physicians could be in-
and it is to their assessment that I now turn. creased, not eroded, by knowledge that physicians will
provide aid in dying when patients seek it.
Potential Bad Consequences of Might Gaylin and his colleagues nevertheless be
Permitting Euthanasia correct in their claim that the moral centre of medicine
Some of the arguments against permitting euthanasia would collapse if physicians were to become killers?
are aimed specifically against physicians, while others This question raises what at the deepest level should be
are aimed against anyone being permitted to perform the guiding aims of medicine, a question that obviously
it. I shall first consider one argument of the former cannot be fully explored here. But I do want to say
sort. Permitting physicians to perform euthanasia, it enough to indicate the direction that I believe an appro-
is said, would be incompatible with their fundamental priate response to this challenge should take. In spelling
moral and professional commitment as healers to care out above what I called the positive argument for vol-
for patients and to protect life. Moreover, if euthanasia untary active euthanasia, I suggested that two principal
by physicians became common, patients would come values—respecting patients' self-determination and pro-
to fear that a medication was intended not to treat or moting their well-being—underlie the consensus that
care, but instead to kill, and would thus lose trust in competent patients, or the surrogates of incompetent
their physicians. This position was forcefully stated in patients, are entitled to refuse any life-sustaining treat-
a paper by Willard Gaylin and his colleagues: ment and to choose from among available alternative
treatments. It is the commitment to these two values in
The very soul of medicine is on trial. . . . This issue guiding physicians' actions as healers, comforters, and
touches medicine at its moral center, if this moral protectors of their patients' lives that should be at the
center collapses, if physicians become killers or are "moral centre" of medicine, and these two values sup-
even licensed to kill, the profession—and, there- port physicians' administering euthanasia when their
with, each physician—will never again be worthy patients make competent requests for it.
of trust and respect as healer and comforter and What should not be at that moral centre is a com-
protector of life in all its frailty. mitment to preserving patients' lives as such, without
regard to whether those patients want their lives pre-
These authors go on to make clear that, while they served or judge their preservation a benefit to them.
oppose permitting anyone to perform euthanasia, their Vitalism has been rejected by most physicians, and
special concern is with physicians doing so: despite some statements that suggest it, is almost cer-
tainly not what Gaylin and colleagues intended. One
We call on fellow physicians to say that they will of them, Leon Kass, has elaborated elsewhere the view
not deliberately kill. We must also say to each that medicine is a moral profession whose proper aim
of our fellow physicians that we will not tolerate is "the naturally given end of health," understood as
Al
II 111 111111
the wholeness and well-working of the human being progress made in securing the rights of patients or their
"for the physician, at least, human life in living bodies surrogates to decide about and to refuse life-sustaining
commands respect and reverence—by its very nature." treatment." This progress has been made against the
Kass continues, "the deepest ethical principle restrain- backdrop of a clear and firm legal prohibition of eutha-
ing the physician's power is not the autonomy or free- nasia, which has provided a relatively bright line limit-
dom of the patient; neither is it his own compassion or ing the dominion of others over patients' lives. It has
good intention. Rather, it is the dignity and mysterious therefore been an important reassurance to concerns
power of human life itself."" I believe Kass is in the about how the authority to take steps ending life might
end mistaken about the proper account of the aims of be misused, abused, or wrongly extended.
medicine and the limits on physicians' power, but this Many supporters of the right of patients or their
difficult issue will certainly be one of the central themes surrogates to refuse treatment strongly oppose eutha-
in the continuing debate about euthanasia. nasia, and if forced to choose might well withdraw their
A second bad consequence that some foresee is that support of the right to refuse treatment rather than
permitting euthanasia would weaken society's commit- accept euthanasia. Public policy in the last 15 years
ment to provide optimal care for dying patients. We has generally let life-sustaining treatment decisions be
live at a time in which the control of health care costs made in health care settings between physicians and pa-
has become, and is likely to continue to be, the dom- tients or their surrogates, and without the involvement
inant focus of health care policy. If euthanasia is seen of the courts. However, if euthanasia is made legally
as a cheaper alternative to adequate care and treatment, permissible greater involvement of the courts is likely,
then we might become less scrupulous about providing which could in turn extend to a greater court involve-
sometimes costly support and other services to dying ment in life-sustaining treatment decisions. Most agree,
patients. Particularly if our society comes to embrace however, that increased involvement of the courts in
deeper and more explicit rationing of health care, frail, these decisions would be undesirable, as it would make
elderly, and dying patients will need to be strong and sound decision-making more cumbersome and diffi-
effective advocates for their own health care and other cult without sufficient compensating benefits.
needs, although they are hardly in a position to do this. As with the second potential bad consequence
We should do nothing to weaken their ability to obtain of permitting euthanasia, this third consideration too
adequate care and services. is speculative and difficult to assess. The feared ero-
This second worry is difficult to assess because sion of patients' or surrogates' rights to decide about
there is little firm evidence about the likelihood of the life-sustaining treatment, together with greater court
feared erosion in the care of dying patients. There are involvement in those decisions, are both possible.
at least two reasons, however, for skepticism about this However, I believe there is reason to discount this gen-
argument. The first is that the same worry could have eral worry. The legal rights of competent patients and,
been directed at recognizing patients' or surrogates' to a lesser degree, surrogates of incompetent patients
rights to forgo life-sustaining treatment, yet there is no to decide about treatment are very firmly embedded
persuasive evidence that recognizing the right to refuse in a long line of informed consent and life-sustaining
treatment has caused a serious erosion in the quality of treatment cases, and are not likely to be eroded by a
care of dying patients. The second reason for skepti- debate over, or even acceptance of, euthanasia It will
cism about this worry is that only a very small propor- not be accepted without safeguards that reassure the
tion of deaths would occur from euthanasia if it were public about abuse, and if that debate shows the need
permitted. In the Netherlands, where euthanasia under for similar safeguards for some life-sustaining treat-
specified circumstances is permitted by the courts, ment decisions they should be adopted there as well.
though not authorized by statute, the best estimate of In neither case are the only possible safeguards greater
the proportion of overall deaths that result from it is court involvement, as the recent growth of institutional
about 2 per cent.16 Thus, the vast majority of critically ethics committees shows.
ill and dying patients will not request it, and so will still The fourth potential bad consequence of permit-
have to be cared for by physicians, families, and others. ting euthanasia has been developed by David Velleman
Permitting euthanasia should not diminish people's and turns on the subtle point that making a new option
commitment and concern to maintain and improve the or choice available to people can sometimes make them
care of these patients. worse off, even if once they have the choice they go on
A third possible bad consequence of permitting to choose what is best for them." Ordinarily, people's
euthanasia (or even a public discourse in which strong continued existence is viewed by them as given, a
support for euthanasia is evident) is to threaten the fixed condition with which they must cope. Making
euthanasia available to people as an option denies them argued, then the court cases permitting such killing
the alternative of staying alive by default. If people have already in effect weakened this prohibition. How-
are offered the option of euthanasia, their continued ever, neither the courts nor most people have seen these
existence is now a choice for which they can be held cases as killing and so as challenging the prohibition of
responsible and which they can be asked by others homicide. The courts have usually grounded patients'
to justify. We care, and are right to care, about being or their surrogates' rights to refuse life-sustaining treat-
able to justify ourselves to others. To the extent that ment in rights to privacy, liberty, self-determination, or
our society is unsympathetic to justifying a severely de- bodily integrity, not in exceptions to homicide laws.
pendent or impaired existence, a heavy psychological Legal permission for physicians or others to per-
burden of proof may be placed on patients who think form euthanasia could not be grounded in patients'
their terminal illness or chronic infirmity is not a suffi- rights to decide about medical treatment. Permitting
cient reason for dying. Even if they otherwise view their euthanasia would require qualifying, at least in effect,
life as worth living, the opinion of others around them the legal prohibition against homicide, a prohibi-
that it is not can threaten their reason for living and tion that in general does not allow the consent of the
make euthanasia a rational choice. Thus the existence victim to justify or excuse the act. Nevertheless, the
of the option becomes a subtle pressure to request it. very same fundamental basis of the right to decide
This argument correctly identifies the reason about life-sustaining treatment—respecting a person's
why offering some patients the option of euthanasia self-determination—does support euthanasia as well.
would not benefit them. Velleman takes it not as a Individual self-determination has long been a well-
reason for opposing all euthanasia, but for restricting entrenched and fundamental value in the law, and so
it to circumstances where there are "unmistakable and extending it to euthanasia would not require appeal
overpowering reasons for persons to want the option to novel legal values or principles. That suicide or at-
of euthanasia," and for denying the option in all other tempted suicide is no longer a criminal offence in vir-
cases. But there are at least three reasons why such re- tually all states indicates an acceptance of individual
striction may not be warranted. First, polls and other self-determination in the taking of one's own life analo-
evidence support that most Americans believe eutha- gous to that required for voluntary active euthanasia.
nasia should be permitted (though the recent defeat of The legal prohibition (in most states) of assisting in
the referendum to permit it in the state of Washington suicide and the refusal in the law to accept the consent
raises some doubt about this support). Thus, many of the victim as a possible justification of homicide are
more people seem to want the choice than would be both arguably a result of difficulties in the legal process
made worse off by getting it. Second, if giving people of establishing the consent of the victim after the fact.
the option of ending their life really makes them worse If procedures can be designed that clearly establish the
off, then we should not only prohibit euthanasia, but voluntariness of the person's request for euthanasia it
also take back from people the right they now have would under those procedures represent a carefully
to decide about-life-sustaining treatment. The feared circumscribed qualification on the legal prohibition of
harmful effect should already have occurred from homicide. Nevertheless, some remaining worries about
securing people's right to refuse life-sustaining treat- this weakening can be captured in the final potential
ment, yet there is no evidence of any such widespread bad consequence, to which I will now turn.
harm or any broad public desire to rescind that right. This final potential bad consequence is the cen-
Third, since there is a wide range of conditions in which tral concern of many opponents of euthanasia and, I
reasonable people can and do disagree about whether believe, is the most serious objection to a legal policy
they would want continued life, it is not possible to permitting it. According to this "slippery slope" worry,
restrict the permissibility of euthanasia as narrowly as although active euthanasia may be morally permissible
Velleman suggests without thereby denying it to most in cases in which it is unequivocally voluntary and the
persons who would want it; to permit it only in cases patient finds his or her condition unbearable, a legal
in which virtually everyone would want it would be to policy permitting euthanasia would inevitably lead
deny it to most who would want it. to active euthanasia being performed in many other
A fifth potential bad consequence of making cases in which it would be morally wrong. To prevent
euthanasia legally permissible is that it might weaken those other wrongful cases of euthanasia we should not
the general legal prohibition of homicide. This prohibi- permit even morally justified performance of it.
tion is so fundamental to civilized society, it is argued, Slippery slope arguments of this form are problem-
that we should do nothing that erodes it. If most cases atic and difficult to evaluate.19 From one perspective,
of stopping life support are killing, as I have already they are the last refuge of conservative defenders of the
II w 11111I
status quo. When all the opponent's objections to the 4. A psychiatric evaluation should ensure that the
wrongness of euthanasia itself have been met, the op- patient's request is not the result of a treatable psy-
ponent then shifts ground and acknowledges both that chological impairment such as depression.2°
it is not in itself wrong and that a legal policy which
resulted only in its being performed would not be bad. These examples of procedural safeguards are all
Nevertheless, the opponent maintains, it should still designed to ensure that the patient's choice is fully
not be permitted because doing so would result in its informed, voluntary, and competent, and so a true
being performed in other cases in which it is not vol- exercise of self-determination. Other proposals for
untary and would be wrong. In this argument's most euthanasia would restrict its permissibility further—for
extreme form, permitting euthanasia is the first and example, to the terminally ill—a restriction that cannot
fateful step down the slippery slope to Nazism. Once be supported by self-determination. Such additional
on the slope we will be unable to get off. restrictions might, however, be justified by concern
Now it cannot be denied that it is possible that for limiting potential harms from abuse. At the same
permitting euthanasia could have these fateful conse- time, it is important not to impose procedural or sub-
quences, but that cannot be enough to warrant pro- stantive safeguards so restrictive as to make euthanasia
hibiting it if it is otherwise justified. A similar possible impermissible or practically infeasible in a wide range
slippery slope worry could have been raised to securing of justified cases.
competent patients' rights to decide about life support, These examples of procedural safeguards make
but recent history shows such a worry would have clear that it is possible to substantially reduce, though
been unfounded. It must be relevant how likely it is not to eliminate, the potential for abuse of a policy
that we will end with horrendous consequences and an permitting voluntary active euthanasia. Any legal-
unjustified practice of euthanasia How likely, and wide- ization of the practice should be accompanied by a
spread, would the abuses and unwarranted extensions well-considered set of procedural safeguards together
of permitting it be? By abuses, I mean the perform- with an ongoing evaluation of its use. Introducing
ance of euthanasia that fails to satisfy the conditions euthanasia into only a few states could be a form of
required for voluntary active euthanasia, for example, carefully limited and controlled social experiment that
if the patient has been subtly pressured to accept it. By would give us evidence about the benefits and harms of
unwarranted extensions of policy, I mean later changes the practice. Even then, firm and uncontroversial data
in legal policy to permit not just voluntary euthanasia, may remain elusive, as the continuing controversy over
but also euthanasia in cases in which, for example, it what has taken place in the Netherlands in recent years
need not be fully voluntary. Opponents of voluntary indicates.2'
euthanasia on slippery slope grounds have not provided
the data or evidence necessary to turn their speculative The Slip into Nonvoluntary
concerns into well-grounded likelihoods. Active Euthanasia
It is at least clear, however, that both the character
and likelihood of abuses of a legal policy permitting While I believe slippery slope worries can largely be lim-
euthanasia depend in significant part on the proced- ited by making necessary distinctions both in principle
ures put in place to protect against them. I will not try and in practice, one slippery slope concern is legitimate.
to detail fully what such procedures might be, but will There is reason to expect that legalization of voluntary
just give some examples of what they might include: active euthanasia might soon be followed by strong
pressure to legalize some nonvoluntary euthanasia of in-
1. The patient should be provided with all relevant competent patients unable to express their own wishes.
information about his or her medical condition, Respecting a person's self-determination and recognizing
current prognosis, available alternative treatments, that continued life is not always of value to a person can
and the prognosis of each. support not only voluntary active euthanasia, but some
2. Procedures should ensure that the patient's request nonvoluntary euthanasia as well. These are the same
for euthanasia is stable or enduring (a brief wait- values that ground competent patients' right to refuse life-
ing period could be required) and fully voluntary sustaining treatment. Recent history here is instructive. In
(an advocate for the patient might be appointed to the medical ethics literature, in the courts since Quinlan,
ensure this). and in norms of medical practice, that right has been
3. All reasonable alternatives must have been explored extended to incompetent patients and exercised by a sur-
for improving the patient's quality of life and reliev- rogate who is to decide as the patient would have decided
ing any pain or suffering. in the circumstances if competent? It has been held
unreasonable to continue life-sustaining treatment that the Netherlands families of incompetent patients have
the patient would not have wanted just because the pa- less authority than do families in the United States to
tient now lacks the capacity to tell us that. Life-sustaining act as surrogates for incompetent patients in making
treatment for incompetent patients is today frequently decisions to forgo life-sustaining treatment.24 From the
forgone on the basis of a surrogate's decision, or less fre- Dutch perspective, it may be we in the United States
quently on the basis of an advance directive executed by who are already on the slippery slope in having given
the patient while still competent. The very same logic that surrogates broad authority to forgo life-sustaining treat-
has extended the right to refuse life-sustaining treatment ment for incompetent persons. In this view, the more
from a competent patient to the surrogate of an incompe- important moral divide, and the more important with
tent patient (acting with or without a formal advance regard to potential for abuse, is not between forgoing
directive from the patient) may well extend the scope of life-sustaining treatment and euthanasia, but instead
active euthanasia. The argument will be: Why continue between voluntary and nonvoluntary performance of
to force unwanted life on patients just because they have either. If this is correct, then the more important issue
now lost the capacity to request euthanasia from us? is ensuring the appropriate principles and procedural
A related phenomenon may reinforce this slippery safeguards for the exercise of decision-making authority
slope concern. In the Netherlands, what the courts by surrogates for incompetent persons in all decisions
have sanctioned has been clearly restricted to voluntary at the end of life. This may be the correct response to
euthanasia. In itself, this serves as some evidence that slippery slope worries about euthanasia.
permitting it need not lead to permitting the nonvolun- I have cited both good and bad consequences
tary variety. There is some indication, however, that for that have been thought likely from a policy change
many Dutch physicians euthanasia is no longer viewed permitting voluntary active euthanasia, and have tried
as a special action, set apart from their usual practice to evaluate their likelihood and relative importance.
and restricted only to competent persons.23 Instead, it Nevertheless, as I noted earlier, reasonable disagree-
is seen as one end of a spectrum of caring for dying ment remains both about the consequences of permit-
patients. When viewed in this way it will be difficult to ting euthanasia and about which of these consequences
deny euthanasia to a patient for whom it is seen as the are more important. The depth and strength of public
best or most appropriate form of care simply because and professional debate about whether, all things con-
that patient is now incompetent and cannot request it. sidered, permitting euthanasia would be desirable or
Even if voluntary active euthanasia should slip into undesirable reflects these disagreements. While my
nonvoluntary active euthanasia, with surrogates acting own view is that the balance of considerations supports
for incompetent patients, the ethical evaluation is more permitting the practice, my principal purpose here has
complex than many opponents of euthanasia allow. been to clarify the main issues.
Just as in the case of surrogates' decisions to forgo life-
sustaining treatment for incompetent patients, so also The Role of Physicians
surrogates' decisions to request euthanasia for incompe-
tent persons would often accurately reflect what the in- If euthanasia is made legally permissible, should phys-
competent person would have wanted and would deny icians take part in it? Should only physicians be per-
the person nothing that he or she would have considered mitted to perform it, as is the case in the Netherlands?
worth having. Making nonvoluntary active euthanasia In discussing whether euthanasia is incompatible with
legally permissible, however, would greatly enlarge the medicine's commitment to curing, caring for, and com-
number of patients on whom it might be performed and forting patients, I argued that it is not at odds with a
substantially enlarge the potential for misuse and abuse. proper understanding of the aims of medicine, and so
As noted above, frail and debilitated elderly people, need not undermine patients' trust in their physicians.
often demented or otherwise incompetent and thereby If that argument is correct, then physicians probably
unable to defend and assert their own interests, may be should not be prohibited, either by law or by profes-
especially vulnerable to unwanted euthanasia. sional norms, from taking part in a legally permissible
For some people, this risk is more than sufficient practice of euthanasia (nor, of course, should they be
reason to oppose the legalization of voluntary eutha- compelled to do so if their personal or professional
nasia. But while we should in general be cautious about scruples forbid it). Most physicians in the Netherlands
inferring much from the experience in the Netherlands appear not to understand euthanasia to be incompat-
to what our own experience in the United States might ible with their professional commitments.
be, there may be one important lesson that we can Sometimes patients who would be able to end
learn from them. One commentator has noted that in their lives on their own nevertheless seek the assistance
II ill IIILL1
of physicians. Physician involvement in such cases may physicians would inevitably be involved in some of the
have important benefits to patients and others beyond important procedural safeguards necessary to a defens-
simply assuring the use of effective means. Historically, ible practice, such as seeing to it that the patient is well
in the United States suicide has carried a strong negative informed about his or her condition, prognosis, and
stigma that many today believe unwarranted. Seeking a possible treatments, and ensuring that all reasonable
physician's assistance, or what can almost seem a phys- means have been taken to improve the quality of the
ician's blessing, may be a way of trying to remove that patient's life. Second, and probably more important, one
stigma and show others that the decision for suicide was necessary protection against abuse of the practice is to
made with due seriousness and was justified under the limit the persons given authority to perform it, so that
circumstances. The physician's involvement provides a they can be held accountable for their exercise of that
kind of social approval, or more accurately helps counter authority. Physicians, whose training and professional
what would otherwise be unwarranted social disapproval. norms give some assurance that they would perform
There are also at least two reasons for restrict- euthanasia responsibly, are an appropriate group of per-
ing the practice of euthanasia to physicians only. First, sons to whom the practice may be restricted.
Notes
1. President's Commission for the Study o f Ethical Problems Tooley, (1983), Abortion and Infanticide (Oxford
in Medicine and Biomedical and Behavioral Research, University Press: Oxford). In my paper "Taking Human
(1983), Deciding to Forego Life-Sustaining Treatment (US Life," (1985), in Ethics 95: 851-65, I argue in more
Government Printing Office Washington, DC); The detail that killing in itself is not morally different from
Hastings Center, (1987), Guidelines on the Termination of allowing to die and defend the strategy of argument
life Sustaining Treatment and Care of the Dying (Indiana employed in this and the succeeding two paragraphs in
University Press: Bloomington); "Current Opinions the text.
of the Council on Ethical and Judicial Affairs of the 9. Dan W. Brock, (1979), "Moral Rights and Permissible
American Medical Association-1989: Withholding Killing," in Ethical Issues Relating to Life and Death, ed. John
or Withdrawing Life-Prolonging Treatment" (1989; Ladd (Oxford University Press: New York), 94-117.
American Medical Association: Chicago); George Annas 10. P Painton and E Taylor, (1990), "Love or Let Die," in Time
and Leonard Glantz, (1986), "The Right of Elderly (19 March): 62-71; Boston Globe/Harvard University Poll,
Patients to Refuse Life-Sustaining Treatment," in Millbank (1991), in Boston Globe (3 Nov.).
Memorial Quarterly 64 (suppl. 2): 95-162; Robert E. 11. James Rachels, (1986), The End of life (Oxford University
Weir, (1989), Abating Treatment with Critically Ill Patients Press: Oxford).
(Oxford University Press: New York); Sidney J. Wanzer, 12. Marcia Angell, (1982), "The Quality of Mercy," in New England
et al., (1984), "The Physician's Responsibility toward Journal of Medicine 306: 98-9; M Donovan, P Dillon, and L
Hopelessly Ill Patients," in New England Medical Journal Mcguire, (1987), "Incidence and Characteristics of Pain in a
310: 955-9. Sample of Medical-Surgical Inpatients," in Pain 30: 69-78.
2. M.A.M. de Wachter, (1989), "Active Euthanasia in the 13. Eric Cassell, (1991), The Nature of Suffering and the Goals of
Netherlands," in jama 262(23): 3315-19. Medicine (Oxford University Press: New York).
3. Anonymous, (1988), "It's Over, Debbie," in JAMA 259: 272; 14. Gaylin, et al., "Doctors Must Not Kill."
Timothy E. Quill, (1990), "Death and Dignity," in New 15. Leon R. Kass, (1989), "Neither for Love Nor Money: Why
England Journal of Medicine 322: 1881-3. Doctors Must Not Kill," in The Public Interest 94: 25-46;
4. Wanzer, et al., (1989), "The Physicianb Responsibility cf. also his Toward a More Natural Science: Biology and
toward Hopelessly Ill Patients: A Second Look," in New Human Affairs (1985; The Free Press: New York), ch. 6-9.
England Journal of Medicine 320: 844-9. 16. Paul J. Van der Maas, et al., (1991), "Euthanasia and Other
5. Willard Gaylin, Leon R. Kass, Edmund D. Pellegrino, and Medical Decisions Concerning the End of Life," in Lancet
Mark Siegler, (1988), "Doctors Must Not Kill," in JAMA 338: 669-74.
259: 2139-40. 17. Susan M. Wolf, (1989), "Holding the Line on Euthanasia,"
6. Bonnie Steinbock, ed., (1980), Killing and Allowing to Die in Hastings Center Report 19(1 [special supplement]): 13-15.
(Prentice Hall: Englewood Cliffs, NJ). 18. My formulation of this argument derives from David
7. Dan W. Brock, (1986), "Forgoing Food and Water: Is It Velleman's statement of it in his commentary on an earlier
Killing?," in By No Extraordinary Means: The Choice to Forgo version of this paper delivered at the American Philosophical
Life-Sustaining Food and Water, ed. Joanne Lynn (Indiana Association Central Division meetings; a similar point was
University Press: Bloomington), 117-31. made to me by Elisha Milgram in discussion on another occa-
8. James Rachels, (1975), "Active and Passive Euthanasia," sion. For more general development of the point see Thomas
in New England Journal of Medicine 292: 78-80; Michael Schelling, (1960), The Strategy of Conflict (Harvard University
Rachels: Active and Passive Euthanasia 309
Press: Cambridge, MA); and Gerald Dworkin, (1988), 21. Richard Fenigsen, (1989), "A Case against Dutch
"Is More Choice Better Than Less?," in The Theory and Practice Euthanasia," in Hastings Center Report 19(1 [special
of Autonomy (Cambridge University Press: Cambridge). supplement]): 22-30.
19. Frederick Schauer, (1985), "Slippery Slopes," in Harvard 22. Allen E. Buchanan and Dan W Brock, (1989), Deciding for
Law Review 99: 361-83; Wibren van der Burg, (Oct. Others: The Ethics of Surrogate Decisionmaleing (Cambridge
1991), "The Slippery Slope Argument," in Ethics 102: University Press Cambridge).
42-65. 23. Van der Maas, et al., "Euthanasia and Other Medical Decisions."
20. There is evidence that physicians commonly fail to diagnose 24. Margaret P Battin, (1990), "Seven Caveats Concerning the
depression. See Robert L Misbin, (1991), "Physicians Aid in Discussion of Euthanasia in Holland," in American Philosophical
Dying," in New England Journal of Medicine 325: 1304-7. Association Newsletter on Philosophy and Medicine 89 (2).
Active and Passive Euthanasia So he asks the doctor for an end to it, and his family
joins in the request.
James Rachels Suppose the doctor agrees to withhold treat-
ment, as the conventional doctrine says he may. The
The distinction between active and passive euthanasia justification for his doing so is that the patient is in
is thought to be crucial for medical ethics. The idea is terrible agony, and since he is going to die anyway, it
that it is permissible, at least in some cases, to with- would be wrong to prolong his suffering needlessly.
hold treatment and allow a patient to die, but it is never But now notice this. If one simply withholds treat-
permissible to take any direct action designed to kill ment, it may take the patient longer to die, and so he
the patient. This doctrine seems to be accepted by most may suffer more than he would if more direct action
doctors, and it is endorsed in a statement adopted by were taken and a lethal injection given. This fact pro-
the House of Delegates of the American Medical As- vides strong reason for thinking that, once the initial
sociation on 4 December 1973: decision not to prolong his agony has been made,
active euthanasia is actually preferable to passive
The intentional termination of the life of one euthanasia, rather than the reverse. To say otherwise
human being by another—mercy killing—is is to endorse the option that leads to more suffering
contrary to that for which the medical profession rather than less, and is contrary to the humanitarian
stands and is contrary to the policy of the Amer- impulse that prompts the decision not to prolong his
ican Medical Association. life in the first place.
The cessation of the employment of extra- Part of my point is that the process of being "al-
ordinary means to prolong the life of the body lowed to die" can be relatively slow and painful,
when there is irrefutable evidence that biological whereas being given a lethal injection is relatively
death is imminent is the decision of the patient quick and painless. Let me give a different sort of ex-
and/or his immediate family. The advice and judg- ample. In the United States about 1 in 600 babies is
ment of the physician should be freely available to born with Down's syndrome. Most of these babies are
the patient and/or his immediate family. otherwise healthy—that is, with only the usual pedi-
atric care, they will proceed to an otherwise normal
However, a strong case can be made against this doc- infancy. Some, however, are born with congenital de-
trine. In what follows I will set out some of the relevant fects such as intestinal obstructions that require oper-
arguments, and urge doctors to reconsider their views ations if they are to live. Sometimes, the parents and
on this matter. the doctor will decide not to operate, and let the infant
To begin with a familiar type of situation, a pa- die. Anthony Shaw describes what happens then:
tient who is dying of incurable cancer of the throat is
in terrible pain, which can no longer be satisfactorily When surgery is denied [the doctor] must try
alleviated. He is certain to die within a few days, even if to keep the infant from suffering while natural
present treatment is continued, but he does not want to forces sap the baby's life away. As a surgeon whose
go on living for those days since the pain is unbearable. natural inclination is to use the scalpel to fight
Source From The New England Journal of Medicine 292 (1975) (2), 78-80. Copyright © 1975 Massachusetts Medical Society. All
rights reserved.
■ Y I
I I .111 111141
off death, standing by and watching a salvage- "kill" it. The fact that this idea leads to such results as
able baby die is the most emotionally exhausting deciding life or death on irrelevant grounds is another
experience I know. It is easy at a conference, in a good reason why the doctrine should be rejected.
theoretical discussion, to decide that such infants One reason why so many people think that there is
should be allowed to die. It is altogether different an important moral difference between active and pas-
to stand by in the nursery and watch as dehydra- sive euthanasia is that they think killing someone is mor-
tion and infection wither a tiny being over hours ally worse than letting someone die. But is it? Is killing,
and days. This is a terrible ordeal for me and the in itself, worse than letting die? To investigate this issue,
hospital staff—much more so than for the parents two cases may be considered that are exactly alike except
who never set foot in the nursery' that one involves killing whereas the other involves let-
ting someone die. Then, it can be asked whether this dif-
I can understand why some people are opposed to ference makes any difference to the moral assessments. It
all euthanasia, and insist that such infants must be is important that the cases be exactly alike, except for this
allowed to live. I think I can also understand why one difference, since otherwise one cannot be confident
other people favour destroying these babies quickly that it is this difference and not some other that accounts
and painlessly. But why should anyone favour letting for any variation in the assessments of the two cases. So,
"dehydration and infection wither a tiny being over let us consider this pair of cases:
hours and days"? The doctrine that says that a baby In the first, Smith stands to gain a large inheritance
may be allowed to dehydrate and wither, but may not if anything should happen to his six-year-old cousin.
be given an injection that would end its life without One evening while the child is taking his bath, Smith
suffering, seems so patently cruel as to require no fur- sneaks into the bathroom and drowns the child, and
ther refutation. The strong language is not intended then arranges things so that it will look like an accident.
to offend, but only to put the point in the clearest In the second, Jones also stands to gain if anything
possible way. should happen to his six-year-old cousin. Like Smith,
My second argument is that the conventional doc- Jones sneaks in planning to drown the child in his
trine leads to decisions concerning life and death made bath. However, just as he enters the bathroom Jones
on irrelevant grounds. sees the child slip and hit his head, and fall face down
Consider again the case of the infants with Down in the water. Jones is delighted; he stands by ready to
syndrome who need operations for congenital defects push the child's head back under if it is necessary, but
unrelated to the syndrome to live. Sometimes, there is it is not necessary. With only a little thrashing about,
no operation, and the baby dies, but when there is no the child drowns all by himself, "accidentally," as Jones
such defect, the baby lives on. Now, an operation such watches and does nothing.
as that to remove an intestinal obstruction is not pro- Now, Smith killed the child, whereas Jones
hibitively difficult. The reason why such operations are "merely" let the child die. That is the only difference
not performed in these cases is, clearly, that the child between them. Did either man behave better, from a
has Down syndrome and the parents and doctor judge moral point of view? If the difference between kill-
that because of that fact it is better for the child to die. ing and letting die were in itself a morally important
But notice that this situation is absurd, no matter matter, one should say that Jones's behaviour was less
what view one takes of the lives and potentials of such reprehensible than Smith's. But does one really want
babies. If the life of such an infant is worth preserving, to say that? I think not. In the first place, both men
what does it matter if it needs a simple operation? Or, if acted from the same motive, personal gain, and both
one thinks it better that such a baby should not live on, had exactly the same end in view when they acted.
what difference does it make that it happens to have an It may be inferred from Smith's conduct that he is a
unobstructed intestinal tract? In either case, the matter of bad man, although that judgment may be withdrawn
life and death is being decided on irrelevant grounds. It or modified if certain further facts are learned about
is the Down syndrome, and not the intestines, that is the him—for example, that he is mentally deranged. But
issue. The matter should be decided, if at all, on that basis, would not the very same thing be inferred about Jones
and not be allowed to depend on the essentially irrelevant from his conduct? And would not the same further
question of whether the intestinal tract is blocked. considerations also be relevant to any modification of
What makes this situation possible, of course, is this judgment? Moreover, suppose Jones pleaded, in
the idea that when there is an intestinal blockage, one his own defence, "After all, I didn't do anything except
can "let the baby die," but when there is no such defect just stand there and watch the child drown. I didn't
there is nothing that can be done, for one must not kill him; I only let him die." Again, if letting die were
Rachels: Active and Passive Euthanasia
in itself less bad than killing, this defence should have I have argued that killing is not in itself any worse
at least some weight. But it does not. Such a "defence" than letting die; if my contention is right, it follows that
can only be regarded as a grotesque perversion of moral active euthanasia is not any worse than passive eutha-
reasoning. Morally speaking, it is no defence at all. nasia. What arguments can be given on the other side?
Now, it may be pointed out, quite properly, that the The most common, I believe, is the following
cases of euthanasia with which doctors are concerned
are not like this at all. They do not involve personal gain The important difference between active and
or the destruction of normal, healthy children. Doctors passive euthanasia is that, in passive eutha-
are concerned only with cases in which the patient's life nasia, the doctor does not do anything to
is of no further use to him, or in which the patient's life bring about the patient's death. The doctor
has become or will soon become a terrible burden. How- does nothing, and the patient dies of whatever
ever, the point is the same in these cases: the bare differ- ills already afflict him. In active euthanasia,
ence between killing and letting die does not, in itself, however, the doctor does something to bring
make a moral difference. If a doctor lets a patient die, about the patient's death: he kills him. The
for humane reasons, he is in the same moral position as doctor who gives the patient with cancer a
if he had given the patient a lethal injection for humane lethal injection has himself caused his patient's
reasons. If his decision was wrong—if, for example, the death; whereas if he merely ceases treatment,
patient's illness was in fact curable—the decision would the cancer is the cause of the death.
be equally regrettable no matter which method was used
to carry it out. And if the doctor's decision was the right A number of points need to be made here. The
one, the method used is not in itself important. first is that it is not exactly correct to say that in pas-
The AMA policy statement isolates the crucial sive euthanasia the doctor does nothing, for he does do
issue very well; the crucial issue is "the intentional one thing that is very important: he lets the patient die.
termination of the life of one human being by an- "Letting someone die" is certainly different, in some
other." But after identifying this issue, and forbidding respects, from other types of action—mainly in that it
"mercy killing," the statement goes on to deny that the is a kind of action that one may perform by way of
cessation of treatment is the intentional termination not performing certain other actions. For example, one
of life. This is where the mistake comes in, for what may let a patient die by way of not giving medication,
is the cessation of treatment, in these circumstances, just as one may insult someone by way of not shaking
if it is not "the intentional termination of the life of his hand. But for any purpose of moral assessment, it
one human being by another"? Of course it is exactly is a type of action nonetheless. The decision to let a
that, and if it were not, there would be no point to it. patient die is subject to moral appraisal in the same
Many people will find this judgment hard to way that a decision to kill him would be subject to
accept. One reason, I think, is that it is very easy to moral appraisal: it may be assessed as wise or unwise,
conflate the question of whether killing is, in itself, compassionate or sadistic, right or wrong. If a doctor
worse than letting die, with the very different ques- deliberately let a patient die who was suffering from a
tion of whether most actual cases of killing are more routinely curable illness, the doctor would certainly be
reprehensible than most actual cases of letting die. to blame for what he had done, just as he would be to
Most actual cases of killing are clearly terrible (think, blame if he had needlessly killed the patient. Charges
for example, of all the murders reported in the news- against him would then be appropriate. If so, it would
papers), and one hears of such cases every day On be no defence at all for him to insist that he didn't "do
the other hand, one hardly ever hears of a case of anything." He would have done something very serious
letting die, except for the actions of doctors who are indeed, for he let his patient die.
motivated by humanitarian reasons. So one learns to Fixing the cause of death may be very import-
think of killing in a much worse light than of letting ant from a legal point of view, for it may determine
die. But this does not mean that there is something whether criminal charges are brought against the
about killing that makes it in itself worse than let- doctor. But I do not think that this notion can be used
ting die, for it is not the bare difference between to show a moral difference between active and passive
killing and letting die that makes the difference in euthanasia. The reason why it is considered bad to be
these cases. Rather, the other factors—the murderer's the cause of someone's death is that death is regarded
motive of personal gain, for example, contrasted with as a great evil—and so it is. However, if it has been
the doctor's humanitarian motivation—account for decided that euthanasia—even passive euthanasia—is
different reactions to the different cases. desirable in a given case, it has also been decided that
uu I I I .111 111111
in this instance death is no greater an evil than the AMA policy statement that I have quoted, they are
patient's continued existence. And if this is true, the endorsing this doctrine as a central point of medical
usual reason for not wanting to be the cause of some- ethics. In that statement, active euthanasia is con-
one's death simply does not apply. demned not merely as illegal but as "contrary to that
Finally, doctors may think that all of this is only for which the medical profession stands," whereas
of academic interest—the sort of thing that phil- passive euthanasia is approved. However, the pre-
osophers may worry about but that has no practical ceding considerations suggest that there is really
bearing on their own work. After all, doctors must no moral difference between the two, considered
be concerned about the legal consequences of what in themselves (there may be important moral dif-
they do, and active euthanasia is clearly forbidden ferences in some cases in their consequences, but,
by the law. But even so, doctors should also be con- as I pointed out, these differences may make active
cerned with the fact that the law is forcing upon euthanasia, and not passive euthanasia, the morally
them a moral doctrine that may well be indefens- preferable option). So, whereas doctors may have to
ible, and has a considerable effect on their practices. discriminate between active and passive euthanasia
Of course, most doctors are not now in the position to satisfy the law, they should not do any more than
of being coerced in this matter, for they do not that. In particular, they should not give the distinc-
regard themselves as merely going along with what tion any added authority and weight by writing it
the law requires. Rather, in statements such as the into official statements of medical ethics.
Notes
1. Shaw, A. 1972. "Doctor, Do We Have a Choice?," in The

New York Times Magazine (30 Jan.): 54.
When Self-Determination Runs Amok individuals may, in the name of their own private, idio-
syncratic view of the good life, call upon others, includ-
Daniel Callahan
ing such institutions as medicine, to help them pursue
that life, even at the risk of harm to the common good.
The euthanasia debate is not just another moral debate, This works against the idea that the meaning and scope
one in a long list of arguments in our pluralistic society. of our own right to lead our own lives must be con-
It is profoundly emblematic of three important turn- ditioned by, and be compatible with, the good of the
ing points in Western thought. The first is that of the community, which is more than an aggregate of self-
legitimate conditions under which one person can kill directing individuals.
another. The acceptance of voluntary active euthanasia The third turning point is to be found in the claim
would morally sanction what can only be called "con- being made upon medicine: it should be prepared to make
senting adult killing." By that term I mean the killing of its skills available to individuals to help them achieve
one person by another in the name of their mutual right their private vision of the good life. This puts medicine
to be killer and killed if they freely agree to play those in the business of promoting the individualistic pursuit
roles. This turn flies in the face of a longstanding effort of general human happiness and well-being. It would
to limit the circumstances under which one person can overturn the traditional belief that medicine should limit
take the life of another, from efforts to control the free its domain to promoting and preserving human health,
flow of guns and arms, to abolish capital punishment, redirecting it instead to the relief of that suffering which
and to more tightly control warfare. Euthanasia would stems from life itself, not merely from a sick body.
add a whole new category of killing to a society that al- I believe that, at each of these three turning points,
ready has too many excuses to indulge itself in that way. proponents of euthanasia push us in the wrong direc-
The second turning point lies in the meaning and tion. Arguments in favour of euthanasia fall into four
limits of self-determination. The acceptance of eutha- general categories, which I will take up in turn: (1)
nasia would sanction a view of autonomy holding that the moral claim of individual self-determination and
Source: From The Hastings Center Report 22, 2 (March/April 1992), 52-5. Reprinted by permission of The Hastings Center and the author.
Callahan: When Self-Determination Runs Amok
well-being; (2) the moral irrelevance of the difference There is another problem as well. If doctors, once
between killing and allowing to die; (3) the supposed sanctioned to carry out euthanasia, are to be them-
paucity of evidence to show likely harmful conse- selves responsible moral agents not simply hired hands
quences of legalized euthanasia; and (4) the compat- with lethal injections at the ready then they must have
ibility of euthanasia and medical practice. their own independent moral grounds to kill those who
request such services. What do I mean? As those who
Self-Determination favour euthanasia are quick to point out, some people
want it because their life has become so burdensome it
Central to most arguments for euthanasia is the prin- no longer seems worth living.
ciple of self-determination. People are presumed to The doctor will have a difficulty at this point. The
have an interest in deciding for themselves, accord- degree and intensity to which people suffer from their
ing to their own beliefs about what makes life good, diseases and their dying, and whether they find life
how they will conduct their lives. That is an important more of a burden than a benefit, has very little directly
value, but the question in the euthanasia context is: to do with the nature or extent of their actual physical
What does it mean and how far should it extend? If it condition. Three people can have the same condition,
were a question of suicide, where a person takes her but only one will find the suffering unbearable. People
own life without assistance from another, that principle suffer, but suffering is as much a function of the values
might be pertinent, at least for debate. But euthanasia of individuals as it is of the physical causes of that
is not that limited a matter. The self-determination in suffering. Inevitably in that circumstance, the doctor
that case can only be effected by the moral and phys- will in effect be treating the patient's values. To be re-
ical assistance of another. Euthanasia is thus no longer sponsible, the doctor would have to share those values.
a matter only of self-determination, but of a mutual, The doctor would have to decide, on her own, whether
social decision between two people, the one to be the patient's life was "no longer worth living."
killed and the other to do the killing. But how could a doctor possibly know that or
How are we to make the moral move from my make such a judgment? Just because the patient said
right of self-determination to some doctor's right to kill so? I raise this question because, while in Holland at
me—from my right to his right? Where does the doc- the euthanasia conference, . . . the doctors present
tor's moral warrant to kill come from? Ought doctors to agreed that there is no objective way of measuring or
be able to kill anyone they want as long as permission is judging the claims of patients that their suffering is
given by competent persons? Is our right to life just like unbearable. And if it is difficult to measure suffering,
a piece of property, to be given away or alienated if the how much more difficult to determine the value of a
price (happiness, relief of suffering) is right? And then patient's statement that her life is not worth living?
to be destroyed with our permission once alienated? However one might want to answer such questions,
In answer to all those questions, I will say this: the very need to ask them, to inquire into the phys-
I have yet to hear a plausible argument why it should ician's responsibility and grounds for medical and moral
be permissible for us to put this kind of power in the judgment, points out the social nature of the decision.
hands of another, whether a doctor or anyone else. Euthanasia is not a private matter of self-determination.
The idea that we can waive our right to life, and then It is an act that requires two people to make it possible,
give to another the power to take that life, requires a and a complicit society to make it acceptable.
justification yet to be provided by anyone.
Slavery was long ago outlawed on the ground that
Killing and Allowing to Die
one person should not have the right to own another,
even with the other's permission. Why? Because it is a Against common opinion, the argument is sometimes
fundamental moral wrong for one person to give over made that there is no moral difference between stop-
his life and fate to another, whatever the good conse- ping life-sustaining treatment and more active forms of
quences, and no less a wrong for another person to killing, such as lethal injection. Instead I would con-
have that kind of total, final power. Like slavery, duel- tend that the notion that there is no morally significant
ling was long ago banned on similar grounds: even free, difference between omission and commission is just
competent individuals should not have the power to wrong. Consider in its broad implications what the
kill each other, whatever their motives, whatever the eradication of the distinction implies: that death from
circumstances. Consenting adult killing, like consent- disease has been banished, leaving only the actions
ing adult slavery or degradation, is a strange route to of physicians in terminating treatment as the cause of
human dignity. death. Biology, which used to bring about death, has
■ Y II 11 I 1 :.111 111111
apparently been displaced by human agency. Doctors disease that brings death when treatment is omitted;
have finally, I suppose, thus genuinely become gods, that is its cause, not the physician's omission. It is a
now doing what nature and the deities once did. misuse of the word killing to use it when a doctor stops
What is the mistake here? It lies in confusing caus- a treatment he believes will no longer benefit the patient
ality and culpability, and in failing to note the way in when, that is, he steps aside to allow an eventually in-
which human societies have overlaid natural causes evitable death to occur now rather than later. The only
with moral rules and interpretations. Causality (by deaths that human beings invented are those that come
which I mean the direct physical causes of death) and from direct killing when, with a lethal injection, we
culpability (by which I mean our attribution of moral both cause death and are morally responsible for it. In
responsibility to human actions) are confused under the case of omissions, we do not cause death even if we
three circumstances. may be judged morally responsible for it.
They are confused, first, when the action of a phys- This difference between causality and culpability
ician in stopping treatment of a patient with an under- also helps us see why a doctor who has omitted a treat-
lying lethal disease is construed as causing death. On the ment he should have provided has "killed" that patient
contrary, the physicianb omission can only bring about while another doctor performing precisely the same
death on the condition that the patient's disease will act of omission on another patient in different circum-
kill him in the absence of treatment. We may hold the stance does not kill her, but only allows her to die. The
physician morally responsible for the death, if we have difference is that we have come, by moral convention
morally judged such actions wrongful omissions. But and conviction, to classify unauthorized or illegitim-
it confuses reality and moral judgment to see an omit- ate omissions as acts of "killing." We call them "killing"
ted action as having the same causal status as one that in the expanded sense of the term: a culpable action
directly kills. A lethal injection will kill both a healthy that permits the real cause of death, the underlying
person and a sick person. A physician's omitted treat- disease, to proceed to its lethal conclusion. By contrast,
ment will have no effect on a healthy person. Turn off the doctor who, at the patient's request, omits or ter-
the machine on me, a healthy person, and nothing will minates unwanted treatment does not kill at all. Her
happen. It will only, in contrast, bring the life of a sick underlying disease, not his action, is the physical cause
person to an end because of an underlying fatal disease. of death; and we have agreed to consider actions of that
Causality and culpability are confused, second, kind to be morally licit. He thus can truly be said to
when we fail to note that judgments of moral respon- have "allowed" her to die.
sibility and culpability are human constructs. By that If we fail to maintain the distinction between killing
I mean that we human beings, after moral reflection, and allowing to die, moreover, there are some disturbing
have decided to call some actions right or wrong, and possibilities. The first would be to confirm many phys-
to devise moral rules to deal with them. When physicians in their already too powerful belief that, when pa-
icians could do nothing to stop death, they were not tients die or when physicians stop treatment because of
held responsible for it. When, with medical progress, the futility of continuing it, they are somehow both mor-
they began to have some power over death, but only its ally and physically responsible for the deaths that follow
timing and circumstances, not its ultimate inevitability, That notion needs to be abolished, not strengthened. It
moral rules were devised to set forth their obligations. needlessly and wrongly burdens the physician, to whom
Natural causes of death were not thereby banished. should not be attributed the powers of the gods. The
They were, instead, overlaid with a medical ethics second possibility would be that, in every case where a
designed to determine moral culpability in deploying doctor judges medical treatment no longer effective in
medical power. prolonging life, a quick and direct killing of the patient
To confuse the judgments of this ethics with the would be seen as the next, most reasonable step, on
physical causes of death which is the connotation of the grounds of both humaneness and economics. I do not
word kill is to confuse nature and human action. People see how that logic could easily be rejected.
will, one way or another, die of some disease; death
will have dominion over all of us. To say that a doctor Calculating the Consequences
"kills" a patient by allowing this to happen should only
be understood as a moral judgment about the licitness When concerns about the adverse social consequences
of his omission, nothing more. We can, as a fashion of of permitting euthanasia are raised, its advocates tend
speech only, talk about a doctor killing a patient "by to dismiss them as unfounded and overly speculative.
omitting treatment he should have provided." It is a On the contrary, recent data about the Dutch experi-
fashion of speech precisely because it is the underlying ence suggests that such concerns are right on target.
Callahan: When Self-Determination Runs Amok
From my own discussions in Holland, and from the I would mention, finally, that the moral logic of
articles on that subject, I believe we can now fully see the motives for euthanasia contain within them the
most of the likely consequences of legal euthanasia. ingredients of abuse. The two standard motives for
Three consequences seem almost certain, in this euthanasia and assisted suicide are said to be our right
or any other country: the inevitability of some abuse of self-determination and our claim upon the mercy
of the law; the difficulty of precisely writing, and then of others, especially doctors, to relieve our suffering.
enforcing, the law; and the inherent slipperiness of the These two motives are typically spliced together and
moral reasons for legalizing euthanasia in the first place. presented as a single justification. Yet if they are con-
Why is abuse inevitable? One reason is that almost sidered independently and there is no inherent reason
all laws on delicate, controversial matters are to some why they must be linked they reveal serious prob-
extent abused. This happens because not everyone will lems. It is said that a competent, adult person should
agree with the law as written and will bend it, or ignore have a right to euthanasia for the relief of suffering.
it, if they can get away with it. From explicit admissions But why must the person be suffering? Does not that
to me by Dutch proponents of euthanasia, and from the stipulation already compromise the principle of self-
corroborating information provided by the Remmelink determination? How can self-determination have any
Report and the outside studies of Carlos Gomez and limits? Whatever the person's motives may be, why are
John Keown, I am convinced that in the Netherlands they not sufficient?
there are a substantial number of cases of nonvoluntary Consider next the person who is suffering but not
euthanasia, that is, euthanasia undertaken without the competent, who is perhaps demented or mentally re-
explicit permission of the person being killed. The other tarded. The standard argument would deny euthanasia
reason abuse is inevitable is that the law is likely to to that person. But why? If a person is suffering but
have a low enforcement priority in the criminal justice not competent, then it would seem grossly unfair to
system. Like other laws of similar status, unless there is deny relief solely on the grounds of incompetence. Are
an unrelenting and harsh willingness to pursue abuse, the incompetent less entitled to relief from suffering
violations will ordinarily be tolerated. The worst thing than the competent? Will it only be fluent, middle-
to me about my experience in Holland was the casual, class people, mentally fit and savvy about working the
seemingly inherent attitude toward abuse. I think that medical system, who can qualify? Do the incompetent
would happen everywhere. suffer less because of their incompetence?
Why would it be hard to precisely write, and then Considered from these angles, there are no good
enforce, the law? The Dutch speak about the require- moral reasons to limit euthanasia once the principle of
ment of "unbearable" suffering, but admit that such a taking life for that purpose has been legitimated. If we
term is just about indefinable, a highly subjective matter really believe in self-determination, then any compe-
admitting of no objective standards. A requirement for tent person should have a right to be killed by a doctor
outside opinion is nice, but it is easy to find complaisant for any reason that suits him. If we believe in the relief
colleagues. A requirement that a medical condition be of suffering, then it seems cruel and capricious to deny
"terminal" will run aground on the notorious difficulties it to the incompetent. There is, in short, no reasonable
of knowing when an illness is actually terminal. or logical stopping point once the turn has been made
Apart from those technical problems there is a down the road to euthanasia, which could soon turn
more profound worry. I see no way, even in principle, into a convenient and commodious expressway.
to write or enforce a meaningful law that can guarantee
effective procedural safeguards. The reason is obvious Euthanasia and Medical Practice
yet almost always overlooked. The euthanasia trans-
action will ordinarily take place within the boundaries A fourth kind of argument one often hears both in
of the private and confidential doctor—patient relation- the Netherlands and in this country is that euthanasia
ship. No one can possibly know what takes place in and assisted suicide are perfectly compatible with the
that context unless the doctor chooses to reveal it. In aims of medicine. I would note at the very outset that
Holland, less than 10 per cent of the physicians report a physician who participates in another person's sui-
their acts of euthanasia and do so with almost complete cide already abuses medicine. Apart from depression
legal impunity. There is no reason why the situation (the main statistical cause of suicide), people commit
should be any better elsewhere. Doctors will have suicide because they find life empty, oppressive, or
their own reasons for keeping euthanasia secret, and meaningless. Their judgment is a judgment about the
some patients will have no less a motive for wanting it value of continued life, not only about health (even if
concealed. they are sick). Are doctors now to be given the right to
II :al 1111:LI
make judgments about the kinds of life worth living but only that suffering which is brought on by illness
and to give their blessing to suicide for those they judge and dying as biological phenomena, not that suffering
wanting? What conceivable competence, technical or which comes from anguish or despair at the human
moral, could doctors claim to play such a role? Are condition.
we to medicalize suicide, turning judgments about its Doctors ought to relieve those forms of suffering
worth and value into one more clinical issue? Yes, those that medically accompany serious illness and the threat
are rhetorical questions. of death. They should relieve pain, do what they can
Yet they bring us to the core of the problem of to allay anxiety and uncertainty, and be a comforting
euthanasia and medicine. The great temptation of presence. As sensitive human beings, doctors should
modern medicine, not always resisted, is to move be prepared to respond to patients who ask why they
beyond the promotion and preservation of health into must die, or die in pain. But here the doctor and the
the boundless realm of general human happiness and patient are at the same level. The doctor may have no
well-being. The root problem of illness and mortality better an answer to those old questions than anyone
is both medical and philosophical or religious. "Why else; and certainly no special insight from his training
must I die?" can be asked as a technical, biological as a physician. It would be terrible for physicians to
question or as a question about the meaning of life. forget this, and to think that in a swift, lethal injection,
When medicine tries to respond to the latter, which medicine has found its own answer to the riddle of
it is always under pressure to do, it moves beyond its life. It would be a false answer, given by the wrong
proper role. people. It would be no less a false answer for patients.
It is not medicine's place to lift from us the They should neither ask medicine to put its own voca-
burden of that suffering which turns on the meaning tion at risk to serve their private interests, nor think
we assign to the decay of the body and its eventual that the answer to suffering is to be killed by another.
death. It is not medicine's place to determine when The problem is precisely that, too often in human
lives are not worth living or when the burden of life history killing has seemed the quick, efficient way to
is too great to be borne. Doctors have no conceivable put aside that which burdens us. It rarely helps, and
way of evaluating such claims on the part of patients, too often simply adds to one evil still another. That is
and they should have no right to act in response to what I believe euthanasia would accomplish. It is self-
them. Medicine should try to relieve human suffering, determination run amok.
Medical Ethics and Double Effect: procedure within palliative care. But sometimes the
amount of sedative needed to control suffering has the
The Case of Terminal Sedation effect of shortening the patient's life. This creates worries
Joseph M. Boyle that the requirements of appropriate palliative care man-
date actions indistinguishable from euthanasia, which
is illegal and morally objectionable to many health care
The ethics of the medical profession incorporates some professionals. Invoking double effect addresses these
application of the moral doctrine of the double effect, worries: the intent of the physician prescribing the life-
particularly to govern medical decisions whose out- shortening analgesics is to control the suffering, not to
comes include shortening a patient's life. Very roughly. shorten life. Evidence of physician intent can be found
double effect provides that it can be morally good to in notations on the patient's chart and in the recorded
shorten a patient's life as a foreseen and accepted but dosages and titration of analgesics. Consequently, this
unintended side effect of an action undertaken for a action is not euthanasia but palliative care.2
good reason, even if it is agreed that intentionally kill- A consensus was reached among a small but rep-
ing the patient or shortening the patient's life is wrong. resentative group of Canadian intensivists and a simi-
The medical profession's use of this moral doctrine has lar group of coroners that this application of double
some support in legal decisions.' effect provided proper ethical guidance concerning
The use of terminal sedation to control the intense terminal sedation. Possibly, therefore, there is consen-
discomfort of dying patients appears to be an established sus or the prospect of the emergence of a consensus
Source: From Theoretical Medicine and Bioethics: Philosophy of Medical Research and Practice 25 (2004) (1), 51-60, Boyle, J.
© 2004, Springer Netherlands.
Boyle: Medical Ethics and Double Effect:The Case of Terminal Sedation
on this application of double effect among intensivists acceptance of the Augustinian limitation on those who
and other physicians, including those with oversight are permitted to kill does not imply the impermissibility
responsibility for deaths related to medical decisions.3 of lethal self-defence, where the death is not intentional.
Whether or not the consensus reported in this study The manuals of moral theology of the nineteenth
can be further validated or extended, it pointedly raises the century refined and formulated Aquinas's reasoning
central ethical issues involved in the prospect that this appli- into a set of rules useful for giving moral advice and
cation of double effect could be an established part of med- conducting casuistry. These rules became known as
ical ethics—and, consequently, legally enforced. Addressing the doctrine of double effect, or the principle of double
these issues is distinct from determining the consensus of effect. They were generally formulated as three or four
practitioners. The central issue is perhaps this: double effect necessary conditions for the permissibility of actions
is a general moral doctrine, not an ad hoc device to deal similar in some of their results to actions that are im-
with terminal sedation or other difficult life and death deci- permissible because of the general categories into which
sions. This doctrine emerged within the casuistry of Roman they fall. Thus, the classic formulation of the Jesuit mor-
Catholicism. How much of this distinctive moral view does alist J.-B. Gury, in the mid-nineteenth century:
the medical profession and the law implicitly accept by
accepting this application of double effect? To answer this It is licit to posit a cause which is either good or in-
question it is useful to consider further the idea of double different from which there follows a twofold effect,
effect and its rationale within Catholic moral theology. one good the other evil, if a proportionately grave
The expression "double effect" was first used by St reason is present, and if the end of the agent is
Thomas Aquinas (1225-74) to refer to the duality of the honorable—that is, if he does not intend the evil.'
results of a single human action. A person performing
an action that is foreseen to have a multiplicity of results Gury elaborated these three conditions into four,
can have very different interests in them—from serious by construing the condition of honourable intention
commitment to bringing about a result to reluctant ac- as two. The first addressed the distinction between a
ceptance of a result that is unwanted but unavoidable. means and a side effect: if the bad effect—that is, the
Since actions are purposeful, at least one of the results of result which would render that action simply wrong
an action must be intended, but others, although know- were it intended—is the means to the good effect, then
ingly and voluntarily caused, can be outside the agent's it cannot be a side effect and is intended. Thus, the
intention; these latter results I will call "accepted side ef- key requirement that the good effect be brought about
fects." Aquinas made this distinction in discussing kill- "immediately," that is, not by means of the bad effect.
ing in self-defence: one who uses lethal force for defence The second of these extrapolated conditions—that one
against attack need not intend the assailant's death; that intend only the good effect—excludes cases in which
can be outside the agent's intention.4 the bad effect is not brought about as a means to the
Following Aquinas, Roman Catholic moral theol- good effect, but is nevertheless intended because it
ogy has attributed a specific, and very important, moral functions as an independent goal. An example would
significance to this distinction between what a person be "bonus" effects—results that emerge as side effects of
intends in acting and what a person accepts as a side bringing about a goal but then recognized as independ-
effect of intentionally acting for another result. The sig- ently useful or beneficial and so (ordinarily) intended.
nificance is this: the factors sufficient to make simply Gury first two conditions—that the "cause" be mor-
wrong actions involving the intention of some result are ally good or indifferent and that there be a proportion-
not also sufficient to make simply wrong actions involv- ately grave reason for doing what brings about evil side
ing accepting, but not intending, a result of the same effects—refer to the further moral considerations that are
kind. In other words, the impermissibility of an action needed for a complete assessment of an action meeting
that is based upon the agentb intention of a certain result the conditions for upright intention. These conditions
does not render impermissible actions having a result address the two areas where Gury thought an action that
of that kind, if the result is not intended, but accepted cleared the intentionally focused conditions might still
as a side effect. In Aquinas's example of killing in self- fail morally. The first area of concern arises from the pos-
defence, the moral issue was framed by the acceptance sibility that, prior to any consideration of further results
of the Augustinian prohibition of killing in self-defence that might be intended or accepted as side effects but not
on the part of private persons, those lacking public au- intended, some actions might be simply wrong. Perhaps
thority to kill. That prohibition, Aquinas maintained, his thought here is that since the movements a person
applied only to intentional killing, not to killing brought chooses for the sake of self-defence have both defensive
about and accepted as a side effect. Consequently, the and destructive results, one can distinguish the chosen
I I ill 111111
performance from the results and ask of it whether that manualist formulations and in Aquinas's reasoning in
performance is morally permissible. In some cases, tell- his classic statement, underlines the fact that double
ing a lie or committing adultery, for example, the action effect does not imply that it is permissible to bring
is impermissible on account of considerations logically about bad results if and only if they are not intended.
prior to those concerning intended or accepted results. In the case of terminal sedation, this condition of
In the case at hand, a physician's prescribing analgesics, proportionality seems to be easily met, and the exist-
described in just that way, is morally indifferent; there- ing consensus assumes that. The need for palliation
fore, the results, intentions, and other circumstances of of some dying patients is substantial and is assumed
this chosen behaviour will determine its permissibility or generally to justify terminal sedation if moral and legal
impermissibility. In this respect the action is unlike acts worries about euthanasia are satisfactorily addressed.9
of adultery or lying, which as so described are wrong. This brief survey of the Catholic sources of double
This condition may be strictly redundant (and per- effect shows what it means, and at least roughly how it
haps also confused) since an intended result is intrinsic works. But the summary does not provide a justifica-
to an action as a chosen performance, as the lying and tion of the special moral significance attributed to the
adultery cases indicate. Nevertheless, this condition distinction between what a person intends in acting
highlights an important aspect of the style of moral and accepts as a side effect (hereafter the intended/
reasoning involved in double effect: if this condition, accepted distinction). That is because there appears to
or either of the intentionally focused conditions is not be no developed justification in the tradition. It seems
met, then the act is simply and indefeasibly impermis- that Aquinas and the theological tradition regarded
sible. These are absolute judgments that cannot be the ethical significance of the intended/accepted dis-
overturned by further considerations of the action's tinction of double effect as simply obvious. Aquinas's
particular circumstances: thus, in Aquinas's example of unargued assertion that what is intended is morally per
killing in self-defence, for any person who is not pub- se and what is outside or beyond the intention is per
licly authorized, intentionally killing the assailant is accidens suggests as much.l°
simply, that is, indefeasibly, wrong; nothing further one Some applications of double effect are certainly
can discover about the action will reverse that moral intuitively compelling, for example, those distinguish-
judgment.6 This is the absolutism of double effect. ing terror bombing in warfare from carefully targeted
The second area of concern is that even if the in- bombing of military targets where some civilians will
tentional conditions are met, and the action is not ex- likely be killed as side effects. But such intuitive convic-
cluded as simply wrong for some prior reason, the full tions are not readily generalized to all cases of double
consideration of its circumstances might still turn up a effect, particularly to those where peoples' normative
morally excluding factor. The requirement of a propor- convictions about the cases distinguished is more vari-
tionately grave reason explicitly addresses this second able and uncertain than in the bombing cases.0 The
area; it presupposes that bringing about as a side effect distinction between terminal sedation and euthanasia
what would be wrong to bring about intentionally is may distinguish just such cases: some may accept ter-
likely harmful or otherwise morally suspect, and so in minal sedation but regard efforts to distinguish it from
need of wider justification. Thus, the requirement is euthanasia to be sleight of hand; some may, for reasons
that any other considerations relevant to the moral as- independent of reasoning about terminal sedation,
sessment of the action should be brought to bear. In accept euthanasia as morally legitimate and so fail to
Aquinas's example of self-defence, two concerns were see the point of introducing double effect's distinctions.
addressed: whether one defending oneself has a duty Moreover, people's convictions about especially
to refrain from harming the assailant, and the extent clear cases are not sufficiently focused to justify the pre-
of the violence of the defensive action. He argued that cise significance double effect requires the intended/ac-
one duty to refrain is generally less pressing than the cepted distinction to bear. Thus, some think that double
duty to protect one's life, and only that level of violence effect implies that the distinction between bad outcomes
needed for the defence is justified.' that are intended and those that are accepted is that
The determination of what constitutes a propor- bringing about the former is, other things being equal,
tionately grave matter is not essential for assessing morally worse than bringing about the latter.12 This is
what is distinctive in the doctrine of double effect. That not obviously true.. the reckless or unjustified inflicting
determination is likely to be as complex and as vari- of harm as a side effect is not clearly worse than inflict-
able in outcome as any other reasoning leading to an ing it intentionally. More importantly, the implication of
all-things-considered assessment of the morality of an double effect is different: namely, that the wrongness of
action!' But the existence of this condition, both in the the former does not guarantee that of the latter.
Boyle: Medical Ethics and Double Effect:The Case of Terminal Sedation
Perhaps the moral significance of this distinction Since some bad side effects are inevitable, a morality
appeared obvious to older Catholic moralists because based on concern for the human good does not justify an
of its analogy to the structure of divine creative activ- impossible prohibition of bringing them about Rather, the
ity: God creates only good, but allows the evils of his relevant moral guidance concemingbringing about bad side
creation—evils he does not intentionally cause but only effects addresses questions such as which bad side effects are
permits as privations that flaw his good creation.° This to be accepted and on whom the harms will fall—matters
analogue suggests a justification. Just as God creates only considered under the proportionality condition of double
what is good, humans should voluntarily pursue in their effect. But if the underlying moral principle is taken to be
actions only what is humanly good. And just as God allegiance to the human good, a persons intentional actions
permits the evils flawing his creation, so humans must can always be aimed at the good, since one can always choose
accept some evil consequences they should not intend. to avoid any that are not Prohibition against intentionally
The idea is that God, if he is to create some uni- harming a human good is not an impossible prohibition.
verse he has good reason to create, must permit the As already noted above, the prohibitions against
evils which inevitably arise as side effects of his creating intentionally harming human goods presupposed by
that universe—for example, the misuse of free will by double effect are taken to be indefeasible or absolute.
rational creatures' immoral choices. Applied to human That puts the morality in which double effect de-
action, the analogy suggests that in acting, humans can veloped at odds with much of current moral opinion.
and should aim exclusively at the good, but that there What now counts as common sense morality largely
will inevitably be some bad side effects of doing that. rejects indefeasible prohibitions of actions of generally
Independently of the suggestive theological an- described kinds; and consequentialism rejects moral
alogy, there is a limitation on the human capacity to absolutes as inadequate devices for promoting the
pursue the good, and that limitation is precisely an good. But if allegiance to the good is morally basic, and
incapacity to avoid evil side effects, not an incapacity if instances of goods are not commensurable in good-
to choose and intend only the good. ness, then rational concern for the good readily justifies
The limitation is this: in all the situations calling absolute prohibitions of intentionally harming them.
for human choice, no matter what a person chooses to For if instances of human goods are incommensurable
do, some instance of a human good will be harmed, de- in their goodness, the respect for each instance of such
stroyed or at least knowingly neglected (hereafter I will goodness is required because there cannot be any good
refer to all such harms as simply "harming a good"). or sum of goods that would capture precisely the good-
Thus, it is beyond human power to act in such a way ness of that instance. On these assumptions, therefore,
that one's action does nothing more than promote absolute prohibitions of harms to the most basic goods
human good; in all human action some instances of alone do justice to the reality of the human good.14 But
human goods are promoted and served while others although double effect presupposes the truth of some
are, at the very least, not promoted (as when a person such indefeasible norms, its function in moral thought
leaves some of her talents undeveloped to pursue a is not to justify them but rather to limit their application
career), or, very often, more or less harmed (as one to intentionally harming the goods of human beings.
risks health for the sake of sports), or in some cases If the preceding account of the justification of double
even destroyed (as when one kills, whether intention- effects use of the intended/accepted distinction is sound,
ally or as a side effect). then within the framework of traditional morality, as
This limitation is essentially a limitation on the understood by the older Catholic moralists, double effect
human capacity to avoid some bad side effects of good is a legitimate moral doctrine. Within that context, it
choices, and not a limitation on the human capacity to plainly is a moral doctrine, not a specifically Catholic or
avoid choosing precisely for the sake of bad goals. In religious doctrine, since it is the implication of the charac-
the choice to act for some goal, namely, in an intentional ter of a limitation in human action and willing joined with
action, it always remains in the agent's power to choose the implications of a kind of goods based ethical theory.
not to do it. So, when something humanly evil— Can the justification be exported from its strictly moral
harming a good—is the intended result of one's action, framework to the regulatory framework of medical ethics?
one always has the choice of not doing that action. But In one respect, double effect cannot be exported
one does not have a choice about whether or not there as is. A person's intentions in doing an action may be
will be some bad side effects of whatever one chooses inferred from the action and other aspects of its context,
to do. Accepting bad side effects, therefore, is unavoid- most importantly, how a person explains his or her ac-
able; choosing to pursue results that involve harming a tions But a person's intentions themselves are at least
good is always avoidable—though often at a high price. in part inaccessible to others, and sometimes difficult
■u I II :III 111111
even for the acting person to articulate accurately and If there are some kinds of behaviour which society
reflectively. Opponents of double effect argue that these or a profession judges unacceptable, then altogether
epistemological difficulties are insurmountable, indeed, banning that behaviour may be tempting. But a general
that they suggest that the basic concepts of double effect ban on behaviour having certain results likely prohibits
cannot be rationally applied." However, difficult appli- too much, for reasons already noted: for example, caus-
cations of double effect can be debated with the possibil- ing death or bodily harm is not reasonably prohibited
ity of confident agreement, and not all applications are by medical ethics when unavoidable, and such results
controversial. Moreover, an individual's efforts to make are unavoidable in some clinical conditions. However,
upright choices will involve personal moral scrutiny and banning intentional killing is not prohibiting too much
discussion with others that can lead to judgments that are in this way.
sufficiently confident to guide choices for many cases." Furthermore, there appear to be good reasons why
But in the context of the regulation of behaviour the medical profession would want to uphold such a
by law or by professional ethics, the third-person per- ban," and, therefore, if practitioners are confident of
spective of judges and juries becomes more central, their ability publicly to determine intent in the relevant
and concerns about people's deepest moral orientations cases, then such a prohibition would be as justified as
less important. Here the relevant volitions are not those the prohibitions in the strictly moral cases. The fact that
inaccessible in a person's heart, but only those of which the ban would not be justified if applied to the accept-
there is evidence accessible to others besides the agent. ance of side effects has, of itself, no tendency to call into
For example, the evidence of intent in terminal sed- question the narrower exclusion of intentional killing.
ation cases is the sum of notations on charts, dosages The logic of double effect, therefore, has applica-
and titration of analgesics, and so on. A physician who tion in medical ethics and the law, quite independ-
would prefer to perform euthanasia but who remains ently of the particular moral framework in which it
constrained in his or her actions by the requirements was developed and has a natural function in moral
of intent as publicly accessible indicates that he or she reasoning. Double effect does not provide the justi-
does not intend to end life, even if he or she wishes to fication of norms excluding intentionally harming a
do that. And that implies that those who co-operate person's good, but reminds us that when such norms
with such a physician's action, are co-operating only in are taken as true or appropriate, whether as moral
a common act of palliative care, not of euthanasia. norms or social regulations, they cannot reasonably
These considerations show that the question about prohibit harming as a side effect. And so, the inference
the exportability of double effect to medical ethics must common in debates about euthanasia is not sound: it is
be reformulated. The inherent limitations involved in not the case that the fact that we accept bringing about
using a moral doctrine for public regulatory purposes death or an earlier death as a side effect of choosing
must be recognized. The question remains: supposing something else gives rational grounds for judging in-
they are recognized, can the moral doctrine be exported? tentional killing to be justified.18
Notes
1. Vacco v. Quill, 117 Supreme Court Reporter 2293 (1997); 8. G. Grisez, (1997), The Way of the LordJesus, Volume 3:
J. Finnis, G. Grisez, and J. Boyle, (2001), "'Direct' and Difficult Moral Questions, Appendix 1: "Human Acts and
Ìndirect': A Reply to Critics of Our Action Theory," in The Moral Judgments" (Franciscan Press: Quincy, IL), 849-70.
Thomist 65: 1-44. 9. Cf. note 2 and note 3, above.
2. L. Hawryluck and W. Harvey, (2000), "Analgesia, Virtue, 10. Cf. note 4, above.
and the Principle of Double Effect," in Journal of Palliative 11. J. Boyle, (1991), "Who Is Entitled to Double Effect?," in
Care 16 (supplement): S24—S30. The Journal of Medicine and Philosophy 16: 475-94.
3. L Hawryluck, W. Harvey, L. Lemieux-Charles, and P 12. Cf. note 11, above.
Singer, (2002), "Consensus Guidelines on Analgesis and 13. J. Boyle, (1997), "Intentions, Christian Morality and Bioethics:
Sedation in Dying Intensive Care Patients," in BMC Medical Puzzles of Double Effect," in Christian Bioethics 3: 87-8.
Ethics [online], available at www.biomedcentral.com/1472- 14. J. Finnis, J. Boyle, and G. Grisez, (1987), "A Sounder
6939/3/3. Theory of Morality," in Nuclear Deterrence, Morality and
4. St Thomas Aquinas, Summa Theologiae, q. 64, a 7. Realism (Oxford University Press: Oxford), 275-96; see
5. J. Boyle, (1980), "Toward Understanding the Principle of also note 11 and note 13, above.
Double Effect," in Ethics 90: 527-38. 15. T. Quill, R Dresser, and D. Brock, (1997), "The Rule of
6. Cf. note 4, above. Double Effect: A Critique of Its Role in End-of-Life Decision
7. Cf. note 4, above. Making," in New EnglandJournal of Medicine 337: 1768-71.
Boyle: Medical Ethics and Double Effect: The Case of Terminal Sedation
16. J. Finnis, G. Grisez, and J. Boyle, (2001), —Direct' and Internal Medicine 159: 545-50; see also Vacco v. Quill, note
'Indirect': A Reply to Critics of Our Action Theory," in The 1, above.
Thomist 65: 1-44; see also note 5, above. 18. Cf. D. Sulmasy and E. Pellegrino, "The Rule of Double
17. D. Sulmasy and E. Pellegrino, (1999), "The Rule of Effect: Clearing Up the Double Talk," note 17, above.
Double Effect: Clearing Up the Double Talk," in Archives of See also note 13, above.
6.3 Physician-Assisted Dying in Canada
Controlling the Risks of PAD still influence an individual's decision by impacting his
or her outlook and perceptions of his or her circum-
British Columbia Supreme Court stances); lack of Information about options; concerns
about burdening, or pressure from, family members;
the physician's influence, particularly in light of the
power differential that exists between physician and
Key areas of risk [748-7541 patient; and society's approval of physician-assisted
Conclusions re: feasibility of controlling the risks dying. Such forces can elude detection.
[798-8541
• Overall conclusion [8831 [7511 Third, Canada says there is a risk of deaths of
individuals with treatable conditions. The desire for
Ix. Safeguards: Feasibility a hastened death is often associated with underlying
conditions that may be amenable to therapeutic inter-
A. Introduction vention, including depression, hopelessness, physical
distress and the perceived difficulty of accessing health
[7481 It is Canada's position [i.e., the position of the care resources to relieve pain or other physical suffering
government of Canada which opposed the legalization
of PAD in the SCCI that without an absolute prohibition [7521 Fourth, Canada refers to the possible deaths of
on physician-assisted suicide, there is a risk of "wrongful ambivalent individuals. The desire for death is often
deaths." Canada identifies the following key areas of risk. transitory Canada says. Individuals frequently change
their minds about suicide, and even life-sustaining
[7491 First, Canada adverts to the deaths of incompe- treatment options, when they have the opportunity.
tent persons. It can be difficult to determine whether
a person is competent to request an assisted death, [7531 Fifth, Canada says there is a risk of deaths of
particularly in the context of a short-term physician— misinformed individuals. It asserts a very real danger
patient relationship. It can be difficult to perform ob- that individuals, during difficult periods of illness or
j ective assessments uninfluenced by assumptions about disability, may choose assisted suicide based on an in-
the quality of the patient's life and the reasonableness of accurate prognosis, and thereby be deprived of a life
her desire for death. It can be just as difficult to assess they did not realize at the time they could have had.
whether and to what extent impaired cognitive function
may have affected a person's decision-making capacity. [7541 Finally, Canada identifies the risk of deaths of
vulnerable individuals, in particular, the elderly and
[7501 Second, Canada identifies the risk of involuntary people with disabilities. [. . .1
deaths. It points to the need to be certain that a pa-
tient's request for an assisted death is voluntary that 1. Competence
is, free from coercion, pressure, undue inducement, [7981 Weighing the evidence as a whole, I [Justice Lynn
and psychological or emotional manipulation. Some of Smith, the trial judge of the BCSCI conclude that it is
the external forces that can influence a patient's deci- feasible for properly-qualified and experienced phys-
sion include illness (such as depression—even where icians reliably to assess patient competence, including
it is not sufficient to impair capacity, depression can in the context of life-and-death decisions, so long as
Source: From Carter v Canada (Attorney General), 2012 BCSC 886. Reprinted with permission.
I II ill 111111
they apply the very high level of scrutiny appropriate Battin, while it is possible for a person who is griev-
to the decision and proceed with great care. E. . .1 ously and irremediably ill to be "suicidal" in the trad-
itional sense, this is not necessarily the case for those
2. Voluntariness seeking physician-assisted death, and it cannot be
[815] Although I accept the evidence of Dr. Gallagher, assumed to be so. Instead, the decisions often reflect
Dr. Chochinov, Professor Helsel and Professor Frazee that long-held, deep-seated values and are rational, con-
influence can be subtle and exercised at an unconscious sultive, informed, and considered.
level, I accept the evidence of Dr. Ganzini and Dr. Don-
nelly that coercion and undue influence can be detected [843] The evidence on the question of ambivalence
as part of a capacity assessment. To be accurate and reli- leads me to the conclusion that it is feasible to screen
able, clinicians who perform such assessments would out from physician-assisted death patients who are am-
have to be aware of the risks of coercion and undue influ- bivalent, by assessing capacity and requiring some time
ence, of the possibility of subtle influence, and of the risks to pass between the decision and its implementation. As
of unconscious biases regarding the quality of the lives of to the argument that it is impossible to screen out people
persons with disabilities or persons of advanced age. who may, in the future, come to change their minds, I
agree with Professor Battin's observations about the need
3. Informed Consent to respect the decisions that people make, even when
[816] I will deal under this heading with two of the future circumstances are not fully certain. [. . .]
concerns that Canada raises: the deaths of individuals
who are misinformed, and the deaths of individuals 5. Vulnerable Individuals
with treatable conditions. [. . .] a) Elderly people
[847] I accept that elderly persons are vulnerable to
[831] The evidence as to informed consent permits me abuse and that the assessment of voluntariness of
to conclude that, just as physicians routinely assess the elderly people must incorporate an understanding
requirements for informed consent in patients seeking of that reality. As discussed earlier, however, there is
or refusing medical treatment, it would be feasible to no evidence that the elderly access physician-assisted
require informed consent for physician-assisted death. dying in disproportionate numbers in permissive juris-
In that context, particular care would be required to dictions (Professor Battin, Dr. Ganzlnl, Dr. van Delden,
ensure a patient is properly informed of her diagnosis Professor Deliens), and Professor Deliens observed that
and prognosis, given the seriousness of the decision. As the number of patients over the age of 80 whose deaths
well, the range of treatment options described would resulted from IAWER [i.e., non-voluntary euthanasia]
have to encompass all reasonable palliative care inter- in Belgium was not disproportionate. [. . .]
ventions, including those aimed at loss of personal
dignity b) People with disabilities
[853]I accept that persons with disabilities face prejudice
4. Ambivalence and stereotyping and that there is a risk of unconscious
[832] In Professor Mishara's view, one of the prime bias about the quality of life of a person with a disability.
characteristics of suicide is that ambivalence is omni- However, while I accept Professor Frazee's evidence and
present. A very large number of people change their take into account the submissions of the defendants and
minds after considering suicide. Even among those the Euthanasia Prevention Coalition, I am not persuaded
who attempt suicide, the vast majority change their that the risks to persons with disabilities are such that
minds after initiating the attempt, he says. The experi- they cannot be avoided through practices of careful and
ence in other jurisdictions, and some experience with well-informed capacity assessments by qualified phys-
patients who initially determined to refrain from life icians who are alert to those risks.
sustaining treatment options, shows how often people
change their minds. C. Addressing the Risks
[854] This review of the evidence permits no conclu-
[833] Expert witnesses called by the plaintiffs, however, sion other than that there are risks inherent in permit-
emphasize that in their opinion the decision-making ting physician-assisted death, and that the utmost care
process for assisted dying cannot be equated with that would be needed in designing and managing a system
involved in suicide outside that context. According to which would allow it, in order to avoid those risks.
those witnesses, such as Professor Werth and Professor [. • .]
Justice: Legislative Background: Medical Assistance in Dying (Bill C-14)
D. Conclusion inherent in permitting physician-assisted death can be

[8831 My review of the evidence in this section, and identified and very substantially minimized through
in the preceding section on the experience in permis- a carefully-designed system imposing stringent limits
sive jurisdictions, leads me to conclude that the risks that are scrupulously monitored and enforced. [. . .1
Legislative Background: Medical the criminal law. From a federal perspective, a reason-
able degree of consistency across and within provinces
Assistance in Dying (Bill C-14) and territories would support the underlying values
Department of Justice and principles of the Canada Health Act—that is, that
all Canadians should have access to needed health care
services, which could include services related to medical
[...] Part 1—Description of Proposed
assistance in dying, without financial or other barriers.
Legislation on Medical Assistance in Dying
The development of the proposed legislation on L Legislative Objectives
medical assistance in dying (Bill C-14) was in- The objectives of the proposed legislation are expressly
formed by the evidence before all levels of court in stated in the preamble of Bill C-14 and include:
the Carter case, by available Canadian and inter-
national research, social science evidence, govern- • recognizing the autonomy of persons who have a
mental reports and parliamentary studies. It was grievous and irremediable medical condition that
also informed by the experience of existing inter- causes them enduring and intolerable suffering to
national medical assistance in dying regimes, as well seek medical assistance in dying;
as by numerous recent consultative activities on • recognizing that robust safeguards, which reflect
such assistance, including the work of the Special the irrevocable nature of ending a life, are essen-
Joint Committee on Physician-Assisted Dying, the tial to prevent error and abuse in the provision of
External Panel on Options for a Legislative Re- medical assistance in dying;
sponse to Carter v Canada, the Provincial-Territorial • affirming the inherent and equal value of every
Expert Advisory Group on Physician-Assisted person's life and avoiding encouraging negative
Dying, the Canadian Medical Association, the Col- perceptions of the quality of life of persons who
lege of Family Physicians of Canada, and the work are elderly, ill or disabled;
of provincial colleges of physicians and surgeons, protecting vulnerable persons from being induced,
among others. in moments of weakness, to end their lives;
Bill C-14 would strike an appropriate balance be- • recognizing that suicide is a significant public
tween the autonomy of those individuals seeking access health issue that can have lasting and harmful ef-
to medical assistance in dying and the interests of vul- fects on individuals, families and communities;
nerable persons and of society, through amendments to • recognizing that those who wish to access medical
the Criminal Code to allow physicians and nurse prac- assistance in dying should be able to do so without
titioners to provide assistance in dying to eligible com- adverse legal consequences on their families;
petent adults in accordance with specified safeguards. • respecting the personal convictions of health care
The legal effect of the new legislation would be to de- providers; and,
criminalize medical assistance in dying and leave further • recognizing that a consistent approach to medical
regulation of the practice to the provinces and territories assistance in dying across Canada is desirable.
(Pi's) should they so choose. Medical assistance in dying
has aspects that fall under both federal and provincial The legislative objectives in the preamble speak
jurisdiction. The criminal law aspects of such assistance to the circumstances for which medical assistance in
fall under exclusive federal jurisdiction and would apply dying would be made available, but they also speak
consistently across the country The Prs can legislate in to the reasons why access would not be permitted
relation to the health care aspects and civil law impli- in other circumstances. For example, the objective
cations so long as PT legislation does not conflict with of recognizing that suicide is a public health issue
Source Minister of Justice and Attorney General of Canada (2016). Reprinted with permission.
II ,a1 III LL.I
helps to explain why medical assistance in dying is colleges of pharmacists). Although the term "phys-
not presently being contemplated for people who are ician" or "doctor" is more plain language in English,
not approaching a natural death. To permit it in cir- the term "medical practitioner" is already used in sev-
cumstances where a person is not approaching natural eral places in the Criminal Code, and so is adopted to
death could be seen as undermining suicide preven- ensure consistency within the Criminal Code. The term
tion initiatives and normalizing death as a solution to "nurse practitioner" is defined in the Bill as a registered
many forms of suffering. nurse who is designated as a nurse practitioner or other
That said, the preamble also recognizes the Govern- equivalent title, and who is entitled to autonomously
ment of Canada's commitment to explore additional cir- make diagnoses, interpret tests, prescribe medications
cumstances in which a person may seek access to medical and treat individuals.
assistance in dying, namely requests by mature minors,
advance requests, and where mental illness is the sole III. Criminal Exemptions
underlying medical condition. These circumstances are Culpable Homicide
complex, and require additional study and consideration. It is a crime to intentionally cause the death of another
As well, the preamble affirms the Government's com- person, even if they consent to die (section 14 of the
mitment to uphold the principles set out in the Canada Criminal Code). The Bill would therefore enact a new
Health Act and to develop non-legislative measures to sup- exemption from criminal liability for culpable homi-
port the improvement of end-of-life care in the country. In cide for medical practitioners and nurse practitioners
this regard, the Minister of Health has offered to work in who provide medical assistance in dying in the form
collaboration with the Yrs on the development of an end- of what is commonly called voluntary euthanasia, i.e.,
of-life care coordination system to provide information on the administration of medication to a person, at their
options and facilitate patient access to care. request, that causes their death (new section 227(1)
of the Criminal Code, in clause 2 of the Bill). The Bill
II. Definitions would also create an additional and related exemp-
The Bill would enact new definitions in the Criminal tion for any other person who would do anything in
Code for the purpose of the national medical assistance order to help a physician or nurse practitioner provide
in dying regime (see Annex I for a clause by clause such assistance. The latter category would include, for
guide to Bill C-14). The proposed new section 241.1 instance, a social worker who is asked to meet with
(clause 3 of the Bill) is a central provision of the regime the patient to help assess the voluntariness of their
as it would define the umbrella term "medical assist- request, a lawyer to a hospital who is asked to review
ance in dying" as encompassing what is commonly the documents to verify compliance with the law, or a
called voluntary euthanasia (i.e., the administration by pharmacist who fills the prescription for the medica-
a medical practitioner or nurse practitioner of medication to be administered by a medical practitioner or
tion that will cause a person's death at their request) nurse practitioner to the patient.
and assisted suicide (i.e., the prescription or provision The Bill would also re-enact the legal rule (section
by a medical practitioner or nurse practitioner of medi- 14 of the Criminal Code—clause 1 of the Bill) that says
cation that a person could self-administer to cause that a person's consent to die is not a defence for some-
their own death). It should be noted that, in the case one who inflicts death on them. This rule was found to
of what is commonly called voluntary euthanasia, the be unconstitutional in Carter. At the same time, the Bill
provision of medical assistance in dying would result would clarify that this legal rule does not apply in the
directly in the patient's death, whereas in the case case of a person who receives medical assistance in dying
of assisted suicide, the provision of such assistance in conformity with the new regime (new subsection
would result in the patient obtaining medication that 227(4) of the Criminal Code—clause 2 of the Bill). In all
they could choose to use—or not—to end their life. other circumstances, it would continue to be a crime to
This difference would be reflected in the new criminal cause the death of a person who consented to die.
exemptions, as set out below.
With respect to health care professionals, who Aiding a Person to Die by Suicide
are regulated under PT responsibility, the new section It is a crime to assist a person to die by suicide,
241.1 would define the terms "medical practitioner" whether or not suicide ensues (paragraph 241 (b) of
and "pharmacist" as those who are entitled to practise the Criminal Code). Assistance can be in the form of
medicine or pharmacology under provincial laws (e.g., providing information about how to end their life,
licensed professionals who are members of provin- or providing a tool or other means that could be
cial colleges of physicians and surgeons or provincial used. The Bill would enact criminal exemptions for
the offence of aiding a person to die by suicide for IV. Eligibility Criteria for Medical Assistance
medical practitioners, nurse practitioners and people in Dying
who would assist them in providing or dispensing Bill C-14 would enact a new section 241.2 of the
medication to eligible persons. The person could self- Criminal Code, which in essence sets out the criminal
administer the medication to cause their own death, rules surrounding the provision of medical assistance
either at that time or at a later time of their choosing in dying. New subsections 241.2(1) and 241.2(2)
(new subsections 241(2) and 241(3) of the Criminal would set out the eligibility criteria for such assist-
Code—clause 3 of the Bill). ance in Canada. Under the proposed legislation,
The Bill recognizes that after a person would medical assistance in dying would be available to a
receive the prescription from the physician or nurse person who meets all of the following criteria (sub-
practitioner (which meets the definition of medical section 241.2(1)):
assistance in dying in the Criminal Code as explained
above), they may still need various forms of assistance • being an adult (at least 18 years old) who is men-
depending on their state of health. For instance, they tally competent ("capable") to make health care
may need someone to collect the medication from the decisions for themselves;
pharmacy to open the bottle that contains such medi- • having a grievous and irremediable medical condi-
cation, or to lift a glass of water to their mouth so that tion (as defined under subsection 241.2(2));
they can swallow it. As these forms of conduct are pro- • making a voluntary request for medical assist-
hibited under the offence of aiding a person to die by ance in dying which does not result from external
suicide, additional exemptions are provided in the Bill pressure;
to address this type of assistance. • giving informed consent to receive medical assist-
Specifically, pharmacists who fill a prescription for ance in dying; and,
the purpose of medical assistance in dying and give the • being eligible for health services funded by a
medication directly to the patient or to another person government.
on their behalf would also be exempted for this conduct
(new subsection 241(4)—clause 3 of the Bill). Similarly, A grievous and irremediable medical condition would
any person who helps the patient to self-administer be expressly defined under the Bill as (subsection
the medication would also be exempted from criminal 241.2(2)):
responsibility (new subsection 24(5)—clause 3 of the
Bill). Any person helping someone to self-administer • having a serious and incurable illness, disease or
the medication should exercise extreme caution, as disability, and,
the decision to self-administer, and the final actions of • being in an advanced state of irreversible decline
doing so, must be those of the patient for whom the in capability; and,
medication was prescribed. In any other circumstance, • experiencing enduring physical or psychological
there would remain a risk of criminal prosecution. suffering, due to the illness, disease, disability or
Finally, the Bill would re-enact the offence of state of decline, that is intolerable to the person
aiding a person to die by suicide, and this conduct and cannot be relieved in a manner that they con-
will continue to be criminal in all circumstances other sider acceptable; and,
than those described in the medical assistance in dying • where the person's natural death has become
regime under the Criminal Code. reasonably foreseeable, taking into account all of
their medical circumstances, without requiring
"Reasonable but Mistaken Belief" a specific prognosis as to the length of time the
For greater certainty, the Bill would codify a common person has left to live.
law principle to clarify that the exemptions would still
apply even if the practitioners, or other exempted in- The proposed eligibility criteria would enable indi-
dividuals, have a reasonable but mistaken belief about viduals who are intolerably suffering, in an advanced
some relevant fact; for example, whether the person state of irreversible decline in capability, and who are
is 18 years of age and is therefore eligible to receive on a trajectory towards their natural death, to have the
medical assistance in dying (new subsections clauses option of a peaceful medically-assisted dying process,
227(3) and 241(6) in clauses 2 and 3 of the Bill). instead of having to endure a painful, prolonged or
These would essentially function as a "good faith" undignified one. It would enable them to make a fun-
defence for those who would participate in medical damentally personal decision concerning their bodily
assistance in dying. integrity, autonomy and dignity, which could also help
uw I I I .111 111111
prevent them from ending their lives prematurely, by access medical assistance in dying if they met all of
providing reassurance that they will have access to the eligibility criteria.
medical assistance in dying at a time when they may The requirement for the person to be eligible to
be unable to end their own life without assistance. In- receive publicly funded health services is intended to
dividuals would have to be able to provide informed prevent foreigners from visiting Canada to obtain med-
consent when eligibility is assessed and confirm their ical assistance in dying.
consent one last time immediately before medical as-
sistance in dying is provided. V. Safeguards
In Carter, the Court expressly stated that the scope In Carter, the Court acknowledged that there are in-
of its declaration was "intended to respond to the fac- herent risks in permitting medical assistance in dying,
tual circumstances" of the case and that it made "no but agreed with thetrialjudge that these risks"can be
pronouncement on other situations where physician- identified and very substantially minimized through
assisted dying may be sought." The proposed eligibility a carefully-designed system imposing stringent limits
criteria would directly respond to the factual circum- that are scrupulously monitored and enforced."
stances raised in Carter. The individuals whose cases As enacted in other jurisdictions that have medical
were considered by the Court were either in physical assistance in dying, robust procedural safeguards
decline and nearing death in the late stages of a fatal are a critical component of any carefully-designed
disease, or were otherwise nearing the end of their lives regime and are essential to prevent error and abuse
while in decline due to a condition that was not fatal from occurring and to protect vulnerable persons.
in itself. This approach would also be consistent with Consistent with the Court ruling and the Special
the Court's comparison of medical assistance in dying Joint Committee's recommendations on safeguards,
with other forms of end-of-life care, i.e., medical as- Bill C-14 would enact mandatory procedural safe-
sistance in dying would become another end-of-life guards that medical practitioners and nurse practi-
option, in addition to palliative care and palliative sed- tioners would be required to follow before providing
ation for instance, for intolerably suffering individuals medical assistance in dying to a person. These safe-
whose deaths are reasonably foreseeable. The Court guards would require the medical practitioner or
also recognized that assisted dying is a complex issue nurse practitioner to: confirm that the person meets
involving various interests, that a number of solutions all eligibility criteria for medical assistance in dying;
were possible, and that Parliament's response would ensure that the person's request was made in writing
receive a high degree of deference. (See Annex B for after the person was informed that his or her natural
more information.) death was reasonably foreseeable and be satisfied
The criterion of reasonable foreseeability of that it was signed and dated in the presence of two
death is intended to require a temporal but flex- independent witnesses; ensure that the person was
ible connection between the person's overall med- informed that they may withdraw their request at
ical circumstances and their anticipated death. As any time; ensure that a second independent med-
some medical conditions may cause individuals to ical practitioner or nurse practitioner provided a
irreversibly decline and suffer for a long period of written opinion confirming the person's eligibility;
time before dying, the proposed eligibility criteria ensure that a period of at least 15 days has elapsed
would not impose any specific requirements in between the moment the written request was signed
terms of prognosis or proximity to death (e.g., a six and the provision of medical assistance in dying
month prognosis as the US states' medical assistance (unless both practitioners agree that death or loss
in dying laws require). The medical condition that of capacity to consent is imminent); and, immedi-
is causing the intolerable suffering would not need ately before providing such assistance, confirm the
to be the cause of the reasonably foreseeable death. person's consent.
In other words, eligibility would not be limited to In order to be considered independent, the wit-
those who are dying from a fatal disease. Eligibil- nesses could not be a beneficiary under the will of that
ity would be assessed on a case-by-case basis, with person or otherwise benefit from their death, or be
flexibility to reflect the uniqueness of each person's involved directly in giving care to the person, among
circumstances, but with limits that require a natural other criteria (new subsection 241.2(5)—clause 3 of
death to be foreseeable in a period of time that is the Bill). With respect to the independence of the first
not too remote. It should be noted that people with medical practitioner or nurse practitioner from the
a mental illness or physical disability would not be second one, the Bill provides that they could not be
excluded from the regime, but would only be able to connected to each other in any way that could impair
their objectivity, such as by being in a business or men- Filing Obligations and Related Offences
toring relationship with each other. They would also To enable the monitoring regime to operate effect-
need to be independent of the patient, in the sense ively, the proposed legislation would also create a
that they could not be a beneficiary under his or her legal obligation for medical and nurse practitioners
will, or be otherwise connected to them in a manner who receive a written request for medical assistance
that could affect their objectivity (new subsection in dying to provide information as prescribed in
241.2(6)). However, the legislation makes clear that regulations (new subsection 241.31(1)). Pharmacists
standard compensation for providing medical services would also be required to provide information in rela-
would not affect the physician or nurse practitioner's tion to their provision of medications used for such
independence. assistance (new subsection 241.31(2)). Bill C-14 pro-
The Bill would make provision for individuals who poses a hybrid offence—punishable by a maximum of
are unable to sign their own request, by enabling them 2 years imprisonment—for failure to comply with the
to ask a capable adult, who is not directly involved in reporting obligations, and for any person who know-
providing health care services to them, to sign and date ingly breaches the regulations. The reporting obli-
their request on their behalf and in their presence (new gations and related offences would be brought into
subsection 241.2(4)). force at a later date than the rest of the Bill, once the
Finally, in order to fall within the exemption, detailed regulations on the monitoring regime would
the medical practitioner or nurse practitioner would be in place.
also be required to act with reasonable knowledge,
care and skill, in accordance with applicable prov- VII. Other Offences
incial laws, rules or standards (new subsection To ensure compliance with the new medical assist-
241.2(7)), and to inform the pharmacist whenever ance in dying legislation, and given the serious and
medication is sought or prescribed for the purposes irrevocable nature of helping people die and the po-
of medical assistance in dying (new subsection tential for criminal liability for medical professionals,
241 .2 (8)). the Bill would enact new hybrid offences for failure to
comply with the mandatory safeguards (new section
VI. Monitoring System 241.3), and for forging or destroying documents re-
As recognized by the Court in Carter, by the Special lated to assistance requests with criminal intent (new
Joint Committe8 report, as well as by many stakehold- section 241.4). For instance, a person might destroy
ers, a pan-Canadian monitoring system to collect and a written medical assistance in dying request in order
analyze data on the provision of medical assistance in to block someone's access to such assistance or to
dying across Canada, to monitor trends and provide interfere with a medical practitioner's ability to rely
information to the public on the implementation of on an exemption, or they might forge the signature of
the new law, would be a critical component of the new a person they were trying to influence to seek assist-
Canadian regime. Such a monitoring system would be ance. These offences would be liable to a maximum
essential to foster transparency and public trust in the term of imprisonment of five years, where prosecuted
system. on indictment, and to a maximum term of 18 months
on summary conviction.
Regulation-Making Power
The proposed legislation would create a power for the VIII. Related Amendments
Minister of Health to make regulations to establish Bill C-14 would enact related amendments to other
a system for monitoring requests for, and the provi- statutes to ensure that recourse to medical assistance
sion of, medical assistance in dying in Canada. Such in dying does not affect pensions under the Pension Act
regulations may include: information to be provided or benefits under the Canadian Forces Members and Vet-
by medical practitioners or nurse practitioners to a erans Re-establishment and Compensation Act. It would
designated body; the form, manner and circumstances also amend the Corrections and Conditional Release Act to
in which the information has to be provided; details ensure that no investigation need be conducted under
about how the data would be analyzed, interpreted, section 19 of that Act as a result of a person dying with
and reported to the public; and rules for the protec- medical assistance.
tion and disposal of such information (new subsec-
tion 241.31(3)—clause 4 of the Bill). The Minister of IX.Parliamentary Review
Health would consult with PTs and others in develop- The Bill includes a mandatory legislative review
ing the regulations. to take place five years after its coming into force.
xThe medical assistance in dying legislation would be X. Areas for Further Study
referred to a designated parliamentary committee of The parliamentary review would be a distinct process
the House of Commons, the Senate or both Houses of from the Government of Canada's commitment to
Parliament. The committee would review all provisions study additional complex circumstances in which a
of Bill C-14 and submit a report to Parliament. It could person may seek access to medical assistance in dying,
also consider any reports that would be published namely requests by mature minors, advance requests,
under the monitoring regime. The parliamentary review and requests where mental illness is the sole under-
could assess whether the new regime is meeting its lying medical condition. The results of these studies
legislative objectives, how medical assistance in dying is could be considered by the parliamentary committee
being implemented across Canada, and identify areas of as part of its broader legislative review of the new
potential changes and improvements, if necessary. regime. F. . .1
Mature Minors, Mental Illness, The Carter decision applies to a "competent adult
person," but does not include a definition of adult. In
Advance Directives, and assessing whether someone is an adult person, an arbi-
Conscientious Objection trary age limit such as 18 years old provides no valid
safeguard. Instead, it is important that willing phys-
This reading consists o f excerpts from two reports com- icians carefully consider the context of each request
piled to compare their recommendations on four issues to determine whether the person has the information
presented below in sections titled "Mature Minors," needed, is not under coercion or undue pressure, and
"Mental Illness," "Advance Directives," and "Conscien- is competent to make such a decision.
tious Objection." Competence and consent to treatment laws vary
The reports are the following: among provinces and territories. In some cases, deci-
sional competence is based on age, but in others, age is
• Provincial-Territorial Expert Advisory Group on irrelevant to whether the patient has the legal authority
Physician-Assisted Dying: Final Report, November to make decisions about his or her treatment.
2015 An ideal regulatory framework for physician-
• Medical Assistance in Dying: A Patient-Centred assisted dying would have uniform eligibility rules.
Approach: Report of the Special Joint Committee Indeed, many provider groups asked the Advisory
on Physician-Assisted Dying, February 2016 Group to recommend a consistent national approach
to eligibility for physician-assisted dying and patient
Below, the reports are referred to as "Expert Advisory advocacy groups also sought an easily understand-
Group" and "Special Joint Committee," respectively. able criterion for accessing the service regardless
of where in the country a patient lives. The federal
government could facilitate this by affirming that
Mature Minors
the definition of "adult" in the context of physician-
Expert Advisory Group (p. 34) assisted dying relates to a patient's ability to give
Adult Person consent rather than a particular age cut-off. That
is, to allow access to patients who are competent,
RECOMMENDATION 17: Access to physician-assisted
regardless of age. [. . .1
dying should not be impeded by the imposition of
arbitrary age limits. Provinces and territories should Special Joint Committee (pp. 18-21)
recommend that the federal government make it D. Age
clear in its changes to the Criminal Code that eligi- The Carter decision dealt with plaintiffs who were adults,
bility for physician-assisted dying is to be based on so no decision was made with regard to minors' eligi-
competence rather than age. bility for MAID. However, as Prof. Pelletier made clear,
Parliament can choose to allow minors to access MAID.'
Source: From Provincial-Territorial Expert Advisory Group on Physician-Assisted Dying Final Report (2015) and from Medical
Assistance in Dying: A Patient-Centred Approach: Report of the Special Joint Committee on Physician-Assisted Dying (2016) 42nd
Parliament, 1st Session.
Expert Advisory Group/Special Joint Committee: Mature Minors, Mental Illness, Advance Directives, and Conscientious Objection
In response to a comment stating that it was up to the who was head of psychiatry at Vancouver's Children's
Committee to determine what the age of consent would Hospital for 30 years, argued for a competence-based
be in relation to a specific offence, Prof. Hogg replied: approach, saying
Yes, I think that's right. The Supreme Court, in its order, I have worked with many teenagers over the years
spoke of a "competent adult person." I don't think it and I have worked with a number who have been
would be open to you, for example, to have 16 as an facing death, and I think they would be competent
age of consent for this purpose, because that would not in the legal sense to consent to physician-assisted
be a competent adult person. Between 18 and 21, I dying as they would be legally competent to agree
would think you would have some leeway within to other kinds of medical care.
the word "adult" to decide that.2
. . . Why would we want teenagers to suffer, but
Certain witnesses, such as the Nova Scotia College of we're prepared to relieve adults of suffering?"
Physicians and Surgeons, chose not to take a position
on this issue, simply asking for greater clarity to be Other witnesses such as Margaret Birrell, Presi-
provided.3 Some witnesses who appeared before the dent of the Alliance of People with Disabilities Who
Committee and the External Panel recommended that Are Supportive of Legal Assisted Dying Society, and Dr.
legislation define an age below which MAID would not John Soles, President of the Society of Rural Physicians
be available (generally 18 but one submission sug- of Canada, were open to minors possibly having access,
gested as old as 25).4 Other witnesses wanted all indi- but felt that this should not be allowed at the present
viduals who are competent to make the decision to be time." Dr. Hartley Stern, Executive Director and CEO
eligible for MAID.5 One witness also flagged the need to of the Canadian Medical Protective Association, said
consider the suffering of children who are not compe- that if "mature minors" are to be entitled to MAID, clari-
tent, though did not go so far as to suggest including fication is needed as to how their competency will be
them in any MAID regime.° To date, Belgium and the assessed." Quebec's An Act respecting end-of-life care re-
Netherlands are the only two jurisdictions that allow stricts "medical aid in dying" to "a person of full age.""
minors access to MAID.7 The Committee understands the concerns of many
Stakeholders who prefer competency-based cri- witnesses regarding the capacity of minors to understand
teria, such as the Provincial/Territorial Expert Advis- the implications of such a serious decision. However, it is
ory Group and the British Columbia Civil Liberties important to remember, as noted in the External Panel's
Association, argue that the trend is toward increased report, that the Supreme Court has stated that minors
recognition of the competence of minors in health care have a right "to a degree of decision-making autonomy
decision-making and that age limits are arbitrary.8 Prof. that is reflective of their evolving intelligence and under-
Pelletier stated that suffering is suffering, regardless of standing."15 Allowing competent minors access to MAID
age, and that there is a risk that the provisions may be would not be eliminating the requirement for compe-
challenged on the basis of section 15 of the Charter tence. Given existing practices with respect to mature
(equality rights) if minors are excluded.9 minors in healthl° and the obvious fact that minors can
The Canadian Paediatric Society advocated against suffer as much as any adult, the Committee feels that it is
including minors, regardless of competence, in any difficult to justify an outright ban on access to MAID for
MAID regime. The organization made this argument for minors. As with issues of mental health, by instituting
a number of reasons, including the lack of evidence appropriate safeguards, health care practitioners can be
before the court in Carter regarding minors; the fact relied upon to identify appropriate cases for MAID and
that an age limit is not arbitrary, and the lack of social to refuse MAID to minors that do not satisfy the criteria.
consensus with respect to MAID for minors. The organ- The Committee acknowledges that a competent
ization also rejected the idea that a constitutional chal- mature minor who has a grievous and irremediable
lenge by excluded minors would clearly be successful. medical condition should not be forced to endure in-
It suggested addressing whether to allow minors to tolerable suffering. Moreover, there are serious ques-
access MAID at a later date, after there has been time to tions whether a restriction of the right to MAID only to
gather data, as was the case in Belgium which legalized competent adults would be consistent with the Charter.
minors' access to MAID in 2014, 12 years after adults However, the Committee realizes that witnesses and
were granted access." briefs received were of differing opinions on the subject
In contrast, Dr. Derryck Smith, Chair of the Phys- of extending the right to MAID to mature minors, and
icians Advisory Council of Dying with Dignity Canada that these differences reflect a divergence of opinion
II .111 111111
among the Canadian public. After reflecting on the excruciating as any suffering that can accompany
issue, the Committee recommends the following: physical illness. Finally, I would argue that exclud-
ing individuals on the basis of mental illness would
RECOMMENDATION 6 violate the charter."
That the Government of Canada implement a
two-stage legislative process, with the first stage A number of witnesses and submissions expressed con-
applying immediately to competent adult persons cern about mental illness in the context of mAiD.2° Dr.
18 years or older, to be followed by a second stage K. Sonu Gaind, President of the Canadian Psychiatric
applying to competent mature minors, coming Association, outlined some of the challenges that will
into force at a date no later than three years after need to be addressed:
the first stage has come into force; and That the
Government of Canada immediately commit to In terms of what is "irremediable," careful con-
facilitating a study of the moral, medical and legal sideration needs to be given about what this means
issues surrounding the concept of "mature minor" in the context of mental illness. Irremediable, of
and appropriate competence standards that could course, cannot simply mean incurable. Many con-
be properly considered and applied to those under ditions in psychiatry and medicine are considered
the age of 18, and that this study include broad- chronic and not curable, but things may be done
based consultations with health specialists, prov- to remediate or improve the situation.21
incial and territorial child and youth advocates,
medical practitioners, academics, researchers, Dr. Tarek Rajji, Chief of Geriatric Psychiatry at the
mature minors families and ethicists before the Centre for Addiction and Mental Health, told the Com-
coming into force of the second stage. mittee that:
Mental Illness
[MI ental illness may be grievous to an individual,
and symptoms can cause enduring psychological
Special Joint Committee (pp. 13-15) and sometimes physical suffering. However,
2. Mental Illness suffering should not be equated with an irremedi-
Because the individuals that brought the case in Carter able nature, and the lack of inevitable and predict-
did not have mental health issues, the Court made no able death by natural history provides us with an
pronouncement with respect to MAID and psychiatric opportunity to deliver recovery-based treatment.
conditions. Jeanette Ettel, Senior Counsel, Human [P] eople with mental illness may be vulner-
Rights Law Section at the Department of Justice, told able to the impact of the social determinants of
the Committee that it was open to the Committee to health. They may live in poverty, have poor hous-
consider whether to include psychiatric illnesses in the ing, and lack social support. These circumstances
conditions that could result in a right to MAID.17 may exacerbate suffering and a person's perception
As was the experience of the External Panel, the that their illness is irremediable . . . with a clinical
Committee heard widely diverging views on how to recovery-based environment, there is always the
address mental health in the context of MAID. Benoit potential for mental illness to be remediable. 22
Pelletier, member of the External Panel and an expert
in constitutional law, noted that the External Panel In response, Professor Downie reminded the Commit-
identified greater support from Canadians for MAID in tee of the following aspect of the Carter judgment:
the context of a physical illness but that, prima facie,
the Carter criteria would also apply to psychiatric "Irremediable," it should be added, does not re-
conditions. 18 Professor Downie and others supported quire the patient to undertake treatments that are
this position: not acceptable to the individual.23
[MI ental illness should not be an exclusion criter- The Committee recognizes that there will be
ion. It was not excluded by the Supreme Court, unique challenges in applying the eligibility criteria for
and not all individuals with mental illness are in- MAID where the patient has a mental illness, particu-
competent. Physicians already routinely determine larly where such an illness is the condition underlying
whether someone is competent, even when they the request. However, where a person is competent
have a mental illness. Furthermore, the suffering and fits the other criteria set out by law, the Commit-
that can accompany mental illness can be as tee does not see how that individual could be denied
a recognized Charter right based on his or her mental are satisfied before accepting a request for MAID and
health condition. Furthermore, we do not understand recommends:
the Carter decision to exclude mental illnesses.
Any individual applying for MAID would need to RECOMMENDATION 4
satisfy all the criteria, including irremediability and That physical or psychological suffering that
capacity. As several witnesses reminded the Committee, is enduring and intolerable to the person in the
health professionals will need to strike an appropriate circumstances of his or her condition should be
balance between the rights of all Canadians to access recognized as a criterion to access medical assist-
this constitutionally protected right, and the protection ance in dying.
of those vulnerable persons who might be coerced into
requesting MAID. Cases involving mental illness may Advance Directives
prove challenging to address for health care practi-
tioners, but the Committee has faith in the expertise Expert Advisory Group (pp. 30-31)
of Canadian health care professionals to develop and Timing of Completion of Declaration
apply appropriate guidelines for such cases. The diffi-
culty surrounding these situations is not a justification
to discriminate against affected individuals by denying RECOMMENDATION 12: Provinces and territories
them access to MAID. The Committee expects that cases should request that the federal government make
where the underlying condition is a mental health con- it clear in its changes to the Criminal Code that at
any time following the diagnosis of a grievous and
dition will be rare, as is the case in other jurisdictions
irremediable condition, a request for physician-
that have legalized MAID.24 A more detailed discussion
assisted dying made through a valid patient
of appropriate safeguards can be found below. The declaration form may be fulfilled when suffering
Committee therefore recommends: becomes intolerable.
RECOMMENDATION 13: Within one year, provinces
RECOMMENDATION 3 and territories, in collaboration with the federal
That individuals not be excluded from eligibility government, should study whether patient dec-
for medical assistance in dying based on the fact laration forms completed prior to the diagnosis
that they have a psychiatric condition. of a grievous and irremediable medical condition
might also be considered valid.
B. Suffering
The Supreme Court did not specify in Carter whether
suffering is limited to physical suffering. Witnesses We have carefully considered the question of when
voiced different opinions, with some advocating for the requirement for competency must be met. We
the inclusion of physical suffering only and others rec- have considered four possibilities and determined that
ommending that psychological suffering be included physician-assisted dying should be permitted in three
as well. 25 The Committee received at least one sub- of those situations.
mission arguing that mental suffering is as severe as Provinces and territories should advise the federal
physical suffering and should not be excluded from government to make clear in the Criminal Code that
MAID.26 In addition, the Supreme Court referred to physician-assisted dying is permissible where:
suffering "from the knowledge that they lack the abil-
ity to bring a peaceful end to their lives at a time and a) the patient is competent at all times from the initial
in a manner of their own choosing," which would be request to the moment of provision of assistance;
psychological in nature.27 The requirements that the b) although competent and having a grievous and ir-
suffering is enduring and intolerable to the person are remediable condition and experiencing intolerable
safeguards to ensure that someone in temporary or suffering at the time of the completion of the pa-
minor pain does not make a rash decision to die. In tient declaration form, the patient lost competence
addition, the suffering must relate to a grievous and between the completion of the patient declaration
irremediable condition. Where mental illness is an form and the provision of assistance., or
issue, Dr. Gaind noted that what is considered endur- c) although competent and having been diagnosed
ing and intolerable suffering may be affected by the with a grievous and irremediable condition, but
mental illness itself.28 The Committee has confidence not yet experiencing intolerable suffering, at the
that health care professionals will proceed cautiously time of the completion of a patient declaration
in such cases, as in all cases, and ensure that all criteria form, the patient lost competence between the
I Iui 111111
completion of the patient declaration form and the intolerable suffering are met, even if the patient will
onset of the enduring intolerable suffering. no longer be competent when those conditions are
met? Our answer is yes.
We note that Carter requires that physician-assisted Particularly for certain degenerative diseases, we
dying be provided to a competent adult. Scenario (a) feel that to require that these patients be competent at
above is clearly allowed under the Supreme Court of the time physician-assisted dying is delivered would
Canada's decision. However, the Supreme Court's deci- force many of them to trigger the act earlier in the
sion is silent on scenarios (b) and (c) above. course of their disease than necessary. Some people
The first situation is the most straightforward: a may be more comfortable with their diagnosis know-
patient who has a grievous and irremediable condition ing that physician-assisted dying will be available to
and is experiencing intolerable suffering and is compe- them after they lose competency. In such cases, phys-
tent until the moment of provision. This person should icians will need to take extra care to ensure that the
qualify for physician-assisted dying. patient understands and freely consents to his or her
The second situation involves a patient who, decision. The physician and patient should discuss the
although competent and having a grievous and normal progression of the disease and how patients
irremediable condition and experiencing intoler- often respond and adapt to the disease. These conver-
able suffering at the time of the completion of the sations should continue throughout the progression of
declaration form, loses competency between the the disease (or up to the point of losing competence)
completion of the form and the provision. In this to assess and confirm that the patient maintains his
case, we believe that the person should be eligible for or her decision with respect to the future provision of
physician-assisted dying. physician-assisted dying.
The third situation is more complex. It may occur
when a patient has been diagnosed with a condition Special Joint Committee (pp. 21-4)
that will remove competency, such as dementia, and E. Advance Request
has concerns about the quality of his or her end-of- The Carter decision dealt with plaintiffs who would
life in that future. The patient is not experiencing remain competent while they faced significant
intolerable suffering, but has been diagnosed with a physical decline. It did not address whether an in-
grievous and irremediable medical condition. Should dividual who is not competent at the time of death
this patient be permitted to complete a patient dec- could identify the circumstances in which he or
laration indicating that he or she would like to receive she would choose MAID in advance. With respect to
physician-assisted dying when certain conditions advance requests for MAID, witnesses and briefs out-
that the patient believes would constitute enduring lined diverging opinions, from recommending not to
Situation Situation Situation
1 2 3
Patient is competent and has been
Patient meets eligibility criteria diagnosed with a grievous
Patient meets eligibility criteria and was competent at the irremediable medical condition
and is competent until the moment of the request, but loses that will remove cognitive
moment of provision. competency between the request capacity potentially before the
and the provision. condition of intolerable suffering
has been met.
Figure 1 Timing of the Completion of a Declaration

allow such requests, to allowing them only after an the advance request have been satisfied. As examples,
individual is diagnosed, to allowing advance requests she listed being bedridden, being unable to feed, wash,
to be written prior to any illness. There was general or shave oneself or being unable to speak for 30 days
agreement however that, if requests are to be allowed or more. The same safeguards for a contemporaneous
in advance, the individual must be competent at the request must be in place for an advanced request to
time the advance request is drafted. confirm informed consent and capacity.34
An advance request could be considered in three Jean-Pierre Menard, representing the Barreau
different situations: du Quebec, noted that the panel of legal experts ap-
pointed by Quebec's government, of which he was a
where a person's request has been accepted but the member, recommended allowing advance requests
individual loses competence before MAID takes place; but that the legislation adopted in Quebec does not
where a person has been diagnosed with a griev- permit them for MAID. He explained that there were
ous and irremediable condition but is not yet ex- a number of questions that were raised about how to
periencing enduring and intolerable suffering and assess competence at the time an advance request is
• prior to diagnosis. made: whether the individual would fully understand
the decisions being made; how to know whether the
Professor Downie recommended that advance requests individual had changed his or her mind; and whether
be permitted in the first two cases, but not the third. a third party could act against the interests of the
She argued that advance requests prevent the suffering patient. He concluded that there was much debate,
of someone who has been approved for MAID but then with valid arguments on both sides, and that a deci-
loses competence and must continue to suffer. It also sion was made in Quebec to prioritize protection of
prevents individuals from ending their lives earlier the vulnerable.35
than they would otherwise in order to avoid losing Jay Cameron from the Justice Centre for Consti-
competence before the suffering becomes intolerable, tutional Freedoms expressed concerns that advance
something which was a major factor in the Carter case.29 requests could result in abuse if a patient becomes in-
Finally, an advance request allows the process to be competent and that it would not be possible to verify
undertaken before the suffering is enduring and intoler- if the request was made under duress. He also argued
able. Otherwise, the person would have to continue to that it is too difficult to know how one will feel once
endure the suffering during the processing of the request in a changed state, such as when one is experiencing
and any waiting period.3° Linda Jarrett and the Hon. the symptoms of dementia.36 Michael Bach, Execu-
Steven Fletcher, both living with disabilities, also told tive Vice-President of the Canadian Association for
the Committee they believed advance requests should Community Living, argued that the requirement for
be respected.3' Other witnesses also voiced support for the suffering of the patient to be intolerable "in the
advance requests.32 Wanda Morris, outgoing CEO of circumstances of his or her condition" bars the use of
Dying with Dignity Canada, argued in a similar vein to advance requests.37 Similarly, Prof. Trudo Lemmens
Professor Downie: from the Faculty of Law and Dalla Lana School of
Public Health at the University of Toronto expressed
In their decision, the Supreme Court justices wrote concern in written submissions after his appearance
that to force someone to choose between under- before the Committee. He felt that someone with de-
going a premature, perhaps violent, death and mentia who is still enjoying life could end up dying
enduring prolonged suffering is a cruel choice. We by MAID because he or she met the criteria related to
submit that unless the committee recommends suffering, such as not recognizing family members,
that informed consent be allowed by advance con- that was included in his or her advance request for
sent, the injustice will continue. MAID.38
Nowhere does this play out more than around Dr. Jeff Blackmer, Vice-President, Medical Profes-
the issue of dementia . . . sionalism at the Canadian Medical Association, noted
I think that what we do will actually be life- that the organization had not consulted its member-
affirming if we are able to provide a clear advance ship on the issue of advance requests because the
consent mechanism.33 issue was not addressed in Carter. However, he did
say that implementing advance directives is "incred-
Ms. Morris explained that objective, verifiable criteria ibly complex and difficult, because it's very hard to
must be included in any such request to assist a health capture all of the nuances and the specifics of a very
care team in assessing whether the criteria outlined in complicated medical condition and intervention ."39
11 111 i 11 II .111 IIILLI
Dr. Douglas Grant, Registrar and CEO of the College

of Physicians and Surgeons of Nova Scotia, without RECOMMENDATION 31: Conscientiously objecting
taking a position on whether advance requests should health care providers should be required to inform
be permitted, also noted that "a myriad of new issues" patients of all end-of-life options, including physician-
would need to be addressed if such requests were assisted dying, regardless of their personal beliefs.
permitted."
The Committee understands these challenges
but is deeply concerned that by exduding individuals patients in a respectful way regardless of the provider's
who want access to MAID but have lost competence, beliefs. Such requirements are often already included in
such individuals will be left to suffer or end their lives professional codes of ethics. For example, the Canadian
prematurely. This situation was exactly what the Carter Medical Association's Code of Ethics requires that phys-
decision sought to avoid. Allowing advance requests also icians provide patients with the information they need
provides comfort to individuals, reducing their psycho- to make informed decisions about their medical care.
logical suffering, knowing that their lives will not end in
a way that is against their wishes.41 Limiting the option
of advance directives to individuals who already have RECOMMENDATION 32: Conscientiously objecting
a diagnosis makes it easier to ascertain that there was health care providers should be required to ap-
informed consent. At that point, the person knows more propriately inform their patients of the fact and
about what he or she may expect in the future to provide implications of their conscientious objection
relevant direction in the request. The same safeguards to physician-assisted dying. Any ongoing treat-
to ensure competence and consent must be in place for ment of the patient must be provided in a non-
advance requests, and consideration could be given to discriminatory manner.
additional safeguards. Thought should be given to en-
couraging and possibly requiring health care practition-
ers to communicate regularly with their patients while Conscientiously objecting providers have a duty
they are still competent to ensure that their advance to inform new patients and, when appropriate, current
requests continue to reflect their wishes. The concerns patients of their position on physician-assisted dying.
of Dr. Blackmer, Dr. Grant and others will need to be This allows patients to make an informed choice with
examined as the system is put in place to minimize the respect to enrolling with or retaining that provider.
risk of abuse and error, but the Committee is confident Regulatory authorities and governments should pro-
that this can and must be done to ensure the autonomy duce educational materials for physicians and patients
of Canadians and the protection of the vulnerable. The to help navigate these discussions. We also encourage
Committee therefore recommends: all physicians to initiate and regularly revisit conversa-
tions with patients around advance care planning, the
RECOMMENDATION 7 goals of care, and the patient's end-of-life wishes.
That the permission to use advance requests for
medical assistance in dying be allowed any time Duty to Care for the Patient
after one is diagnosed with a condition that is
reasonably likely to cause loss of competence or
RECOMMENDATION 33: Conscientiously objecting
after a diagnosis of a grievous or irremediable
health care providers should be required to either
condition but before the suffering becomes in-
provide a referral or a direct transfer of care to
tolerable. An advance request may not, however, another health care provider or to contact a third
be made, prior to being diagnosed with such a party and transfer the patient's records through the
condition. The advance request is subject to the system described in Recommendation 4.
same procedural safeguards as those in place for
contemporaneous requests.
To facilitate access to physician-assisted dying,
Conscientious Objection conscientiously objecting health care providers must
refer, transfer care or contact a third party in a timely
Expert Advisory Group (pp. 44-7) manner and transfer records to another health care
Duty to Inform provider or third party, as required. There is no ex-
All end-of-life options (including palliative care and pectation that they will assess the patient's eligibility for
physician-assisted dying) should be discussed with physician-assisted dying.
We heard from some stakeholders who believed Duties of Institutions

that a referral in the context of physician-assisted dying All institutions have certain responsibilities i n common,
implied that the conscientiously objecting physician many of which align with our recommendations related
had assessed the patient, found them eligible for to physicians.
physician-assisted dying and was transferring care for
the purposes of providing the service. While this is an
RECOMMENDATION 34: All institutions should be
option available to those who object to the provision of
required to inform patients/residents of any Institu-
service, we did not want to imply that assessment is an
tional position on physician-assisted dying, includ-
obligation, or that this was the only option available to ing any and all limits on its provision.
providers.
When we discuss a direct transfer of care, we
envisage a situation where a patient has requested This recommendation will ensure that patients have
physician-assisted dying and has already discussed clarity on what is permitted within the facility. They
end-of-life options with their health care provider. We can then make informed decisions with respect to
recognize that some providers view a transfer of care as whether to enter or remain in the facility.
morally preferable to referral because, unlike referral,
it is taken to neither explicitly nor implicitly affirm the
service sought by the patient. A health care provider RECOMMENDATION 35: Provinces and territories
could transfer the patient to another health care pro- should prohibit any requirement by institutions that
vider for the assessment and treatment of the patient's patients give up the right to access physician-assisted
medical condition and, if the patient meets the eligibil- dying as a condition of admission.
ity criteria, provision of physician-assisted dying. The
receiving health care provider must be someone who is
willing and able to accept the person as a patient and
RECOMMENDATION 36: Provinces and territories
does not conscientiously object to physician-assisted
should prohbit any requirement by institutions that
dying. Following the transfer of care, the receiving
physicians refrain from the provision of physician-
health care provider would follow the pathway out- assisted dying external to the non-participating institu-
lined earlier in this report. tion. In addition, employment conditions or privileges
For those conscientiously objecting health care should not be negatively impacted in any way.
providers who are not willing to provide a direct
transfer of care on conscience or religious grounds,
we provide the option of contacting and transfer- Taken together, the two recommendations above limit
ring the patient's records through a publicly-funded the power of institutions to restrict the autonomy of pa-
system designed to ensure that patients are able to tients who may wish to seek physician-assisted dying at
access a health care provider willing to accept them a later date, and providers, who may wish to provide
as a patient for assessment and, if the eligibility cri- physician-assisted dying outside a faith-based institution
teria are met, to provide physician-assisted dying. If that does not offer physician-assisted dying. They repre-
this system is not in place, conscientiously objecting sent an appropriate balance between patient and physician
providers should be required to make a direct transfer autonomy, patient access, and institutional autonomy.
of care to a health care provider who is willing and
able to accept the person as a patient and does not Duties of Non Faith-Based Institutions
conscientiously object to physician-assisted dying. It
is imperative that the burden of the transfer to an-
RECOMMENDATION 37: Non faith-based institu-
other physician, institution or third party not fall on tions, whether publicly- or privately-funded, must
the patient seeking the care. not prevent physician-assisted dying from being
As described in Recommendation 4, we recom- provided at their facilities.
mend a system where regional health authorities de-
velop a publicly-funded care coordination system to
help facilitate access to physician-assisted dying. One Governments have significantly more levers to in-
of this system's key functions should be to receive re- fluence the policies and practices of institutions that are
quests from unwilling or unable health care providers funded in whole or in part by public funds. However, as
and transfer patients to providers who are willing and a point of principle, we believe that physician-assisted
able to receive new patients. F. . .1 dying should be available wherever people are living and
II ui IIILLJ
dying. This includes privately-funded institutions. Rec- witnesses who appeared before the Committee, and
ognizing that not all institutions will feel capable of pro- briefs/letters that were submitted to the Committee,
viding physician-assisted dying for a variety of reasons, discussed the extent to which health care practition-
including size, geography ar d mandate, we concluded ers should be able to refuse to participate in MAID for
that institutions should be permitted to offer a patient reasons of conscience.' No one was of the opinion that
transfer to another institution, as long as the receiving in- a health care practitioner should be obliged to perform
stitution can and will provide a health care provider who MAID. As the Supreme Court of Canada stated in Carter,
is willing and able to accept the person as a patient, to "Rin our view, nothing in the declaration of invalidity
assess whether the criteria for access to physician-assisted which we propose to issue would compel physicians to
dying have been met, and provide physician-assisted provide assistance in dying."44
dying where the criteria for access have been met. It was argued by some witnesses that strong protec-
tions for health care practitioners who refuse to participate
Duties of Faith-Based Institutions for reasons of conscience need to be put in place, includ-
ing the possibility that such protection be established in
legislation45 Other witnesses were concerned about the
RECOMMENDATION 38: raith-based institutions effect a practitioner's refusal to participate in MAID would
must either allow physician-assisted dying within have both on the individual who was seeking an assisted
the institution or make arrangements for the safe death and on the availability of MAID more broadly. As
and timely transfer of the patient to a non-objecting
Vyda Ng from the Canadian Unitarian Council told the
institution for assessment and, potentially, provision
Committee, " [i] t's very much in keeping with Canadian
of physician-assisted dying. The duty of care must
be continuous and non-discriminatory.
values to put the needs and wishes of Canadians ahead
of the values of individual doctors and institutions, and
to respect each person's dignity at the most traumatic
Faith-based institutions have a duty to care for period of their lives."46 Some witnesses and submissions
and not abandon the patients within their institution. to the Committee recommended that a practitioner who
While they should not be recuired to provide access to conscientiously objects to MAID should be required to
physician-assisted dying, they must still ensure access for provide an effective referral or transfer of care for their
patients who wish to seek it. When a patient makes a patient,47 while some felt that referring the individual to
request for physician-assisted dying, faith-based institu- a third-party organization should be sufficient.48 Joanne
tions should be required to either allow for the assess- Klineberg from the Department of Justice noted that
ment and provision of physician-assisted dying within provinces and territories "have legislation and policies in
the institution or make arrangements for an effective relation to the rights of physicians to refuse to partake
transfer of the patient to a non-objecting institution. This in certain types of medical practices, so it is definitely
transfer must also include the transfer of all relevant re- something that the provinces and territories already are
cords, and must be made to a non-objecting institution responsible for."49 In Quebec, a physician must notify a
where the patient's medical condition can be assessed and designated individual if he or she refuses to participate
treated by a health care provider who is willing and able in MAID so that a willing physician may be identified.5°
to assess whether the patient meets the eligibility criteria The Committee notes that the Supreme Court of
for physician-assisted dying and, if so, can provide assist- Canada in Carter stated that the Charter rights of patients
ance. If a safe and timely transfer to a non-objecting insti- and physicians will need to be reconciled.5' The Com-
tution is not possible, the objecting institution must allow mittee believes that having health care professionals who
an outside health care provide: to assess the patient and, conscientiously object to MAID provide an effective refer-
if the eligibility criteria for physician-assisted dying are ral for a patient who seeks MAID is an appropriate balan-
met, to provide assistance. The receiving outside health cing of the rights of patients and the conscience rights of
care provider would follow the pathway for physician- physicians. The Committee therefore recommends:
assisted dying as outlined previously in this report.
RECOMMENDATION 10
Special Joint Committee (pp. 25-27) That the Government of Canada work with
the provinces and territories and their medical
A. Conscientious Objection to Participating in regulatory bodies to establish a process that
Medical Assistance in Dying respects a health care practitioner's freedom of
The External Panel's report noted that "the medical conscience while at the same time respecting the
profession is divided over the issue of MAID."42 Many needs of a patient who seeks medical assistance
in dying. At a minimum, the objecting prac- argued, and the Committee also believes, that if a
titioner must provide an effective referral for health care facility is publicly funded, it must provide
the patient. MAID.53 The difficulty in transferring a patient from one
facility to another was highlighted.54
Witnesses and briefs also addressed whether a The Committee recommends therefore:
health care facility should be permitted to refuse to
either provide MAID or to allow MAID to be provided on RECOMMENDATION 11
its premises. One witness told the Committee that in That the Government of Canada work with the
Quebec, hospices (which the witness stated are largely provinces and territories to ensure that all publicly
privately funded) sought and received an exemption funded health care institutions provide medical as-
from having to provide MAID.52 A number of witnesses sistance in dying.
References
1. PDAM, Evidence, 26 January 2016, 1805 (Pelletier). had historically narrowed dramatically when applied to
2. PDAM, Evidence, 25 January 2016, 1240 (Hogg). children. However the common law has more recently
3. PDAM, Evidence, 2 February 2016, 1940 (Grant). abandoned the assumption that all minors lack decisional
4. See, for example, PDAM, Evidence, 4 February 2016, 1640 capacity and replaced it with a general recognition that
(Carmela Hutchison, President of Disabled Women's children are entitled to a degree of decision-making
Network of Canada) (DAWN); PDAM, Evidence, 4 February autonomy that is reflective of their evolving intelligence
2016, 1920 (Bach). Regarding the suggestion of age 25, see and understanding. This is known as the common law
Colette Squires, Physician Assisted Dying Public Consultation, "mature minor" doctrine. As the Manitoba Law Reform
January 30, 2016, in Langley, B.C., submission to the Com- Commission noted, this doctrine is "a well-known, well-
mittee, p. 4. accepted and workable principle which . . . raise's] few
5. Provincial-Territorial Report, Recommendation 17. Also difficulties on a day-to-day basis" (Minors' Consent to
see, for example, PDAM, Evidence, 2 February 2016, Health Care (1995), Report #91, at p. 33). The doctrine
1735 (Dr. Derryck Smith, Chair of Physicians Advisory addresses the concern that young people should not
Council, Dying with Dignity Canada). automatically be deprived of the right to make decisions
6. PDAM, Evidence, 2 February 2016, 1735 (Smith). affecting their medical treatment. It provides instead that
7. PDAM, Evidence, 18 January 2016, 1415 (Klineberg). the right to make those decisions varies in accordance
8. PDAM, Evidence, 26 January 2016, 1835 (Gibson); PDAM, with the young person's level of maturity, with the degree
Evidence, 1 February 2016, 1100 (Pastine). to which maturity is scrutinized intensifying in accord-
9. PDAM, Evidence, 26 January 2016, 1810 (Pelletier). ance with the severity of the potential consequences of
10. PDAM, Evidence, 3 February 2016 (Mary Shariff, Asso- the treatment or of its refusal
ciate Professor of Law and Associate Dean Academic, 17. PDAM, Evidence, 18 January 2016, 1530 (Jeanette Ettel,
University of Manitoba, Canadian Paediatric Society). Senior Counsel, Human Rights Law Section, Depart-
11. PDAM, Evidence, 2 February 2016, 1815 (Smith). ment of Justice).
12. PDAM, Evidence, 4 February 2016, 1730 (Margaret Bir- 18. PDAM, Evidence, 26 January 2016, 1750 (Benoit Pelletier,
rell, President, Alliance of People with Disabilities Who Member, External Panel on Options for a Legislative Re-
Are Supportive of Legal Assisted Dying Society); 4 Feb- sponse to Carter v. Canada).
ruary 2016, 1900 (Dr. John Soles, President, Society of 19. PDAM, Evidence, 28 January 2016, 1850 (Downie). See
Rural Physicians of Canada). also, for example, Centre for Inquiry, p. 4, and PDAM,
13. PDAM, Evidence, 4 February 2016, 1915 (Dr. Hartley Evidence, 26 January 2016, 1900 (Taylor).
Stem, Executive Director and CEO, the Canadian Med- 20. See, for example, Living with Dignity, Recommendations
ical Protective Association). for the Special Joint Committee on Physician Assisted Dying,
14. An Act respecting end-of-life care, section 5. p. 3, written submission to the Committee; Derek B.M.
15. A.C. v Manitoba (Director of Child and Family Services), Ross & Johnathan R. Sikkema, Christian Legal Fellow-
2009, sec 30, para. 69. ship, Submission of the Christian Legal Fellowship to the Spe-
16. In A.C. v Manitoba (Director of Child and Family Services), cial Joint Committee on Physician-Assisted Dying, 1 February
2009 scc 30, the Supreme Court of Canada discusses 2016, p. 4.
the ability of minors to consent to medical treatment in 21. PDAM, Evidence, 27 January 2016, 1935 (Dr. K. Sonu
the context of protective legislation that allows a court Gaind, President, Canadian Psychiatric Association).
to authorize treatment for a child when it deems it to 22. PDAM, Evidence, 3 February 2016, 1805 (Dr. Tarek Rajji,
be in the child's best interest. At para. 46, Justice Abella Chief, Geriatric Psychiatry, Centre for Addiction and
(for the majority) states: The latitude accorded to adults Mental Health).
at common law to decide their own medical treatment 23. Carter, 2015, para. 127.
H Ju 111111
24. Regarding Belgium, for example: neuropsychiatric 38. Trudo Lemmens, Response to Comments Made during
disorders were 1.2% of cases in 2004/05, 2.8% (or 58 the Committee Hearings of January 28, 2016, 2 February
cases) in 2010/11 and 3.7% (or 67 cases) in 2013/14 2016, submission to the Committee.
according to Trudo Lemmens, Why Canada Should Avoid 39. PDAM, Evidence, 27 January 2016, 1750 (Blackmer).
a Belgian-Style Regulatory Regime for Physician Assisted 40. PDAM, Evidence, 2 February 2016, 1925 (Grant). See also,
Dying, Memorandum for the Special Joint Committee on for example, PDAM, Evidence, 3 February2016, 1850 (Radji).
Physician-Assisted Dying, written submission to the Com- 41. Provincial-Territorial Report, p. 31.
mittee, p. 6. 42. External Panel Report, p. 98.
25. See, for example, PDAM, Evidence, 4 February 2016, 43. The issue of freedom of conscience of pharmacists was
1705 (Somerville), regarding limiting MAID to physical raised by the Canadian Pharmacists Association (CPbA),
suffering; and PDAM, Evidence, 1 February 2016, 1155 PDAM, Evidence, 27 January 2016 (Phil Emberley, Can-
(Wanda Morris, Dying with Dignity), regarding psycho- adian Pharmacists Association).
logical suffering. 44. Carter, 2015, para. 132.
26. Marcia Hogan, Brief to the Joint Committee on Phys- 45. See, for example, PDAM, Evidence, 27 January 2016,
ician Assisted Dying, submission to the Committee, p. 2. 1725 (Branigan); PDAM Evidence, 3 February 2016,
27. Carter, 2015, para. 14. 1905 (Hashmi); 1705 (Collins); Letter to Minister Wilson-
28. PDAM, Evidence, 27 January 2016, 1935 (Gaind). Raybould and Minister Philpott, Canadian Conference of
29. Carter, 2015, para. 57-58. Catholic Bishops, 20 January 2016.
30. PDAM, Evidence, 28 January 2016, 1850 (Downie). 46. PDAM, Evidence, 3 February 2016, 1705 (Ng).
31. PDAM, Evidence, 28 January 2016, 1735 (Jarrett); 1805 47. See for example, PDAM, Evidence, 26 January 2016,
(Fletcher). 1900 (Gibson); PDAM, Evidence, 1 February 2016, 1125
32. See, for example, PDAM, Evidence, 1 February 2016, (Moms); Rhonda Morison, submission to the Commit-
1105 (Paterson); PDAM, Evidence, 4 February 2016, tee, 30 January 2016.
1655 (Angus Gunn, Counsel, Alliance of People with 48. PDAM, Evidence, 1 February 2016, 1105 (Paterson), submis-
Disabilities Who Are Supportive of Legal Assisted Dying sion to the Committee, Vivre dans la Dignite 2016; Ellen
Society); Centre for Israel and Jewish Affairs, Brief for Agger, submission to the Committee, 1 February 2016.
the Special Joint Committee on Physician Assisted Dying, 49. PDAM, Evidence, 18 January 2016, 1535 (Klineberg).
submission to the Committee, p. 4-5. 50. An Act Respecting End-of-Life Care, RSQ c S-32.0001,
33. PDAM, Evidence, 1 February 2016, 1120 (Morris). section 31.
34. Ibid., 1155. 51. Carter, 2015, para. 132.
35. PDAM, Evidence, 25 January 2016, 1255 (Menard). 52. PDAM, Evidence, 2 February 2016, 1845 (Baxter).
36. PDAM, Evidence, 1 February 2016, 1205 (Jay Cameron, 53. See, for example, PDAM, Evidence, 3 February 2016, 1700
Barrister and Solicitor, Justice Centre for Constitutional (Ng); British Columbia Humanist Association, Allow Assisted
Freedoms). Dying for All Who Choose It. A Brief for the Special Joint Com-
37. PDAM, Evidence, 4 February 2016, 1925 (Bach). See also, mittee on Physician-Assisted Dying, 25 January 2016; Lori
for example, POAM, Evidence, 4 February 2016, 1930 Goodwin, submission to the Committee, 30 January 2016.
(Gerald Chipeur, Lawyer, as an individual). 54. PDAM,Evidence, 2 February 2016, 1850 (Baxter).
6.4 The Concept of Death and Its Practical Implications
Defining Death is a matter of importance. It is, however, surprisingly

difficult to identify that time. The aim of this essay is to
Mister Browne explain why there is a problem and to canvass the main
solutions that have been offered.
The end of life is death, and death is typically thought
to be a time of great behavioural significance. For ex- Definitions of Death
ample, it is normally though: that we can do certain
things such as routinely remove all life-support, mine The earliest understanding of death was in terms
an organ donor's body for transplantable organs, or of the loss of heartbeat and respiration. There was
bury a person after they are dead but not before, and always a problem of saying how long these func-
thus that determining when exactly a person is dead tions must be absent before a person is dead, and
This article was written specifically for this book.

Browne: Defining Death
since that exact interval was not (and still is not) technical, legal, or moral? The literature on this is large
known, death was defined as the "prolonged" ab- and inconclusive, and it is worrisome that a key term
sence of these functions. This gave rise to worries that cannot be eliminated remains undefined.
about premature burial, but the real demand for A fourth definition of death has been added to the
greater precision came from the invention of venti- list in the past 15 years—cardiocirculatory death for
lators and the advent of transplantation. Ventilators DCD—and is valid only for what is known in Canada
made it possible to sustain cardiorespiratory func- as "donation after cardiocirculatory death (DCD)"
tion long after it was no longer spontaneous. This and elsewhere as "non-heart-beating organ donation
provoked the question of whether such a patient is (NHBoD)." The aim of this protocol is to increase the
alive or a corpse being ventilated, and the possibil- supply of transplantable organs by enabling death to
ity of transplantation made it important to know. be declared before whole-brain death is diagnosed and
If the person could be declared dead while on the thus enlarging the donor pool. In DCD, after the patient
ventilator, transplantation could begin immediately or family has consented to the procedure, the ventilator
with organs kept in the freshest possible condition. is removed, and once heartbeat and respiration cease,
If not, the ventilator would have to be disconnected the patient is continuously observed for two to ten (the
and the patient declared dead by cardiocirculatory interval varies from centre to centre, the most common
criteria before organs could be excised. being five) minutes. If there has been no pulse for that
The whole-brain definition of death was invented time, death is declared, and transplantation starts to
to solve this problem. On this view, death occurs when take place.
the upper brain (which sponsors consciousness) and
lower brain (which controls heartbeat and respiration) Selecting a Definition
have permanently ceased to function. This means that
once a person no longer has spontaneous respiration That is the menu of choices of when to declare death.
and heartbeat, they are dead. Death can thus be de- To have a uniform definition of death—i.e., a single
clared either clinically by the permanent cessation of all-purpose definition of death for a society—one of
spontaneous heartbeat and respiration or neurologic- the first three of these (the last is only for DCD) must
ally by the irreversible cessation of all brain function. be chosen, but it is not clear how to select one. It is
Patients on ventilators with artificially supported sometimes suggested that a committee of physicians
respiration and heartbeat could thus be declared or health care professionals should choose, but the
dead. This allowed for an uncontroversial termina- decision is not a matter for empirical investigation. Sci-
tion of treatment and (more importantly) facilitated ence can tell when cardiorespiratory, whole-brain, and
transplantation. higher-brain death occur, but it cannot tell us which
Thus, whole-brain death became the standard one should be picked as the death of a human being.
definition of death. But it had critics. Some argued Deciding which of these candidates is the time of death
that it diagnosed death too soon on the ground that is, rather, more like deciding what the age of majority
a person with artificially supported heartbeat and is, and one may suggest that the question should be de-
respiration was still alive, perhaps not alive in a state cided in the same way. We decide on the age of major-
worth preserving, but alive nonetheless and not ity by deciding when it is appropriate to allow people
dead until all spontaneous and artificially supported to vote, drive, drink in public, etc. and identify that
cardiorespiratory function had irreversibly ceased. time as that age. Similarly, one may contend that we
Others argued that the whole-brain definition pegs should proceed by determining when death-behaviour
the time of death too late. On this view, a person is appropriate—i.e., behaviour that is appropriate if
is dead as soon as the possibility of consciousness and only if a person is dead—and identify that time as
is permanently lost, and that occurs as soon as all the time of death.
higher-brain function has irreversibly shut down. This, however, does not yield a clear answer.
Other brain activity is no more relevant than con- Death-behaviour covers a wide range, includ-
tinued cellular activity in non-vital parts of the body ing disparate things such as when it is possible to
is for the whole-brain account. routinely withdraw all health care, take organs for
All the above definitions define death as an ir- transplantation, use the body for experimental or
reversible state, but "irreversibility" is an undefined teaching purposes, begin an autopsy, initiate burial
term, and it is not easy to define it. Does it mean that or cremation procedures, prosecute a crime as
the state will not reverse itself or cannot be reversed, murder, inherit an estate, and remarry without di-
and if the latter, is the impossibility logical, empirical, vorce. It is not clear that all these items of behaviour
MI Bli II Al II Al 111111
become appropriate at the same time. Indeed, it is Death?" cited in the section "Defining Death" in the
pretty clear that they do not. But if they do not, Suggested Further Readings) has proposed that an in-
this way of selecting between the candidates for the dividual could authorize organ retrieval under general
time of death will not work. Moreover, even if all anaesthetic without first undergoing an orchestrated
death-behaviour did coalesce at a single point, the withdrawal of life support in which the patient is de-
original aim in seeking a definition of death would clared dead. On this view, organ procurement would
be lost. That aim was to define death so that death- be controlled by considerations of consent and harm
behaviour could be initiated, and on this account to the donor rather than the concept of death. When
we have to know when death-behaviour is appropri- the patient does not want to live in that state and
ate before we can define death. wants to be an organ donor, this could be viewed as a
This difficulty in choosing between candidates species of euthanasia—euthanasia by organ donation,
for the time of death has led some to abandon a search or if one prefers, organ donation by euthanasia.
for a single uniform definition of death and accept
instead different definitions for different purposes.
Death in Canada
The age of majority differs depending on whether we
are talking about the right to drive, drink in public, This ends our quick and very incomplete tour of the
vote, marry, live on one's own, and so forth, and the hard and complex question of defining death. The
suggestion is that we can have different definitions of debate over whether to seek a uniform definition of
death for different items of death-behaviour. This has death or individualize death, and what would be the
in turn led to a more radical suggestion—namely, to best way to provide either, continues without a clear
allow individuals to pick th e times that they would end in sight. The Canadian Law Reform Commis-
like various items of death-behaviour initiated for sion in 1981 recommended that Canada adopt the
them rather than have them designated by the state. whole-brain definition, but that was not accepted, and
Robert Veatch, for example, has suggested that soci- Canada remains without a legislated definition. None-
ety adopt whole-brain deat:i as a default definition theless, in Canada (and the world generally) death is
but allow individuals to deviate from that by select- understood either as whole-brain death—diagnosed
ing other definitions of death for different death- by its cardiorespiratory criteria for the vast majority of
behaviour. On his proposal, :ndividuals would not be patients, its neurological criteria for organ donors on
able to choose a definition of death earlier than the ventilation—or the modified cardiocirculatory criteria
irreversible cessation of consciousness or later than used for DCD.
the loss of all circulatory and respiratory functions,
but within that range they should be free to say what Further Reading
death-behaviour can be visited on them and when.
An even more radical view has also been pro- The following readings will be useful to those who
posed. On this view, we should give up the so-called would like to pursue the topic of defining death.
"dead donor rule." Rather than adjust the definition of References to the articles by the authors mentioned
death to allow for death-behaviour to occur at differ- here are in the Suggested Further Readings under the
ent times for different purposes, the suggestion is to heading "Defining Death." Robert Truog provides an
uncouple the definition of death from its traditional explanation and critique of whole brain death and
death-behaviour. If it can be determined what can be gives an exceptionally clear survey of the debate on
done to whom and when, legislation permitting those defining death. James L Bernat brings us to the fron-
items of behaviour could be passed without using the tier of the development of DCD and identifies issues
word "death" at all. For example, instead of saying that are yet to be resolved and yet to come. Readings
that transplant proceedings can only begin when a representing the main positions on defining death
person is dead (or dead in the sense that would li- are: "Higher-Brain Death," Green and Wikler; "The
cense transplantation), we could say that transplanta- Individualization of Death," Emanuel; "Eliminating
tion cannot begin until the heart has stopped beating the Concept of Death," Dworkin. The Canadian ver-
with or without mechanical assistance (or whatever sion of DCD is stated by the Canadian Council for
conditions one thinks apt). The same would apply Donation and Transplantation (ccDT) and the Amer-
for deciding when it is appropriate to distribute an ican version by the Institute of Medicine (IOM). The
estate, remarry without committing bigamy, and all reading by Gail A. Van Norman identifies some of
the other items of traditional death-behaviour. In this the many medical and moral complexities that arise
spirit, Robert D. Truog ("Is It Time to Abandon Brain with DCD.
Cases
6.5 Cases
Case 1 artificial respiration be discontinued was "real and

enlightened."
Nancy B.: Withdrawing
The judge (Dufour J.) posed himself the questions:
Life-Sustaining Treatment "Can the conduct of a physician who stops the respiratory
support treatment of his patient at the freely given and in-
formed request of the patient, and so that nature may take
Nancy B. (her last name was never disclosed) was a
its course, be characterized as unreasonable? Or does such
25-year old woman who suffered from Guillain-Barre conduct denote wanton and reckless disregard?' Dufour
syndrome, a disabling neurological disorder. In Janu-
J. answered both questions, "I do not believe so." The re-
ary 1991, she was informed that her condition was ir- sulting judgment was that, on expiration of the time for
reversible. Her respiratory muscles had atrophied, and appeal from the judgment, the court (1) permits Nancy
degeneration of the motor nerves left her tied to her B.'s attending physician to stop the respiratory support
hospital bed, entirely dependent for survival on mech-
treatment when Nancy B. so desires and (2) permits the
anical ventilation by intubation.
physician to request from the hospital the necessary assist-
In the months following this, Nancy B. deter-
ance in circumstances such as these so that everything can
mined that she was unwilling to survive in her intub-
take place in a manner respecting her dignity.
ated, dependent existence. She was not diagnosable Dufour J. found that the circumstances of Nancy
as clinically depressed but was distressed that life
B.'s proposed death would constitute neither homicide
could afford her nothing more than limited head
nor suicide but rather natural death. In support of this,
movement and utter dependency for comfort on Dufour J. cited the extremely influential New Jersey Su-
others and a machine. Her intellectual faculties were
preme Court case of Re Conroy to the effect that: "Declin-
intact, and she was lucid and logical in expression of
ing life-sustaining medical treatment may not properly
her wish to be disconnected from the mechanical res-
be viewed as an attempt to commit suicide. Refusing
pirator. She knew that the consequence would be that medical intervention merely allows the disease to take
she would die in a very short time. Her mother said its natural course; if death were eventually to occur, it
that her family had come to accept her wish, and the
would be the result, primarily, of the underlying disease,
Quebec Superior Court judge (who himself attended and not the result of a self-inflicted injury"
her in hospital) found that Nancy B.'s preference that
Source: Taken from Bernard M. Dickens, "Medically The citation for the case is: Nancy B. v. Hotel-Dieu de
Assisted Death: Nancy B. v. Hotel-Dieu de Quebec," McGill Quebec [1992] R.J.Q. 361, 86D.L.R. (4th) 385 (Sup. Ct.)
LawJournal 38 (1993), 1053-70.
Case 2 healing from his prior surgeries had become impos-

Dr Nancy Morrison: Nonvoluntary sible. Pus continuously oozed from a gap in the chest
wall, which could not be covered. Estimates suggest
Active Euthanasia of an Adult that there were approximately 10 chest tubes, stomach
tubes, and Ns connecting Mr Mills's body to machines
In 1996, 65-year-old Halifax, Nova Scotia, resident Paul and equipment. A DNR (Do Not Resuscitate) order was
Mills was suffering from terminal esophageal cancer that put in place, and on 9 November, when it became clear
had required previous removal of his esophagus and that there was no hope of recovery the family agreed to
repositioning of his stomach in an attempt to repair the withdraw active life support in favour of palliative care.
gap. By mid-October of 1996, Mr Mills had experienced On the morning of 10 November, Mr Mills remained
nine additional operations at two different hospitals due in his heavily sedated state and was still receiving narcot-
to a severe post-surgical infection and was profoundly ics for pain. His feed tubes and antibiotics were with-
depressed, expressing a wish to die. Between 15 October drawn, and he was given increasing doses of Dilaudid
and 6 November, his weight dropped by 19 kg (42 lbs), for pain. At approximately 12:30 p.m., the respirator
and the infection had developed to such a degree that was withdrawn; however, the patient did not then die
as expected but rather struggled for air for quite some In May 1997, a hospital physician who had seen an
time. Pus oozed from his incisions as he laboured for internal document about the case and was afraid that
breath. His attending nurse, Elizabeth Bland-Maclnnes, the hospital would attempt to cover it up reported the
described his struggle for air as "a horrible and hideous incident to the police as an instance of active eutha-
scene" and said that "it was beyond the shadow of a nasia. This resulted in Dr Morrison being charged
doubt the worst death I've ever witnessed." The only with first-degree murder on 6 May. In February 1998,
HCP consistently present at the bedside during these Judge Hughes Randall declined to commit Dr Morrison
hours, Bland-Maclnnes had no doubt that Mr. Mills was to stand trial. He noted that Mr Mills had been given
suffering. However, the icu resident, Dr Cohen, only extraordinary amounts of Dilaudid, morphine, and
described Mr Mills as "in distress" and "apparently in other painkillers prior to Dr Morrison's involvement in
discomfort" but would not commit to stating that the his treatment. These potentially lethal drugs had been
patient was conscious and therefore aware and suffering. legally administered earlier in an unsuccessful attempt to
There is speculation that Mr Mills lost consciousness relieve Mr Mills's apparently intense suffering after being
when the tubes were removed that morning and there- removed from the respirator and could have been the
fore some question as to whether he was experiencing cause of death rather than the potassium chloride Dr Mor-
an agonizing death or not. Whether or not Mr Mills was rison injected. Because Dr Morrison used drugs that have
conscious and requesting assistance to die is relevant to no pain-killing, anaesthetic, or sedative properties, Nova
determining whether this was a case of voluntary as op- Scotia's College of Physicians and Surgeons stated that the
posed to nonvoluntary active euthanasia. use of these drugs had no place in Mr Mills's medical man-
What is clear is that Nurse Bland-Maclnnes was agement.2 In delivering their formal reprimand to Dr Mor-
deeply disturbed by Mr Mills's apparent distress. She rison, the college labelled her actions as "inappropriate
communicated to Dr Nancy Morrison her exasperation and outside the bounds of acceptable medical practice."1
that sedatives and narcotics were ineffective. Reports The college would have preferred Dr Morrison to have
indicate that the nurse begged Dr Morrison to help chosen one or more of these alternatives: (1) consulting
end the patient's suffering. At 2:52 pm, Dr Morrison with other intensive care physicians; (2) ensuring correct
injected nitroglycerin into the Iv line in an unsuccessful functioning of the Iv; (3) using alternative medications;
attempt to decrease blood pressure and thereby end the (4) re-establishing a supplementary airway; or (5) con-
patient's suffering. Seven minutes later, Dr Morrison re- tinuing with the chosen course of treatment, knowing that
turned to inject Mr Mills with potassium chloride, and death was imminent.2 This reprimand now appears on Dr
within minutes he was dead. Morrison's formal discipline record held at the college.
Notes
1. Barney Sneiderman, and Raymond Deutscher, "Dr. Nancy 2. College of Physicians and Surgeons of Nova Scotia,
Morrison and Her Dying Patient: A Case of Medical "Morrison Accepts College Reprimand" (1999) [online
Necessity" (2002), www.umanitoba.ca/centres/ethics/ press release), www.cpsns.ns.ca/pr/morrison_3_30_99
articles/BarneyART3.pdf (accessed 11 December 2007). .html (accessed 16 December 2007).
Case 3 swallow—her throat had to be manually massaged

Tracy Latimer: Nonvoluntary Active to allow her to ingest food. Her full-brain impair-
ment left Tracy with the mental age of a four- to five-
Euthanasia of a Minor month-old infant and a tendency toward seizures.
Seizure medication reduced the number of seizures
On 23 November 1980, T-acy Latimer of Wilkie, to approximately five per day but decreased Tracy's
Saskatchewan, suffered from oxygen deprivation respiration and digestive systems, further burdening
during her birth, which resulted in an extreme form these critically over-taxed systems and diminishing
of cerebral palsy, an incurable, permanent brain Tracy's quality of life.
injury typically affecting muscle control and move- By the age of four, increasing muscle tension and
ment.' Tracy was a spastic quadriplegic, unable to atrophy caused Tracy continuous pain, which was
walk, talk, or feed herself; indeed, she could not even treated with her first of three surgeries. It was at this
Cases 343
time the family discovered that Tracy could not take fact, constitute a mutilation of their daughter.3 In con-
analgesics stronger than regular Tylenol because of sideration of her situation, Mr Latimer decided that
concerns about interactions with her anti-seizure and Tracy's death was a better alternative than constant
anti-convulsion medications. This would have sig- surgery and unremitting pain. On Sunday, 24 October
nificant implications for Tracy's quality of life as her 1993, while his family was at church, Robert Latimer
condition deteriorated. A second surgery, in 1990, placed 12-year-old Tracy in the cab of his pickup
was meant to relieve the 10-year-old Tracy's muscle truck and piped in carbon monoxide fumes from the
tension, redistribute strain, and alleviate pressure on truck's exhaust, killing his daughter by carbon mon-
her hips to minimize the potential for hip dislocation; oxide poisoning.
however, the procedure was only minimally success- Mr Latimer was found guilty of the second-
ful and relieved pressure for just a few months. She degree murder of his daughter both at trial and on
developed scoliosis—curvature of the spine—and her appeal. The jury recommended that he be eligible
hip became dislocated as muscles were pulled away for parole after one year rather than have to serve
from the bones. Problems with eating and digestion, the mandatory minimum sentence for second-degree
including vomiting, developed as the spine and vital murder-10 years before parole is considered. In
organs were compressed by the scoliosis, and the fre- December 1997, Justice Ted Noble granted Latimer
quency of her seizures increased.' a constitutional exemption from the minimum sen-
In 1992, the scoliosis had developed to 75 per cent tence, saying that imposing the minimum sentence
off perpendicular. Tracy required a third surgery, to on Latimer in this case of mercy killing, as opposed
have a steel rod inserted on either side of her spine, to cold-blooded killing, would constitute "cruel and
held secure by holes drilled into her pelvic bones. unusual punishment." In November 1998, the Sas-
This surgery, like the previous one, relieved her katchewan Court of Appeal overturned Noble's ruling
symptoms for only a few months and introduced a and imposed the mandatory sentence: 25 years in
whole new set of problems. Since the steel rods made prison with no chance of parole for 10 years. In De-
Tracy's body rigid, there were few positions in which cember 2007, after having served 10 years in prison,
she could rest. She developed painful pressure sores Latimer was refused day parole. The parole board
and skin deterioration, had difficulty sleeping, and reported that they were "struck by Latimer's lack of
was losing weight. insight into the crime he had committed. . . . Latimer
In 1993, as Tracy's condition deteriorated and should stay in prison and receive more counselling."4
the physicians recommended more surgeries, Robert Latimer had told the parole board that he felt no guilt
Latimer, Tracy's father, became increasingly concerned for killing his daughter and had stated repeatedly
about his daughter's quality of life. A proposed fourth during the hearing that it was the right thing to do.
operation would involve surgically removing Tracy's When asked if he had the moral authority to take
upper thigh bone, a procedure that the family was someone's life, Latimer replied that the laws were less
told would be excruciatingly painful to endure, with important than the welfare of his daughter, who was
the pain continuing long afterward. Additional sur- in a lot of pain. In February 2008, after an appeal
geries would be required to mitigate the pain Tracy spearheaded by the BC Civil Liberties Association, a
felt in other joints. Tracy would have to endure all of review board reversed the parole board's decision and
this with inadequate pain management. The Latimers granted Robert Latimer day parole in 2010; he can
felt that further surgery would be futile and would, in now apply to travel outside Canada.
Notes
1. Ontario Foundation for Cerebral Palsy "What Is Cerebral 4. CBC News, "Parole Board Denies Latimer's Bid for Partial
Palsy?" (2007), www.ofcp.on.ca/aboutcp.html##A. Freedom" (2007), www.cbc.ca/canada/story/2007/12/05/
2. Robert Latimer, "Tracy's Illness" (2007), Latimer-parole-mtg.html.
www.robertlatimer.net/story/tracysillness.htm.
3. Supreme Court of Canada, 2001, R. v Latimer, 12001]
1 SCR 3, 2001 5CC 1, http://scc.lexum.umontrealca/
en/2001/2001scc1/2001sccl.html.
II All 1111:111
Case 4 present patients with options, particularly those pa-

Mr McCullough: Recommending tients who have reached the point of despondency, to
allow patients to make well-informed choices about
Voluntary Passive Euthanasia their future. Both McCulloughs saw it as an issue of
planting seeds of fear in elderly patients: "How many
In September 2003, 85-year-old Burnaby, BC, resident aged patients over the past year have been slipped this
Carl McCullough went to his local newspaper recount- information?"'
ing his concern that health cave providers were encour- In an environment where the need for medical
aging him to take his own life. Three times a week, for resources exceeds supply, one way to reduce the drain
seven years, Mr McCullough had been having kidney on resources is to exclude some categories of patients,
dialysis treatments at a local hospital, treatments that such as the elderly or addicts, from receiving medical
hadn't always been easy for him to endure. On two treatment—even if doing so hastens death. HcPs may
separate occasions during dialysis, as Mr McCullough see the disclosure of the option to withhold or withdraw
complained about how hard he was finding the treat- treatment in order to hasten death and end suffering as
ments that day, two separate HCPs offered him substan- providing information necessary to a patient's informed
tially the same unsolicited advice: "If I was getting fed consent. On the other hand, however, it opens the pos-
up with things, I could take this way out: just don't sibility that patients may perceive, rightly or wrongly,
come to dialysis."' When Mr McCullough's son, Davin, that they are being pressured to "let nature take its
discussed his father's concerns with hospital staff and course" so they'll die sooner and cease to be a burden.
administrators, he was told that it was important to This appeared to be the concern for the McCulloughs.
Note
1. Julie MacLellan, "A Question o f Life o r Death?" i n Burnaby:

NOW 27 September 2003, 1, •4.
Case 5 that if she'd been able to commit suicide with assist-

Elizabeth and Eric MacDonald: ance in Canada, she would have lived longer: "she
would have held on a bit longer because she was ter-
Assisted Suicide rified of being trapped in her own body and being
unable to travel."2 Ms MacDonald's feelings were made
On 8 June 2007, 38-year-old Elizabeth MacDonald of clear in a letter she left in which she expressed her
Windsor, Nova Scotia, died at .3 clinic in Zurich, Switz- frustration with Canadian law: "It is intolerable and
erland, after drinking a fatal dose of barbiturates she unacceptable that I cannot be assisted to die here in
received with the assistance of the Swiss suicide group Canada, in my own home, in my own bed, surrounded
Dignitas. Staff at the clinic provided Ms MacDonald by those I love."2
with the glass of barbiturates and told her that if she When Ms MacDonald's obituary later publicly
drank it, she would die. She said, "I understand that" thanked the staff at the Dignitas clinic, Alex Schaden-
and drank it without hesitation. She died in her hus- berg, a member of the Canadian-based Euthanasia
band's arms shortly afterward. Prevention Coalition (EPc) contacted the police, re-
Ms MacDonald had been suffering from multiple questing an investigation into whether Ms MacDon-
sclerosis, an autoimmune disease affecting the central ald had been counselled to commit suicide while on
nervous system, which can result in a multitude of Canadian soil. At that time, Canadian law punished
mobility problems, spasticity, and tremors, as well as assisting suicide with up to 14 years in prison, whereas
cognitive and emotional issues, including depression. Swiss law allowed assisted suicide if it was done for
Ms MacDonald had a severe form of the disease and unselfish reasons. (Assisted suicide is now called PAD in
had been wheelchair-bound, unable to move. Recently, Canada and became legal in 2015.) Attention turned to
her throat had begun to paralyze.' Her condition was her husband, Eric MacDonald, a retired Anglican min-
so distressing to her that she had already attempted to ister who had accompanied her to Switzerland and was
commit suicide a year earlier. Her husband believed present at her death. However, Canadian prosecutors
Cases
concluded that accompanying someone to the place of when evidence of wrongdoing occurs. He criticized Mr
their suicide and being with them during the act is not MacDonald for accompanying his wife to her suicide:
the same thing as "aiding and abetting" a suicide; with "The woman was not terminally ill. She had MS. She
no evidence that Eric counselled Elizabeth to commit had a disability. He [MacDonald] might consider it to
suicide, no charges were laid against him. be a loving act, but really what she didn't need was
Schadenberg responded to the case by reminding death. I consider the act of her husband to be an aban-
us that Canadian laws are designed to protect the vul- donment of her needs in this situation. This is not a
nerable and should remain unchanged and enforced supportive act."2
Notes
1. CBC News, "RCMP t o Question Man Who Took Ill Wife to 2. John Jalsevac, CBC News, "No Charges Laid in Nova
Commit Suicide Overseas," 27 June 2007, www.cbc.ca/ Scotia Assisted Suicide Case," CBC News 7 July 2007,
canada/story/2007/06/27/suicide-assisted.html. wwwlifesite.net/ldn/2007/jul/07070511.htm.
6.6 Study Questions
1. What argument can be given to show that physicians who WLST do not kill their
patients? What reply can be made to this?
2. How does the DDE allow physicians to perform terminal sedation but still oppose PAD?
Does the DDE succeed in reconciling the prohibition of PAD and permissibility of ter-
minal sedation? Is there a good reason to hold that physicians must never intentionally
kill their patients?
3. What is the main philosophical argument for legalizing PAD? What is the Constitu-
tional argument? What are the main arguments against legalizing PAD? Do you think
PAD should be legalized?
4. What reasons can be given to extend Bill C-14 to allow for advance directives and be
available to mature minors and (through substitute decision-makers) incapable chil-
dren and adults? What reasons can be given for requiring that death be foreseen? Do
you think those reasons are sufficient? Why or why not?
5. What are the two approaches to special safeguards articulated by the SCC? What are
the arguments for and against each? Which do you prefer, and why?
6. What obligations, if any, do health care providers who have conscientious objection to
participating in PAD have to help patients who want PAD to access it? Should institu-
tions that are publicly funded and have conscientious objection to PAD be required
to allow it on their premises, or should they be allowed to transfer patients to other
hospitals to have it carried out?
7. What is cardiocirculatory death, whole-brain death, and higher-brain death? Will any
of these serve as a uniform definition of death for our society? Should Canada have
a uniform definition of death, and if so, what? Or should the definition of death be
individualized, and if so, how? What do you think of DCD?
WLST and Refusal ofTreatment

Beauchamp, Tom L., and Robert Veatch, eds. 1996. Ethical Issues in Death and Dying. Upper
Saddle River, NJ: Prentice-Hall.
Bernat, James L., Bernard Gert, and R. Peter Mogielnicki. 1993. "Patient Refusal of Hydra-
tion and Nutrition: An Alternative to Physician-Assisted Suicide or Voluntary Active
Euthanasia." Archives of Internal Medicine 153 (27 December): 2723-8.
• El 11 .111 111111
DeGra.zia, David. 1992. "On the Right of Nondangerous' Incompetent Patients to Leave
Psychiatric Units against Medical Advice." Contemporary Philosophy 14 (September): 1-5.
Gostin, Lawrence 0.1991. "Life and Death Choices after Cruzan." Law, Medicine & Health
Care 19: 9-12.
Ross, Lainie Friedman. 1998. Children, Families, and Health Care Decision Making. New York:
Oxford University Press.
Sullivan, Mark D., and Stuart J. Youngner. 1994. "Depression, Competence, and the Right to
Refuse Livesaving Medical Treatment." American Journal of Psychiatry 151 (July): 971-8.
Weir, Robert F, and Charles Peters. 1997. "Affirming the Decisions Adolescents Make
about Life and Death." Hastings Center Report 27 (November/December): 29-40.
Terminal Sedation and Palliative Care

Bennett, J.F. 1981. Morality and Consequences." In S.M. McMurrin, ed., The Tanner Lec-
tures on Human Values 1981 II. Salt Lake City: University of Utah Press.
Cantor, Norman L. 2001. "Glucksberg, the Putative Right to Adequate Pain Relief, and
Death with Dignity." Journal of Health Law 34: 301-33.
Kamm, Frances M. 1999. "Physician-Assisted Suicide, the Doctrine of Double Effect, and
the Ground of Value." Ethics 109 (April): 586-605.
Quill, Timothy E., et al. 2000. "Palliative Treatments of Last Resort: Choosing the Least
Harmful Alternative." Annals of Internal Medicine 132: 488-93.
Quill, Timothy E., and Ira R. Byock 2000. "Responding to Intractable Terminal Suffering:
The Role of Terminal Sedation and Voluntary Refusal of Food and Fluids: Position
Paper." Annals of Internal Medicine 132: 408-14.
Truog, Robert, et al. "Barbiturates in the Care of the Terminally Ill." New England Journal of
Medicine 327: 1678-82.
Physician-Assisted Death
Battin, Margaret P, Rosamond Rhodes, and Anita Silvers, eds. 1998. Physician Assisted Sui-
cide: Expanding the Debate. New York: Routledge.
Beauchamp, Tom L., ed. 1996. Intending Death: The Ethics of Assisted Suicide and Euthanasia.
Upper Saddle River, NJ: Prentice-Hall.
Bickenbach, J. 1998. "Disability and Life-Ending Decisions." In M. Battin, R. Rhodes,
and A. Silvers, eds., Physician-Assisted Suicide: Expanding the Debate, 123-32. London:
Routledge.
Dworkin, Gerald, Raymond G. Frey, and Sissela Bok. 1998. Euthanasia and Physician-Assisted
Suicide: For and Against. New York: Cambridge University Press.
Kamisar, Yale. 1958. "Some Nonreligious Views against Proposed `Mercy-Killing' Legislation."
Minnesota Law Review 42 (6):324-32.
Koop, C. Everett, and Edward R. Grant. 1986. "The 'Small Beginnings' of Euthanasia."
Journal of Law, Ethics & Public Policy 2: 607-32.
Quill, Timothy E. 1998. "The Debate over Physician-Assisted Suicide: Empirical Data and
Convergent Views." Annals of Internal Medicine 128 (April): 488-93.
Rachels, James. The End of Life: Euthanasia and Morality. 1986. Oxford University Press.
Schaffner, Kenneth E 1988. "Recognizing the Tragic Choice: Food, Water, and the Right to
Assisted Suicide." Critical Care Medicine 16: 1063-8.
Sneddo a, Andrew 2006 "Equality, Justice, and Paternalism: Recentering the Debate about
Physician-Assisted Suicide." Journal of Applied Philosophy 23 (4).
Stingl, Michael, ed. 2010. The Price of Compassion: Assisted Suicide and Euthanasia. Broadview
Press.
Sumner, L.W. 2011. Assisted Death: A Study in Ethics and Law. Oxford University Press.
Thomson, Judith Jarvis. 1999. "Physician-Assisted Suicide: Two Moral Arguments." Ethics
109: 497-518.
Velleman, J. David. 1992. "Against the Right to Die." Journal of Medicine and Philosophy 17:
665-81.
Williams, Glanville. 1957. The Sanctity of Life and the Criminal Law, ch. 8. New York: Alfred
A. Knopf.
. 1958. -Mercy-Killing Legislation—A Rejoinder." Minnesota Law Review 43 (1): 1-12.
Physician-Assisted Death in Canada: Government Documents
(1) Before Carter

Law Reform Commission of Canada. 1982. Euthanasia, Aiding Suicide and Cessation of Treat-
ment. Working Paper 28.
. 1983. Euthanasia, Aiding Suicide and Cessation of Treatment. Final Report.
National Assembly of Quebec. 2003. Bill 52: Act Respecting End-of-Life Care.
Royal Society of Canada. 2011. End-of-Life Decision Making.
Special Senate Committee on Euthanasia and Assisted Suicide. 1995. Of Life and Death—
Final Report.
(2) After Carter

Canadian Medical Association. July 2014. A Canadian Approach to Assisted Dying: CMA
MemberDialogue:SummaryReport.https://www..cma.ca/Assets/assets-library/document/
en/advocacy/Canadian-Approach-Assisted-Dying-e.pdf.
.End-of-LifeCare—ANationalDialogue: CMAMemberConsultationReport. July2014.
haps://www..cma.ca/Assets/assets-library/document/en/advocacy/end-of-life-care-
report-e.pdf.
College of Family Physicians of Canada. September 2015. A Guide for Reflection on Eth-
ical Issues Concerning Assisted Suicide and Voluntary Euthanasia. cfpc Task Force
on End-of-Life Care. http://www.cfpc.ca/uploadedFiles/Health Policy/ PDFs/Guide for
%20Euthanasia EN Final.pdf.
House of Commons and Senate, Special Joint Committee on Physician-Assisted Dying.
February 2016. Medical Assistance in Dying: A Patient-Centered Approach. httpi/
www.parl. gc.ca/HousePublications/Publication. aspx?Docid=8120006&Language=E&
Mode=l&Par1=42&Ses=1.
Provincial-Territorial Expert Advisory Group on Physician-Assisted Dying. 30 November
2015. Final Report. http://www.health.gov.on.ca/en/news /bulletin/2015/docs/eagreport
20151214 en.pdf.
Defining Death
Bernat, James L 14 August 2008. "The Boundaries of Organ Donation after Circulatory
Death." New England Journal of Medicine 359 (7): 669-75.
Dworkin, Roger. 1973. "Death in Context." Indiana Law Journal 48: 623-46.
Emanuel, Linda. 1995. "Reexamining Death: The Asymptotic Model and a Bounded Zone
Definition." Hastings Center Report 25 (4): 27-35.
Green, Michael, and Daniel Wikler. 1980. "Brain Death and Personal Identity." Philosophy
and Public Affairs 9: 105-33.
Institute of Medicine. 2000. Non-Heart-Beating Organ Transplantation. Washington, DC:
National Academy Press: 2000.
Law Reform Commission of Canada. 1979. Criteria for the Determination of Death. Working
Paper 239.
President's Commission for the Study of Ethical Problems in Medicine and Biomedical
and Behavioural Research. 1981. Defining Death: Medical, Legal and Ethical Issues in the
Definition of Death. Washington, DC: U.S. Government Printing Office.
Shermie, S.D., Al Baker, G. Knoll, W Wall, G. Rocker, D. Howes, et al. 2006. National
Recommendations for Donation after Cardiocirculatory Death in Canada. Canadian
Medical Association Journal 175 (8 Suppl): S1—S24.
Truog, Robert. 1997. "Is It Time to Abandon Brain Death?" Hastings Center Report 27 (1):
29-37.
Van Norman, Gail A. 2003. "Another Matter of Life and Death: What Every Anesthesiolo-
gist Should Know about the Ethical, Legal, and Policy Implications of the Non-Heart-
Beating Cadaver Organ Donor." Anesthesiology 98: 763-73.
Delivery o Heakh Care
and Resource AHocatIon
7.1 Introduction
The topic of resource allocation is commonly divided into three areas. These are
macroallocation, which concerns how much of society's resources should go to health macroallocation the fair
care; mesoallocation, which concerns what areas of health care should receive what pro- allocation of a society's
resources for health care
portion of those resources; and microallocation, which addresses the question of which
under conditions of scarcity.
patients should receive health care when not all can. Together, these areas represent aspects
of the problem of distributive justice or how to fairly allocate scarce resources in a society.
mesoallocation the fair
We first describe macro, meso, and microallocation in more detail to give a clearer over-
allocation of scarce health
view of the topic of distributive justice and then discuss related issues of fair allocation of care resources within a
health care resources. specific health care region or
Macroallocation questions occur at a societal level and involve two basic questions. institution.
The first is about the relative value of health care. Health care is a scarce good that com-
petes for resources with a number of other scarce goods such as education, social services, microallocation the fair
roads, and public safety. How much of the available resources should go to health care? We allocation of scarce resources
cannot answer this question without understanding the health care needs of a community to specific individuals.
and the costs of providing it. It follows that macroallocation decisions must be informed by
answers to general questions about how much money should be available to fund specific distributive justice the fair
areas of health care, including acute care, preventative care, public health, long-term care, allocation of scarce resources.
rehabilitation, community care, payment of health care workers and administrators, and so
on. Macroallocation decisions are, therefore, profoundly difficult to make with precision.
The second macroallocation question is about the delivery of health care. Should there
be a one-tier health care system with a public or government payer that tries to serve all
necessary health care needs and does not allow individuals to purchase more? Or should
there be a public tier available to everyone and an option to buy more privately if one can
afford it? Or should the government stay out of the health care business and let health care
be distributed by the free market?
Mesoallocation questions occur at an institutional level. Once a health care region rationing addressing
or public or private institution has its budget, how should it spend it? For instance, how scarcity of health care
much should a hospital spend on emergency care, the icu, surgery? And within those resources by eliminating
services or treatments
divisions further questions arise, such as what support is to be given to urology, cardiol-
offered, or excluding classes
ogy, neurology etc. And since there is never enough to give every patient everything they of individuals from health
want or need when they want or need it, rationing at a meso level is needed. Sometimes services or treatments, or
this involves eliminating classes of service or treatments—for example, not offering all prioritizing access.
II 'MI
7 Delivery of Health Care and Resource Allocation
emergency services or leaving certain types of surgery to other institutions. Sometimes the
remedy involves prioritizing the availability of care, as when hospitals impose guidelines or
wait-lists to provide care in proportion to need. Such decisions merge substantially into the
sphere of microallocation. Thus, one of the great health care allocation issues of our time—
rationing (though the word is studiously avoided)—falls under the topic of mesoallocation
M:.croallocation concerns making choices between particular patients (versus classes
of patients as in mesoallocation). The topic arises because there are never enough resources
to meet all the health care needs. Even if society forgot about all other goods such as roads
and education and devoted all of its resources to health care, there would still be a gap
between supply and demand. And since more and more can be continually done to provide
for better health care, this gap seems permanent. Thus, hard choices between patients are
triage the assignment inevitable when health care providers have to practise triage and decide which patients
of degrees of urgency to will receive organ transplants or ventilators in an influenza epidemic, as well as make many
wounds, injuries, or illnesses to
more, less dramatic decisions. The central ethical question provoked here is how to make
decide the order of treatment
of a large number of patients these decisions. Should they be made on purely medical criteria, or should they also take
under conditions of scarcity of into account, where possible, social criteria (such as being a teacher versus a street person
medical resources. or single versus married with children), age, and responsibility for the medical condition?
Macroallocation
The two leading questions under this topic are (to repeat) how much of society's resources
should go to health care and how those resources should be delivered. How we answer
these questions depends entirely on whether we want to live in an egalitarian, libertar-
ian, or liberal society. The problems of resource allocation thus quickly lead us to the
basic question of social philosophy—how should we live together?—and it is not easy to
choose .netween the alternatives. Indeed, it is not clear that there is a single right answer to
the question. However, once we answer that question, we will have answers to the main
questions of macroallocation. As we will see, an egalitarian society will spend generously
on health care to cover all its citizens with a high level of care and is usually assumed to
favour a one-tier public delivery system. A libertarian society will leave it to a free market
to decide how much (or little) should be spent on health care. We now turn to examine
these alternatives more closely.
Egalitarian
egalitarian theory of Canada has elements of an egalitarian social philosophy. The moral basis of the egalitar-
justice every person is ian theory of justice is the view that all people are equally valuable. Thus, since health is
equally valuable, and society
necessary for a good life, and health care for health, all people have an equal right to health
has an obligation to provide
them with the basic resources
care. This means that society has an obligation to provide it or see that it is provided. This
for a good life. does not mean that society has an obligation to provide its citizens with equal health, for
delivering equal health is not within the domain of social control. Nor does it mean that
there is an obligation on the part of society to provide citizens with all the health care they
need to achieve a basic level of health, for conditions of scarcity may make this impossible.
Rather, it means that a modern, developed society has an obligation to secure enough
resources to provide its citizens with a decent minimum of health care. In "The Right to a
Decent Minimum of Health Care" (see page 367), Allen E. Buchanan explains further what
this obligation does and does not involve.
The basic elements of the egalitarian approach to delivery of health care are articulated
in the main articles of the Canada Health Act, which can be summarized as follows:
1. Public administration: All medically necessary services offered by doctors and hospi-
tals are insured by a single public payer.
Macroallocation
2. Universality: Everyone who is covered by the health care system is covered in the same
manner and under the same terms.
3. Accessibility: There are no financial barriers (e.g., user fees, extra billing) to accessing
health care services.
Articles 2 and 3 amount to the egalitarian commitment that essential health care
services are available on the basis of need and need alone; article 1 is advanced as the most
efficient means to give that commitment practical effect.
Libertarian
The mirror image of the egalitarian system is the libertarian. While the egalitarian libertarian theory of
seeks to maximize equality, the libertarian seeks to maximize freedom. This is a view justice every person has a
right to private property and
made famous (though not invented) by Robert Nozick in Anarchy, State, and Utopia'
to enter into any agreements
and is based on two natural rights—i.e., rights that all human beings have as human they want to. Government can
beings and which are impervious to political interference. The first is a right to any- interfere only to prevent force
thing one justly acquires. One can justly acquire something by either working for it or or fraud.
inheriting it from someone who has justly acquired it. The second is a right to enter
into just agreements. If I have something you want (e.g., health care) and you have natural rights rights that
something I want, (e.g., money), we can agree to exchange those things and thus buy all human beings have just
and sell health care. Libertarians hold that any system of health care delivery (like the h virtue of being human
Canadian system) that does not permit these rights is unjust. They also hold that it is and which are not subject to
political control.
unjust to help those who cannot afford health care through public taxation (as, again,
the Canadian system does), for the public does not owe anyone in need anything and
thus cannot be forced to help.
Application of the libertarian philosophy would almost certainly result in unequal
access to health care, but it does not follow that it is therefore unjust. For (so the argument
goes) any pattern of distribution that is reached by just steps must be just. Indeed, in the
libertarian view, any attempt to interfere with a pattern of distribution based on libertarian
principles will necessarily be unjust, for that will violate either the right to keep what one
has justly acquired or the right to enter into just agreements. Libertarians will thus leave it
to the free market to allocate health care. Pure libertarians will leave it entirely to the market
and rely on the compassion of individuals and groups (versus the government) to care for
those who cannot care for themselves, but some libertarians accept a basic government-
supplied safety net.
Liberal
Both the egalitarian and libertarian view can claim the backing of a theory of distributive
justice (albeit strikingly different ones), and both have a point. A liberal view seeks to com- liberal theory of justice
bine the two views by advocating a delivery system that provides for a public tier of health a social view that combines
egalitarian and libertarian
care as good as communal resources can afford and a private tier that will allow those who
elements. In the context of
have resources to buy enhanced services and avoid a wait-time. This is the system in place resource allocation, it means
in many European and developed countries, and some aspects of it are reflected in the US two-tier health care.
health care system as amended under President Barack Obama. The liberal system, it is
argued, is the fair compromise between irreconcilable views that are both attractive. It is
also sometimes argued that everyone will benefit from two-tier health care. Those who can
afford enhanced care or quick access to health care will benefit. Those who cannot afford to
buy health care in the private tier will also benefit because the private tier will shorten wait
lists in the public tier. Thus, everyone will benefit, albeit not benefit equally. Egalitarians
often reply by saying that the private tier will cause the public tier to deteriorate, because
allowing some to buy care in a private tier reduces pressure to maintain and improve qual-
ity in the public tier. It is also often argued that even if everyone got better care in a two-tier
MI RI II 11 II 1111.1J
system, two-tier health care should still be resisted because having one health care for the
rich and another for the poor will undermine social solidarity.
The Romanow Report

The reading by Roy Romanow, "Sustaining Medicare: The Commission on the Future of
Health Care in Canada" (see page 357) is part of a royal commission report, Building on
Values, published in 2002.2 The commission's mandate was to review the Canadian public
health care system and make recommendations to improve its quality and sustainability
without compromising the principle that an individual's financial resources should not
determine access to services.
Romanow does not argue for the superiority of the Canadian system over its rivals. His
main aim is to show that the system can be sustained, where "sustainability means ensur-
ing that sufficient resources are available over the long term to provide timely access to
quality services and address Canadians' evolving health needs" (see page 357). Romanow
thinks that the Canadian health care system should continue to be financed by progressive
taxation and higher taxes if necessary. He rejects a number of suggestions about alternative
ways of paying for health care services (e.g., user fees, extra billing, medical savings ac-
counts) on the ground that they will deter individuals from seeking health care and make
access dependent on ability to pay rather than on need alone.
The Romanow reading is included in this chapter because it expresses thoughtful
support for the Canadian health care system and offers a plan that remains the current
blueprint for maintaining and sustaining it. It remains an open question whether the Can-
adian health care system is sustainable and whether a more liberal two-tier system, such
as those in other modern industrial democracies outside of North America, do a better job
of providing decent minimum care for all their residents. Many of these communities with
two-tier systems consistently rank ahead of Canada in provision of health care.3
Mesoallocation
How Much for Health Care?

The Romanow report and egalitarian and liberal approaches to health care all broadly agree
on the appropriateness of providing at least a decent minimum of health care for all mem-
bers of a community. But what is lacking from these discussions so far is an actual model
for determining how much of society's resources should be spent on health care, how much
should go to specific areas of health care, and how to make the hard rationing decisions
that run throughout health care. In short, we need a decision procedure for answering
these macro- and mesoallocation questions in practical terms and putting the idea of a
decent minimum of health care into effect. Ronald Dworkin's "Justice and the High Cost
of Health" (see page 372) attempts to address this issue. Although Dworkin is addressing
a US audience, his arguments are general and relevant to all modern developed countries.
Dworkin begins by noting that no society or individual could ever afford to fund
"the rescue principle": the idea that society is obligated to allocate whatever resources
are needed to save a life whatever the cost. If that is correct, a decision process is needed
to figure out how to weigh competing claims on scarce societal resources and to make
difficult decisions about not only which lives to save when all cannot be but also how
to allocate resources to health care generally. The procedure Dworkin proposes is the
"prudent insurance" test. This test attempts to combine the most compelling elements of
the egalitarian, liberal, and libertarian approaches to address the problem of macroalloca-
tion. The test asks us to consider what prudent persons (called "prudent insurers") would
decide to spend on their own health care if, as individuals, they were buying insurance
under fair free-market conditions, did not know what special medical needs they were
Mesoallocation
likely to have (nor did insurers know this), knew all that is currently known about pos-
sible medical conditions and the efficacy of treatments for them, and had significant but
limited financial resources to spend.
Dworkin contends that how much money prudent insurers would spend on health
care insurance provides the model for how many resources society should allocate for
health care. He also contends that prudent insurers would not choose to buy "rescue prin-
ciple" insurance because its expense would be prohibitive, requiring that other important
personal needs and wants could not be served (for example, education, food, shelter, and
entertainment). Dworkin proposes that how much money should be allocated toward a
basic minimum of health care and how it should be spent could be determined by es-
tablishing an agency (or perhaps a sort of citizens' assembly) where average citizens and
others with relevant expertise could attempt to apply the prudent insurance test. Dworkin's
prudent insurance test is thus an all-purpose test of how society should allocate resources.
How much of society's total resources should go to health care, how those resources should
be allocated to specific areas of health care, and how rationing decisions should be made are
all determined by estimating what prudent insurers would spend to insure their health care.
How Do We Measure and Prioritize Health Benefits?

One problem that is not discussed in Dworkin's hypothetical insurance scheme is how to
measure and prioritize health outcomes so that whatever money is allocated for health care
is used efficiently to preserve and promote health. Ethicists and health care economists have
come up with a variety of models that would be of use to the members of Dworkin's citizens'
assembly. Bjarne Robberstad's "QALYs vs DALYs vs LYs Gained: What Are the Differences, and
What Difference Do They Make for Health Care Priority Setting?" (see page 382) sets out the
main models and critically discusses them. All these models, however, have their limitations.
Quality and Adjusted Life Years (QALYs), Disability Adjusted Life Years (DALYs), and Life
Years Saved (LYs) are all types of cost-benefit analyses. The classic approach to cost-benefit
analysis states that resources should be used in ways that produce the greatest economic bene-
fit. This is widely rejected for health care as not being sufficiently egalitarian. It would mean
health resources would go first to individuals with the greatest capacity to produce wealth in
a society, typically the richest and most powerful persons in society. But there are also specific
complaints of unfairness to be made to each of the other metrics. If we allocated resources to
produce the most LYs, this would tend to overlook the claims of persons with disabilities and
those with chronic illnesses. Using QALYs and DALYs would address these deficiencies to some
degree because they attempt to measure quality of life and the burdens of disability but at the
expense of LYs and other considerations, including what to do when maximizing QALYs and
minimizing DALYs conflict. The fact is that there is no acknowledged metric for fairly assessing
the benefits of different health outcomes. This is perhaps an argument for having a transpar-
ent democratic process for making macro and mesoallocation decisions like the one suggested
by Dworkin, since a fair decision-making process establishes some legitimacy for decisions
that everyone will have to acknowledge at the outset are bound to be imperfect.
The Rescue Principle Rejected Again

Tony Hope's "Why Undervaluing 'Statistical' People Costs Lives" (see page 377) is a deeper
investigation into the problems with using the rescue principle (he calls it "the rule of
rescue") as a basis for allocating scarce resources. Hope's article describes how the rescue
principle is reflected in many health care decisions for allocating scarce resources, espe-
cially in emergency care, and how it can undermine fair and just treatment of individuals.
In particular, Hope describes how large amounts of scarce resources tend to be allocated to
actual individuals who are in dire straits at the expense of using those resources to save the
lives of other, and often a greater number of, "statistical" individuals (that is, individuals
who are not currently being treated but who may be affected by decisions to allocate large
amounts of scarce resources to rescue others).
Ethicists are widely agreed that arguments like Hope's against the rescue principle are
compelling, However, accepting the weight of those arguments has radical implications
for the way health care resources are allocated, particularly at the level of mesoallocation.
A lesser role for the rescue principle would mean fewer resources for emergency and acute
care and more for preventative care, since this would mean more life years saved and higher
quality of life generally. Hope notes that despite the apparently compelling nature of these
argummts from distributive justice, there is no country in the world that has attempted to
put them into effect. See the Hope article for a discussion of why this has been the case. It is
interesting to consider whether, contrary to appearance, the rescue principle represents a
failure of empathy or compassion, in particular for those "statistical" persons whose needs
are jus: as real, though not immediately seen or felt, as those needing immediate rescue.
Microallocation
Even if questions about macroallocation and mesoallocation could be fully and fairly ad-
dressed, there would still not be enough resources to meet everyone's medical needs. Such
predictable shortages mean that decisions would still have to be made at an individual
level about who will or will not receive a scarce health care resource. Microallocation is
concerned with making these decisions fairly. More bluntly, microallocation is concerned
with choosing between particular patients when there are not sufficient resources for each
one's medical needs to be met. The central question is who should get treatment when not
everyone can. These decisions can involve life and death questions, such as who should re-
ceive vital organ transplants or ventilation in an influenza epidemic, as well as less dramatic
but very important interventions, such as access to speech language pathology, psychiatry,
and rehabilitation services. Sometimes the exclusion is permanent, sometimes it is a matter
of prioritizing patients into a waiting list. We can also note that microallocation questions
typically arise only where health care is a public good—i.e., a good to which everyone in
society has an equal claim. Where health is not such a good, as in private systems, it can
(except perhaps in special conditions of scarcity) be distributed in any way the owner of
health care wants, such as for money or to friends and relatives.
Complex Criteria
There are three main approaches to the question o f microallocation. The first is a "complex
criteria .system." All complex criteria systems first exclude patients who will not benefit
from the treatment at all. Then they apply a number of criteria to select the candidate or
candidates who will receive the treatment. Complex criteria systems vary in the criteria
they use, but a standard list will include (1) how much the patient will benefit in terms of
quantity and quality of life; (2) whether the patient has dependents; (3) the patient's past
contributions to society, (4) the patient's expected future contributions to society, (5) the
patienth age; and (6) whether the patient's behaviour is responsible for his or her illness. If
there are still more patients than can be provided with resources, random selection ("first
come first served" or a lottery) is used to make the final selection. This is the kind of system
proposed (with some wrinkles of his own and a scoring system for comparing candidates)
by Nicholas P Rescher in "The Allocation of Exotic Life-Saving Therapy" (see page 391).
Random Selection
The second method of microallocation is by a "random selection system." Random se-
lection systems also begin by excluding patients who are not expected to benefit from
Microallocation
the treatment but then studiously turn a blind eye to other factors used in complex
criteria systems. The main virtue of such a system is that it preserves the inherent
dignity and worth of all human beings by giving all an equal opportunity to live.
Need and need alone determines access to health care. Random selection also has
the advantage of simplifying decision-making by eliminating the need to assemble
a committee to evaluate patients on the basis of difficult-to-apply and potentially
discriminatory criteria.
Modified Random Selection

There is a further question of whether random selection is otherwise satisfactory as it
stands. It is sometimes argued that fairness requires that choosing between patients take
into account the age of the patient (age-based rationing) and whether his or her behaviour
is responsible for the illness (responsibility for condition). Thus, we get a third method of
microallocation that lands between complex criteria and random selection methods.
Age-Based Rationing
The problem that provokes the proposal to ration care on the basis of age is that as a
population ages, and as medicine can increasingly help the elderly live longer and better
lives, it is impossible to give both the elderly and non-elderly the best possible health
care. Thus, we must either ration care for the elderly or disadvantage the young. In "Just
Caring: In Defense of Limited Age-Based Healthcare Rationing" (see page 399), Leonard
M. Fleck canvasses the main arguments for and against age-based rationing, and we age-based rationing
will look at age-based rationing and holding individuals responsible for their condition rationing based on age
and age alone. As used in
in turn.
the literature, it means that
Many bioethicists hold that age-based rationing is straightforward discrimination, no if an elderly person and a
different in principle from treating people differently on the basis of race or gender. If an non-elderly person each has
elderly person and a non-elderly person each have essentially the same medical problem essentially the same medical
requiring the same life-preserving treatment, then (so the argument goes) they must have problem, needs life-preserving
treatment, and has the same
an equal right to receive that treatment. Elderly Canadians have contributed to universal
prognosis, the treatment can
health care for most or all of their working lives, and so they are now entitled to the health be denied the elderly person
care they need. on the basis of age.
Many other bioethicists, however, argue that fairness requires age-based rationing,
and Fleck identifies two different ways in which this has been argued. Sometimes the
judgment of fairness rests on the fact that the old have had a chance to experience life,
while the young have not had the same chance. The argument is that this difference in
opportunity should be equalized by giving the young priority in receiving life-preserving
care. This argument is known in the literature as the "fair innings" argument. On this
view, everyone is entitled to a fair and equal shot at life, and a just health care system will
try to make it so.
A second argument for age-based rationing begins with a thought-experiment. The
central character in this experiment is the "prudent insurer" described earlier. (This, you
will recall is a person who wants to buy insurance, is knowledgeable about what illnesses
may befall human beings and what can medically be done about them, but has limited
funds and does not know his state of health.) We are now asked to suppose that the
prudent insurer is given a choice between a policy that will provide life-preserving care
in his middle years and one that will provide for that care in his old age. There can be
(as Dworkin claims) no doubt that he would choose the one that will protect him in his
middle years. But if so, the prudent insurer must be in favour of age-based rationing,
because to choose to have life-preserving treatment in his middle years rather than old
age is just another way of choosing to have age-based rationing. Since everyone planning
what health care should be delivered to him or her over a lifetime can also be expected to
prefer life-preserving treatment in the middle years rather than old age, everyone could
II URI II HUI
be expected to want age-based rationing. But if everyone wants age-based rationing, this
rationing must be fair.
Responsibility for Condition

The second deviation from random selection—responsibility for one's condition—is based
luck egalitarianism on luck egalitarianism. Applied to health care, this is the view that individuals are not
individuals are responsible responsible for health problems that are a result of brute bad luck, and so they should be
for consequences if and only
treated by the common pool of resources that make up the health care system. However, if
if they are the result of their
voluntary choices.
they brought the condition on themselves by behaviour they could reasonably have been
expected to know would put them at risk (e.g., smoking, drinking, overeating, or extreme
sports), it is only fair, in conditions of scarcity, that they should either not be treated or have
a lower priority. Alvin H Moss and Mark Siegler in "Should Alcoholics Compete Equally
for Liver Transplantation?", Morten Ebbe Juul Nielsen and Martin Marchman Andersen
in "Should We Hold the Obese Responsible? Some Key Issues," and Katherine Browne in
"Voluntary Sterilisation and Access to IVF in Quebec" consider how luck egalitarianism
fares in limiting treatment in these cases. (See the Suggested Readings for the references.)
The main problem with appeals to luck egalitarianism is how to be able to say when
choices are voluntary. This problem surfaces in two ways. First, there is a question of
whether anyone can ever be held responsible for his or her actions. This rests on the pos-
determinism every sibility that determinism is true. Determinism is the view that every event has a cause. On
event is brought about and this view, all our behaviour can be explained in terms of causes: our behavior is caused by
necessitated by prior causes
our choices, our choices by our decisions, our decisions by our desires, our desires by our
that occur according to the
laws of nature.
charac:er, and our character by our genetics and environment. Given that our genetics and
environment lie beyond our control, the unbroken causal chain stretching from them to
our behaviour entails that, for anything we do, we could not have done otherwise. But if
so, individuals cannot be held responsible for their actions.
One need not, however, resort to metaphysical speculations to be reluctant to hold
individuals such as alcoholics responsible for their condition. There are a number of more
mundane reasons that we can appeal to. Social determinants of behaviour such as parents,
upbringing, poverty, peer pressure, educational and vocational opportunities, coupled per-
haps with a genetic predisposition to alcoholism cast doubt on how big a role voluntary
choice plays in becoming an alcoholic. It is thus problematic to invoke luck egalitarianism
to give alcoholics second-rate treatment because they are responsible for their condition.
The same can be said for other patients such as smokers, substance abusers, the obese, and
those who have brought their condition on themselves by their choices All these types of
patients may require special handling, but it is hard to see how they can, without consider-
able further argument, be punished or otherwise disadvantaged because they are respon-
sible for their behaviour.
Increasing Resources
One way to address problems of scarce resources in health care is to attempt to increase
those resources. This can be done in a variety of ways that include increasing taxes or
growing an economy so that more money is available to fund health care. Sometimes,
however, non-monetary resources are what is scarce—for example, blood or organs. In
Canada. it is illegal to sell or buy these resources, and so the health care system relies fully
on the altruism of fellow citizens to provide them. The question arises, then, as to whether
it should be possible to purchase them from individuals to increase the availability of these
resources and to help sick individuals get access to potentially life-saving surgeries in more
timely ways. In "Human Organs, Scarcities, and Sale: Morality Revisited" (see page 407),
R.R. Kishore focuses mainly on whether individuals should be permitted to sell one of their
kidneys to an individual who needs it. Kishore argues that concerns that sale of human
Increasing Resources
organs violates human dignity are mistaken and that concerns about abuse, exploitation,
and difficulties obtaining genuinely informed consent can be addressed by regulation.
Notes
1. Robert Nozick, Anarchy, State, and Utopia (Cambridge, MA: Harvard University Press, 1974).
2. Roy Romanow, Building on Values: The Future of Health Care in Canada—Final Report
(Ottawa: Royal Commission on the Future of Health Care in Canada, 2002).
3. Conference Board of Canada, 2012, "International Ranking: Health" (2012)
http://www.conferenceboard.ca/hcp/details/health.aspx.
7.2 Access to Health Care in Canada
Sustaining Medicare: The Commission interpretations and misinterpretations. Moreover,

much of the recent debate on health care has focused
on the Future of Health Care in Canada on one aspect only—namely costs. People conclude
Roy Romanow that the system is not sustainable because it costs too
much money it takes too large a proportion of gov-
The heart of the Commission's mandate was to make ernments' budgets, or it is an impediment to lowering
recommendations "to ensure the long-term sustainabil- taxes. There are others who argue that the problem with
ity of a universally accessible, publicly funded health the system is the way it is organized and the inefficien-
system." The rationale behind this mandate was quite cies that result. Reorganize the system, they argue, and
simple. For a number of years now, Canadians have there is more than enough money to meet our needs.
been told by some of their governments and a number Still other voices have argued that the only problem
of health policy experts that the system popularly with the system is the lack of money provided in recent
known as medicare is no longer "sustainable." years. Restore and increase the financial resources, they
At the same time, the Commission's extensive argue, and all will be well.
consultations with Canadians and its comprehensive In the Commission's view, this narrow focus on
research program clearly indicate that Canadians want money is inadequate and does not help inform the de-
the system to be sustainable, not only for themselves bates or enable an overall assessment of whether or not
but for future generations of Canadians. They want it to Canada's health care system is sustainable.
change, and to change in some very fundamental and Instead, the Commission takes the view that:
important ways. But they also want it to endure and,
indeed, to thrive. Sustainability means ensuring that sufficient re-
Is it possible to reconcile these two perspectives? sources are available over the long term to pro-
The place to start is with a clear understanding of what vide timely access to quality services that address
makes a system sustainable and what needs to be done Canadians' evolving health needs.
to ensure that Canada's health care system is sustain-
able in the future. For many years, health policy experts have focused
on three essential dimensions that are each key to sus-
What Is Sustainability?
taining the health care system:
In some ways, the word "sustainability" both illumin- • services—A more comprehensive range of neces-
ates and obscures the debate. It is a word that is im- sary health care services must be available to meet
mediately understandable and yet open to multiple Canadians' health needs. The services must be of
Source: From Building on Values: The Future of Health Care in Canada (Health Canada, 2002). Reproduced with the permission of the
Minister of Public Works and Government Services Canada, 2008.
MI NI II 11 II all 1111411
358 7 Delivery of Health Care and Resource Allocation
a high quality and accessible on a timely basis. system: public health programs aimed at the preven-
This aspect of sustainability involves looking at tion of illness such as the immunization of children;
the changing ways health care services are deliv- visits to family physicians, pediatricians, or gyne-
ered, whether they are accessible for Canadians, cologists; diagnostic tests; and day surgery. Moving
and whether they are efficiently and effectively across the spectrum, we find the complex and intense
delivered. care that requires the increasing use of advanced
needs—The health care system must meet Can- technology as well as highly trained specialists and
adians' needs and produce positive outcomes not large support teams. In addition, long-term or con-
only for individual Canadians but also for the tinuing care is typically provided in nursing homes
population as a whole. This dimension examines or other specialized residential settings for people
how Canada's health care outcomes measure up who require ongoing medical attention and support
to other countries, identifying disparities in the but who do not need to be treated in hospitals. Pal-
health of different Canadians and looking at trends liative care is provided to people who are dying and
in health. is available in hospitals, hospices, and, to a grow-
• resources—This includes not only financial re- ing extent, at home. Home care is an increasingly
sources but also the required health care providers important component of health care that can allow
and the physical resources (facilities, equipment, people to avoid hospitalization or recover at home
technology, research, and data) that are needed to following a shorter hospital stay. At any point along
provide the range of services offered. the spectrum, people can and frequently do receive
prescription drugs.
There is no "invisible hand" that silently and un- The key question in terms of sustainability is
obtrusively keeps these elements in balance. Decisions whether this vast continuum of services provided
about providing adequate resources imply that there is in Canada's health care system meets the needs of
political support by governments and by Canadians to Canadians, is accessible, and can be adapted in the
continue supporting the system through public funds future to meet the changing needs of Canadians. . . .
and public oversight. Maintaining the balance is, in
fact, a deliberate act of will on the part of society and, Private For-Profit Service Delivery: The Debate
thus, it is the overall governance of the system at all One of the most contentious issues facing Canadians
levels that ultimately decides how these elements are is the extent to which the private sector should be
balanced. . . . involved in delivering health care services. Currently,
provincial and territorial governments provide cover-
Health and Health Care Services
age for a range of services, and those services can be
delivered in any number of ways. Almost all Canadian
Canada health care system provides a range of servi- hospitals are not-for-profit institutions and, in most
ces, some of which are covered by the Canada Health provinces, are operated by regional health authorities.
Act and the well-known five principles, some that are Most physician services are delivered by what are ef-
covered by provinces and territories, and some that fectively owner-operated small businesses ranging from
are provided through the private sector. In large part, single-physician practices to multi-provider clinics that
provinces and territories are responsible for organizing may include a range of health care providers. Large for-
and delivering health care services to people across the profit corporations deliver a narrower range of services
country Since medicare was first established, there including laboratory services and continuing and long-
have been considerable changes in both the scope of term care.
health care services provided in Canada and the differ- In the face of continuing pressures on the health
ent ways they are organized and delivered from rela- care system, some argue that more private for-profit
tively large regional health authorities to small clinics service delivery ought to be introduced in order to
or doctors' offices. bring more resources, choice, and competition into the
Services offered in our health care system can be Canadian health care system and to improve its effi-
differentiated by their complexity and intensity: the ciency and effectiveness. Others argue as strongly that
more or less specialized nature of interventions to main- the private sector should be completely excluded from
tain or restore health and the number of qualified health health care delivery, suggesting that private for-profit
personnel needed to see the interventions through. delivery runs counter to Canadians' values, is inequit-
At one end of the spectrum are a wide variety of able, and is less cost-effective than public delivery in
services that are covered by the public health care the long run.
Roma now: Sustaining Medicare:The Commission on the Future of Health Care in Canada
To try to make sense of this debate, it is import- much harder to find a replacement once public facili-
ant to distinguish between two types of services: direct ties have stopped providing the services—the capacity
health care services such as medical, diagnostic, and that existed in the public system will have been lost.
surgical care; and ancillary services such as food prep- Some suggest that private for-profit delivery is
aration, cleaning, and maintenance. An increasing more efficient than not-for-profit delivery (Gratzer
proportion of ancillary services provided in Canada's 1999 and 2002). Given that most of the private facili-
not-for-profit hospitals are now contracted out to forties currently operating and being planned focus only
profit corporations. Canadians seem to find this role for on providing a limited range of services, there are some
private sector companies acceptable and some studies important concerns that must be addressed in terms of
suggest that these enterprises achieve economies of how these facilities interact with the more comprehen-
scale (McFarlane and Prado 2002). Ancillary services sive public system. In effect, these facilities "cream-off'
are relatively easy to judge in terms of quality—the those services that can be easily and more inexpensively
laundry is either clean or it is not, the cafeteria food provided on a volume basis, such as cataract surgery or
is either good or it is not. Consequently, it is relatively hernia repair. This leaves the public system to provide
easy to judge whether the company is providing the the more complicated and expensive services from
service as promised. Also, there is a greater likelihood which it is more difficult to control cost per case. But
that there are competitors in the same business to if something goes wrong with a patient after discharge
whom hospitals can turn for laundry or food services from a private facility—as a result, for example, of a
if their current contractor is unsatisfactory. post-operative infection or medical error—then the pa-
In terms of direct health care services, the pre- tient will likely have to be returned to a public hospital
cise number of for-profit facilities delivering direct for treatment insofar as private facilities generally do
health care services is unknown. One estimate in 1998 not have the capacity to treat individuals on an inten-
(Deber et al.) suggested that there were 300 private sive care basis. Thus, the public system becomes liable
for-profit clinics in Canada delivering many diagnostic for the care triggered by a poor-quality outcome within
and therapeutic services formerly provided in hospi- a private facility, yet under current arrangements there
tals, including abortions, endoscopies, physiotherapy, is no way for the public system to recover those costs
new reproductive technologies, and laser eye surger- from the private facility. In other words, the public
ies. In addition, there are a growing number of small system is required to provide a "back-up" to the private
private for-profit hospitals or stand-alone clinics in facilities to ensure quality care.
some provinces providing more complex surgeries, Proponents of for-profit care may insist that the
some requiring overnight stays. These facilities vary quality of care is not an issue, but there is evidence
considerably in terms of the number of services they from the United States to suggest that the non-profit
offer and their ownership structure. Furthermore, sector tends to have better-quality outcomes than the
some provinces have expressed an interest in contract- for-profit sector in such things as nursing home care
ing out an increasing number of surgical services to (Harrington 2001; Marmor et al. 1987) and managed
private for-profit hospitals and clinics in the hope of care organizations and hospitals (Kleinke 2001; Gray
realizing efficiencies. 1999). More recently, a comprehensive analysis of the
Unlike ancillary services, direct health care servi- various studies that compare not-for-profit and for-
ces are very complex and it is difficult to assess their profit delivery of services concluded that for-profit hos-
quality without considerable expertise. Indeed, the ef- pitals had a significant increase in the risk of death and
fects of poorly provided service may not be apparent also tended to employ less highly skilled individuals
until some time after the service has been delivered, as than did non-profit facilities (Devereaux et al. 2002).
in the event of a post-operative complication. This is For those reasons, the Commission believes a line
what most clearly distinguishes direct health care servi- should be drawn between ancillary and direct health
ces from ancillary services—a poorly prepared cafeteria care services, and that direct health care services
meal may be unpleasant, but poor quality surgery is should be delivered in public and not-for-profit health
another matter altogether. It is also unlikely that there care facilities.
would be a significant number of competitors able to There are, however, several grey areas around the
offer health care services if a given for-profit provider is issue of private for-profit delivery. First, diagnostic
unsatisfactory. There simply is not a significant surplus services have expanded considerably in the past few
of health care administrators or providers waiting in years and, in many cases, these services are provided
the wings to take over service delivery in a hospital. in private facilities under contracts with regional
Thus, if services are of poor quality it is going to be health authorities or provincial governments. Much
I I 111 111111
of this involves relatively routine procedures such as changing with new advances in medicine and, as a
laboratory tests and x-rays triat can be done with little result, the biggest growth in services is outside of hos-
delay or wait on the part of the patient. But there ap- pital and physician services. Subsequent chapters will
pears to be a growing reliance on the private provision show that there is tremendous growth in home care
of more advanced and expensive diagnostics such as and that prescription drugs have become the fastest
MRIs (magnetic resonance imaging), for which the growing part of the health care system. Canadians also
waiting times in the public: system can be frustrat- are only too well aware of the fact that services are not
ingly long because of what appears to be an under- always available on a timely basis. In areas like diagnos-
investment in such technology within the public tic services and some surgeries, people sometimes wait
system. The growth of private advanced diagnostic too long for access to the services they need. People in
facilities has permitted individuals to purchase faster rural and remote communities also have problems in
service by paying for these services out of their own accessing services. In spite of what appears to be almost
pocket and using the test results to "jump the queue" overwhelming support for primary health care, only
back into the public system for treatment. While this limited progress has been made in extending primary
is not currently a common occurrence, Canadians health care across the country. All of these issues apply
made it clear to the Commission that they are deeply in every province and territory. The conclusion, then,
concerned about the prospect of this becoming rou- on services and sustainability is that more needs to be
tine (Commission 2002a). done to ensure timely access to quality services. The
Medicare rests on the principle that an individual's answer, however, is not to look to the private sector
financial resources should not determine access to for solutions. Instead, governments should seek the
services. In the Commission's view, governments have best solutions within the public system and ensure
a responsibility to guarantee that the public system that adequate resources are available and services are
has sufficient resources to ensure appropriate access accessible to all.
to advanced technology. Increased investment within The Commission is strongly of the view that a
the public system for new diagnostic technology can properly funded public system can continue to provide
remove the temptation to "game" the system by indi- the high-quality services to which Canadians have
viduals and health care providers through the private become accustomed. Rather than subsidize private fa-
purchase of diagnostic tests :hat could allow them to cilities with public dollars, governments should choose
jump the queue. to ensure that the public system has sufficient capacity
The second grey area is services provided to work- and is universally accessible. In addition, . . . any deci-
ers' compensation clients wim job-related injuries and sions about expanding private for-profit delivery could
illnesses. Because of the belief that it is important to have implications under international trade agreements
get these people back to work quickly, these clients get that need to be considered in advance.
preferential treatment in accessing diagnostic and other
health care services over those whose illness or injury Needs and Sustainability
is not work-related or who may not be formally em- The second key dimension of assessing sustain-
ployed. . . . This current exception under the Canada ability is needs, namely: does the health care system
Health Act should be reconsidered. adequately meet Canadians' needs? The answer is a
The third grey area is contracting out of surgical qualified yes. Canada's health outcomes compare fa-
services. In some cases, regional health authorities have vourably with other countries and evidence suggests
contracted with private for-profit facilities that provide that we are doing a good job of addressing factors that
specific surgeries such as cataract and some day surger- affect the overall health of Canadians. There are, how-
ies. Again, there is no clear evidence that this practice is ever, areas where there is room for improvement. And
more efficient or less costly than providing the services there are serious disparities in both access to health
in an adequately resourced not-for-profit facility. care and health outcomes in some parts of Canada.
Clearly, more needs to be done to reduce these dis-
Services and Sustainability parities and also to address a number of factors that
Services are the first element in our definition of sus- affect Canadians' health such as tobacco use, obesity,
tainability The previous information suggests that and inactivity . . . The other conclusion is that aging
there are complex, and sometimes confusing, relation- is not the ominous threat to future sustainability of
ships between the federal, provincial, and territorial our system that some would suggest. Aging will chal-
governments. Much has changed since medicare was lenge and add costs to our health care system, but
first introduced. The range of services is growing and those costs can be managed, particularly if we begin
Roma now: Sustaining Medicare: The Commission on the Future of Health Care in Canada
to prepare and make adjustments to anticipate the relies almost entirely on taxes to fund hospital and
impact of an aging population. physician services, co-payments and user fees for
these services are common in most OECD countries.
Resources in the System: The Case of Funding At the same time, Canada relies more heavily on pri-
As was noted at the outset of this chapter, the third vate insurance and out-of-pocket payments for health
major component of the definition of sustainability care services that are not covered by the Canada
relates to the availability of necessary resources. The Health Act.
health care system needs a variety of resources in order . . . Dental services, for example, are almost en-
to deliver services and meet the health care needs of the tirely funded (94 per cent) through private insurance
population. That includes not only financial resources and direct fees in Canada but are often part of public
but also human and physical resources such as equip- coverage in many western European countries.
ment, facilities, and technology . . . In comparison with selected countries, only Japan
However, the primary focus of much of the debate and Australia have higher levels of out-of-pocket ex-
about sustainability has been about money. Questions penditures than Canada while in the United Kingdom,
about the increasing costs of health care, who pays for Sweden, the Netherlands, Germany, and France all
what aspects of the health care system, and whether we have substantially lower levels of out-of-pocket pay-
will be able to afford the health care system in the future ments. This is because the fees charged in those
have played a significant part in the debates about countries are low and represent a relatively small
medicare's sustainability. The debate has centred on proportion of the real cost of the services provided.
whether there is too little public money in the system, Canadians, however, pay relatively high co-payments
whether there should be different ways of raising those and deductibles for prescription drugs and health
public funds, and whether the system as we know it is services outside the CHA and this results in Canada
"affordable" any longer. Because other chapters do not having a higher percentage of out-of-pocket payments
deal with these issues in detail, the remainder of this than other countries.
chapter addresses the fiscal questions directly, begin- Even though the co-payments and deductibles
ning with how Canada's funding for health care com- are high, the percentage of out-of-pocket payments
pares with other countries, whether other options for in Canada accounts for a relatively small percent-
funding should be considered, and the relative shares age of the total costs of health care services and is
paid by different governments. lower than the OECD average. Canada, like most of
the wealthier OECD countries including the United
Canada's Reliance on Taxes States, relies primarily on funding provided through
Canadians pay directly or indirectly, for every aspect of governments or through insurers. In high-income
our health care system through a combination of taxes, countries, what we call "third-party" payments (i.e.,
payments to government, private insurance premiums, payments made by governments or insurers) make
and direct out-of-pocket fees of varying types and up between 80 and 90 per cent of health expendi-
amounts. Some have suggested that Canada relies too tures (OECD 2002). In less wealthy OECD countries,
heavily on taxation to support its health care system. however, there tends to be a much higher reliance on
. . . Seven per cent of the total funding for Canada's out-of-pocket payments.
health care comes from taxation In countries such as
Germany Japan, France, and the Netherlands, the ma- The Balance between Public and Private Funding
jority of funding for health care comes from social in- of Health Care
surance premiums in the form of employment payroll There is some debate in Canada about the appropri-
taxes. In most developed countries (other than those ate balance between public and private funding for
that rely heavily on social insurance), between 70 and health care. Recently, a number of Canadian providers,
80 per cent of total health care is funded through the scholars, and journalists argued in favour of a greater
taxation system (Mossialos et al. 2002). . . . [II t is hard private role in funding Canadian health care on the as-
to conclude that Canada depends too heavily on taxes sumption that Canadian health care spending is overly
to support health care. weighted to the public side (Gratzer 2002). However, a
comparison with other industrialized countries shows
Use of Private Insurance and Out-of-pocket that Canada is hardly an exception in terms of the
Payments public share of total health expenditures. The United
One area where Canada differs from most OECD coun- Kingdom, Sweden, Germany, France, Japan, and Aus-
tries is in co-payments and user fees. While Canada tralia all have larger public health care sectors than
I I ill IIILLI
Canada, while the Netherlands' public share is slightly access and equity. A number of the most common pro-
lower than Canada's. . . What is truly noteworthy is posals are critically examined below.
the extent to which these countries' public health care
expenditures resemble each other. User Fees and Out-of-Pocket Co-payments
While most wealthy countries rely heavily on User fees are definitely a "hot button" issue for many
public funding for health care, private insurance Canadians. While many are opposed to user fees be-
plays a significant role in funding health care in the cause they discourage poorer people from accessing
United States. Private insurance in the United States health care services, others see user fees as a necessary
is supported by tax breaks known as "tax expendi- way of either raising additional funds for health care
ture subsidies." These subsidies exist, but to a much or curbing abuse of the health care system. Interest-
lesser extent, in all the comparison countries. Since ingly, during the Citizens' Dialogue sessions held by the
these subsidies are not generally included when public Commission, the interest in user fees was not aimed
health care expenditures are tallied, they are difficult at raising more revenue for the system but at curbing
to trace and are therefore referred to as "covert" ex- what some participants felt was abuse and unnecessary
penditures (Mossialos and Dixon 2002). In fact, tax use of the system (Commission 2002a).
subsidies play an enormous role in providing health There is overwhelming evidence that direct char-
care coverage in the United States. When these tax ges such as user fees put the heaviest burden on the
breaks are taken into account, the public share of poor and impede their access to necessary health care.
health care spending in the United States increases This is the case even when low-income exemptions are
to nearly 60 per cent of its total health care spending in place. The result may be higher costs in the long run
(Woolhandler and Himmelstein 2002). This changes because people delay treatment until their condition
the common perception tha: the United States has a gets worse. In addition, user fees and payments also
predominantly private system of health care. involve significant administrative costs that directly
Even without including tax subsidies, the extra- reduce the modest amount of revenue generated from
ordinarily high level of total health care spending in the fees (Evans 2002a; Evans et al. 1993; Barer et al.
the United States translates into far more spending per 1993, 1979; as 2001).
capita than in Canada and the other OECD countries. One of the key features of the Canada Health Act
This has been described as tantamount to paying for was its effective ban on user fees for hospital and phys-
national health insurance and, in return, getting a frag- ician services. Given what we know about the impact
mented system with significant gaps in coverage—the of even relatively low user fees, the Commission feels
worst of both worlds. While the United States' "health that this was the right decision then and remains the
care system is usually portrayed as largely private," right decision today.
a more apt description is "[plublic money, private
control" (Woolhandler and Himmelstein 2002: 22). Medical Savings Accounts
Indeed, the larger the public share of health care finan- Perhaps no recent suggestion for raising additional rev-
cing beyond tax expenditure subsidies, the more total enue has attracted as much attention as medical savings
health expenditures are capable of being controlled. accounts, in part because they seem to address some
In contrast, the larger the pr.vate share of health care of the criticisms of user fees. Medical savings accounts
financing, the more difficult it is to control health care (MSAs) can be designed in a number of different ways
expenditures (Majnoni d'Intignano 2001). but the fundamental concept is that individuals are al-
lotted a yearly health care allowance and they can use it
Alternative Funding Sources to "purchase" health care services (Gratzer 1999, 2002;
In recent years, a number c f suggestions have been Migue 2002; Ramsay 2002). If they have funds left in
made that Canada should consider alternative ways of their MSA allowance at the end of the year, depending
paying for health care services. These proposals may be on how the plan is designed, they may be able to keep
a reaction to the fact that people see costs increasing, the funds or save them for future years when their
are worried about sustainability, and question whether health care costs may be higher.
we should change the current funding system to look MSAs are intended to provide patients with more
for additional sources of revenue. Undeniably, each of control and to inject market forces into the organiza-
these proposals has some potential to raise additional tion and delivery of health care services. They pro-
money to fund the health care system. But some pose vide patients with an incentive to "shop" for the best
problems in terms of the impact they would have on services and best prices, and to avoid unnecessary
Roma now: Sustaining Medicare: The Commission on the Future of Health Care in Canada
treatments, particularly if they get to keep any sur- cover a portion of the cost of the health services they
plus in their account at the end of the year. If the received.
costs of health care services people use in a year are On the positive side, this approach would raise
higher than their yearly allowance, they would be re- additional revenues. People would know the costs of
quired to pay all or a portion of the additional costs, the services they received, and any additional taxes
depending on how the plan was designed. Most MSA would be based on their ability to pay. On the other
proposals discussed in Canada involve a so-called hand, the approach could potentially bankrupt people
corridor where people pay some of the cost of health who had chronic health conditions or who suffered a
care expenses above their annual allowance up to a catastrophic illness or injury. To address this concern,
certain point before catastrophic coverage funded the amount of the co-payment or additional taxes a
entirely by government would cover any remaining person paid could be capped at a certain percentage of
costs (Mazankowski 2001). his or her income and very low-income people could
Because medical savings account approaches are be exempt (Aba and Mintz 2002).
relatively new, we know very little about their effects Even with these conditions, there are concerns
and the literature to date is contradictory . . . The with this approach. Fundamentally, it means that if
limited evidence available suggests that medical sav- people are sick or injured, they will be taxed more
ings accounts have a number of shortcomings that and pay more for health care. This is counter to
have been understated or ignored by their proponents the basic premise in Canada's health care system
(Maynard and Dixon 2002; Shortt 2002; Hurley 2000, that access should be determined only by need and
2002; Barr 2001). Overall, mSAs are based on the as- not by ability to pay. As in the case of mSAs or user
sumption that the use of necessary health care services fees, it may result in people not using needed health
is highly discretionary, when this is almost invariably care services, a phenomenon that has been seen in
not the case. a number of European systems (cES 2001). It also
mSAs are unlikely to effectively control overall raises the question of whether middle- and higher-
spending on health care (Forget et al. 2002). Most income earners, who currently pay the bulk of the
health care costs are incurred by a small proportion of costs of a universal health care system, will eventu-
people who have very high health care needs and they ally become dissatisfied when they also have to pay
will continue to spend a lot regardless of whether or even more at tax time based on their use of the health
not they have an MSA... care system.
MSAs may compromise equity in access to health
care services. If individuals are required to pay once Public—Private Partnerships
they have used all of their MSA allowance, it could While different options like user fees, taxable benefits,
cause hardships for people with lower incomes or medical savings accounts are designed to provide
or higher health care needs due to chronic or life- more private payments for health services, other ap-
threatening conditions. This is precisely the reason proaches such as public—private partnerships (P3s)
why Canada's medicare system was introduced—to are being considered as a way of supporting capital
avoid a situation where wealthy people could get projects. P3s involve a number of different options
access to all the health care services they needed and including long-term outsourcing contracts, joint
poor people could not. ventures, strategic partnerships, or private financing
models. In the United Kingdom, under private finan-
Tax-Based Co-payments, Tax Credits, and Deductibles cing initiatives (PH), private-sector firms are awarded
A number of recent articles have focused on the long-term contracts to design, build, finance, and
use of the tax system as a way of increasing private operate hospitals.
payment in the health care system (Aba et al. 2002; While P3s may be a useful means of bringing the
Aba and Mintz 2002; Reuber and Poschmann 2002). innovation of the private sector to bear, they are not
The simplest way of doing this would be to include without their critics. In many cases, governments find
publicly provided health care services as a taxable P3s attractive because the private-sector company as-
benefit on individuals' annual income tax returns sumes the heavy capital costs of a project and gov-
(Kent 2000). People would get something like a T4-H ernments are only required to pay "rental fees" over
showing the cost of the health services they received the longer term. Unfortunately, while P3s may cost
in a year. This amount would be added to their tax- governments and taxpayers less in the short term,
able income and they would pay additional taxes to these arrangements often cost more in the longer term
II 111111
(Sussex 2001). The rental costs charged to govern- hospital and physician services. There also continues to
ments must be high enough to allow the private-sector be a strong consensus among Canadians that "ability to
partner to recoup its costs and make a profit for its pay" should not be the predominant factor in how we
shareholders. The cost of borrowing is often higher for fund key aspects of our health care system. Canadians
the private sector than for governments. And P3s often want necessary hospital and physician services to be
have higher administration costs. Critics also suggest fully funded through our taxes. This may be because
that the quality of private, for-profit-run facilities can our tax-funded, universal health care system provides a
be lower than publicly run facilities and that, in some kind of "double solidarity" It provides equity of fund-
cases, these arrangements have resulted in beds being ing between the "haves" and the "have-nots" in our
closed and staff being reduced (Pollack et al. 2001). society and it also provides equity between the healthy
This is not to say that P3s are without a place (for ex- and the sick. . . .
ample, in the case of health information systems), but
they are no panacea, and their use and value need to Health as a Major Contributor to the Economy
be carefully considered. Discussions about health care are most often focused
on costs while, in fact, health care is also a major con-
Should Canada Consider Alternative tributor to Canada's economy and economies around
Funding Schemes? the world.
Each of the alternative options outlined above would According to American economist William Nordhaus,
raise more money for the health care system or free up the "medical revolution over the last century appears
money for governments to spend on other priorities to qualify, at least from an economic point of view, for
such as lowering taxes or paying down debt. However, Samuel Johnson's accolade as 'the greatest benefit to
many of the options also compromise the principles mankind— (2002: 38). This increase in economic value
and values on which Canadians built the health care comes from numerous directions including improve-
system. Some of the options would simply shift the ments in:
burden of health expenditures from the public purse
to individuals and would ultimately undermine the • basic knowledge from the germ theory of disease
equity that currently exists in both funding and access at the beginning of the twentieth century to the
to needed health care services. more recent DNA revolution;
Through the Commission's consultations, Canadians • public health capital and infrastructure;
indicated that they were willing to pay more in taxes • diagnostic tools and processes;
to sustain the health care system, but only if changes • logistics in terms of obtaining critical care
are made to improve the current system. Consist- (e.g., emergency response); and
ent with this view, some have suggested a dedicated • treatment technologies and protocols including
tax for health care. This could take a number of pharmaceuticals.
different forms. At one end of the spectrum is what
public finance experts call a hypothecated tax—a In the early 1990s, rising health care costs were
single-purpose tax that is formally separated from all seen in many countries as an obstade to balancing
other revenue streams in a special fund similar to the budgets and cutting taxes. This created the view that
Canada or the Quebec Pension Plans. At the other health care costs were a threat to future national com-
end of the spectrum, a health tax or premium could petitiveness. But based on Nordhaus's calculations, it
be established, but the money flows into the general appears that health care spending contributed at least
revenue funds of governments. Both may satisfy the as much to the American economy as spending on all
public's desire to ensure some degree of transparency other consumption expenditures combined. Canadian
and accountability but they provide less than perfect economist Tom Courchene (2001) has made a similar
solutions in other respects. . . . argument about viewing health care expenditures as a
Based on evidence both in Canada and inter- dynamic investment in the economy rather than simply
nationally, progressive taxation continues to be the as consumption.
most effective way to fund health care in Canada. From Health care investments not only lead to longer
what the Commission heard from Canadians through and more productive working lives on an individual
the Citizens' Dialogue and other consultations, the basis; properly targeted public health care invest-
large majority of Canadians do not want to see any ments can also provide countries with a competitive
change in the single-payer insurance principle for core advantage. According to the Canadian Council of Chief
Romanow: Sustaining Medicare:The Commission on the Future of Health Care in Canada
Executives' submission to the Commission (2002: 2), Resources and Sustainability

"Canada's business leaders have been strong support- What conclusions can we draw about resources and
ers of Canada's universally accessible public health care sustainability? Canada's spending on health care is
system" because it provides a "significant advantage in comparable with other OECD countries although we
attracting the people and investment that companies spend considerably less per capita than the United
need to stay competitive." Indeed, the "big three" auto- States. All OECD countries are facing increasing
makers (Ford, General Motors, and Daimler-Chrysler) health care costs and experience suggests that the
recently signed joint letters with their largest union, the wealthier the country, the more it spends on health
Canadian Auto Workers, expressing support for Can- care. Some suggest that Canada relies too heavily on
ada's publicly funded health care system and noting taxation, and yet, comparisons show that we are not
that it provides an important competitive advantage to much different from other countries. A look at vari-
the Canadian auto and auto-parts industries relative to ous alternative ways of funding health care shows
their American counterparts. In short, it is more eco- that each option raises a number of problems and
nomical for the employers to pay taxes in support of many would simply shift the burden of funding from
medicare than to be forced to buy private health insur- governments to individual Canadians. At the same
ance for their workers. time, there are serious problems in the balance be-
It is also true that health care is what economists tween federal and provincial—territorial funding for
call a superior good in that, as individuals, we tend to health care, and health care is taking up an increas-
spend progressively more on health care than other ing proportion of provincial budgets. Later chapters
goods and services as our incomes go up. Based on a of this report address specific ways in which steps
series of international studies summarized by Gerdtham can be taken to control rising costs, especially for
and Jonsson (2000), higher income is the single most prescription drugs. But the reality is that health care
important factor determining higher levels of health costs are likely to continue to increase and choices
spending in all countries. Indeed, the more econom- have to be made about how those costs will be man-
ically developed the country, the more pronounced aged. Overwhelmingly, Canadians told the Commis-
the effect (Scheiber and Maeda 1997). According to sion that they are prepared to pay more for health
Reinhardt et al. (2002: 171), per capita GDP is with- care to ensure the system's sustainability, provided
out doubt "the most powerful explanatory variable for the system is prepared to change to meet their needs
international differences in health spending." and expectations.
References
Aba, S., W.D. Goodman, and J.M. Mintz. 2002. Fund- Canadian Council of Chief Executives. 2002. Written sub-
ing Public Provision of Private Health: The Case for a mission to the Commission on the Future of Health Care
Copayment Contribution through the Tax System. C.D. in Canada, entitled "Shared Enterprise: Sustaining and
Howe Institute Commentary 163. C.D. Howe Institute: Improving Health Care for Canadians."
Toronto. as (College des Economistes de la Sante). 2001. "Utilisation
Aba, S., and J. Mintz. 2002. "Should Public Health Care Fees Imposed to Public Health Care System Users in
Benefits Be Included as Part of Taxable Income?" Paper Europe." Proceedings of the Workshop on 29 November
presented at the Roundtable on Financing Options for organised for the Commission on the Future of Health
the Commission on the Future of Health Care in Canada. Care in Canada College des Economistes de la Sante: Paris.
24 May, C.D. Howe Institute, Toronto. Commission. 2002a. Report on Citizens' Dialogue on the Future of
Barer, M., V. Bhatia, G.L. Stoddart, and R.G. Evans. 1993. Health Care in Canada. Prepared for the Commission on the
The Remarkable Tenacity of User Charges: A Concise Future of Health Care in Canada by J. Maxwell, K. Jackson,
History of the Participation, Positions, and Rationales and B. Legowski (Canadian Policy Research Networks),
of Canadian Interest Groups in the Debate over "Direct S. Rosell and D. Yankelovich (Viewpoint Learning), in col-
Patient Participation" in Health Care Financing. Centre laboration with P.-G. Forest and L Lozowchuk (Commis-
for Health Services and Policy Research, University of sion on the Future of Health Care in Canada). Commission
British Columbia: Vancouver. on the Future of Health Care in Canada: Saskatoon.
Barr, M.S. 2001. "Medical Savings Accounts in Singapore: Courchene, T. 2001. A State of Minds: Toward a Human Capital
A Critical Inquiry," in Journal of Health Politics, Policy and Future for Canadians. Institute for Research on Public
Law 26 (4): 709-26. Policy: Montreal.
O 111 II 11
II ill 111111
Deber, R. 2002. "Delivering F.:ealth Care Services: Public, Maynard, A., and A. Dixon. 2002. "Private Health Insurance
Not-for-Profit, or Private?" Discussion Paper prepared and Medical Savings Accounts: Theory and Experience,"
for the Commission on the Future of Health Care in Funding Health Care: Options for Europe, ed. E. Mossialos,
in Canada. A. Dixon, J. Figueras, and J. Kutzin. Open University
Devereaux, P.J., P.T.L. Choi, C. Lacchetti, B. Weaver, H.J. Press: Buckingham, 109-27.
Schunemann, T. Haines, J.N. Lavis, B.J.B. Grant, D.R.S. Mazankowski, D. 2001. [Alberta 2001.] A Framework for
Haslam, M. Bhandari, T. Sullivan, D.J. Cook, S.D. Walter, Reform. Report of the Premier's Advisory Council on Health.
M. Meade, H. Khan, N. Ehatnagar, and G.H. Guyatt. D. Mazankowski, Chair. Premier's Advisory Council on
2002. "A Systematic Review and Meta-Analysis of Studies Health: Edmonton.
Comparing Mortality Rates of Private For-Profit Hospitals Migue, Jean-Luc. 2002. "Funding and Production of Health
and Private Not-For-Profit Hospitals," in ow 166 (11): Services: Outlook and Potential Solutions." Discussion
1399-1406. Paper prepared for the Commission on the Future of
Evans, R.G. 2002a. "Financing Health Care: Taxation and Health Care in Canada.
the Alternatives," in Funding Health Care: Options for Mossialos, E., and A. Dixon. 2002. "Funding Health Care: An
Europe, ed. E Mossialos, A. Dixon, and J. Figueras. Open Introduction," in Funding Health Care: Options for Europe
University Press: Buckingham, 31-58. ed. E Mossialos, A. Dixon, J. Figueras, and J. Kutzin.
- . 2002b. "Raising the Money: Options, Consequences Open University Press: Buckingham, 1-30.
and Objectives for Financing Health Care in Canada." Mossialos, E., A. Dixon, J. Figueras, and J. Kutzin, eds. 2002.
Discussion Paper prepared for the Commission on the Funding Health Care: Options for Europe. Open University
Future of Health Care in Canada. Press: Buckingham.
Evans, R.G., M.L. Barer, G.I. Stoddart, and V Bhatia. 1993. OECD (Organisation for Economic Co-operation and De-
It's Not for the Money, It's the Principle: Why User Charges velopment). 2002. OECD Health Data 2002b. A Compara-
for Some Services and Not Others? Centre for Health Servi- tive Analysis of 30 Countries. Organisation for Economic
ces and Policy Research, University of British Columbia: Co-operation and Development: Paris.
Vancouver. Pollack, A., J. Shaoul, D. Rowland, and S. Player. 2001.
Forget, E.L., R. Deber, and L.L. Roos. 2002. "Medical Sav- "Public Services and the Private Sector: A Response t o the
ings Accounts: Will They Reduce Costs?," in CMAJ 167 IPPR." Working paper. Catalyst: London.
(2): 143-7. Ramsay, C. 2002. "A Framework for Determining the Extent
Gratzer, D., ed. 2002. Better Medicine: Reforming Canadian of Public Financing of Programs and Services." Discus-
Health Care. ECW Press: Toronto. sion Paper prepared for the Commission on the Future of
Gratzer, D. 1999. Code Blue: Reviving Canada's Health Care Health Care in Canada.
System. ECW Press: Toronto. Reinhardt, U.E., P.S. Hussey, and G. F. Anderson. 2002.
Gray, B.H. 1999. The Empirical Literature Comparing For- "Cross-National Comparisons of Health Systems Using
Profit and Nonprofit Hospitals, Managed Care Organ- OECD Data, 1999," in Health Affairs 21 (3): 169-81.
izations and Nursing Homes: Updating the Institute of Reuber, G.L., and F. Poschmann. 2002. "Increasing Patient
Medicine Study. Coalition for Nonprofit Healthcare: Incentives to Improve the Financial Stability of the
Washington, DC. Health Care System." Paper presented 24 May at the
Harrington, C. 2001. "Residential Nursing Facilities in the Roundtable on Financing Options for the Commission
United States," in smj 323: 507-10. on the Future of Health Care in Canada. C.D. Howe
Hurley, J. 2002. "Medical Savings Accounts Will Not Advance Institute.
Canadian Health Care Objectives," in ow 167 (2): 152-3. Scheiber, G., and A. Maeda 1997. "A Curmudgeon's Guide to
Kent, T. 2000. What Should Be Done about Medicare? Caledon Financing Health Care in Developing Countries," in "In-
Institute of Public Policy: Ottawa. novation in Health Care Financing," ed. G.J. Schieber. World
Kleinke, J.D. 2001. Oxymorons: The Myth of a U.S. Health Care Bank Discussion Paper No. 365. World Bank Washington,
System. Jossey-Bass: San Francisco. DC, 1-38.
McFarlane, L., and C. Prado. 2002. The Best-Laid Plans: Health Shortt, S.E.D. 2002. "Medical Savings Accounts in Publicly
Care's Problems and Prospects. McGill-Queen's University Funded Health Care Systems: Enthusiasm Versus Evi-
Press: Montreal. dence," in ami 167 (2), 159-62.
Majnoni d"Intignano, B. 2001. Sante et economie en Europe. Sussex, J. 2001. The Economics of the Private Finance In-
Presses Universitaires de France: Paris. itiative in the NHS. The Office of Health Economics:
Marmor, T.R. M. Schlesinger, and R.W. Smithey. 1987. London.
"Nonprofit Organizations and Healthcare," in The Non- Woolhandler, S., and D.U. Himmelstein. 2002. "Paying
profit Sector: A Research Handbook, ed. W.W. Powell. Yale for National Health Insurance-and Not Getting It," in
University Press: New Haven, 1-35. Health Affairs 21 (4): 88-98.
Buchanan: The Right to a Decent Minimum of Health Care
73 The Right to Health Care, Macroallocation, and Setting Priorities
The Right to a Decent Minimum right is understood instead as the right to everything
which can be done to achieve some significant level
of Health Care of health for all, then the claim that there is such a
Allen E. Buchanan right becomes implausible simply because it ignores
the fact that in circumstances of scarcity the total social
The Assumption That There Is a Right expenditure on health must be constrained by the need
to a Decent Minimum to allocate resources for other goods.
Though the concept of a right is complex and con-
A consensus that there is (at least) a right to a decent troversial, for our purposes a partial sketch will do. To
minimum of health care pervades recent policy debates say that person A has a right to something, x, is first of all
and much of the philosophical literature on health care. to say that A is entitled to x, that xis due to him or her.
Disagreement centres on two issues. Is there a more This is not equivalent to saying that if A were granted
extensive right than the right to a decent minimum of x it would be a good thing, even a morally good thing,
health care? What is included in the decent minimum or that x is desired by or desirable for A. Second, it is
to which there is a right? usually held that valid right-claims, at least in the case
of basic rights, may be backed by sanctions, including
Preliminary Clarification of the Concept coercion if necessary (unless doing so would produce
extremely great disutility or grave moral evil), and
Different theories of distributive justice may yield dif- that (except in such highly exceptional circumstances)
ferent answers both to the question "Is there a right failure of an appropriate authority to apply the needed
to a decent minimum?" and to the question "What sanctions is itself an injustice. Recent rights-theorists
comprises the decent minimum?" The justification a have also emphasized a third feature of rights, or at
particular theory provides for the claim that there is a least of basic rights or rights in the strict sense: valid
right to a decent minimum must at least cohere with right-claims "trump" appeals to what would maximize
the justifications it provides for other right-claims. utility, whether it be the utility of the right-holder, or
Moreover, the character of this justification will deter- social utility. In other words, if A has a right to x, then
mine, at least in part, the way in which the decent min- the mere fact that infringing A's right would maximize
imum is specified, since it will include an account of overall utility or even A's utility is not itself a sufficient
the nature and significance of health care needs. To the reason for infringing it.' Finally, a universal (or general)
extent that the concept of a decent minimum is theory- right is one which applies to all persons, not just to
dependent, then, it would be naive to assume that a certain individuals or classes because of their involve-
mere analysis of the concept of a decent minimum ment in special actions, relationships, or agreements.
would tell us whether there is such a right and what The second feature—enforceability—is of crucial
its content is. Nonetheless, before we proceed to an importance for those who assume or argue that there is a
examination of various theoretical attempts to ground universal right to a decent minimum of health care. For,
and specify a right to a decent minimum, a preliminary once it is granted that there is such a right and that such
analysis will be helpful. a right may be enforced (absent any extremely weighty
Sometimes the notion of a decent minimum is ap- reason against enforcement), the claim that there is a
plied not to health care but to health itself, the claim universal right provides the moral basis for using the
being that everyone is entitled to some minimal level, coercive power of the state to assure a decent minimum
or welfare floor, of health. I shall not explore this vari- for all. Indeed, the surprising absence of attempts to
ant of the decent minimum idea because I think its justify a coercively backed decent minimum policy by
implausibility is obvious. The main difficulty is that arguments that do not aim at establishing a universal
assuring any significant level of health for all is simply right suggests the following hypothesis: advocates of a
not within the domain of social control. If the alleged coercively backed decent minimum have operated on
Source: From President's Commission: Securing Access to Health Care, vol. 2 (Washington, DC: US Government Printing Office, 1983).
Reprinted by permission of the author.
11 311 MILLI
the assumption that such a policy must be based on a have been expended to guarantee equal access to that
universal right to a decent minimum. The chief aim of level, individuals are forbidden to spend any of their
this article is to show that this assumption is false. resources for services not available to all. Granted that
I think it is fair to say that many who confidently individuals are allowed to spend their after-tax incomes
assume there is a (universal) right to a decent minimum on more frivolous items, why shouldn't they be allowed
of health care have failed to appreciate the significance to spend it on health? If the answer is that they should
of the first feature of our sketch of the concept of a be so allowed, as long as this does not interfere with
right. It is crucial to observe that the claim that there the provision of an adequate package of health care
is a right to a decent minimum is much stronger than services for everyone, then we have retreated from the
the claim that everyone ought to have access to such a strong equal access principle to something very like the
minimum, or that if they did :Lt would be a good thing, principle of a decent minimum. If, on the other hand,
or that any society which is capable, without great sac- we set the level of services guaranteed for all so high as
rifice, of providing a decent minimum but fails to do so to eliminate the problem of persons seeking extra care
is deeply morally defective. None of the latter assertions beyond this level, this would produce a huge drain on
implies the existence of a right, if this is understood as total resources, foreclosing opportunities for producing
a moral entitlement which ought to be established by important goods other than health care.
the coercive power of the state if necessary . . . So both the recognition that health care must com-
pete with other goods and the conviction that beyond
The Attractions of the Idea of a some less than maximal level of publicly guaranteed
Decent Minimum services individuals should be free to purchase addi-
tional services point toward a more limited right than
There are at least three features widely associated with the strong access principle asserts. Thus, the endorse-
the idea of a right to a decent minimum which, together ment of a right to a decent minimum may be more of a
with the facile consensus that vagueness promotes, recognition of the implausibility of the stronger right to
help explain its popularity over competing conceptions equal access than a sign of any definite position on the
of the right to health care. First, it is usually, and quite content of the right to health care.
reasonably, assumed that the idea of a decent minimum A third attraction of the idea of a decent minimum
is to be understood in a society-relative sense. Surely is that since the right to health care must be limited
it is plausible to assume that, as with other rights to in scope (to avoid the consequences of a strong equal
goods or services, the content of the right must depend access right), it should be limited to the "most basic"
upon the resources available in a given society and services, those normally "adequate" for health, or for
perhaps also upon a certain consensus of expectations a "decent" or "tolerable" life. However, although this
among its members. So the first advantage of the idea aspect of the idea of a decent minimum is useful be-
of a decent minimum, as it is usually understood, is cause it calls attention to the fact that health care needs
that it allows us to adjust the level of services to be are heterogeneous and must be assigned some order of
provided as a matter of right to relevant social condi- priority, it does not itself provide any basis for deter-
tions and also allows for the possibility that as a society mining which are most important.
becomes more affluent the floor provided by the decent
minimum should be raised. The Need for a Supporting Theory
Second, the idea of a decent minimum avoids the In spite of these attractions, the concept of a right to a
excesses of what has been called the strong equal access decent minimum of health care is inadequate as a moral
principle, while still acknowledging a substantive uni- basis for a coercively backed decent minimum policy
versal right. According to the strong equal access prin- in the absence of a coherent and defensible theory of
ciple, everyone has an equal right to the best health justice. Indeed, when taken together they do not even
care services available. Aside from the weakness of the imply that there is a right to a decent minimum. Rather,
justifications offered in support of it, the most implaus- they only support the weaker conditional claim that if
ible feature of the strong equal access principle is that there is a right to health care, then it is one that is more
it forces us to choose between two unpalatable alterna- limited than a right of strong equal access, and is one
tives. We can either set the publicly guaranteed level of whose content depends upon available resources and
health care lower than the level that is technically pos- some scheme of priorities which shows certain health
sible or we can set it as high as is technically possible. In services to be more basic than others. It appears, then,
the former case, we shall be committed to the uncom- that a theoretical grounding for the right to a decent
fortable conclusion that no matter how many resources minimum of health care is indispensable. . . .
My suggestion is that the combined weight of The Harm Prevention argument for traditional
arguments from special (as opposed to universal) rights public health services, however, may be elaborated in
to health care, harm prevention, prudential arguments a way that brings them closer to arguments for a uni-
of the sort used to justify public health measures, and versal right to health care. With some plausibility one
two arguments that show that effective charity shares might contend that once the case has been made for
features of public goods (in the technical sense) is suf- expending public resources on public health measures,
ficient to do the work of an alleged universal right to a there is a moral (and perhaps Constitutional) obligation
decent minimum of health care. to achieve some standard of equal protection from the
harms these measures are designed to prevent. Such an
Arguments from Special Rights argument, if it could be made out, would imply that
The right-claim we have been examining (and find the availability of basic public health services should
unsupported) has been a universal right-claim: one not vary greatly across different racial, ethnic, or geo-
that attributes the same right to all persons. Special graphic groups within the country.
right-claims, in contrast, restrict the right in question
to certain individuals or groups. Prudential Arguments
There are at least three types of arguments that can Prudent arguments for health care services typically
be given for special rights to health care. First, there are emphasize benefits rather than the prevention of
arguments from the requirements of rectifying past or harm. It has often been argued, in particular, that the
present institutional injustices. It can be argued, for ex- availability of certain basic forms of health care make
ample, that American blacks and Native Americans are for a more productive labour force or improve the fit-
entitled to a certain core set of health care services owing ness of the citizenry for national defence. This type of
to their history of unjust treatment by government or argument, too, does not assume that individuals have
other social institutions, on the grounds that these injus- moral rights (whether special or universal) to the ser-
tices have directly or indirectly had detrimental effects vices in question.
on the health of the groups in question. Second, there It seems very likely that the combined scope of
are arguments from the requirements of compensation the various special health care rights discussed above,
to those who have suffered unjust harm or who have when taken together with harm prevention and pru-
been unjustly exposed to health risks by the assignable dential arguments for basic health services and an
actions of private individuals or corporations—for in- argument from equal protection through public health
stance, those who have suffered neurological damage measures, would do a great deal toward satisfying the
from the effects of chemical pollutants. health care needs which those who advocate a uni-
Third, a strong moral case can be made for special versal right to a decent minimum are most concerned
rights to health care for those who have undergone ex- about. In other words, once the strength of a more
ceptional sacrifices for the good of society as a whole— pluralistic approach is appreciated, we may come to
in particular those whose health has been adversely question the popular dogma that policy initiatives de-
affected through military service. The most obvious signed to achieve a decent minimum of health care for
candidates for such compensatory special rights are all must be grounded in a universal moral right to a
soldiers wounded in combat. decent minimum. This suggestion is worth considering
because it again brings home the importance of the
Arguments from the Prevention of Harm methodological difficulty encountered earlier. Even if,
The content of the right to a decent minimum is typ- for instance, there is wide consensus on the considered
ically understood as being more extensive than those judgment that the lower health prospects of inner-city
traditional public health services that are usually justi- blacks are not only morally unacceptable but an injus-
fied on the grounds that they are required to protect tice, it does not follow that this injustice consists of
the citizenry from certain harms arising from the inter- the infringement of a universal right to a decent min-
actions of persons living together in large numbers. imum of health care. Instead, the injustice might lie in
Yet such services have been a major factor—if not the the failure to rectify past injustices or in the failure to
major factor—in reducing morbidity and mortality achieve public health arrangements that meet a reason-
rates. Examples include sanitation and immuniza- able standard of equal protection for all.
tion The moral justification of such measures, which
constitute an important element in a decent minimum Two Arguments for Enforced Beneficence
of health care, rests upon the widely accepted Harm The pluralistic moral case for a legal entitlement to a
(Prevention) Principle, not upon a right to health care. decent minimum of health care (in the absence of a
■ IN I II I I III 111111
universal moral right) may be strengthened further Social critics on the left often argue that in a highly
by non-rights-based arguments from the principle competitive acquisitive society such as ours it is naive
of beneficence.' The possibility of making out such to think that the sense of beneficence will win out over
arguments depends upon the assumption that some the urgent promptings of self-interest. One need not
principles may be justifiably enforced even if they are argue, however, that voluntary giving fails from weak-
not principles specifying valid right-claims. There is ness of the will. Instead one can argue that even if each
at least one widely recognized class of such principles individual recognizes a moral duty to contribute to the
requiring contribution to the production of "public aid of others and is motivationally capable of acting on
goods" in the technical sense (for example, tax laws that duty, some important forms of beneficence will not
requiring contribution to national defence). It is char- be forthcoming because each individual will rationally
acteristic of public goods that each individual has an conclude that he should not contribute.
incentive to withhold his contribution to the collect- Many important forms of health care, especially
ive goal even though the net result is that the goal will those involving large-scale capital investment for tech-
not be achieved. Enforcement of a principle requiring nology, cannot be provided except through the contri-
all individuals to contribute to the goal is necessary butions of large numbers of persons. This is also true
to overcome the individud.'s incentive to withhold of the most important forms of medical research. But if
contribution by imposing penalties for his own fail- so, then the beneficent individual will not be able to act
ure to contribute and by assuring him that others will effectively, in isolation. What is needed is a coordinated
contribute. There is a spec:al subclass of principles joint effort.
whose enforcement is justified not only by the need
to overcome the individua:.'s incentive to withhold First Argument
compliance with the principle but also to ensure that There are many ways in which I might help others
individuals' efforts are appropriately coordinated. For in need. Granted the importance of health, provid-
example, enforcing the rule of the road to drive only ing a decent minimum of health care for all, through
on the right not only ensures a joint effort toward large-scale collective efforts, will be a more important
the goal of safe driving but also coordinates indi- form of beneficence than the various charitable acts
viduals' efforts so as to make the attainment of that A, B, and C, which I might perform independently,
goal possible. Indeed, in the case of the "rule of the that is, whose success does not depend upon the
road" a certain kind of coordinated joint effort is the contributions of others. Nonetheless, if I am ration-
public good whose attainment justifies enforcement. ally beneficent I will reason as follows: either enough
But regardless of whether the production of a public others will contribute to the decent minimum project
good requires the solution of a coordination problem to achieve this goal, even if I do not contribute to it; or
or not, there may be no right that is the correlative not enough others will contribute to achieve a decent
of the coercively backed obligation specified by the minimum, even if I do contribute. In either case, my
principle. There are two arguments for enforced be- contribution will be wasted. In other words, granted
neficence, and they each depend upon both the idea the scale of the investment required and the virtually
of coordination and on certain aspects of the concept negligible size of my own contribution, I can disregard
of a public good. the minute possibility that my contribution might
Both arguments begin w.th an assumption reason- make the difference between success and failure. But
able libertarians accept: there is a basic moral obliga- if so, then the rationally beneficent thing for me to
tion of charity or beneficence to those in need. In a do is not to waste my contribution on the project of
society that has the resources and technical knowledge ensuring a decent minimum but instead to undertake
to improve health or at least to ameliorate important an independent act of beneficence, A, B, or c —where
health defects, the application of this requirement of I know my efforts will be needed and efficacious. But
beneficence includes the prcvision of resources for at if everyone, or even many people, reason in this way,
least certain forms of health care. If we are sincere, we then what we each recognize as the most effective form
will be concerned with the efficacy of our charitable of beneficence will not come about. Enforcement of a
or beneficent impulses. It is all well and good for the principle requiring contributions to ensuring a decent
libertarian to say that voluntary giving can replace the minimum is needed.
existing array of government entitlement programs, but The first argument is of the same form as standard
this possibility will be cold comfort to the needy if, for public goods arguments for enforced contributions to
any of several reasons, voluntary giving falters. national defence, energy conservation, and many other
goods, with this exception. In standard public goods sense of beneficence, or because they reason as I did in
arguments, it is usually assumed that the individual's the First Argument, or for some other reason.
incentive for not contributing is self-interest and that it Both arguments conclude that an enforced decent
is in his interest not to contribute because he will be minimum principle is needed to achieve coordin-
able to partake of the good, if it is produced, even if ated joint effort. However, there is this difference. The
he does not contribute. In the case at hand, however, Second Argument focuses on the assurance problem,
the individual's incentive for not contributing to the while the first does not. In the Second Argument all that
joint effort is not self-interest, but rather his desire to is needed is the assumption that rational beneficence
maximize the good he can do for others with a given requires assurance that enough others will contribute.
amount of his resources. Thus if he contributes but the In the First Argument the individual's reason for not
goal of achieving a decent minimum for all would have contributing is not that he lacks assurance that enough
been achieved without his contribution, then he has still others will contribute, but rather that it is better for him
failed to use his resources in a maximally beneficent way not to contribute regardless of whether others do or not.
relative to the options of either contributing or not to Neither argument depends on an assumption of
the joint project, even though the goal of achieving a conflict between the individual's moral motivation of
decent minimum is attained. The rationally beneficent beneficence and his inclination of self-interest. Instead
thing to do, then, is not to contribute, even though the the difficulty is that in the absence of enforcement, in-
result of everyone's acting in a rationally beneficent way dividuals who strive to make their beneficence most
will be a relatively ineffective patchwork of small-scale effective will thereby fail to benefit the needy as much
individual acts of beneficence rather than a large-scale, as they might.
coordinated effort. A standard response to those paradoxes of ration-
ality known as public goods problems is to introduce
Second Argument a coercive mechanism which attaches penalties to non-
I believe that ensuring a decent minimum of health contribution and thereby provides each individual with
care for all is more important than projects A, B, or C, the assurance that enough others will reciprocate so
and I am willing to contribute to the decent minimum that his contribution will not be wasted and an effective
project, but only if I have assurance that enough others incentive for him to contribute even if he has reason to
will contribute to achieve the threshold of investment believe that enough others will contribute to achieve
necessary for success. Unless I have this assurance, I the goal without his contribution. My suggestion is that
will conclude that it is less than rational—and perhaps the same type of argument that is widely accepted as a
even morally irresponsible—to contribute my resour- justification for enforced principles requiring contribu-
ces to the decent minimum project. For my contributions toward familiar public goods provides support for
tion will be wasted if not enough others contribute. If I a coercively backed principle specifying a certain list
lack assurance of sufficient contributions by others, the of health programs for the needy and requiring those
rationally beneficent thing for me to do is to expend who possess the needed resources to contribute to the
my "beneficence budget" on some less-than-optimal establishment of such programs, even if the needy have
project A, B, or C, whose success does not depend on no right to the services those programs provide. Such
the contribution of others. But without enforcement, I an arrangement would serve a dual function: it would
cannot be assured that enough others will contribute, coordinate charitable efforts by focusing them on one
and if others reason as I do, then what we all believe set of services among the indefinitely large constel-
to be the most effective form of beneficence will not lation of possible expressions of beneficence, and it
be forthcoming. Others may fail to contribute either would ensure that the decision to allocate resources to
because the promptings of self-interest overpower their these services will become effective.
Notes
1. Ronald Dworkin, (1977), Taking Rights Seriously (Harvard health care specifically, see Allen E. Buchanan, (1982),
University Press: Cambridge, MA), 184-205. "Philosophical Foundations of Beneficence," in Beneficence
2. For an exploration of various arguments for a duty of and Health Care, ed. Earl E. Shelp (Reidel Publishing Co.:
beneficence and an examination of the relationship Dordrecht, Holland).
between justice and beneficence, in general and in
II ill 111111
Justice and the High Cost of Health The rescue principle does have something helpful,
though negative, to say about the other question of jus-
Ronald Dworkin tice, which is how health care should be distributed. It
says that if rationing is necessary, it should not be done,
1. . .1 II as it now largely is in the United States, on the basis of
For millennia doctors have paid lip service, at least, to money. But we need more positive advice: What should
an ideal of justice in medi:ine which I shall call the the basis of rationing be? The egalitarian impulse of the
rescue principle. It has two connected parts. The first principle suggests that medical care should be distrib-
holds that life and health are, as Rene Descartes put it, uted according to need. But what does that mean—
chief among all goods: everything else is of lesser im- how is need to be measured? Does someone "need" an
portance and must be sacrificed for them. The second operation that might save his life but is highly unlikely
insists that health care must be distributed on grounds to do so? Is someone's need for life-saving treatment
of equality: that even in a society in which wealth is very affected by the quality his life would have if the treat-
unequal and equality is otherwise scorned, no one must ment were successful? Does the age of the patient
be denied the medical care :le needs just because he is matter—does someone need or deserve treatment less
too poor to afford it. These are understandable, even at seventy than at a younger age? Why? How should we
noble, ideals. They are grounded in a shared human balance the need of many people for relief from pain
understanding of the horror of pain, and, beyond that, or incapacity against the need of fewer people for life-
of the indispensability of life and health to everything saving care? At one point the procedures of an Oregon
else we do. The rescue principle is so ancient, so intui- commission appointed to establish medical priorities
tively attractive, and so widely supported in political ranked tooth-capping ahead of appendectomy, because
rhetoric that it might easily be thought to supply the so many teeth can be capped for the price of one oper-
right standard for answering questions about rationing. ation. Why was that so clearly a mistake?
In fact, however, the rescue principle is almost We need a different, more helpful statement of
wholly useless for that purpose, and the assumption that ideal justice in health care, and we should start by
it sets the proper standard for health-care reform has done noticing one problem that seems to make reform man-
more harm than good. The principle does offer an answer datory. Why does America spend so much—so much
to the question of how mucl-. America should spend on more than other nations—on medicine? In large part
health care overall: it says we should spend all we can because individual decisions about how much health
until the next dollar would buy no gain in health or life care to buy are made by patient and doctor but paid for
expectancy at all. No sane society would try to meet that by a third party, the insurance company, so that those
standard, any more than a sane person would organize who make the decisions have no direct incentive to save
his life on that principle. In past centuries, however, there money. Insurance premiums are tax-deductible, more-
was not so huge a gap between the rhetoric of the rescue over, and an employer's contribution is not treated as
principle and what it was medically possible for a com- part of the employee's taxable income. So health insur-
munity to do. But now that science has created so many ance makes patients insensitive to cost at the moment
vastly expensive forms of medical care, it is preposterous of decision, and the real price of that insurance is sub-
that a community should treat longer life as a good that it sidized by the nation. People would probably spend
must provide at any cost—even one that would make the less on their own or their family's care if they had to
longer lives of its people lives barely worth living. pay the actual cost themselves, at the expense of other
So the rescue principleb answer to the question of goods and opportunities they might also want or want
how much a society should spend on health care over- their families to have.
all must be rejected as incredible. Once that answer is Of course, in the long run most people do pay the
rejected, the principle has no second-best or fallback true costs of their health care, but they do so indirectly
level of advice: it simply is silent. That is worse than un- and unwisely, because employer contributions and tax
helpful, because it encouragers the idea that justice has funds could be used to buy what they would choose to
nothing to say about how much a society should spend have if they made the choice themselves: better schools
on health care as against other goods, like education for their children, for example, or economic invest-
or controlling crime or material prosperity or the arts. ments and programs that would improve America's
Source: From Sovereign Virtue: The Theory and Practice of Equality, Chapter 8, pp. 309-18 (Harvard University Press, 2002).
Reprinted with permission.
Dworkin: Justice and the High Cost of Health
competitiveness and give them greater job security. effects of particular medical procedures—everything,
Our medical expenditures are therefore irrational: the in other words, that good doctors know—is generally
system makes choices for people that they would not known by the public at large as well. Third, imagine
make for themselves, and the result is that our col- that no one—including insurance companies—has any
lective expenditures are too high—measured, as they information available about how likely any particular
should be, by how much care we really want, taken person is to contact any particular disease or to suffer
together, at the price we really want to pay. any particular kind of accident. No one would be in
Conservative economists seize on this fact: they a position to say, of himself or anyone else, that that
say we should create a free market in health care by person is more or less likely to contract sickle-cell
removing all tax benefits and subsidies so that people anemia or diabetes, or to be the victim of violence in
can have only the care they can afford. While that is, the street, than anyone else.'
of course, an unacceptable solution, it is important The changes I am asking readers to imagine are
to see why. It is unacceptable for three reasons. First, very great, but they are not, I think, beyond the reach
wealth is so unfairly distributed in America that many of the imagination. Now suppose that health-care de-
people would be unable to buy any substantial health cisions in this transformed community are left simply
insurance at market rates. Second, most people have to individual market decisions in as free a market as
very inadequate information about health risks and we can imagine so that doctors and hospitals and
medical technology; they do not know what the risk of drug companies are free to charge whatever they wish.
breast cancer is before the age of fifty, for example, or Medical care is not provided by the government for
how many years having routine mammography before anyone, nor are medical expenses or health-insurance
that age would add to their life expectancy. Third, in premiums tax-deductible. There is no need to sub-
an unregulated market, insurance companies would sidize medical care in any such way, because people
charge some people higher premium rates because they have enough resources to buy for themselves the
were greater health risks (as, indeed, many insurance medical care they decide is appropriate. What kind
companies now do) so that people with a poor health of health-care institutions would actually develop in
history, or who were members of ethnic groups par- such a community? Would most people join health
ticularly susceptible to certain diseases, or who lived in maintenance organizations that provided care by staff
areas where the risk of violent injury was greater would doctors at a relatively inexpensive rate? Would any
be charged prohibitive rates. substantial number choose more expensive insurance
This analysis points to a more satisfactory ideal of arrangements that allowed more freedom of choice in
justice in health care—the "prudent insurance" ideal. doctors or hospitals? Would the average plan or policy
We should allocate resources between health and other provide coverage for routine medical examinations or
social needs, and between different patients who need diagnostic screenings? What kind, how often, and at
treatment, by trying to imagine what health care would what age? How many plans or policies would provide,
be like if it were left to a free and unsubsidized market at appropriately high rates or premiums, experimental
and if the three deficiencies I have just described were or very expensive or high-risk or low-expected-benefit
somehow corrected. So try to imagine that America is procedures of different kinds? How much of its aggre-
transformed in three ways. Suppose, first, that the dis- gate resources would the community devote to medical
tribution of wealth and income is as fair as it possibly care through these various individual decisions?
can be. In my view, that means that the resources people It is impossible to answer these questions with any
can initially command, in making their decisions about precision.2 But we can nevertheless make two crucial
education, work, and investment, are as nearly equal claims about justice. First, whatever that transformed
as possible; but you should imagine an economic community actually spends on health care in the aggre-
distribution that is fair according to your own views, gate is the morally appropriate amount for it to spend:
whatever these are. (I shall assume, however, that on it could not be criticized, on grounds of justice, for
your views, as on mine, the wealth of everyone in a fair spending either too much or too little. Second, how-
society would be much closer to the average than is ever health care is distributed in that society is just for
true in America now: the great extremes between rich that society: justice would not require providing health
and poor that mark our economic life now would have care for anyone that he or his family had not purchased.
largely disappeared.) These claims follow directly from an extremely appeal-
Second, imagine that America has also changed so ing assumption: that a just distribution is one that well-
that all the information that might be called state-of- informed people create for themselves by individual
the-art knowledge about the value and cost and side choices, provided that the economic system and the
II III 111111
distribution of wealth in the community in which these character, after he entered the late stages of Alzheimer's
choices are made are themselves just.3 disease or other form of irreversible dementia. The
These important conclusions help us to decide money spent on premiums for such insurance would
what health care we should aim to provide for everyone have been better spent, no matter what happens, in
in our own, imperfect, and unjust community. We can making life before dementia more worthwhile. Of
speculate about what kind of medical care and insur- course, most prudent people would want to buy insur-
ance it would be prudent for most Americans to buy for ance to provide custodial care, in conditions of dignity
themselves if the changes I :Imagined had really taken and adequate comfort, if they became demented; pro-
place, and we can use those speculations as guidelines viding for such care would be much less expensive than
in deciding what justice requires now—in deciding, for providing for life-saving treatment—for example, renal
example, which medical tests and procedures the Na- dialysis or an organ transplant—if it were needed.5
tional Health Board should decide are "necessary and Now consider a somewhat more controversial sug-
appropriate" if the Health Security Act is passed. gestion. In most developed countries, a major fraction
Consider a twenty-five-year-old with average of medical expense—over a quarter of Medicare pay-
wealth and prospects and state-of-the-art knowledge of ments in the United States, for example—is spent on
medicine. Suppose he can choose from a wide variety people in the last six months of their lives. Of course,
of possible arrangements to provide for the health care doctors do not always know whether a particular pa-
he might want, under various contingencies, over the tient will die within a few months no matter how much
course of his life. What arrar gements would it be pru- is spent on his care. But in many cases, sadly, they do
dent for him to make?4 He might be tempted, initially, know that he will. Most young people on reflection
to buy insurance providing every form of treatment or would not think it prudent to buy insurance that could
care that might conceivably be beneficial for him under keep them alive, by expensive medical intervention,
any circumstance. But he would soon realize that the for four or five more months at the most if they had
cost of such wildly ambitious insurance would be already lived into old age. They would think it wiser to
prohibitive—he would have nothing left for anything spend what that insurance would cost on better health
else—and decide that prudence required a much less care earlier, or on education or training or investment
comprehensive insurance program. that would provide greater benefit or more important
Of course, what is prudent for someone depends security. Of course, most people would want to live
on that person's own individual needs, tastes, person- those additional months if they did fall ill: most people
ality, and preferences, but we can nevertheless make want to remain alive as long as possible, provided they
some judgments with confidence that they would fit remain conscious and alert and the pain is not too
the needs and preferences of most contemporary Amer- great. But prudent people would nevertheless not want
icans. We can be confident, for example, about what to guarantee those additional months at the cost of
medical insurance it would not be prudent for most sacrifices in their earlier, vigorous life, although, once
people to buy, because some insurance would be a mis- again, they would certainly want insurance to provide
take no matter what happened in the future, including the much less expensive care that would keep them as
the worst outcome. It woule be irrational for almost comfortable and as free of pain as possible.
any twenty-five-year-old to insure himself as to provide We can use these assumptions about what most
for life-sustaining treatment if he falls into a persistent people would think prudent for themselves, under
vegetative state, for example. The substantial sum he fairer conditions than those we now have, as guides
would have to spend in insurance premiums, year by to the health care that justice demands everyone have
year, to provide that coverage would be much better now. If most prudent people would buy a certain level
used in other ways to enhance his actual, conscious of medical coverage in a free market if they had average
life. Even someone who lived only a few months after means—if nearly everyone would buy insurance cover-
purchasing the insurance before he fell into a vegetative ing ordinary medical care, hospitalization when neces-
state would have made, in retrospect, a mistake, giving sary, prenatal and pediatric care, and regular checkups
up resources that could have made his short remaining and other preventative medicine, for example—then
conscious life better in order to buy a longer uncon- the unfairness of our society is almost certainly the
scious state. reason some people do not have such coverage now. A
We can enlarge this claim to include dementia as universal health-care system should make sure, in all
well as unconsciousness: it would not be prudent, for justice, that everyone does have it.
almost anyone, to purchase insurance providing for On the other hand, if even under fair condi-
expensive medical intervention, even of a life-saving tions very few prudent people would want to insure
themselves to a much higher level of coverage—if, as guaranteed even if it offers only the barest hope. Most
I said, very few people would insure to provide life- potential parents would decide, I believe, that it would
saving care when demented or heroic and expensive be better for them and their families to buy the first
treatment that could prolong their lives only by a few policy and to use the premiums they would save each
months, for instance—then it is a disservice to justice to year to benefit their healthy children in other ways—to
force everyone to have such insurance through a man- provide better routine medical care, or better housing,
datory scheme. Of course, any judgment about what or better education, for example—even though they
most prudent people would do is subject to exceptions: would be giving up the chance for a desperate gamble
some people have special preferences and would make to save a defective child if they ever had one.
very different decisions from those that many other Any public body charged with overseeing the
people would. Some people might think, even on re- distribution of health care would have to decide what
flection, that guaranteeing a few extra months of life at medical procedures are "necessary and appropriate"
the end was worth great sacrifice earlier, for example.° and thus should be part of a comprehensive package
But it seems fair to construct a mandatory coverage of benefits everyone is guaranteed. Some of these de-
scheme on the basis of assumptions about what all but cisions would be particularly difficult: deciding when
a small number of people would think appropriate, al- very expensive diagnostic techniques or experimental
lowing those few who would be willing to spend more organ transplants with a low chance of success are ap-
on special care to do so, if they can afford it, through propriate, for example. Such decisions must of course
supplemental insurance. be based on the best and latest medical evidence and
If we substituted the prudent insurance approach must constantly be reviewed as that evidence changes.
for the rescue principle as our abstract ideal of justice But they, too, should be guided by the standard of in-
in health care, we would therefore accept certain limits dividual prudence: Would it make sense for someone
on universal coverage, and we would accept these to insure himself when young to guarantee a vastly ex-
not as compromises with justice but as required by it. pensive blood test which would improve the diagnosis
Expensive treatment for unconscious or demented or of a heart attack by a very small percentage of accuracy
terminally ill patients would be relatively easy cases if he should ever have doubtful symptoms of cardiac
to decide if we adopted that approach. Other deci- disease? Or to provide a risky, expensive, and probably
sions would be more difficult to make, including, for ineffective bowel and liver transplant if doctors decided
example, heart-wrenching decisions about the care it would give him a small chance to live?
of babies born so deformed or diseased that they are The rescue principle insists that society provide
unlikely to live more than a few weeks even with the such treatment whenever there is any chance, however
most heroic and expensive medical intervention. A few remote, that it will save a life. The prudent insurance
years ago doctors in Philadelphia separated newborn principle balances the anticipated value of medical
Siamese twins who shared a single heart, though the treatment against other goods and risks: it supposes
operation would certainly kill one baby and give the that people might think they lead better lives overall
other only a one-in-a-hundred chance of surviving for when they invest less in doubtful medicine and more
long, and though the total cost was estimated to be a in making life successful or enjoyable or in protecting
million dollars. (The twins' parents had no medical themselves against other risks, including economic
insurance, but Indiana, where they lived, paid $1,000 ones, that might also blight their lives. An agency
a day toward the cost, and the Philadelphia hospital might well decide that while prudent people would
absorbed the rest.) The chief surgeon justified the pro- provide their family with the prenatal and well-child
cedure by appealing to the rescue principle: "There has care that so many Americans lack and would insure
been a unanimous consensus," he said, "that if it is pos- against serious medical risks at all stages of their lives
sible to save one life, then it is worth doing this." by providing tested and reasonably effective treatment
But the different standard I am defending would should they need it, they would forgo heroic treatment
probably have recommended against the operation. of improbable value if they needed it in return for more
Suppose people of average wealth, when they marry, certain benefits like education, housing, and economic
are offered the opportunity to buy one of two insurance security. If so, then justice demands that a universal
policies: the first provides that if any of their children health scheme not provide such treatment.
is born with a life-threatening defect, neonatal treat- In summary, the prudent insurance test helps to
ment will be covered only if it offers a reasonable (say, answer both questions of justice I mentioned at the
25 percent) chance of success, and the second—much beginning How much should America spend overall
more expensive—provides that such treatment will be on its health care, and how should that health care
M MI
I 1 311 lIlli1
be distributed among its citizens? The test asks what different circumstances, and these are also judgments
people would decide to spend on their own medical that a prudent insurer would be required to make.
care, as individuals, if they were buying insurance The prudent insurance test also makes plain why
under fair free-market conditions, and it insists, first, it is so important to consult public opinion before
that we as a nation should spend what individuals rationing decisions are made. Since rationing should
would spend, collectively, under those conditions; and, reflect not just technical cost-benefit calculations but
second, that we should use that aggregate expenditure also the public's sense of priorities, consultation is es-
to make sure that all have now, as individuals, what sential. When Oregon sought to establish priorities
they would have then. in health care under Medicaid, it organized a series
Of course some of the decisions I have been of "town meetings" and a "parliament" to discuss the
discussing would be made differently by different matter, and though the meetings were criticized by
people trying to apply the prudent insurance test. It some because they were attended by very few of the
is very important that any agency charged with those poor whose health .care was being debated, the meet-
decisions should be made up of representatives of dif- ings were nevertheless valuable sources of information
ferent groups that might be expected to make such about what those who did participate thought would
judgments differently; it should have doctors and be prudent insurance decisions.
health-care specialists, of course, but it should also Still, no matter how much information an agency
have ordinary people of various ages drawn from dif- seeking to apply the prudent insurance test is able to
ferent parts of the country and, if possible, different gather, its results must be provisional, open to revision
ways of life. Such an agency could draw on the ex- on the basis of further evidence of public preference
perience of countries with "single-payer" government- as well as of medical technology and experience.
run health services which have had systematically to Clinton's failed health-care plan would have allowed
ration health care. people covered by the plan to purchase supplemental
In Britain, for example, doctors in the national health insurance at market prices, with no tax deduc-
health system have been forced to allocate scarce re- tion or subsidy. That provision would have fitted the
sources like renal dialysis machines and organs for prudent insurance approach particularly well.' If, after
transplant, and they have worked out informal guide- an agency has established a basic coverage package, a
lines that take into account a potential recipient's age, very substantial number of people of average income
general health, life quality, and prospects, as well buy supplemental insurance, in spite of its expense,
as prospects for adequate care by family or friends. the basic package should be expanded. If most men of
Though this supposed cost-benefit test is different from average wealth bought supplemental coverage provid-
the prudent insurance test, the decisions doctors have ing yearly prostate examinations beginning at a younger
made under the former presumably reflect their judg- age than the basic package specified, for example, the
ments, guided by experience, about the relative value prudent insurer test would require that the age speci-
of different kinds of treatment at different ages and in fied in the basic package be lowered. F. . .1
Notes
E. . .1 the sacrifice in giving up sex is so much greater than that

1. 1 am ignoring a n important issue that my argument in giving up smoking?
raises but that I will not pursue here. Is it right, in the 2. It does seem likely that even though the members of the
hypothetical exercise I am ccnstructing, to exclude imagined community would begin by making individual
information relating risk of disease to voluntarily chosen insurance decisions, they would soon develop, through
behavior? Should insurance companies be in a position these individual decisions, collective institutions and
to charge cigarette smokers cr mountain climbers higher arrangements like cooperative insurance purchasing
premiums, for example? That seems reasonable. But if agencies or pools, because these would provide economic
so, what counts as voluntary behavior? Should sexual advantages in a free market among people of roughly
behavior of a particular kind be treated as voluntary equal wealth. The result of the process might very well
for this purpose? It would seem wrong for insurance be something functionally very close to that proposed in
companies to charge active male homosexuals higher Clinton's plan.
premiums because they are considered more likely to 3. My claim needs minor qualification. Even in the imagined
contract AIDS. Is this because sexual preference is less community, some paternalistic interference might be
under people's control than nicotine addiction? Or because necessary to protect people from imprudent insurance
decisions, particularly when they are young. And some be wrong, as a matter of religious principle, to decline an
constraints might be necessary to provide adequate expensive and arduous operation that he could afford in
resources for later generations. order to extend his life a few months might nevertheless,
4. That is different from asking what any particular twenty- with perfect consistency, think it prudent not to pay for
five-year-old would in fact do, because many people, the insurance, over his lifetime, that would enable him
particularly when young, do not make prudent decisions. to afford it. He might think it more sensible to use the
They do not, that is, make the decisions that best serve money that such insurance would cost to provide better
the plans, convictions, tastes, and preferences they would medical treatment for himself or his family earlier, or
find, on reflection, that they already have. It is prudent, better education, or some other good or opportunities his
of course; to provide for change—any prudent long-term convictions also deemed important. If so, then the prudent
insurance policy is written in general clauses rather than insurance test offers no reason why a national health
precise details of treatment and is open to revision year scheme should make available for him what he would not
by year. have insured to provide for himself.
5. See Ronald Dworkin, life's Dominion (New York Alfred A. 7. The rescue principle might hold it unjust for anyone to
Knopf, 1993), chap. 9. buy better medical care than everyone can have and that
6. People with certain religious convictions might make that medical care outside a system of universal coverage should
choice, for example. But it is worth noticing that even therefore be abolished or, as it is in Canada, seriously
those Catholics and others who think that it is always discouraged. But the prudent insurance approach begins
wrong to refuse available life-prolonging treatment do not in a different idea: that no one can complain, on grounds
necessarily think that all possible sacrifices must be made, of justice, that he has less of something than someone else
in advance, to insure that very expensive life-prolonging does so long as he has all he would have if society were
treatment is available. Someone who thought it would overall just.
Why Undervaluing "Statistical" a life? Is the answer "whatever it takes," or should there
be a limit? When is the chance of success too low even
People Costs Lives to attempt a costly rescue operation?
Tony Hope Let me pose a more general question. What is the
cash value of a human life? This question is a disturb-
Whether happiness be or be not the end to which ing one to ask but, paradoxically, there are situations
morality should be referred—that it should be re- where avoiding the question may cost lives, and allo-
ferred to an end of some sort, and not left in the cating scarce medical resources is one of them.
dominion of vague feeling or inexplicable internal There is no health care system in the world that
conviction, that it be made a matter of reason and has sufficient money to provide the best possible treat-
calculation, and not merely of sentiment, is essen- ment for all patients in all situations, not even those
tial to the very idea of moral philosophy . . . that spend relatively large sums on health care. New
J.S. Mill, London and and better treatments are being developed all the time.
Westminster Review, 1838 On average, in the UK, about three new medicines are
licensed each month. Almost all have some benefit over
The Cash Value of Life
existing treatments and some will extend people's lives.
Many of these new medicines are expensive. When is
In January 1997 Tony Bullimore was attempting to the extra benefit worth the extra cost? This question
sail round the world in the Vendee Globe race. He had must be asked by all health care systems, whether pri-
reached the dangerous and cold waters of the Southern vate systems, such as "managed care" in the US, or pub-
Ocean, over 1,500 miles south of the Australian coast, licly funded systems, such as the British National Health
when his boat was capsized by hurricane force winds Service. If the best treatment cannot always be provided,
and enormous waves. He spent four days trapped then choices have to be made. The general question of
under its hull before he was rescued in the largest and how our limited health care resources should be distrib-
most expensive such operation ever undertaken by the uted is one of the most important in medical ethics. The
Australian defence forces. How much money should a quality and quantity of thousands of people's lives will
civilized society be prepared to spend in order to save be affected by the answers that we give.
Source From Medical Ethics: A Very Short Introduction, pp. 26-41 (Oxford University Press, 2004). Reprinted with permission.
II AI 1111.1i
Quality of Life cholesterol or to people with the brain tumour. In each

case people stand to die prematurely and in each case
Some medical treatments have little or no effect on life- the treatment increases the chance that they will live for
span but improve quality of life: hip replacement for longer. What we should do, therefore, is to spend the
osteoarthritis is an example. One rather deep problem money so that we can "buy" as many life years as possible.
that faces us in thinking about the right way to distribute By doing this we are treating everyone fairly we are valu-
health resources is how we compare and evaluate the ing one year of life equally, regardless of whose life it is.
relative importance of improving quality of life vis-à-vis
extending it. I am not going to tackle this issue, nor the
The Distribution Problem
problems associated with the measurements of quality
of life in the first place. I will focus exclusively on life- Even amongst people (like me) who are attracted by
extending treatments since there are more than enough this approach, there is an issue that needs to be faced:
problems in thinking about allocating resources to these the "distribution problem." Take a look at the three
treatments alone. There are many examples of life- interventions described in Table 1.
extending treatments. Surgery for appendicitis extends Suppose that all these interventions cost the
life because without such surgery most people would die. same and that we can only afford one of them. Sup-
Breast cancer screening can extend life because early de- pose further that the distributions are as follows. The
tection and treatment can increase life-span. High blood two people who are benefited by intervention 3 will
pressure increases the risk of death from heart attack and enjoy 8 more years of life each. Of the ten people who
stroke. Treatment that lowers blood pressure reduces, are benefited by intervention 1, the average benefit is
although it does not eliminate, this risk. Renal dialysis 3.5 years, and the range is 2-4 years. Of the fifteen
keeps those people alive whose kidneys no longer func- people who are benefited by intervention 2, the average
tion adequately. Each year of dialysis is a year more life. benefit is 2 years, and the range is 1 to 3 years. Which
of the three interventions should we go for?
If we think that what we should do is to "buy"
In Control of a Budget
the maximum number of life years that we can (the
Imagine that you are in charge of a health service for a maximization view) then we should put our money
particular population. You have a limited budget—you into intervention 1 because we buy 35 life years, which
cannot afford the best treatment for all of the people is more than we will get if we spend the money on
all of the time. You have decided how to spend most either of the other two interventions. Some might argue
of your budget, and you have a few hundred thousand that intervention 2 is preferable because we help more
pounds left uncommitted. You sit down with your ad- people (15 as opposed to 10) although each person
visers to consider the best way of spending this last re- gains fewer extra years of life. Still others might argue
maining tranche of money. There are three possibilities, that intervention 3 is the best option because the two
and you must choose one of them. The possibilities are: people who are helped receive a really significant gain
(eight years of life) whereas no one gains more than
1) a new treatment for bowel cancer that gives the four years of life with either of the other two options.
relevant patients a small but significant chance of The question of whether it is only the total number of
increased life-expectancy; life years that matters, or whether the way in which
2) a new drug that lowers the chance of death from those years are distributed between people is import-
heart attack in people with genetically induced ant, is known as "the distribution problem." Those
raised blood cholesterol; who reject the maximization view have to specify how
3) a new piece of surgical kit that ensures a lower they balance the value in helping more people, but
mortality from a particularly difficult kind of brain each gaining relatively less, against the value in helping
surgery. fewer people, but each gaining relatively more. Except
at extremes I am generally happy to go with maximiz-
On what basis do you choose between these ing the total number of life years and not worry too
possibilities? much about their distribution.
One approach that has a lot going for it is to say.
there is no good reason to prefer one person's year of life Table 1 Choosing between Three Interventions
over another person's, or to give any priority to people Intervention 1 benefits 10 people total life years gained: 35
who would benefit from the bowel cancer treatment over Intervention 2 benefits 15 people total life yeas gained: 30
people who have the genetically induced high blood Intervention 3 benefits 2 people total life years gained: 16
Hope: Why Undervaluing "Statistical" People Costs Lives
In being generally happy with using resour- some quite independent cause before any benefit from
ces to maximize total number of life years I am in a A could be gained. There are also problems in calculat-
minority—and no health care system in the world ing the cost per life year gained when at least some of
behaves remotely in this way. One problem with my the costs of the intervention are borne years before the
position (the maximization view) takes us right back benefits are seen. This is because of monetary infla-
to Tony Bullimore and his attempt to sail round the tion. Both these effects can be allowed for in the calcu-
world. My position gives no moral weight to what has lation of cost per life year gained. Having made such
been called "The Rule of Rescue"—and yet this rule allowances, there seems no good reason to value the
seems, intuitively to be right. saving of life years in the future any less than saving
life years now.
The Rule of Rescue The second difference between the interventions
is that B will almost certainly save the lives of the
The "rule of rescue" is relevant to a situation where relevant patients, but A only has a low probability of
there is an identified person whose life is at high risk. doing so. Thus B might be seen as giving greater bene-
There exists an intervention ("rescue") which has a fit to individuals than A. I will argue, in a moment, that
good chance of saving the person's life. The value that this is false.
is at the heart of "the rule of rescue" is this: that it is The third difference is that intervention B benefits
normally justified to spend more per life year gained identifiable people. Intervention A benefits a propor-
in this situation than in situations where we cannot tion of patients within a group (e.g., those with raised
identify who has been helped. blood pressure), but we cannot know who within the
Consider two hypothetical, but realistic, situations group will benefit (although we may know the likely
in health care. proportion that will benefit).
According to the rule of rescue it may be right for a
Intervention A (Saves Anonymous "Statistical" Lives) health care system to fund intervention B but not inter-
A is a drug which will change the chance of death by vention A, even though B is more expensive in terms
a small amount in a large number of people. For ex- of life years gained. For example, the rule of rescue
ample, out of every 2,000 people in the relevant group, would provide justification for spending more per life
if A is not given, then 100 people will die over the next year gained on treatments such as renal replacement
few years. If A is given, then only 98 will die. Although therapy than on treatments like statins.
we know that drug A will prevent deaths, we do not In practice this is exactly what health care systems
know which specific lives will be saved. Drug A is do. The British National Health Service provides renal
cheap—the cost per life year gained is £20,000. One dialysis at costs over £50,000 per life year gained,
example of a medical treatment like this is treatment whilst paying for statins only for those with very high
that lowers moderately raised blood pressure. Another cholesterol levels. This is despite the fact that treatment
example is a class of medicines known as statins that with statins for those with moderately raised choles-
lower blood cholesterol. Lowering blood pressure, and terol levels would cost only about £10,000 per life year
lowering cholesterol, reduce risk of heart attack, stroke, gained. In other words, if the money spent on some
and death. people for renal dialysis were, instead, spent on some
people with moderately raised cholesterol, five times
Intervention B (Rescues an Identified Person) as many life years could be gained. But we don't do
B is the only effective treatment for an otherwise life- it—because we would feel that we had condemned the
threatening condition. Those with the condition face person needing dialysis to death, whereas all we would
a greater than 90 per cent chance of death over the be doing in the case of statins is slightly lowering an
next year if not given B. If given B, then there is a good already quite small chance of death.
chance of cure—say 90 per cent. B is expensive. The The most powerful reason in support of the rule of
cost per life year gained is £50,000. Renal (kidney) rescue is that in the typical case, the identified person,
dialysis is an example of this type. like Tony Bullimore, stands to gain a significant in-
There are three, potentially relevant, differences crease in chance of life, whereas in the typical case
between intervention A and intervention B. The first of saving anonymous "statistical" lives no one stands
is that B saves lives within the next year, whereas the to gain more than a small decrease in probability of
benefits of A are not realized for many years. This death. I will put this argument in favour of the rule of
difference has some moral relevance. A few of those rescue as strongly as I can. I will then say why I do not
who might benefit from intervention A will die from agree with it.
• II 'SI 1111.13
The Strongest Argument in Favour (because we had moderately raised cholesterol levels) it
of the Rule of Rescue would be reasonable, and not extraordinarily altruistic,
for us to prefer that the money go not to provide us
Premature death is, normally, a very significant harm with statins but towards the cost of renal dialysis for
indeed. But a very small chance of premature death someone who would otherwise die. From the point of
is by no means a great harm—and we cannot claim view of those who have to decide how limited health
that we need something which reduces by a very small care resources should be distributed, it certainly seems
amount the chance of premature death. All of us in our better to keep a few people alive (who would otherwise
lives trade small increases ir. the chance of premature certainly die) than to reduce only slightly the chance of
death against really quite small benefits. Consider the death of a large number of people, particularly when
Sunday morning cyclist. the risk of premature death is fairly low anyway.
The Sunday Morning Cyclist Back to the Distribution Problem

On Sunday mornings I cycle along the busy Banbury The rule of rescue seems to be a particular example of
Road in my home town of Oxford to buy a newspaper. the distribution problem. Most people reject maximiz-
In doing this I am putting myself at (what I hope is) ing life years gained (which would favour paying for
only a small extra risk of premature death. I am trading statin treatment). Essentially, the intuitive appeal is as
this extra risk against the pleasure and value of read- follows: it is better to provide a great benefit (continuing
ing the Sunday morning paper. In balancing these two I life in people who would otherwise certainly die) to a
find that the pleasure of the paper—a really rather small few people than a trivial benefit (a small reduction in
pleasure in my life—outweighs the extra risk of pre- chance of premature death) in a large number of people.
mature death. There seems nothing irrational in this. A
very small chance of a terrible harm is itself only a small Why I Disagree with the Rule of Rescue
negative weight easily outweighed by other benefits.
Most of us will take these small risks not only for Despite the strong intuitive appeal of the rule of rescue,
our own benefit but for the benefit of others. Consider and the arguments that I have outlined in favour of it,
the friend's job application. I stick by my preference for maximizing benefit. I will
argue for my position by considering a counter-example
The Friend's Job Application to this conclusion: the case of the trapped miner.
Suppose that a friend is applying for a job which he is
keen to get. To meet the application deadline it has to The Case of the Trapped Miner
be in the postbox today. Owing to a severe bout of in- Consider the case of the trapped miner (see box).
fluenza, my friend cannot post it himself. To help him I Suppose that the facts are these (perhaps not entirely
cycle to his house to collect the application and post it. realistic). There is a small risk of death to those in the
Again, this action increases by a very small amount my rescue party, and this risk varies according to the size
chance of premature death. This is easily outweighed of the rescue party. If there were 100 rescuers, there
by the value of helping my friend. would be a 1:1,000 chance for each rescuer of death. If
With these considerations in mind I will propose there were 1,000 rescuers, each would face a 1:2,000
an argument in favour of a health care system paying chance of death. If 10,000 rescuers, then each would
for a "rescue" intervention of type B (at, for example, face a 1:5,000 chance of death. If 100,000 rescuers
a cost of £50,000 per life year gained) whilst refusing (an extraordinarily large rescue party—but this is a
to pay for an anonymous "statistical" intervention of "thought experiment" to test a theoretical point)—then
type A (at, for example, a cost of only £20,000 per each would face a 1:10,000 risk.
life year gained). I will make the argument using the Thus, the larger the size of the rescue party, the
cholesterol-lowering drugs (statins) as an example of smaller the risk of death faced by each individual res-
the anonymous "statistical" intervention and renal di- cuer. It is also the case, however, that the larger the size
alysis as an example of the rescue intervention. of the rescue party, the more people are likely to die in
Those who would benefit from treatment with the rescue attempt. With a rescue party of 100,000, each
statins gain very little—a very small reduction in the member of the rescue party faces a very small risk of
risk of premature death. The "friend's job application" death—well within the risks that we normally take for
shows that we readily risk small changes in the chance much less important gains than saving a life. However,
of premature death, even for benefits to other people. with such a rescue party, about ten people are likely to
If we ourselves stood to gain from the statins treatment die in order to save the life of the one trapped miner.
Hope: Why Undervaluing 'Statistical" People Costs Lives
The Case of the Trapped Miner

A miner lies trapped following an accident. Without rescue he will die. Given a sufficiently large rescue party, the
miner can be saved.
Take a moment to consider the following questions:
1. Do you think you should join the rescue party if you faced a 1:10,000 risk of death in so doing?
2. Is there any further key information you need to know before you can answer the first question?
If we assume that most people are altruistic at least will be lost in rescuing the one miner. I would be much
to a small extent, and most people will accept a very happier (from the moral point of view) volunteering for
small level of risk of personal death in order to save an- the first rescue party than the second.
other's life, and if we assume, further, that most people,
given the choice, would like to face as low a personal Back to Health Care
risk of death as possible, then respecting the wishes of Let us reconsider statins and renal dialysis. It is not
each potential member of the rescue party would have clear that those who could benefit from the anonym-
the following result. The wishes of potential members ous "statistical" intervention (e.g., statins) have vol-
of the rescue party would be most respected by putting untarily agreed to forgo their treatment in order for
together an enormous rescue party in order to save the identifiable patients to receive expensive life-extending
trapped miner—at the expense of many lives. treatment. A health care system that spends more per
Thus, if the issue of rescue is seen simply as a year of life gained on rescue treatments (such as renal
question of balancing individual risks for each rescuer dialysis) than on "statistical" treatments is effectively
against the benefit to the individual of being rescued, volunteering those who would benefit from the pre-
then it would seem right to pursue a policy which over- ventive treatment to take part in a "rescue party" for
all was very costly in terms of lives lost. those requiring the rescue treatment. Because of lim-
Suppose that a senior army officer will lead the ited resources, any health care system, in making deci-
rescue. If that army officer were to coordinate the sions about treatments which extend people's lives, has
rescue, with the foreseeable result that more people to extend some people's lives at the expense of other
would die in the attempt to rescue than would be saved people's lives. In the absence of a clear mandate from
by the rescue, then the army officer might reasonably the group of people who stand to lose by a particular
be criticized, even if the rescue party were made up decision, it seems to me that the core principle must
entirely of volunteers who knew and accepted the risk be that those decisions should be taken which overall
to themselves. He would have been responsible for a maximize the number of life years gained. And even if
rescue operation that caused, and had been expected there were such a clear mandate (which there is not) it
to cause, more deaths amongst the rescuers than the remains questionable, as with the army officer leading
number of people who were rescued. Leading such a the rescue operation with fully informed volunteers,
rescue even with fully informed volunteers is highly whether it would be right for a health care system to
problematic from a moral point of view. let more die to save fewer.
Further Key Information A Counter-intuitive Conclusion

Let me return to the second question I asked about But can we accept this conclusion? Let's go back to
the case of the trapped miner is there any further key Tony Bullimore and the dramatic and successful rescue
information you need before answering the first ques- undertaken by the Australian defence forces. Only a
tion? I think you should know not only your personal stone-hearted theorist could read Bullimore's account
risk in joining the rescue party but also the size of the and conclude that it was wrong to mount such a rescue.
rescue party. Because if the rescue party needs only The Australian defence forces were right to spend mil-
ten people and each member has a risk of 1:10,000 of lions of tax-payers' dollars. In the same way, it is right
dying, then the miner's life will be saved with (almost for a society to spend £50,000 a year to keep a patient
certainly) no loss of life. But if the rescue party needs alive on renal dialysis. How could we stand by and say
to be 100,000 strong, then almost certainly many lives to a patient: we could keep you alive for many years,
II 5111 111111
but we will not provide the necessary money—we have failure? The answer, I think, is not that we should become
other priorities. And how could we say this to the rela- stone-hearted logicians and refuse to attempt the rescue
tives who would be bereaved? of Bullimore or to provide renal dialysis. It is right that
This seems very different from the situation of the our moral imagination and our human sympathy are
patient with moderately raised cholesterol. Without awakened. What we should learn from the logic of the
treatment the chances are that the person will not have case of the trapped miner is that our moral imagination
a heart attack and die. By refusing the treatment we are must also be awake to the sadness of lives cut short, and
not condemning him to death as we are the person who relatives bereaved, because we did not provide treatment
needs renal dialysis. for moderately raised cholesterol. Deaths are not less sig-
But the logic of the case of the trapped miner nificant because we cannot put a face or a name to the
refutes this. It is true that if we do not provide treat- person whose life could have been saved.
ment for the raised cholesterol we will not know which Health care is good value for money. The lesson
specific people die as a result of lack of treatment, nor we should learn from our empathy for those in need of
which relatives have been bereaved. But we do know rescue is to widen our moral imaginations. We rightly
that there are such people. respond to the person in distress by being prepared
to spend money to save a life. We should respond in
Enlarging Our Moral Imagination the same way to prevent "statistical" deaths, for such
So how do we square the circle? What do we learn from deaths are real people and the friends and relatives who
our empathy with Tony Bullimore or a person with renal are left behind mourn in just the same way.
QALYs vs DALYs vs LYs Gained: In this situation, with health challenges that are
immense and budgets that are extremely scarce, giving
What Are the Differences, and What priority to the "wrong" programs may have large impli-
Difference Do They Make for Health cations for health system performance. The opportun-
Care Priority Setting? ity cost of wasting 10 USD in a country like Tanzania
is more than one year of human life. An example is im-
Bjarne Robberstad munisation against hepatitis B, which is not routinely
done in a majority of sub-Saharan African countries.4
Introduction
This intervention has been demonstrated to cost as
little as 8 USD per life year saved.5. Putting weight on
The total population of Africa is some 655 million people, cost-effectiveness seems like a good idea in this situa-
who experience an annual loss of about 358 million dis- tion, because when resources are very scarce it seems
ability adjusted life years (DALYs). This is almost half a DALY reasonable to make the most of it. Economic evalua-
per capita per year, with HIV/ADS, childhood diarrhoea, tion, like cost-effectiveness analysis (CEA), has therefore
measles, malaria, and respiratory infections among the been assigned a key role since the World Development
most important causes of disease burden.' The life ex- Report 1993.6
pectancy in the region is generally short, with 47 years in For CEA to serve a meaningful role in comparing
Tanzania and 37 years in Malawi as examples.' Effective the relative value of different health interventions,
treatment or prevention is available for most of the above it is necessary to apply a commonly agreed measure
conditions,' but health care budgets that are extremely of population health. The objectives of this paper are
scarce represent effective barriers against improved therefore to present the major differences between
population health. Generally for the region, health care three commonly applied health measures: life years
budgets are in the range 3-20 USD per capita per year, gained Os gained), quality adjusted life years (QALYs),
while for e.g., Tanzania the health budget in 2002 was and disability adjusted life years (DALYs). Furthermore,
7 USD per capita3 To put this into perspective, the public it will be explored what difference these alternative
health care expenditures in Norway were 3366 USD per measures may have for health care priority setting in
capita last year,3 almost 500 times higher. sub-Saharan Africa.
Source: From Norsk Epidemiologi 1`

-•, 2 (2005), 183-191. Reprinted with permission.
Robberstad: QALYs vs DALYs vs LYs Gained
Life Years Gained are few value choices involved. The perhaps most im-
portant disadvantage is that the method ignores the ob-
Traditionally, the impact of health care has been meas- vious fact that health is more than merely staying alive.
ured in terms of its effect on mortality,' for example, The method will fail to acknowledge health improve-
deaths averted. A potential drawback of using deaths ments such as improved physical ability, reduced
averted to measure health effects is that the method neuropsychological stress, and reduced chronic pain.
doesn't distinguish between children, adults and the This makes it difficult to compare outcomes of health
elderly. This is not unproblematic since most people, interventions that target e.g., chronic conditions with
if forced to make a choice, would prefer to save e.g., a interventions targeting fatal conditions. Furthermore,
5-year-old infant rather than a person aged 95. if only the mortality aspect is quantified and included
Life years gained is a modified mortality measure in the denominator of the cost-effectiveness ratio, LYs
where remaining life expectancy is taken into account. gained represent an intrinsic bias against conditions
This method accrues more weight to young target that are largely non-fatal like for example poliomyelitis,
populations, because saving the life of an infant yields onchocercosis (river blindness) and schizophrenia.
more life years than saving the life of an old person. Life
years are calculated as the remaining life expectancy at Taking Morbidity into Account
the point of each averted death. Life expectancies may
be taken from life tables that are specific for each setting It is non-controversial that it is necessary to say some-
or standardized across larger regions. The choice of life thing about the majority of people who do not suffer
table is not uncontroversial, as a life table with high life premature death.° Albert Schweitzer once said that
expectancies will yield more life years and render inter- "Pain is a more terrible lord of mankind than even
ventions more attractive than life tables with shorter death itself."14 It is therefore, at one level or another,
life expectancies. necessary to take people's morbidity into account. This
recognition has led to the development of measures
Discounting of Health Effects that simultaneously seek to capture gains from reduced
morbidity and reduced mortality The two dominating
LYs gained can be discounted, to adjust for time differ- such summary measures are the Quality Adjusted Life
ences, or presented undiscounted. Common practice Years (QALYs) and the Disability Adjusted Life Years
is to discount health outcomes at a rate between 3 and (DALYs). QALYs and DALYs represent an implicit trade-
5%." Discounting accrues less weight to future than off between quantity for quality of well-being. Such
to immediate health outcomes, which is illustrated in trade-offs are well known from many aspects of life,
Figure 1. With a 3% discount rate, for example, half the but many are unfamiliar with the concept of trading off
weight is accrued to outcomes 23 years into the future quality for quantity of years of life and health.
compared to immediate outcomes, while a 5% discount William Somerset Maugham, in the novel "As a
rate reduces the weight to half in about 14 years. There Lily in the Field," describes a young banker in London
is a relatively extensive amount of literature discuss- (Thomas Wilson) who found the prospective of doing
ing whether or not the practice of discounting future the same kind of work, day by day, for the rest of his
health effects is appropriate. For example, discounting life rather depressing. Thomas Wilson once went on a
has been criticized for discriminating future genera- holiday to Capri in Italy, and later he kept daydream-
tions and for being immoral.1° Others claim that for the ing about the swimming, the wine yards, the hillsides
sake of consistency, discounting of health as well as in moonlight, the sea and piazza in the evening. At
costs is necessary."2 This debate will not be pursued the age of 35 he quits his job, sells his house and uses
further in this paper, because discounting is not what all his assets to invest in a life pension. The life pen-
makes LYs, QALYs and DALYs different. Nevertheless, the sion will provide him with sufficient means to rent
choice of discount rate is important because it may de- a house at Capri, employ a servant to tend for him,
termine the mix between typical curative interventions, enough for tobacco, good food, and a book from time
like improved case management of malaria, and typical to time for a period of 25 years. Rather than living, in
preventive interventions, like hepatitis B vaccination. his view, a miserable life as a banker for the remaining
life time, he decides to live like "a lily in the field" for
Advantages and Disadvantages of LY Gained 25 years, till he is 60, and then put an end to it. In other
words, Thomas Wilson makes a plan that enables him
LYs saved is as indicated a relatively easy and transpar- to trade-off a long and miserable life for a shorter but
ent method for measuring population health, and there qualitatively superior life.
II MN
1.0
0.8 0%
a6
0.4
10 20 30 40 50
Years from now
Figure 1 The Effect of Discounting on the Weight of Future Health Effects
Quality Adjusted Life Years (QALYs) (i) that the burden calculated for like health outcomes
should be the same, and (ii) that the non-health charac-
The story of Thomas Wilson brings us to the QALY, teristics that should be considered should be restricted
which was first introduced by Klarman et al.,15 al- to age and sex.14 Other characteristics, like income,
though the concept is more commonly associated education, ethnicity, etc., should in other words not be
with an applied paper by Weinstein and Stason.16 In taken into account.
QALYs, premature death is combined with morbidity These propositions represent intentions of cre-
by attaching a weight to each health state such that ating a methodology that treat people as equally as
value 0 represents death, while value 1 represents full possible. However, some would say that more weight
health. The number of QALYs for a health profile is should be put on the diseases of the disadvantaged,
found by multiplying the health related quality of life and that failing to do so is inequitable. Anand and
weight (HRQof..) of the health state with the duration Hanson, for example, claim that disease weights
of the health state. Figure 2 illustrates a person who should take account of the way in which individual
gets some disability at the age of 10, lives with the and social resources can compensate for the level of
condition for 35 years, and suffers premature death disability experienced.17 For example, being blind in
at the age of 45. If the life expectancy is 60 years, Niger is worse than being blind in the UK, because the
and the health related quality of life weight associated structural interventions in the UK make the disability
with the condition is 0.75, the (undiscounted) life- less severe than in Niger.18 Blindness, according to
time QALYs of this person are 1.0 x 10 (QALYs before this view, should therefore be weighted more heavily
onset of disease) + 0.75 x 35 (QALYs during disease) in Niger than in the UK.
= 36.25. The QALY loss would be 1.0 x 60 (QALYs in Like the QALY, the DALY measure facilitates com-
case of full health for full life expectancy) — 36.25 = parisons of all types of health outcomes by attaching
23.75. Like the LY measure, QALYs can be presented disease weights where value 0 represents full health
discounted or undiscounted, although the former is and value 1 represents death. Note that these disease
most common. weights are the opposite of the HRQ0L weights in the
QALY. A DALY can therefore be seen as an inverse QALY.
Disability Adjusted Life Years (DALYs) Figure 2 illustrates that while QALYs are a good, DALYs
are something one seeks to avert. In the following sec-
The DALY concept was introduced by the World Bank tion the statement that a DALY is an inverse QALY will be
in the World Health Report 1993.6 Two important somewhat modified because of different weighting and
propositions are underlying the DALY concept, namely elicitation techniques.
0.0 In the time trade-off (Tro) methodology people

DALYs
are asked to choose between two certain alternatives
involving trade-off between quantity and quality of
life. For example, they may be asked how many years
3461am as easia
of perfect health they would consider equally good
Health-related qu
as living 10 years with psoriasis. TTO questions are

the most common technique for elicitation of HRQoL
weights in QALYs, although VAS and standard gamble
QALYs
(see below) are also sometimes used.
Person trade-off (PTo) is structurally similar to
0.0 1.0
Onset of Death LE 60 no, but here people are asked to trade off extending
disability 45 years years the lives of people with full health vs improving the
10 years health expectancy of people with some disability from
Age
suboptimal to perfect health. This method was used to
Figure 2 Relationship between QALYs and elicit disease weights for the DALY methodology from
DALYs When the DALYs Are Not Age Weighted a panel of health experts, while the HRQoL weights in
QALYs are usually found by interviewing lay people
and patients.
Weighting of Health Outcomes In the standard gamble (SG) approach, respondents
are asked to choose between the certainty of an inter-
In the above section it has been shown that the differ- mediate health state and the uncertainty of a treatment
ent health measures LYs, QALYs and DALY5 are weighted with two possible outcomes, where one of the out-
in different ways. They are all adjusted for the life ex- comes is more attractive than the certain outcome and
pectancy of people affected, giving more weight to the the other is less attractive (e.g., death). The inclusion
young. Furthermore, they are all usually discounted, of uncertainty makes SG more consistent with standard
giving more weight to immediate over distant out- economic utility theory than the other methods, but it
comes. However, only DALY5 and QALYs are weighted is also probably the cognitively most demanding of the
to account for disease severity in order to summarise techniques for respondents to answer.
mortality and morbidity The following section will Disease weights or HRQoL weights may easily
partly be devoted to explaining how such weights influence priority decisions. It is therefore of interest
can be elicited. Furthermore, DALY5 are so-called age whether or not the choice of elicitation technique is
weighted, and this also calls for some explanation. likely to influence the weights. From a theoretical point
Finally, it has been proposed that such health measures of view, one may expect that weights elicited with SG
should incorporate that extra weight or priority should are higher than the VAS because of risk aversion. One
be given to the most severely ill. This last concept has may also expect TTO weights to be higher than VAS
yet not been implemented in any of our outcome can- weights because of time preferences.
didates in applied research, but the argument has some These expectations were confirmed by Salomon
intuitive appeal, and at the end of this section some and Murray, who in a sample of 69 individuals found
advantages and problems with such severity weighting that PTO weights were higher than SG, which again
will briefly be discussed. were higher than TTO, while weights produced by VAS
were the lowest.19 In other words, VAS seemed to weight
conditions as more severe than the other alternatives.
Elicitation of Weights
While there was significant disagreement between the
Several methods are being used to elicit health related methods on the level of severity in this study, there was,
quality of life or disability weights. The simplest ap- however, considerable agreement in the rank orderings
proach is the so-called Visual Analogue Scale (VAS), where of the 12 conditions.19 Currently there is no agreement
respondents are asked to rate health states on a scale. on which method should be preferred for eliciting dis-
This scale often has the range from zero to one, where ease or quality of life weights.19,2°
zero is the worst possible outcome (assumed to be
death) and one is the best possible (perfect health). VAS Age Weighting
is a simple method but has been strongly criticised for
not being choice based, in other words, that it does not Age weighting is a concept unique to the DALY meth-
involve a trade-off between quantity for quality of years. odology and is intended to account for the fact that
Ai I I JAI mw
people are supported by others during infancy and at Simple Utilitarian Ageism
an advanced age but support others during adulthood.
This notion is called welfare interdependence. Welfare
interdependence does not imply that the time lived at •_
different ages are more or less important to those indi-

viduals but that the social vi.ue is different.14
The age weighting of DALYs therefore accrue more Age
weight to the life years of breast-feeders and caregivers Productivity Ageism

and can be illustrated by a hump-shaped curve which
starts at zero for newborns, peaks at 25, and gradually
declines throughout adulthood without ever reaching
zero. It can be argued that welfare interdependence is
an arbitrary choice for an age-weighting function. In
fact there exist several alternative ways to specify age Age
weights depending on different perceptions on fairness.
Egalitarian Ageism
I will mention two alternatives to the DALY age weights.
The first alternative is the so-called simple utilitar-
ian ageism,21' 22 or horizontal age weights. Horizontal
age weights accrue equal weight to all years and is
what is being used in the traditional QALY measure and
non-age-weighted DALYs. The last alternative is funded
Age
on the fair innings argument2' and is called egalitarian
ageism.22,24 According to this argument everyone is Figure 3 Three Alternative Age-Weighting
entitled to a similarly long and healthy life, and the im- Functions
plication is that anyone failing to achieve this has been
cheated, while anyone getting more is living on bor-
rowed time. Z5 The elderly according to this argument, equity-weighted QALYs has b y Alan Williams been pro-
have already had their fair share of life and should be posed as ". . . the best way to integrate efficiency and
given gradually less weight in the distribution of scarce equity in health care."25 And furthermore that equity
health care resources. These three alternative age- weights ". . . rules out the giving of priority to things
weighting functions are illustrated in Figure 3, but it that do no good, which is the danger with equalization
should be noted that several sub-versions of the prin- policies." Williams argued for a version of the fair in-
ciples exist 21 nings, where the quality of a person's life is important
Tsuchia has reviewed the empirical evidence on as well as the length of it.25
age weights and sums up that (i) there is no support There are two major barriers to implementing
for a horizontal age-weight profile (current QALY prac- equity-weighted QALYs or DALYs. First, there exists no
tice) and (ii) there is broad agreement that the profile agreement as to over which notion of disease severity
declines after middle age, but (iii) there is some dis- one should pursue equity. Is it, for example, acute ill-
agreement on whether middle age or childhood is the health, or is it lifetime ill-health?26 Secondly, empirical
peak.22 Despite the agreement that the young should evidence including both estimation of equity weights
have some priority over the old, as Williams points out, and implementation of them in QALY maximization
the consensus is likely to evaporate when the quanti- exercises is scarce. Bleichrodt and colleagues estimated
tative issue is addressed as to how much priority the equity weights in two Dutch populations. When they
young should have.25 incorporated these weights in QALY calculations, they
found that this in fact changed the ranking of 12 health
interventions.27 These findings suggest that the issue of
Equity Weighting
equity weighting needs further research, although it is
There is considerable empirical evidence suggesting likely that such an enterprise will be quite data hungry.25
that people in general are inequality averse in the sense
that they hold the view that more weight should be What Difference Do They Make?
allocated to those that are most severely ill. However,
equity weighting is currently not standard procedure After having briefly presented some major differences be-
either in DALYs, or QALYs or LY1; gained. Introduction of tween LYs gained, QALYs and DALYs, it is time to approach
Robberstad: QALYs vs DALYs vs LYs Gained 387
the second objective of this paper: What difference do therefore increase the likelihood of implementation,
they make for health care priority setting? I will first ad- all other things being equal.
dress the question by looking at a small selection of the
literature in the area. For example, it has been argued "DALY Weights Are Unfair"
that quality adjustment of life years makes little differ-
ence for priority setting. Others have argued that indeed There have been some concerns on whether or not the
it makes a difference, and moreover, that the difference weighting of DALYs is fair. With the exception of the age
is unfair or discriminating. Secondly, I will address the weights, this critique also applies to QALYs. Anand and
question by looking into two applied cases that I know Hanson have claimed that "Using the DALY framework
well, namely, my own studies on the cost-effectiveness of for resource allocation may lead to consequences that
zinc as adjunct therapy in childhood diarrhoea28 and on are at odds with principles of equity." They are wor-
pharmaceutical preventive cardiology29 ried that life-extension of disabled people avert fewer
DALYs, and therefore count less, than life-extension
of able-bodied people." Hypertension management
"Quality Adjustments Make Little Difference"
of people with polio would, for example, avert fewer
Chapman and colleagues concluded that quality ad- DALYs than hypertension management of people with-
justments make little difference for priority setting after out such a chronic condition because each life year
having considered 63 CEAs reporting both costs per gained will be given less weight. The disability weight
QALY and costs per LY gained. They therefore concluded of polio is 0.369.14
that in most cases findings can be reported as costs per This critique seems to be formally correct but in-
LY rather than the technically more challenging costs correctly applied. When DALYs, or QALYs, are used to
per QAa.3° This conclusion is partly based on the find- estimate burden of disease, or in cost-effectiveness an-
ing that quality adjustment of life years did not sub- alysis to inform resource allocation, no such distinction
stantially change the estimated level cost-effectiveness. is made between disabled and able-bodied people. The
More specifically, they found that switching from QALY cost-effectiveness of e.g., preventive cardiology is done
to a pure mortality measure like as gained altered the assuming that people have no unrelated conditions,
conclusion of whether or not to support implementa- like polio. The critique of Anand and Hanson therefore
tion of health programs only in 18.5% of the cases.3° does not seem to apply to DALYs and QALYs the way
This is of course interesting, but if one believes that they are used in economic evaluations.
morbidity should be taken into account, it may also be Anand and Hanson are also critical of age weighting
interpreted as a simple measure like LY gained leading and argue that there are no reasons for valuing time lived
to the wrong conclusion in almost one-fifth of the cases. at different ages differently and that a principle of uni-
Furthermore, Chapman and colleagues investi- versalism would argue strongly for a common intrinsic
gated the number of times the ratio difference (Costs/ valuation of human life, regardless of the age at which
LY — Costs/QALY) was positive, zero or negative. They it is lived." Others are less worried by this kind of dis-
found that 60% of the ratio differences were positive.30 crimination, because of the fact that we may all experi-
A positive ratio difference is the expected finding ence each age level. The discriminatory effect of setting
and implies that the cost-effectiveness ratios become unequal age weights is therefore ethically very differ-
more favourable when life years are quality adjusted. ent from the setting of weights according to ethnicity,
Furthermore, they found that 3% of the ratio differen- sex, education or income."
ces were zero, implying that morbidity is not an issue
for these conditions, while 36% of the ratio differences "DALY Weights Are Discriminating"
were negative.30 The latter of these observations means
that quality adjustment has reduced the denominator It has been argued that disability weights are discrimin-
in the cost-effectiveness ratio. This is conceivable only ating against developing countries. In the DALY manual,
in two situations: (i) when a treatment has side effects disability weights for clinical conditions are reported
affecting the quality of life sufficiently to out-crowd both before and after treatment, based on the general
its life-extending effect (e.g., cancer treatment when idea that a disease is less burdensome if it is treated.
prognosis is poor) or (ii) when the health improve- In a paper exploring whether DALY weights can influ-
ment of a treatment is out-crowded by shortened life ence global priority setting, Arnesen and Kapiriri claim
expectancy. The main lesson, nevertheless, seems to that since few of the communicable diseases, which are
be that quality adjustments increase the attractiveness typical developing country conditions, are reported
of interventions in terms of cost-effectiveness and with reduced disease weights after treatment, the DALY
II ,911 II
methodology is discriminating against developing opposite directions with respect to the burden on the
countries." Arnesen and Kapiriri are also critical of the youngest. The aggregate effect of age weights on chil-
general level of disease weights between communic- dren, including chronic, recurrent, fatal and non-fatal
able and non-communicable diseases, indicating that conditions, remains uncertain. However, if we believe
the non-communicable diseases generally should have that mortality is more important for burden of disease
been weighted more heavily.3. than morbidity, age weights will lead to more resource
Furthermore, Arnesen ar d Kapiriri claim that dis- being allocated to interventions targeting children.
ease weights discriminate against the young, because
non-communicable diseases, which are a relatively The Effect of Weighting in Two Applied
higher burden in the elderly population, generally are Economic Evaluations
assigned higher weights than the communicable dis-
eases, which are relatively n- ore important in young From the above theoretical considerations, it is useful
populations. In particular, they question that a condi- to turn our attention towards two applied economic
tion like developmental disability due to malnutrition, evaluations and consider how disability weighting,
primarily affecting children, has a disease weight of age weighting and discounting influence the cost-
only 0.024, while unipolar major depression, which effectiveness and attractiveness in these cases. The
is a typical adult condition, ;las a weight of 0.600." scenarios of these two studies are different over several
The young are also discriminated against, in their view, dimensions and therefore useful to illustrate what dif-
because of the age weights.' ference different outcome measures may have for health
It is evidently correct that the disease weights of the care priority setting. The first of these is an economic
DALY methodology need to be looked further into, with evaluation of zinc as adjunct therapy in diarrhoea,28
respect both to elicitation techniques, level of disease while the second is a study of the cost-effectiveness of
burden and change of weights after treatment. Several pharmaceutical interventions to prevent cardiovascular
examples given by Arnesen and Kapiriri illustrate this, disease in sub-Saharan Africa.29
for example, that amputation of a thumb has been as- Both of the two economic evaluations include a
signed higher disease weight than amputation of an number of treatment alternatives. In Table 1, the cost-
arm. 14,31 Without further empirical evidence, there are effectiveness ratios (cERs) for two arbitrary example
few reasons either to support or to dismiss Arnesen interventions are presented, zinc and oral rehydration
and Kapiriri's general critique of the level of disability solution in diarrhoea management and aspirin and a di-
weights for different groups of diseases. Nevertheless, it uretica given to patients with very high cardiovascular
is a main point that since health care resources are gen- risk Results are first presented as USD per LY gained,
erally distributed within and not between countries, secondly as USD per non-age-weighted DALY, which
the claim that people with communicable diseases and roughly corresponds to QALYs, and finally as USD per
developing countries are discriminated against due to standard age-weighted DALY. In all cases, both costs
disease-weighted DALYs remains undocumented. and health effects are discounted using a rate of 3%.
Arnesen and Kapiriri also conclude that age The CER of zinc is 74 USD per LY gained, which
weights lead to the disease burden carried by the improves slightly to 70 USD per non—age-weighted
youngest age groups are given considerably less weight DALY (or QALY) by including a disease weight of 0.119
and, hence, less priority than it would without these
choices.31 This conclusion is true, but only in the case
of non-fatal and non-chronic conditions. For fatal or
chronic conditions, on the other hand, it can be dem- Table 1 The Impact on the Cost-Effectiveness
onstrated that age weights favour people aged less Ratio of Using LY Gained, QALYs (Non-Age-
than 35-45 years. The reason for this is that for fatal Weighted DALYs) and Standard Age-Weighted
or chronic conditions, the DALY calculations of young DALYs in Interventions Targeting Childhood
people will include the peak of the hump shaped age- Diarrhoea and Preventive Cardiology
weighting function. This is in fact illustrated in Figure
Aspirin and
1 in Arnesen and Kapiriri's own paper," where different Zinc in diuretica in
combinations of discounted/undiscounted and age- childhood preventive
weighted/non-age-weighted DALYs are presented. How- diarrhoea cardiology
ever, the authors fail to sufficiently pull forward this USD per LY gained 74 55
finding and base their conclusion on the joint effect of USD per DALY (no age weights) 70 46
USD per DALY (age weighted) 66 56
age weighting and discounting, which in fact pulls in
for diarrhoea. When the outcome in addition is age undocumented, in particular the claim that age weights
weighted to standard DALYs, the CER further improves discriminate against children.
to 66 USD per DALY (see Table 1). In the case of diar- It is undoubtedly true that age weighting and
rhoea, it therefore turns out that both disease weighting disease weighting have ethical implications that are
and age weighting makes the intervention more cost- challenging, as critics frequently have pointed out. We
effective and therefore more attractive for implementa- have for example seen that the method used to elicit
tion. This result supports the view that age weighting is disease weights may influence the morbidity estimates,
likely to favour interventions targeting children. with visual analogue scales (vAS) seemingly weighting
While diarrhoea is a disease group typically affecting conditions as more severe than the other methods.
young children, preventive cardiology is foremost rel- If comparison is made across diseases where different
evant for adults aged 40 and higher. In this case, the CER elicitation techniques for disease severity have been
improved from 55 USD per LY to 46 USD per non-age- used, this may lead to erroneous priority setting deci-
weighted DALY (or QALY) by including disease weights of sions. The direction of this error is, however, depend-
0.268 and 0.490 for stroke and coronary heart disease, ent on the circumstances in each case.
respectively. This improvement is rather large because The important question is not whether these weights
the disease weights for stroke and coronary heart disease and value choices may influence resource allocation.
are quite high and because people who survive such at- Because dearly, making a difference is what they are in-
tacks can be expected to live with a disability for their tended to do. A weighting principle that doesn't create
remaining lives. Inclusion of the morbidity component winners, as well as losers, is inconceivable. One must
therefore makes a larger difference in cardiology than in therefore ask whether the weights brought about by QALYs
diarrhoea, where each case has an expected duration of and DALYs are contributing to improved population health
only 7 days28 and the disease weight is relatively small. and reduced health inequalities in sub-Saharan Africa.
While age weighting made diarrhoea management All the measures—LYs, QALY and DALY—clearly
more attractive, the opposite is the case for preventive lead to more weight being accrued to the young over
cardiology. For this intervention, the positive effect of the elderly compared to a simple measure of mortality
disability weighting is out-crowded by age weighting, like deaths averted. This is in accordance with both an
resulting in a CER of 56 USD per DALY. This is because objective of improved population health and common
cardiovascular patients are relatively old and already notions of equality. The effects on population health and
have passed the "hump" of the age-weight curve and health inequalities of moving from LYs gained to QALY or
are facing declining age weights for the remaining DALY are perhaps less obvious. It might be argued that
life span. This supports the undisputed view that age in settings with extreme resource scarcity with many
weights are disfavouring the elderly. conditions that can be treated cheaply and effectively,
crude estimates of population health are sufficient to
Concluding Remarks inform priority setting. It may further be argued that
more subtle analysis, such as the QALY and DALY, will
In this paper some of the differences between the not give more real information than such crude meas-
health measures life years gained, QALYs and DALYs are ures. Many doubt that it at all is possible to translate
briefly presented. To wrap up, one main difference is fatal and nonfatal health outcomes into one numerical
that whereas LYs gained is a pure measure of mortality unit. How, asks Hilda Bastian, do you decide which
the two latter seek to incorporate morbidity by doing problems and limitations are worse than others, and
disability and quality of life weighting. A major differ- give them a numerical weight?32 Arnesen and Kapiriri
ence between QALY and DALY is that the latter measure worry that valuable epidemiological information may
in addition incorporates age weights. lose importance if it is mixed with value choices of
It has also been discussed what difference these unknown validity.31 Taken together, such concerns cast
alternative measures of health improvement may have doubt on the value of summary health measures like the
for priority setting. Claims that health related quality of QALY and DALY in improving population health.
life weighting has little impact on priority setting have On the other hand, the number of even basic
been briefly presented and discussed. Furthermore, health services that cannot be financed in settings with
worries that DALY weights are unfair and discriminat- extreme resource scarcity is very large. This, I believe,
ing have been briefly presented. Some of the critique increases the importance of developing and applying
is evidently reasonable; in particular there seems to be methodology that is as objective and consistent as pos-
a need to further improve and justify the DALY disease sible. Obviously, there are challenging value choices
weights. Other parts of the critique are in my opinion embedded in these methods, but trade-offs between
I II :III MILLI
quantity for quality of health need to be made in public mortality, it is possible to have a fair amount of resources
health systems, they are in fact being made, and they allocated to rehabilitation of chronic conditions and
will continue to be made, with implicit or explicit marginal patient groups.33
value judgements. The adva :nage of health outcomes Irrespective of these views, in order for the policy
combining mortality and morbidity, like the DALY, is mix to be perceived fair (equitable) as well as effective
that they can be used to inform such trade-offs in a in improving population health within limited budgets,
fairly transparent and consistent way. Many feel that a continuous public scrutiny and evaluation of the
only through quantification of morbidity, as well as applied weights and value choices needs to be made.'
References
1. WHO. World Health Report 2002. Geneva: World 15. Klarman H, Francis J, Rosenthal G. Cost-effectiveness
Health Organization, 2002. analysis applied to the treatment of chronic renal dis-
2. Lopez AD, Ahmad OB, Guillot M, Inoue M, Ferguson ease. Med Care 1968; 6 (1): 48-54.
BD, Salomon JA. Life Tables for 191 countries for 2000: 16. Weinstein MC, Stason WB. Hypertension: A Policy Per-
Data, Methods, Results. Geneva: Evidence and Informa- spective. Cambridge, MA: Harvard University Press, 1976.
tion for Policy (EIP). World Health Organization, 2001. 17. Anand 5, Hanson K. Disability-adjusted life years: a
GPE Discussion Paper No. 90. critical review. J Health Econ 1997; 16 (6): 685-702.
3. WHO. World Health Report 2005. Geneva World 18. Reidpath D, Allotey PA, Kouame A, Cummins RA. Meas-
Health Organization, 2005. uring health in a vacuum: examining the disabilityweight
4. who.int IHomepage on the internet). World Health of the DALY. Health Policy Plan 2003; 18 (4): 351-6.
Organization, 2004. Vaccines, Immunization and 19. Salomon JA, Murray CJL. A multi-method approach to meas-
Biologicals. (Accessed 09.11.2004, at www.who.int/ uring health-state valuations. Health Econ 2004; 13: 281-90.
vaccines-surveillance/StatsAndGraphs.htm). 20. Nord E. Cost-Value Analysis in Health Care. Cambridge:
5. Miller MA, McCann L. Policy analysis of the use of hepa- Cambridge University Press, 1999.
titis B, haemophilus influelzae type B, streptococcus 21. Tsuchiya A. QALYs and ageism: Philosophical theories
pneumoniae-conjugate and rotavirus vaccines in national and age weighting. Health Econ 2000; 9: 57-68.
immunization schedules. Health Econ 2000; 9: 19-35. 22. Tsuchiya A. Age-related preferences and age weighting
6. World Bank. World Development Report 1993: Investing health benefits. Soc Sci Med 1999; 48: 267-76.
in Health. New York Oxford University Press, 1993. 23. Harris J. QALYfying the value of life. J Med Ethics 1987;
7. McKeown T. The direction of medical research. Lancet 13: 119-20.
1979; 2 (8155): 1281-4. 24. Nord E, Street A, Richardson J, Kuhse H, Singer P. The
8. Smith DH, Gravelle H. The practice of discounting in significance of age and duration of effect in social evalua-
economic evaluation of health care interventions. Int J tion of health care. Health Care Anal 1996; 4 (2): 103-11.
Technol Assess Health Care 2001; 17: 236-43. 25. Williams A. Intergenerational equity: An exploration of
9. Walker D, Kumaranayake L. Allowing for differential the "fair innings" argument. Health Econ 1997; 6: 117-32.
timing in cost analyses: discounting and annualization. 26. Cookson R, Dolan P. Principles of justice in health care
Health Policy Plan 2002; 17 (1): 112-18. rationing. J Med Ethics 2000; 26: 323-9.
10. Parfit D. Reasons and Persons. Oxford: Clarendon Press, 27. Bleichrodt H, Doctor J, Stolk E A nonparametric elicita-
1984. tion of the equity-efficiency trade-off in cost-utility an-
11. Viscusi WK. Discounting health effects for medical alysis. J Health Econ 2005; 24 (4): 655-78.
decisions. In: Sloan FA, ed. Valuing Health Care Costs, 28. Robberstad B, Strand T, Black RE, Sommerfelt H. Cost-
Benefits, and Effectiveness of Pharmaceuticals and Other effectiveness of zinc as adjunct therapy for acute child-
Medical Technologies. Cambridge: Cambridge Univer- hood diarrhea in developing countries. Bull WHO 2004;
sity Press, 1995: 125-47. 82 (7): 523-31.
12. Weinstein MC, Stason WB. Foundations of cost- 29. Robberstad B, Hemed Y, Norheim OE Cost-effectiveness of
effectiveness analysis for health and medical practices. N medical interventions to prevent cardiovascular disease in
Engl J Med 1977; 296: 716-21. a sub-Saharan African country-the case of Tanzania. In:
13. Dolan P. The measurement of health-related quality of Robberstad B. Economic evaluation of health interventions
life for use in resource allocation decisions in health care. in sub-Saharan Africa Applied economic evaluations and
In: Culyer AJ, Newhouse JP, eds. Handbook of Health studies on time preferences for health in Tanzania Bergen:
Economics. Amsterdam: Elsevier, 2000: 1723-60. Dissertation for the PhD degree, University of Bergen, 2005.
14. Murray C, Lopez A. The Global Burden of Disease. A 30. Chapman RH, Berger M, Weinstein MC, Weeks JC,
Comprehensive Assessment of Mortality and Disability Goldie 5, Neumann PJ. When does quality-adjusting
from Diseases, Injuries, and Risk Factors in 1990 and Pro- life-years matter in cost-effectiveness analysis? Health
jected to 2020. The Harvard School of Public Health, 1996. Econ 2004; 13: 429-36.
31. Amesen T, Kapiriri L Can value choices in DALYs influence 33. Jelsma J, Weerdt WD, Cock PD. Disability Adjusted Life
global priority setting? Health Policy 2004; 70: 137-49. Years (DALYs) and rehabilitation. Disabil Rehab 2002; 24
32. Bastian H. A consumer trip into the world of the DALY (7): 378-82.
calculations: An Alice in Wonderland experience. Reprod 34. Sen A. Development as Freedom. Oxford: Oxford Uni-
Health Matters 2000; 8 (15): 113-15. versity Press, 1999.
7.4 Microallocation: Decisions at the Bedside
The Allocation of Exotic Medical is at hand. The problem then may be formulated as fol-
lows: how is one to select within the pool of afflicted
Life-Saving Therapy patients the ones to be given the Err treatment in ques-
Nicholas P. Rescher tion; how to select those "whose lives are to be saved"?
Faced with many candidates for an ELT process that can
be made available to only a few, doctors and medical
L The Problem
administrators confront the decision of who is to be
Technological progress has in recent years transformed Oven a chance at survival and who is, in effect, to be
the limits of the possible in medical therapy. However, condemned to die.
the elevated state of sophistication of modern medical As has already been implied, the "heroic" variety of
technology has brought the economists' classic problem spare-part surgery can pretty well be assimilated to this
of scarcity in its wake as an unfortunate side product. paradigm. One can foresee the time when heart trans-
The enormously sophisticated and complex equipment plantation, for example, will have become pretty much
and the highly trained teams of experts requisite for a routine medical procedure, albeit on a very limited
its utilization are scarce resources in relation to poten- basis, since a cardiac surgeon with the technical com-
tial demand. The administrators of the great medical petence to transplant hearts can operate at best a rather
institutions that preside over these scarce resources small number of times each week and the elaborate
thus come to be faced increasingly with the awesome facilities for such operations will most probably exist
choice: Whose life to save? on a modest scale. Moreover, in "spare-part" surgery
A (somewhat hypothetical) paradigm example of there is always the problem of availability of the "spare
this problem may be sketched within the following set parts" themselves. A report in one British newspaper
of definitive assumptions: we suppose that persons in gives the following picture: "Of the 150,000 who die of
some particular medically morbid condition are "mor- heart disease each year [in the UK], Mr Donald Long-
tally afflicted"; it is virtually certain that they will die more, research surgeon at the National Heart Hospital
within a short time period (say, 90 days). We assume [in London] estimates that 22,000 might be eligible for
that some very complex course of treatment (e.g., a heart surgery. Another 30,000 would need heart and
heart transplant) represents a substantial probability of lung transplants. But there are probably only between
life prolongation for persons in this mortally afflicted 7,000 and 14,000 potential donors a year."' Envisaging
condition. We assume that the facilities available in this situation in which at the very most something like
terms of human resources, mechanical instrumental- one in four heart-malfunction victims can be saved, we
ities, and requisite materials (e.g., hearts in the case of clearly confront a problem in ELT allocation.
a heart transplant) make it possible to give a certain A perhaps even more drastic case in point is af-
treatment—this "exotic (medical) lifesaving therapy," forded by long-term hemodialysis, an ongoing process
or ELT for short—to a certain, relatively small number by which a complex device—an "artificial kidney
of people. And finally we assume that a substantially machine"—is used periodically in cases of chronic
greater pool of people in the mortally afflicted condition renal failure to substitute for a non-functional kidney
Source: From Ethics 79,3 (1969), 173-86. Reprinted with permission.

II 1111:111
in "cleaning" potential poisons from the blood. Only a life-or-death situations. This is an issue in which many
few major institutions have chronic hemodialysis units, interested parties have a substantial stake, including
whose complex operation is an extremely expensive the responsible decision-maker who wants to satisfy
proposition. For the present and the foreseeable future his conscience that he is acting in a reasonable way
the situation is that "the number of places available for Moreover, the family and associates of the man who is
chronic hemodialysis is hopelessly inadequate."2 turned away—to say nothing of the man himself—have
The traditional medical ethos has insulated the the right to an acceptable explanation. And indeed even
physician against facing the very existence of this the general public wants to know that what is being
problem. When swearing the Hippocratic Oath, he done is fitting and proper. All of these interested parties
commits himself to work for the benefit of the sick in are entitled to insist that a reasonable code of operating
"whatsover house I enter."3 In taking this stance, the principles provides a defensible rationale for making
physician substantially renounces the explicit choice the life-and-death choices involved in ELT.
of saving certain lives rather than others. Of course,
doctors have always in fact nad to face such choices II.The Two Types of Criteria
on the battlefield or in times of disaster, but there the
issue had to be resolved hurriedly, under pressure, and Two distinguishable types of criteria are bound up in
in circumstances in which the very nature of the case the issue of making ELT choices. We shall call these
effectively precluded calm deliberation by the decision- Criteria of Inclusion and Criteria of Comparison, respect-
maker as well as criticism by others. In sharp contrast, ively. The distinction at issue here requires some ex-
however, cases of the type we have postulated in the planation. We can think of the selection as being made
present discussion arise predictably, and represent by a two-stage process: (1) the selection from among
choices to be made deliberately and "in cold blood." all possible candidates (by a suitable screening process)
It is, to begin with, appropriate to remark that this of a group to be taken under serious consideration as
problem is not fundamentally a medical problem. For candidates for therapy, and then (2) the actual singling
when there are sufficiently many afflicted candidates out, within this group, of the particular individuals to
for ELT then—so we may assume—there will also whom therapy is to be given. Thus the first process nar-
be more than enough for whom the purely medical rows down the range of comparative choice by elimin-
grounds for ELT allocation are decisively strong in any ating en bloc whole categories of potential candidates.
individual case, and just about equally strong through- The second process calls for a more refined, case-by-
out the group. But in this circumstance a selection of case comparison of those candidates that remain. By
some afflicted patients over and against others cannot means of the first set of criteria one forms a selection
ex hypothesi be made on the basis of purely medical group; by means of the second set, an actual selection
considerations. is made within this group.
The selection problem, as we have said, is in sub- Thus what we shall call a "selection system" for
stantial measure not a medical one. It is a problem for the choice of patients to receive therapy of the ELT type
medical men, which must somehow be solved by them, will consist of criteria of these two kinds. Such a system
but that does not make it a medical issue—any more will be acceptable only when the reasonableness of its
than the problem of hospital building is a medical issue. component criteria can be established.
As a problem it belongs to the category of philosophical
problems—specifically a problem of moral philosophy III. Essential Features of an Acceptable
or ethics. Structurally, it bears a substantial kinship with ELT Selection System
those issues in this field that revolve about the notorious
whom-to-save-on-the-lifeboat and whom-to-throw-to- To qualify as reasonable, an ELT selection must meet
the-wolves-pursuing-the-sled questions. But whereas two important "regulative" requirements: it must be
questions of this just-indicated sort are artificial, hypo- simple enough to be readily intelligible, and it must be
thetical, and far-fetched, the ELT issue poses a genuine plausible, that is, patently reasonable in a way that can
policy question for the responsible administrators in be apprehended easily and without involving ramified
medical institutions, indeed a question that threatens to subtleties. Those medical administrators responsible
become commonplace in the fo:1-eseeable future. for ELT choices must follow a modus operandi that vir-
Now what the medical administrator needs to tually all the people involved can readily understand
have, and what the philosopher is presumably ex of- to be acceptable (at a reasonable level of generality, at
ficio in a position to help in providing, is a body of any rate). Appearances are critically important here. It
rational guidelines for making choices in these literally is not enough that the choice be made in a justifiable
Rescher: The Allocation of Exotic Medical Life-Saving Therapy
way, it must be possible for people—plain people—to the institutionalization of medicine, moving it away
"see" (i.e., understand without elaborate teaching or from its original status as an art practised by an indi-
indoctrination) that it is justified, insofar as any mode of vidual practitioner.
procedure can be justified in cases of this sort.
One "constitutive" requirement is obviously an es- B.The Progress-of-Science Factor
sential feature of a reasonable selection system: all of The needs of medical research can provide a second
its component criteria—those of inclusion and those valid principle of inclusion. The research interests of
of comparison alike—must be reasonable in the sense the medical staff in relation to the specific nature of
of being rationally defensible. The ramifications of this the cases at issue is a significant consideration. It may
requirement call for detailed consideration. But one of be important for the progress of medical science—
its aspects should be noted without further ado: it must and thus of potential benefit to many persons in the
be fair—it must treat relevantly like cases alike, leaving future—to determine how effective the ELT at issue is
no room for "influence" or favouritism, etc. with diabetics or persons over 60 or with a negative
RH factor. Considerations of this sort represent another
IV. The Basic Screening Stage: Criteria type of legitimate factor in ELT selection.
of Inclusion (and Exclusion) A very definitely borderline case under this head
would revolve around the question of a patient's will-
Three sorts of considerations are prominent among the ingness to pay, not in monetary terms, but in offering
plausible criteria of inclusion/exclusion at the basic himself as an experimental subject, say by contracting
screening stage: the constituency factor, the progress- to return at designated times for a series of tests sub-
of-science factor, and the prospect-of-success factor. stantially unrelated to his own health, but yielding data
of importance to medical knowledge in general.
A. The Constituency Factor
It is a "fact of life" that ELI can be available only in the C. The Prospect-of-Success Factor
institutional setting of a hospital or medical institute It may be that while the ELT at issue is not without some
or the like. Such institutions generally have normal effectiveness in general, it has been established to be
clientele boundaries. A veterans' hospital will not highly effective only with patients in certain specific cat-
concern itself primarily with treating non-veterans, a egories (e.g., females under 40 of a specific blood type).
children's hospital cannot be expected to accommodate This difference in effectiveness—in the absolute or in
the "senior citizen," an army hospital can regard col- the probability of success—is (we assume) so marked as
lege professors as outside its sphere. Sometimes the to constitute virtually a difference in kind rather than in
boundaries are geographic—a state hospital may admit degree. In this case, it would be perfectly legitimate to
only residents of a certain state. (There are, of course, adopt the general rule of making the ELI at issue avail-
indefensible constituency principles—say race or reliable only or primarily to persons in this substantial-
gion, party membership, or ability to pay; and there are promise-of-success category. (It is on grounds of this
cases of borderline legitimacy, e.g., sex.4) A medical in- sort that young children and persons over 50 are gener-
stitution is justified in considering for ELI only persons ally ruled out as candidates for hemodialysis.)
within its own constituency, provided this constituency We have maintained that the three factors of con-
is constituted upon a defensible basis. Thus the hemo- stituency, progress of science, and prospect of success
dialysis selection committee in Seattle "agreed to con- represent legitimate criteria of inclusion for ELI selec-
sider only those applications who were residents of the tion. But it remains to examine the considerations which
state of Washington. . . . They justified this stand on legitimate them. The legitimating factors are in the final
the grounds that since the basic research . . . had been analysis practical or pragmatic in nature. From the prac-
done at . . . a state-supported institution, the people tical angle it is advantageous—indeed to some extent
whose taxes had paid for the research should be its first necessary—that the arrangements governing medical
beneficiaries."5 institutions should embody certain constituency princi-
While thus insisting that constituency considera- ples. It makes good pragmatic and utilitarian sense that
tions represent a valid and legitimate factor in Eur progress-of-science considerations should be operative
selection, I do feel there is much to be said for mini- here. And, finally, the practical aspect is reinforced by
mizing their role in life-or-death cases. Indeed a refusal a whole host of other considerations—including moral
to recognize them at all is a significant part of medical ones—in supporting the prospect-of-success criterion.
tradition, going back to the very oath of Hippocrates. The workings of each of these factors are of course
They represent a departure from the ideal arising with conditioned by the ever-present element of limited
II all IIILLI
availability. They are operative only in this context, that obviously matters that deserve to be given weight in the
is, prospect of success is a leg:timate consideration at all ELT selection process. Other things being anything like
only because we are dealing with a situation of scarcity. equal, the mother of minor children must take priority
over the middle-aged bachelor.
V.The Final Selection Stage:
Criteria of Selection D.The Potential Future-Contributions Factor
(Prospective Service)
Five sorts of elements must, as we see it, figure primar- In "choosing to save" one life rather than another, "the
ily among the plausible criteria of selection that are to society," through the mediation of the particular med-
be brought to bear in further screening the group con- ical institution in question—which should certainly
stituted after application of the criteria of inclusion: the look upon itself as a trustee for the social interest—is
relative-likelihood-of-success factor, the life-expectancy clearly warranted in considering the likely pattern of
factor, the family-role factor, the potential-contributions future services to be rendered by the patient (adequate
factor, and the services-rendered factor. The first two recovery assumed), considering his age, talent, train-
represent the biomedical aspect, the second three the ing, and past record of performance. In its allocations
social aspect. of ELT, society "invests" a scarce resource in one person
as against another and is thus entitled to look to the
A.The Relative-Likelihood-of-Success Factor probable prospective "return" on its investment.
It is clear that the relative likelihood of success is a It may well be that a thoroughly egalitarian society
legitimate and appropriate factor in making a selec- is reluctant to put someone's social contribution into
tion within the group of qualified patients that are to the scale in situations of the sort at issue. One popular
receive ELT. This is obviously one of the considerations article states that "the most difficult standard would be
that must count very significantly in a reasonable se- the candidate's value to society," and goes on to quote
lection procedure. someone who said: "You can't just pick a brilliant
The present criterion is of course closely related painter over a laborer. The average citizen would be
to item C of the preceding section. There we were quickly eliminated."6 But what if it were not a brilliant
concerned with prospect-of-success considerations painter but a brilliant surgeon or medical researcher
categorically and en bloc. Here at present they come that was at issue? One wonders if the author of the
into play in a particularized case-by-case comparison obiter dictum that one "can't just pick" would still feel
among individuals. If the therapy at issue is not a once- equally sure of his ground. In any case, the fact that the
and-for-all proposition and requires ongoing treatment, standard is difficult to apply is certainly no reason for
cognate considerations must be brought in. not attempting to apply it. The problem of ELT selection
Thus, for example, in the case of a chronic ELT is inevitably burdened with difficult standards.
procedure such as hemodialysis it would clearly make Some might feel that in assessing a patient's value
sense to give priority to patients with a potentially re- to society one should ask not only who if permitted to
versible condition (who would thus need treatment for continue living can make the greatest contribution to
only a fraction of their remaining lives). society in some creative or constructive way, but also
who by dying would leave behind the greatest burden
B.The Life-Expectancy Factor on society in assuming the discharge of their residual
Even if the ELT is "successful" in the patient's case he responsibilities? Certainly the philosophical utilitarian
may, considering his age and/or other aspects of his would give equal weight to both these considerations.
general medical condition, look forward to only a very Just here is where I would part ways with orthodox
short probable future life. This is obviously another utilitarianism. For—though this is not the place to do
factor that must be taken into account. so—I should be prepared to argue that a civilized so-
ciety has an obligation to promote the furtherance of
C.The Family-Role Factor positive achievements in cultural and related areas even
A person's life is a thing of importance not only to if this means the assumption of certain added burdens.'
himself but to others—friends, associates, neigh-
bours, colleagues, etc. But his (or her) relationship E. The Past Services-Rendered Factor (Retrospective
to his immediate family is a thing of unique intimacy Service)
and significance. The nature of his relationship to his A person's services to another person or group have
wife, children, and parents, and the issue of their fi- always been taken to constitute a valid basis for a claim
nancial and psychological dependence upon him, are upon this person or group—of course a moral and
Rescher: The Allocation of Exotic Medical Life-Saving Therapy 395
not necessarily a legal claim. Society's obligation for distasteful problems must be faced, since a failure to
the recognition and reward of services rendered—an choose to save some is tantamount to sentencing all.
obligation whose discharge is also very possibly con- Unpleasant choices are intrinsic to the problem of ELT
ducive to self-interest in the long run—is thus another selection; they are of the very essence of the matter.'°
factor to be taken into account. This should be viewed But is reference to all these factors indeed inevit-
as a morally necessary correlative of the previously able? The justification for taking account of the medical
considered factor of prospective service. It would be factors is pretty obvious. But why should the social
morally indefensible of society in effect to say: "Never aspect of services rendered and to be rendered be taken
mind about services you rendered yesterday—it is into account at all? The answer is that they must be
only the services to be rendered tomorrow that will taken into account not from the medical but from the
count with us today" We live in very future-oriented ethical point of view. Despite disagreement on many
times, constantly preoccupied in a distinctly utilitarian fundamental issues, moral philosophers of the present
way with future satisfactions. And this disinclines us day are pretty well in consensus that the justification of
to give much recognition to past services. But parity human actions is to be sought largely and primarily—
considerations of the sort just adduced indicate that if not exclusively—in the principles of utility and of
such recognition should be given on grounds of equity. justice." But utility requires reference of services to be
No doubt a justification for giving weight to services rendered and justice calls for a recognition of services
rendered can also be attempted along utilitarian lines. that have been rendered. Moral considerations would
("The reward of past services rendered spurs people on thus demand recognition of these two factors. (This,
to greater future efforts and is thus socially advanta- of course, still leaves open the question of whether
geous in the long-run future.") In saying that past servi- the point of view provides a valid basis of action: Why
ces should be counted "on grounds of equity"8—rather base one's actions upon moral principles?—or, to put
than "on grounds of utility"—I take the view that even it bluntly—Why be moral? The present paper is, how-
if this utilitarian defence could somehow be shown ever, hardly the place to grapple with so fundamental
to be fallacious, I should still be prepared to maintain an issue, which has been canvassed in the literature of
the propriety of taking services rendered into account. philosophical ethics since Plato.)
The position does not rest on a utilitarian basis and so
would not collapse with the removal of such a basis.9 VI. More Than Medical Issues Are Involved
As we have said, these five factors fall into three
groups: the biomedical factors A and B, the familial An active controversy has of late sprung up in med-
factor c, and the social factors D and E. With items A ical circles over the question of whether non-physician
and B the need for a detailed analysis of the medical laymen should be given a role in ELT selection (in
considerations comes to the fore. The age of the patient, the specific context of chronic hemodialysis). One
his medical history, his physical and psychological con- physician writes: "I think that the assessment of the
dition, his specific disease, etc., will all need to be taken candidates should be made by a senior doctor on the
into exact account. These biomedical factors represent [dialysis] unit, but I am sure that it would be helpful
technical issues: they call for the physicians' expert to him—both in sharing responsibility and in avoiding
judgment and the medical statisticians' hard data. And personal pressure—if a small unnamed group of people
they are ethically uncontroversial factors—their legit- [presumably including laymen] officially made the final
imacy and appropriateness are evident from the very decision. I visualize the doctor bringing the data to the
nature of the case. group, explaining the points in relation to each case,
Greater problems arise with the familial and social and obtaining their approval of his order of priority.',12
factors. They involve intangibles that are difficult to Essentially this procedure of a selection commit-
judge. How is one to develop subcriteria for weighing tee of laymen has for some years been in use in one
the relative social contributions of (say) an architect or of the most publicized chronic dialysis units, that of
a librarian or a mother of young children? And they in- the Swedish Hospital of Seattle, Washington.° Many
volve highly problematic issues. (For example, should physicians are apparently reluctant to see the choice
good moral character be rated a plus and bad a minus of allocation of medical therapy pass out of strictly
in judging services rendered?) And there is something medical hands. Thus in a recent symposium on the
strikingly unpleasant in grappling with issues of this sort "Selection of Patients for Haemodialysis,"14 Dr Ralph
for people brought up in times greatly inclined toward Shakman writes: "Who is to implement the selection?
maxims of the type "Judge not!" and "Live and let live!" In my opinion it must ultimately be the responsibility
All the same, in the situation that concerns us here such of the consultants in charge of the renal units. . . . I can
II j11 IIILLI
see no reason for delegating this responsibility to lay resolved. The instrumentalities of ELr have been cre-
persons. Surely the latter would be better employed ated through the social investment of scarce resources,
if they could be persuaded to devote their time and and the interests of the society deserve to play a role
energy to raise more and more money for us to spend in their utilization. As representatives of their social
on our patients."5 Other contributors to this sym- interests, lay opinions should function to complement
posium strike much the same note. Dr F.M. Parsons and supplement medical views once the proper arena
writes: "In an attempt to overcome . . . difficulties in of medical considerations is left behind.'8 Those phys-
selection some have advocated introducing certain icians who have urged the presence of lay members on
specified lay people into the discussions. Is it wise? selection panels can, from this point of view, be recog-
I doubt whether a committee of this type can adjudi- nized as having seen the issue in proper perspective.
cate as satisfactorily as two medical colleagues, particu- One physician has argued against lay representa-
larly as successful therapy involves close cooperation tion on selection panels for hemodialysis as follows: "If
between doctor and patient."m And Dr M.A. Wilson the doctor advises dialysis and the lay panel refuses,
writes in the same symposium: "The suggestion has the patient will regard this as a death sentence passed
been made that lay panels should select individuals for by an anonymous court from which he has no right of
dialysis from among a group who are medically suit- appeal."19 But this drawback is not specific to the use
able. Though this would relieve the doctor-in-charge of a lay panel. Rather, it is a feature inherent in every
of a heavy load of responsibility, it would place the selection procedure, regardless of whether the selec-
burden on those who have no personal knowledge tion is done by the head doctor of the unit, by a panel
and have to base their judgments on medical or social of physicians, etc. No matter who does the selecting
reports. I do not believe this would result in better among patients recommended for dialysis, the feelings
decisions for the group or improve the doctor—patient of the patient who has been rejected (and knows it)
relationship in individual cases."" can be expected to be much the same, provided that he
But no amount o f flag waving about the doctor's recognizes the actual nature of the choice (and is not
facing up to his responsibility—or prostrations before deceived by the possibly convenient but ultimately poi-
the idol of the doctor—patient relationship and reluc- sonous fiction that because the selection was made by
tance to admit laymen into the sacred precincts of the physicians it was made entirely on medical grounds).
conference chambers of medical consultations—can In summary, then, the question of ELr selection
obscure the essential fact that ELT selection is not a would appear to be one that is in its very nature heav-
wholly medical problem. When there are more than ily laden with issues of medical research, practice, and
enough places in an ELT program to accommodate all administration. But it will not be a question that can
who need it, then it will clearly be a medical question be resolved on solely medical grounds. Strictly social
to decide who does have the need and which among issues of justice and utility will invariably arise in this
these would successfully respond. But when an admit- area—questions going outside the medical area in
ted gross insufficiency of places exists, when there are whose resolution medical laymen can and should play
10 or 50 or 100 highly eligible candidates for each place a substantial role.
in the program, then it is unrealistic to take the view
that purely medical criteria can furnish a sufficient basis VII. The Inherent Imperfection (Non-Optimality)
for selection. The question o ELT selection becomes of Any Selection System
serious as a phenomenon of scale—because, as more
candidates present themselves, strictly medical factors Our discussion to this point of the design of a selec-
are increasingly less adequate as a selection criterion tion system for ELT has left a gap that is a very
precisely because by numerical category-crowding there fundamental and serious omission. We have argued
will be more and more cases whose "status is much the that five factors must be taken into substantial and
same" so far as purely medical considerations go. explicit account:
The ELI- selection problem clearly poses issues that
transcend the medical sphere because—in the nature a. Relative likelihood of success. Is the chance of the
of the case—many residual issues remain to be dealt treatment's being "successful" to be rated as high,
with once all of the medical questions have been faced. good, average, etc.?2°
Because of this there is good reason why laymen as b. Expectancy of future life. Assuming the "success" of
well as physicians should be involved in the selection the treatment, how much longer does the patient
process. Once the medical considerations have been stand a good chance (75 per cent or better) of
brought to bear, fundamental social issues remain to be living—considering his age and general condition?
c. Family role. To what extent does the patient have basis of any general abstract considerations of reason-
responsibilities to others in his immediate family? ableness. Within broad limits, a variety of resolutions
d. Social contributions rendered. Are the patient's past are all perfectly acceptable—so that no one procedure
services to his society outstanding, substantial, can justifiably be regarded as "the (uniquely) best" be-
average, etc.? cause it is superior to all others.21
e. Social contributions to be rendered. Considering his
age, talents, training, and past record of perform- VIII.A Possible Basis for a Reasonable
ance, is there a substantial probability that the Selection System
patient will—adequate recovery being assumed—
render in the future services to his society that can Having said that there is no such thing as the optimal selec-
be characterized as outstanding, substantial, aver- tion system for ELT, I want now to sketch out the broad
age, etc.? features of what I would regard as one acceptable system.
The basis for the system would be a point rating.
This list is clearly insufficient for the construction of a The scoring here at issue would give roughly equal
reasonable selection system, since that would require weight to the medical considerations (A and B) in
not only that these factors be taken into account (some- comparison with the extramedical considerations
how or other), but—going beyond this—would specify (C = family role, D = services rendered, and E = services
a specific set of procedures for taking account of them. to be rendered), also giving roughly equal weight to the
The specific procedures that would constitute such a three items involved here (C, D, and E). The result of
system would have to take account of the interrelation- such a scoring procedure would provide the essential
ship of these factors (e.g., B and E), and to set out exact starting point of our ELT selection mechanism. I delib-
guidelines as to the relevant weight that is to be given erately say "starting point" because it seems to me that
to each of them. This is something our discussion has one should not follow the results of this scoring in an
not as yet considered. automatic way. I would propose that the actual selection
In fact, I should want to maintain that there is no should only be guided but not actually be dictated by
such thing here as a single rationally superior selection this scoring procedure, along lines now to be explained.
system. The position of affairs seems to me to be some-
thing like this: (1) It is necessary (for reasons already IX.The Desirability of Introducing an
canvassed) to have a system, and to have a system that Element of Chance
is rationally defensible, and (2) to be rationally defens-
ible, this system must take the factors A—E into sub- The detailed procedure I would propose—not of course
stantial and explicit account. But (3) the exact manner as optimal (for reasons we have seen), but as eminently
in which a rationally defensible system takes account acceptable—would combine the scoring procedure just
of these factors cannot be fixed in any one specific discussed with an element of chance. The resulting se-
way on the basis of general considerations. Any of the lection system would function as follows:
variety of ways that give A—E "their due" will be accept-
able and viable. One cannot hope to find within this 1. First the criteria of inclusion of Section iv above
range of workable systems some one that is optimal in would be applied to constitute a first-phase selection
relation to the alternatives. There is no one system that group—which (we shall suppose) is substantially
does "the (uniquely) best"—only a variety of systems larger than the number n of persons who can ac-
that do "as well as one can expect to do" in cases of tually be accommodated with ELT.
this sort. 2. Next the criteria of selection of Section v are brought
The situation is structurally very much akin to to bear via a scoring procedure of the type described in
that of rules of partition of an estate among the rela- Section vm. On this basis a second-phase selection group
tions of a decedent. It is important that there be such is constituted which is only somewhat larger—say by a
rules. And it is reasonable that spouse, children, par- third or a half—than the criticalnumber n at issue.
ents, siblings, etc., be taken account of in these rules. 3. If this second-phase selection group is relatively
But the question of the exact method of division—say homogeneous as regards rating by the scoring
that when the decedent has neither living spouse nor procedure—that is, if there are no really major dis-
living children then his estate is to be divided, divid- parities within this group (as would be likely if the
ing 60 per cent between parents, 40 per cent between initial group was significantly larger than n)—then
siblings versus dividing 90 per cent between parents, the final selection is made by random selection of n
10 per cent between siblings—cannot be settled on the persons from within this group.
II 311 111111
This introduction of the element of chance—in what handling of like cases over the widest possible area that
could be dramatized as a "lottery of life and death"— seems reasonable in the circumstances.
must be justified. The fact is that such a procedure Third (and perhaps least), such a recourse to
would bring with it three substantial advantages. random selection does much to relieve the administra-
First, as we have argued above (in Section vii), any tors of the selection system of the awesome burden of
acceptable selection system is inherently non-optimal. ultimate and absolute responsibility.
The introduction of the element of chance prevents These three considerations would seem to build up a
the results that life-and-death choices are made by the substantial case for introducing the element of chance into
automatic application of an ac mittedly imperfect selec- the mechanism of the system for ELT selection in a way
tion method. limited and circumscribed by other weightier consider-
Second, a recourse to chance would doubtless make ations, along some such lines as those set forth above.22
matters easier for the rejected patient and those who have It should be recognized that this injection of
a specific interest in him. It would surely be quite hard man-made chance supplements the element of natural
for them to accept his exclusion by relatively mechanical chance that is present inevitably and in any case (apart
application of objective criteria in whose implementation from the role of chance in singling out certain persons
subjective judgment is involved. But the circumstances as victims for the affliction at issue). As F.M. Parsons
of life have conditioned us to accept the workings of has observed: "any vacancies [in an ELT program—
chance and to tolerate the element of luck (good or bad): specifically hemodialysis] will be filled immediately
human life is an inherently contingent process. Nobody, by the first suitable patients, even though their claims
after all, has an absolute right to ELT—but most of us for therapy may subsequently prove less than those of
would feel that we have "every bit as much right" to it other patients refused later."23 Life is a chancy business
as anyone else in significantly similar circumstances. and even the most rational of human arrangements can
The introduction of the element of chance assures a like cover this over to a very limited extent at best.
Notes
1. Christine Doyle, (1967), "Spate-Part Heart Surgeons," in 6. Lawrence Lader, (1968), "Who Has the Right To Live?," in
the Observer (12 May), 623. Good Housekeeping (Jan.): 144.
2. J.D.N. Nabarro, "Selection of Patients for Haemodialysis," 7. This approach could thus be continued to embrace the
in British Medical Journal, There are about 30 new cases previous factor, that of family role, the preceding item.
per million of population—only 10 per cent of these can 8. Moreover a doctrinaire utilitarian would presumably be
for the foreseeable future be accommodated with chronic willing to withdraw a continuing mode of ELT such as
hemodialysis. Kidney transplantation—itself a very tricky hemodialysis from a patient to make room for a more
procedure—cannot make a more than minor contribution promising candidate who came to view at a later stage and
here. As this article goes to press, I learn that patients who could not otherwise be accommodated. I should be
can be maintained in home didysis at an operating cost unwilling to adopt this course, partly on grounds of utility
about half that of maintaining them in a hospital dialysis (with a view to the demoralization of insecurity), partly on
unit (roughly an $8,000 minimum). In the United States, the non-utilitarian ground that a "moral commitment" has
around 7,000 patients with terminal uremia who could been made and must be honoured.
benefit from hemodialysis evolve yearly. As of mid-1968, 9. Of course the difficult question remains of the relative
some 1,000 of these can be accommodated in existing weight that should be given to prospective and
hospital units. By June 1967, a world-wide total of some retrospective service in cases where these factors conflict.
120 patients were in treatment by home dialysis. (Data There is good reason to treat them on a par.
from a forthcoming paper, "Home Dialysis," by C.M. Conty 10. This in the symposium on "Selection of Patients for
and H.V. Murdaugh. See also R.A. Baillod, et al., (1965), Haemodialysis," in British Medical Journal (11 March
"Overnight Haemodialysis in the Home," in Proceedings of 1967): 622-4. F.M. Parsons writes: "But other forms of
the European Dialysis and Transplant Association 6: 99ff. selecting patients [distinct from first come, first served]
3. For the Hippocratic Oath, see .,Iippocrates: Works (Loeb are suspect in my view if they imply evaluation of man
ed.; London, 1959), I, 298. by man. What criteria could be used? Who could justify
4. Another example of borderline legitimacy is posed by an a claim that the life of a mayor would be more valuable
endowment "with strings attacned," e.g., "In accepting this than that of the humblest citizen of his borough? Whatever
legacy the hospital agrees to admit and provide all needed we may think as individuals none of us is indispensable."
treatment for any direct descer.dant of myself, its founder." But having just set out this hard-line view he immediately
5. Shana Alexander, (1962), "They Decide Who Lives, Who backs away from it: "On the other hand, to assume that
Dies," in life, 53 (9 Nov.), 10/-25 (see 107). there was little to choose between Alexander Fleming and
Adolf Hitler . . . would be nonsense, and we should be factor here The man who gives up takes not his life alone
naive if we were to pretend that we could not be influenced but (figuratively speaking) also that of the person he
by their achievements and characters if we had to choose replaced in the treatment schedule.
between the two of them. Whether we like it or not we 21. To say that acceptable solutions can range over broad
cannot escape the fact that this kind of selection for long- limits is not to say that there are no limits at all. It is an
term haemodialysis will be required until very large sums obviously intriguing and fundamental problem to raise the
of money become available for equipment and services [so question of the factors that set these limits. This complex
that everyone who needs treatment can be accommodated)." issue cannot be dealt with adequately here. Suffice it to
11. The relative fundamentality of these principles is, however, say that considerations regarding precedent and peopl6
a substantially disputed issue. expectations, factors of social utility, and matters of
12. J.D.N. Nabarro, op. cit., 622. fairness and sense of justice all come into play.
13. See Shana Alexander, op. cit. 22. One writer has mooted the suggestion that: "Perhaps the right
14. British Medical Journal (11 March 1967): 622-4. thing to do, difficult as it may be to accept, is to select [for
15. Ibid., 624. Another contributor writes in the same hemodialysis) from among the medical and psychologically
symposium: "The selection of the few Ito receive qualified patients on a strictly random basis" (S. Gorovitz,
hemodialysis) is proving very difficult—a true 'Doctor's 1966, "Ethics and the Allocation of Medical Resources," in
Dilemma'—for almost everybody would agree that this must Medical Research Engineering 5: 7). Outright random selection
be a medical decision, preferably reached by consultation would, however, seem indefensible because of its refusal to
among colleagues" (Dr F.M. Parsons, ibid., 623). give weight to considerations which, under the circumstances,
16. "The Selection of Patients for Haemodialysis," op. cit. (note deserve to be given weight. The proposed procedure of
10 above), 623. superimposing a certain degree of randomness upon the
17. Dr Wilson's article concludes with the perplexing rational-choice criteria would seem to combine the advantages
suggestion—wildly beside the point given the structure of of the two without importing the worst defects of either.
the situation at issue—that "the final decision will be made 23. "Selection of Patients for Haemodialysis," op. cit., 623. The
by the patient." But this contention is only marginally question of whether a patient for chronic treatment should
more ludicrous than Parson's contention that in selecting ever be terminated from the program (say, if he contracts
patients for hemodialysis "gainful employment in a well cancer) poses a variety of difficult ethical problems with
chosen occupation is necessary to achieve the best results" which we need not at present concern ourselves. But it
since "only the minority wish to live on charity" (ibid.). does seem plausible to take the (somewhat anti-utilitarian)
18. To say this is of course not to deny that such questions view that a patient should not be terminated simply
of applied medical ethics will invariably involve a because a "better qualified" patient comes along later on. It
host of medical considerations; it is only to insist that would seem that a quasi-contractual relationship has been
extramedical considerations will also invariably be at issue. created through established expectations and reciprocal
19. M.A. Wilson, "Selection of Patients for Haemodialysis," op. understandings, and that the situation is in this regard
cit., 624. akin to that of the man who, having under-taken to sell
20. In the case of an ongoing treatment involving complex his house to one buyer, cannot afterward unilaterally undo
procedure and dietary and other mode-of-life this arrangement to sell it to a higher bidder who "needs it
restrictions—and chronic hemodialysis definitely falls worse" (thus maximizing the overall utility).
into this category—the patient's psychological makeup, 24. I acknowledge with thanks the help of Miss Hazel
his willpower to "stick with it" in the face of substantial Johnson, reference librarian at the University of Pittsburgh
discouragements, will obviously also be a substantial Library, in connection with the bibliography.
Just Caring: In Defense of Limited Daniel Callahan' and the other by Norman Daniels.'
These books ignited a firestorm of criticism,3 best cap-
Age-Based Healthcare Rationing tured in the claim that any form of age-based healthcare
Leonard M. Fleck rationing was fundamentally ageist, discriminatory
in a morally objectionable sense. That is, the elderly
Health Law had equal moral worth and an equal right to life as the
nonelderly. If an elderly and nonelderly person each
The debate around age-based healthcare rationing was had essentially the same medical problem requiring the
precipitated by two books in the late 1980s, one by same medical treatment, then they had an equal right
Source: From Cambridge Quarterly of Healthcare Ethics 10, 1 (2010), 27-37. Reprinted with permission.
Y 111 II 11 II 4111 HILL1
to receive that treatment no matter what the cost of that of life-prolonging resources they would allocate to
treatment. Alternatively, if cost was an issue because the their future possible elderly selves (especially if only
benefits of the treatment were too marginal, then both very marginal benefits could be achieved regarding
the elderly and nonelderly patients requiring that treat- length or quality of life) and allocate those resources
ment ought to be denied it. If there were something to younger stages of their life in order to diminish their
absolutely scarce about the treatment, then some fair risks of premature death or disability. This strategy was
process would have to be used to make an allocation intended to defuse the criticism that his views were
decision (and that fair process could not use some age ageist, but it is a very broad general argument.
cutoff among the allocation criteria). John Harris has introduced into the literature the
"fair innings" argument.4 The general idea is that in-
The Beginning of the Age-Based dividuals who have achieved a normal life span have
Rationing Controversy had their "fair innings," and, consequently, they ought
not to demand extra innings, especially if making such
Callahan and Daniels may have opened themselves up demands denied younger individuals the opportunity
to these criticisms because their defense of age-based to have their fair innings This broad argument has
rationing was too broad arid unqualified. Callahan, intuitive moral legitimacy for many.5 However, Harris
for example, can be fairly read as saying that beyond himself is a critic of this argument (and its ageist im-
some age (75-80) individuals ought to be denied very plications) except in a couple of very narrowly framed
expensive life-prolonging medical care. He appealed to circumstances 6
the notion of a "natural life span" as a reference point
for justifying this admonition. Once individuals had My Key Claim
had the opportunity to achieve what it was reasonable
to achieve over a natural life span, they no longer had a In this essay I want to defend a "limited" form of age-
strong moral claim to expensive life-prolonging resour- based rationing. One implication of this claim is that
ces, especially if such claims threatened the financial I do not regard age-based rationing as intrinsically
resources needed to afford younger members of society morally objectionable (unjust). There are some cir-
the same opportunities already enjoyed by the now cumstances in which an age-based rationing decision
elderly. Callahan's hope was that a cultural shift in social would be unjust but other circumstances in which
expectations would freely accomplish this result rather such a decision would be just, or at least "just enough."
than coercive government policies. Callahan did not In other words, it will sometimes be true that it will
try to make subtle distinctions among life-threatening be morally permissible to deny an elderly individual
circumstances faced by the elderly. Two 80-year-olds a costly life-prolonging intervention that we would be
with the same life-threatening cardiac condition and morally obligated to provide a nonelderly individual
surgical need might have radically different prospects in those same medical circumstances, and the primary
for long-term survival (1 year vs. 10 years), but neither reason for that denial will be the age of that individual as
would have a stronger claim to those medical resources opposed to the probability of success of that intervention, or
because both had exceeded a natural life span. degree of effectiveness, or comorbidities, or any other mor-
Daniels offered a very different (but very general) ally and medically relevant considerations.
argument in defense of age-based rationing, the "pru- If age-based healthcare rationing has such a mor-
dential life span account." What he wanted to avoid ally obnoxious quality about it (but we need to find ef-
was a situation in which the healthcare needs of the fective ways to control escalating healthcare costs), why
young were pitted against the healthcare needs of the should we not identify other approaches to rationing
elderly. He asks us to imagine ourselves (in a Rawlsian- justly that avoid using any age-based reference points?
like thought experiment) as prudent young adults who The short answer I would offer to that question is that
have to allocate limited healthcare resources across our we would have to tolerate significant injustices in our
entire life span without knowing what specific health healthcare system if we avoided altogether some forms
problems we might encounter that would threaten us of age-based rationing.
with premature death unless we had assured access to
some expensive lifesaving intervention. What he rea- Social Origins of the Age-Based Rationing
sonably assumes is that prudent individuals (in this Problem
state of healthcare ignorance) would want to maximize
the likelihood of their achieving old age, and, con- In brief, here are the social, political, and economic
sequently, they would reduce substantially the amount circumstances that prompted Callahan and Daniels
Fleck: Just Caring: In Defense of Limited Age-Based Healthcare Rationing
to write their books in the late 1980s. First, the cur- a threat to the just healthcare claims of the nonelderly
rent elderly population in the United States represents because so little of the costly high-tech medicine we
about 13% of the overall population, but approxi- take for granted today had been invented or dissemin-
mately 35% of all health dollars spent in the United ated. But that dissemination has taken place over the
States are devoted to meeting the healthcare needs of past 40 years. The overall result is that the elderly are
the elderly. The statistical fact is that, as things are now, living longer with a greater burden of chronic illness
the elderly consume a hugely disproportional share of that requires constant costly medical assessment and
health resources compared to the nonelderly. management.11)
Second, in the United States and other advanced The Medicare and Medicaid programs (the latter
nations in the West, we are just beginning to experience covering long-term care for the impoverished elderly)
the post—World War II "baby boom" reaching retire- represent public expenditures, which are adding to the
ment age. In the United States this is a cohort of about current and projected deficit problems of the United
78 million individuals who will raise the fraction of our States. As I write we are in the middle of wrestling
population over age 65 to about 21% in the year 2030.7 with health reform, the hope being that virtually all the
If that future generation were to consume healthcare currently uninsured in the United States (47 million)
resources at the same level as the current generation would be provided with a substantial package of health
of the elderly, we would have extraordinarily difficult benefits, equal at least to what we provide Medicare
moral and political and economic challenges in meet- beneficiaries. But the total cost of the package to the fed-
ing those demands without sacrificing the just claims eral government is now estimated at $1.2 trillion over a
(of many sorts) of the nonelderly. 10-year period. It is unthinkable that these costs would
But, third, it is very unlikely that those future simply be added to the federal deficit. That means sav-
demands would be so restrained because of the con- ings have to be achieved elsewhere (read: Medicare)
tinued proliferation of increasingly expensive medical from other federal healthcare expenditures in order
technology throughout our healthcare system (as well to cover at least part of those costs. Critics will be
as the healthcare systems of other advanced nations).8 prompted to ask whether that is fair to the elderly. The
The likelihood is that per capita healthcare spend- obvious retort is to ask whether the status quo (47 mil-
ing on the elderly will increase very significantly and lion uninsured) is fair to the nonelderly.
make even more difficult the resulting moral and pol- To a large extent the elderly have access to what-
itical problems. In 1993 the Medicare program for the ever the healthcare system has available for meeting
elderly (which only covers about half the healthcare their health problems, no matter how costly, no matter
expenses of the elderly) spent $150 billion. In 2009 how marginally beneficial." The relatively well-off
the projections are for $503 billion, and in 2018 the elderly can demand these extraordinarily expensive
projections are for $932 billion (still 12 years out from cancer drugs that yield only extra weeks or months of
having the entire baby boom generation reach age 65).9 life whereas the uninsured nonelderly with early-stage
The conclusion I would draw from these three points cancers will have no assured access to less expensive,
is that whatever the data were that prompted Callahan more effective cancer treatments that can either save
and Daniels to propose age-based rationing in the late their lives or prolong them for very long periods of
1980s, those data have become substantially worse time. Many of these individuals will unnecessarily die
since then, thereby justifying examining age-based ra- prematurely.' A number of recent medical journal
tioning again. articles have been discussing a relatively dramatic in-
crease in the number of individuals in their 80s and
Medicare and Its Unintended Consequences 90s who are having complicated surgical procedures
for their heart problems." The logic behind these ef-
There were good reasons for creating the Medicare pro- forts is nicely captured by Shim:
gram in the United States in 1965. Relatively few of the Together, the hope of medical intervention, the
elderly then could afford to purchase health insurance difficulty of prognostication and the actual gains of life
from private insurance companies after they had retired that a great many cardiac procedures confer, contribute
(if insurance would even be offered to them) because to the ethical viability and imperative of treating risk.
they were seen as being at high risk for serious health For practitioners and patients alike, understandings of
problems. This was properly seen as a justice issue the mandate to treat at ever-older ages contribute, prag-
corrected by the creation of the Medicare program. matically, to the elimination of any significant delibera-
In short, what we created was national health insur- tion about whether or not to treat. Instead, standard
ance for the elderly. At the time this did not represent practice replaces choice."
II JII 111111
The point is that the availability of the technology that might cost $100,000 for a course of treatment that
for prolonging life in the very old, together with assured could yield only a few extra months of life.
funding for its use in the very old, and the perception Yet another argument invoked in these circum-
of a likely net benefit create a clinical and (unreflective) stances is that denying the terminally ill elderly
moral imperative for its use that undermines our soci- life-prolonging medical resources is just a form of
ety's capacity (and willingness) to meet the just claims involuntary euthanasia. For those who propound this
of the uninsured nonelderly. These are the clinical and line of argument the offering of hospice instead of the
social facts that generate a need for a renewed moral life-prolonging care these patients want is simply a
appraisal of the legitimacy of some forms of age-based more socially genteel form of involuntary euthanasia.
healthcare rationing. The alternative is the continued Finally, John Harris defends what he calls an "age
toleration of the multiple, diffuse, invisible injustices indifference" principle as a reference point for con-
that are inflicted on the nonelderly regarding their demning virtually all forms of age-based rationing.
healthcare needs.' He writes, "An individual's entitlement to the concern,
respect, and protection of the community, as expressed
Arguments against Age-Based Rationing in its public health care system does not vary with age
or life expectancy. I believe this principle is the only
Let us briefly rehearse the major moral arguments that one consistent with justice." Harris then adds as a sup-
speak against the legitimacy of any form of age-based portive argument, "Those who believe in discrimin-
rationing. Baruch Brody cor tends that allowing any ating in favor of the young or against the old must
elderly individual to die when we have the capacity believe that in so far as murder is an injustice, it is less
to prolong that individual's life (and that individual of an injustice to murder the old than the young." But
wants that intervention) represents a failure to fulfill he adds that our common law tradition does not see
our "duty to rescue."16 He gives special moral weight things that way. Murder is just as bad, whether an in-
to the fact that we are talking about identifiable indi- dividual is denied 40 additional years of life or a single
viduals who are patients of physicians (as opposed to hour of life.''
statistical lives that might be sacrificed as a result of a
social nonfunding decision). We will spend millions In Defense of Limited Age-Based Rationing
of dollars to try to save the life of some hapless hiker
in the Arctic wilderness, and, therefore, according to What we need to notice with regard to the above argu-
Brody, we have at least as strong an obligation to save ments is that they are perfectly general. They are not
older patients who are conveniently at hand in hospital aimed at any particular types or circumstances in which
beds. That duty to rescue does not have an age cutoff, age-based rationing might be considered. All age-based
which would represent a failure to accord equal respect rationing is judged to be fundamentally flawed. Here,
to older individuals whose lives were in peril. however, are some examples that ought to precipitate
Someone with a Rawlsian bent might be inclined some critical reflection.
to argue that the terminally ill elderly are among the We should start by taking note of the fact that
"medically least well-off," and consequently, justice re- subtle age rationing is already part of accepted (and
quires that they be provided with more in the way of morally defensible) medical practice. We typically
medical resources, not less (as would be characteristic limit lots of screening tests by age, such as prostate
of age-based rationing decisions). cancer screening. We stop doing screening mammo-
Again, with reference to the terminally ill elderly grams beyond age 70. We focus certain public health
someone could invoke a "last chance therapy" argu- programs, such as antismoking campaigns, on the
ment: It is one thing to insist that someone with a relatively young, believing (correctly) that we are much
life-threatening medical need accept a less expensive more likely to achieve substantial positive results there
effective intervention for their medical problem rather than in someone with a 30-year habit of smoking.18 We
than a more expensive medical intervention that might use 10-year duration hip prostheses in patients over
be only marginally better. But when someone has only age 80 (as opposed to much more costly longer dur-
one therapeutic alternative or death, then we are mor- ation prostheses) when it is unlikely those patients will
ally obligated to provide that alternative no matter what survive more than 10 years due to assorted comorbid-
the cost. The suppressed premise in this argument is ities. This is morally warranted so long as the resources
that human life is priceless (no matter what the age saved are reallocated to meet higher priority healthcare
of a person), and, hence, it is morally wrong to deny needs of the elderly. Still, my critic might argue that
individuals with a terminal cancer these cancer drugs these examples are not persuasive enough because they
Fleck: Just Caring: In Defense of Limited Age-Based Healthcare Rationing 403
arenot weighty enough; age-based rationing matters or Hunter's syndrome. These are metabolic disorders
most when life itself is at stake. So we shall consider that will result in very premature death. Today we have
some of those cases. the drugs imiglucerase and idursulfase, each of which
costs $300,000 per person per year. But they are effect-
The DeBakey Case ive in preventing very premature death for these young
First, consider the case of Dr. Michael DeBakey, the afflicted individuals. These are last chance therapies;
famed cardiac surgeon who pioneered a number of im- there are no alternatives. Further, these children can be
portant surgical procedures. He diagnosed himself with regarded as being among the medically least well-off.
an aortic aneurysm at age 98. He knew that in theory it What makes the critical moral difference in these cases
was surgically repairable, but he feared the brain injury is that these interventions are very effective in sustain-
such surgery could occasion. His family and colleagues ing both length and quality of life, and these individ-
convinced him to have the surgery. The surgery was uals are very far from having had their fair innings. The
successful, but Dr. DeBakey spent eight months in same cannot be said for Dr. DeBakey.
the hospital at a cost of more than $1 million (and he My critic might respond that this is not really a
survived for a year after that). Our question is: Would matter of age-rationing rather, what matters is whether
Dr. DeBakey have been treated unjustly if he had been a life-sustaining intervention is going to be sufficiently
denied that surgery, largely because of his advanced age effective in a very old individual. So the proverbial
and the likely excess costs associated with a prolonged vigorous elderly individual (Hazel Homer [age 99]
hospitalization? in note 13) who gets 5 extra years of life from an ex-
Several of the critical arguments advanced above pensive cardiac intervention has a just claim because
could be invoked to justify providing Dr. DeBakey with that intervention proved to be so effective. However, I
the surgery. That is, someone could argue there was a would argue that is too quick a moral judgment. What
duty to rescue, or that this was a last chance interven- can be argued instead is that there are compelling con-
tion, or that Dr. DeBakey was among the medically siderations of compassion that would warrant giving
least well. However, the very generality of each of these Hazel her bimodal cardiac resynchronization device
rationales has the morally problematic implication of at $60,000. Compassion must not be confused with
justifying virtually every costly marginally beneficial life- justice, nor will compassion trump justice if a compas-
prolonging intervention for every patient on Medicare, sionate response to the health needs of the very old
thereby adding to the problem of escalating health costs threatens to fail to meet the just health claims of the
and diminishing the likelihood of health reform that relatively young. We can illustrate this last point with
would provide health insurance for America's uninsured. our second critical example.
Invoking the "duty to rescue" argument makes
moral sense when the numbers are small and the cir- The Case of the Artificial Heart
cumstances rare, even if the costs per rescue are extra- In the United States we are currently clinically testing
ordinarily high. But in the world of life-prolonging totally implantable artificial hearts [Timis]. These de-
medical technology we have today, the opportunities vices may be regarded as last chance therapies for pa-
for rescue are ubiquitous among the elderly, and the tients in end-stage heart failure. Each year in the United
likelihood is that a majority of Medicare recipients States 550,000 individuals enter early-stage heart fail-
would need to be rescued at great expense several times ure. It is expected that improved models of this device
before disease overmastered them. This is a radically will yield 5 extra years of life expectancy on average
different situation from the "usual" circumstances in for patients who would otherwise die of heart failure.
which we invoke the "duty to rescue." The command Computer modeling suggests that we could implant as
of public resources by the Medicare elderly, especially many as 350,000 of these devices per year at a cost
under this alleged ethical obligation, is very much a of $300,000 each. That would add $105 billion to the
threat to the just claims of the uninsured and margin- cost of healthcare in the United States, roughly 70% of
ally insured nonelderly. which would be added to the cost of Medicare. About
A similar analysis will apply if the ethical argument 100,000 of these devices would be needed by those
involves an appeal to last chance therapies or to the below age 70. If the societal decision were made that
status of being medically least well-off. There is a range we could afford only 100,000 of these devices, do we
of circumstances in which these moral considerations want to argue that they ought to be distributed by lot-
will be legitimate and determinative of a just outcome. tery over the elderly and nonelderly (so long as minimal
For example, a relatively small number of children are criteria for likelihood of success are met)? Or would we
born each year in the United States with either Gaucher's have good reasons for saying that the nonelderly had
N NI ill IIILL1
the stronger just claims because they had not yet had age indifference would be a moral requirement. The
their fair innings?" Should we be morally comfortable 60-year-old and the 80-year-old would both be justly
with accepting Harris's age indifference principle as denied an ICD if the relevant medical test indicated it
justification for using a lottery to distribute access to was extremely unlikely the ICD would make a differ-
TIAHs? If we were, then a 40-year-old in heart failure ence. In both cases the marginal benefit is too small
facing death in 6 months with a prospect of 20-plus to justify the cost for either individual. Higher priority
years survival with a TIAH would be entitled to no more health needs of both the elderly and nonelderly could
a chance for that TIAH than a 90-year-old in heart failure be better served by the achieved savings.
with only a 2-year prospective gain in life expectancy. Harris invokes the analogy with murder as a basis
This does not look like a morally plausible result that for rejecting all age-based rationing that involves loss of
most of us would be comfortable endorsing. potential life-years. However, this is an inapt analogy.
We must ask ourselves how the 90-year-old man- The murderer is violating the liberty rights of his victim.
aged to achieve that age. He might have gotten there His victim would have gone on living happily but for his
because substantial and costly healthcare resources got murderous intervention. His victim does not require any
him there, the very sorts of resources that the 40-year- expensive social resources for continued survival. No ques-
old might now be denied. Alternately he might have tion of distributive justice is raised in the case of murder.
simply been the beneficiary of an excellent genetic But that is precisely the moral issue at stake with respect to
endowment (whereas our 40-year-old was the unfortu- the problem of healthcare rationing. We have only limited
nate recipient of a deficient genetic endowment). From resources to meet virtually unlimited healthcare needs.
the perspective of healthcare justice, what would justify What criteria may we use to determine which needs are
society conferring additional life-years on the fortunate fairly met or unmet? If we spend $300,000 to give a child
90-year-old at the expense of life-years lost by the 40-year- with Gaucher's an extra year of life, are we equally morally
old? Neither egalitarian nor utilitarian considerations obligated to spend that same amount to give an 80-year-
would warrant such an allocation. old an extra week of life, or an extra day of life, as Harris's
Harris's view is that both such individuals equally use of the murder analogy seems to imply? This is neither
desire to experience "the rest of their lives" and this just nor reasonable. The loss of that extra week of life is
is what requires equal moral respect. But that expres- rightly regarded as being unfortunate, not unjust.
sion of desire looks more like plain self-interest than There are numerous reasonable and morally rel-
any consideration having substantial moral weight. If evant considerations that might be invoked for making
we perform a Rawlsian-like mental experiment with fair healthcare rationing decisions at the level of social
two 20-year-olds and tell them that one of them will policy. These considerations will often need to be bal-
have the fate of our 90-year-old whereas the other will anced against one another in complex circumstances.
have the fate of our 40-year-old but we must decide We have no reason to believe there exists one "perfectly
now which of those two future possible selves will have just" balancing that could be achieved through human
assured access to the TIAH, I am confident that the im- reason in all these circumstances.2° Numerous options
partial answer that would emerge would be that the for allocating artificial hearts can be reasonable and
40-year-old gets the TIAH. "just enough." But it would be unwise and unfair if
Harris is correct that there are numerous rationing such allocation decisions were left to the moral intui-
scenarios in which age indifference would be a moral tions of tens of thousands of individual physicians, all
requirement. But that does not imply that age indiffer- vulnerable to the normal range of personal biases.
ence is required in every rationing scenario. Implant-
able cardiac defibrillators (Icps) cost $40,000 each. Rationing and Rational Democratic
They are intended to prevent sudden death from an Deliberation
arrhythmia In the United States we may be implanting
200,000 per year. However, 8].% of these devices never The better option, as I have argued at length in a recent
fire over a 5-year period, at which point batteries need book,21 would be a public process of rational democratic
to be replaced at a cost of $20,000. There are now tests deliberation. This would not be a Rawlsian mental ex-
that can determine with 98.7% accuracy which patients periment; this is a real world option. The vast majority
will likely not suffer an arrhythmic event over the next of us in middle age are completely ignorant of our future
2 years. We could reduce by 70,000 the number of de- possible health needs, and in that respect we are capable
vices implanted and save $3 billion per year, but there of the moral impartiality required by these choices. Fur-
would have been 800 (in theory) preventable deaths ther, we are mindful of what we regard as reasonable fi-
as a trade-off for saving those dollars. In this scenario nancial expenditures of our money (taxes) for healthcare.
Fleck: Just Caring: In Defense of Limited Age-Based Healthcare Rationing
That creates the budgets (limits) within which we can serious influenza pandemic, comparable in severity to the
collectively set healthcare priorities and determine what 1918 Spanish flu that may have killed 50 million people.
rationing protocols we are willing to impose upon our If we did not have enough respirators to save the lives of
future possible selves that we judge to be "fair enough." all who would otherwise be virtually certain to die, then
If we decide that we are willing to pay for only it would be morally defensible (not unjust) if we were to
100,000 artificial hearts and that these should not be deny respirators to those over the age of 70. The substan-
available at social expense to those over age 70, we tive moral justification for such a choice would be the
have made an age-based rationing decision. But it is fair innings argument. The procedural moral justification
not a decision we are imposing on others we deem un- would be a rational democratic deliberative process that
worthy of equal moral respect. And it is a judgment we endorsed such a policy, even if the endorsement was as
make with the understanding that this represents a last the least unjust option available. We might wonder how
chance therapy for a future possible version of myself. this could be fair to those who are currently in their 60s
We recognize that we are telling future caregivers that or 70s. I will mention that I am 65 years old, but I would
we have freely given up a right to be rescued with an certainly endorse this policy if any alternative arrange-
artificial heart at age 77 because we have chosen, in ment put my children at risk of being denied a life-saving
effect, to allocate limited and expensive life-prolonging respirator (say, as a result of a lottery process). I suspect
healthcare resources to earlier stages in our life when this same commitment would be endorsed by the vast
they would do much more good and be judged to be majority of my fellow nearly senior citizens.
much more valuable to us. Such judgments represent
fair terms of cooperation, commendable moral reci- Conclusion
procity. They cannot properly be described as nonvo-
luntary forms of euthanasia." What we are defending in this essay is the need for sur-
If we collectively decided, we could pick age 75 gically precise age-based rationing protocols that have
instead of age 70, or we could choose age 80 as an ab- been fairly and democratically legitimated in a range of
solute limit for an artificial heart at social expense. We rationing circumstances where an age-based criterion
could choose not to use any age at all, relying instead would yield a more just outcome. We are not denying
upon some complex mix of clinical criteria related to that the greater health needs of the elderly justly com-
probability of survival for some number of years. But mand more social resources to meet those needs. But
that would create considerable fuzziness and uncer- we are saying that there must be just limits to meeting
tainty (and consequent risk to fairness) in actual prac- those needs. We must also note the morally relevant
tice. This is what Callahan refers to as the "ragged edge" point that some of the savings that result from age-
problem.23 The virtue of a definite age limit in this based rationing protocols redound to the benefit of the
context is that it creates a bright line that might better elderly in the form of health services that yield more
protect fairness. What a fair deliberative process allows health good for them at a lower cost. Purchasing very
us to say at the very least is that choosing a specific age expensive marginally beneficial life-prolonging health-
beyond which individuals would not have a claim to an care can adversely affect the just claims of the elderly
artificial heart at social expense is not unjust. to needed healthcare as much as the just claims of the
Another circumstance in which we may have to nonelderly. This last point applies as much to the US
invoke some form of age-based rationing would be a healthcare system as to European healthcare systems.
Notes
1. Callahan D. Setting Limits: Medical Goals in an Aging Society. Forge Press; 2000; Moody H. Ethics in an Aging Society.
New York: Simon and Schuster; 1987. Baltimore, MD: Johns Hopkins University Press; 1992;
2. Daniels N. Am I My Parents' Keeper? An Essay on Justice Kilner J. Life on the Line: Ethics, Aging, Ending Patients'
between the Young and the Old. New York: Oxford Lives, and Allocating Vital Resources. Grand Rapids, MI:
University Press; 1988. W.B. Eerdmans Publishing Co.; 1992; Smeeding T. Should
3. This could be a very long list, but we will simply list Medical Care Be Rationed by Age? Totowa, NJ: Rowman
more prominent critiques. See Binstock RH, Post SG, and Littlefield; 1987.
eds. Too Old for Health Care? Controversies in Medicine, 4. Harris J. The Value of Life. London: Routledge and Kegan
Law, Economics, and Ethics. Baltimore, MD: Johns Paul; 1985.
Hopkins University Press; 1991; Moody H. Aging: 5. Williams A. Intergeneration equity: An exploration of the
Concepts and Controversies. Thousand Oaks, CA: Pine "fair innings" argument. Health Economics 1997;6:117-32.
■Y 1 I II .111 111111
6. Harris J. Cardiac surgery in the elderly. Heart of it Updating the Institute of Medicine analysis on the
1999;82:119-20. impact of uninsurance on mortality; 2008 Jan; available at
7. Europe is facing a comparable problem with the aging httpl/www.urban.orgiurl.cfm?1D5411588&renderforprint=1
out of its population. "In Europe, predictions from the (last accessed 19 July 2009).
Statistical Office of the European communities (EURO- 13. See, for example, Huber CH, Goeber V, Berdat P, Carrel T,
STAT) estimate that from 2010 to 2030 the population Eckstein E Benefits of cardiac surgery in octogenarians-A
of octogenarians in 25 countries of the European Union postoperative quality of life assessment. European Journal
will increase by 57%. By 2050, the population of those of Cardio-Thoracic Surgery 2007;31:1099-105; Filsoufi F,
aged 80 and older will increase by 180%." See Cvachovec Rahmanian PB, Castillo JG, Chikwe J, Silvay G, Adams
K. Coronary artery bypass sargery in the very old: Light DH. Results and predictors of early and late outcomes
at the end of the tunnel or a dead-end road? Journal of of coronary artery bypass graft surgery in octogenarians.
Cardiothoracic and Vascular Anesthesia 2007;21:781-3 at Journal of Cardiothoracic and Vascular Anesthesia
pp. 781-2. 2007;21:784-92. See also the case of Hazel Homer
8. As numerous policy analysts have noted, advancing who, at age 99, received an advanced pacemaker and
medical technologies are the primary driver of the defibrillator to assist a failing heart. She is alive today at
problem of escalating healthcare costs. See Bodenheimer 104; Hartocollis A. Rise seen in medical efforts to treat the
T. High and rising health care costs. Part 1: Seeking an very old. New York Times 2008 Jul 18.
explanation. Annals of Internal Medicine 2005;142:847-54; 14. Shim JK, Russ AJ, Kaufman SR. Risk, life extension and the
Bodenheimer T. High and rising health care costs. Part pursuit of medical possibility. Sociology of Health and Illness
2: Technologic innovation. Annals of Internal Medicine 2006;28:479-502, at 496.
2005;142:932-7. See also Aaron HJ, Schwartz WB, Cox 15. For a profoundly effective understanding of what it
M. Can We Say No? The Chal'enge of Rationing Health Care. means to be uninsured, the following Kaiser Family
Washington, DC: Brookings Institute; 2005. In 1993 Foundation website ought to be visited for in-depth
in the United States we were spending $912 billion on interviews with a number of such families. These are the
healthcare. The estimate for 2009 is that we will spend kinds of injustices that are essentially morally invisible at
$2.5 trillion on healthcare and $4.4 trillion in 2018 (or a broader social level. See Shirk M. In Their Own Words:
about 20.3% of our projected Gross Domestic Product The Uninsured Talk about Living without Health Insurance;
at that time). See Sisko A, Truffer C, Smith S, Keehan available at httpl/www.kff.org/uninsured/2207-index.cfm
S, Cylus J, Poisal J, et al. Health spending projections (last accessed 19 July 2009).
through 2018: Recession effects add uncertainty to the 16. Brody B. The macro-allocation of health care resources.
outlook. Health Affairs 2009;28:w346-57. In Sass HM, Massey R, eds. Health Care Systems: Moral
9. See Note 8, Sisko et al. 2009. Conflicts in European and American Public Policy. Dordrecht:
10. Almost 82% of the elderly have at least one chronic Kluwer Academic Publishers; 1988:213-36.
condition, 65% have at least two chronic conditions, and 17. See note 6, Harris 1999:119.
25% have four or more chronic conditions. The reader 18. See the fuller exposition of this argument in Dey I, Fraser
should think of heart disease or cancer or lung diseases N. Age-based rationing in the allocation of health care.
or kidney disease or diabetes or stroke or dementias Journal of Aging and Health 2000;12:511-37.
or arthritis or sensory deficits, and so forth. Although 19. It should be noted that this particular issue will be just as
many of these conditions are ultimately fatal, the period much an issue in Europe as in the United States. Simply
of time for which individuals can survive has increased having a universal healthcare system does not address this
dramatically as a result of costly contemporary medicine. issue as a justice issue.
This is reflected in the Medicare cost statistics cited 20. This is what John Rawls refers to as the "burdens of
earlier. See Wolff J, Starfield B, Anderson G. Prevalence, judgment." Our moral arguments and moral theories will
expenditures and complications of multiple chronic often not be powerful enough in complex circumstances
conditions in the elderly. Archives of Internal Medicine to yield a uniquely correct moral response. See his Political
2002;162:2269-76. Liberalism. New York: Columbia University Press; 1993.
11. This sentence needs a qualification because the Medicare 21. See my book Just Caring: Health Care Rationing and
program has significant copayments and deductibles. Democratic Deliberation. New York: Oxford University
Those serve as an effective barrier for the elderly in Press; 2009, especially Chapter 5, for a fuller exposition of
roughly the lower half of the income spectrum. They the role of rational democratic deliberation in yielding just
cannot demand everything, no matter what the cost. healthcare rationing policies.
More economically well-off elderly have supplementary 22. I address this issue more comprehensively in an
insurance that eliminates those barriers and allows them earlier essay: Just caring: Assisted suicide and health
to demand whatever they see as medically beneficial for care rationing. University of Detroit Mercy Law Review
themselves. 1995;72:873-99.
12. The most often cited estimate of that annual number in the 23. See Callahan D. What Kind of Life: The Limits of Medical
United States is 22,000 premature deaths attributable to Progress. New York: Simon and Schuster; 1990, especially
being uninsured. See Dorn S. Uninsured and dying because Chapter 2.
Kishore: Human Organs, Scarcities, and Sale: Morality Revisited
7.5 Increasing Resources
Human Organs, Scarcities, and Sale: allowing altruistic donations from strangers have not
resolved the problem. Organ scarcity continues to pre-
Morality Revisited vail, leading to inequitable therapeutic dispensation,
R.R. Kishore escalating costs, trade, crime, and premature death. In
India there are periodic reports of organ trafficking in-
The recent exposure of an international racket in organ volving clinicians, managers of clinical centres, middle
trafficking, extending from Brazil to South Africa, has men, and even state officials; several cases are at present
prompted me to write this paper.' Kidney vendor under active investigation or at trial. The "worldwide
Alberty Jose da Silva and the American woman who shortage of kidneys from cadavers has resulted in il-
bought the organ were both in the same boat. To many licit organ sales and even kidnapping and murder of
the process may sound iniquitous and even sinful but, children and adults to 'harvest' their organs."4 Millions
in fact, it is fair and natural and is consistent with of people are suffering, not because the organs are not
normal human behaviour. available, but because "morality" does not allow them
The shortage of available organs is a global feature to have access to the organs.
of organ transplantation and has been a challenge The question is what is good and what is bad. How
almost since its inception. In the US, as at 10 July 2004, are we to measure the moral content of a particular
86,173 people are on the nation's organ transplant act? Morality is always contextual. It depends on how
waiting list and on average 17 patients die every day and in what context we interpret values. The famous
while awaiting an organ—one person every 85 minutes. Roman physician Galen—for example, did most of his
On average, 115 people are added to the nation's organ anatomy research on pigs and dogs as it was regarded
transplant waiting list each day—one every 13 minutes. as immoral to dissect humans at that time,' but, sub-
In 2001, 6,251 individuals died on the US organ sequently, dissection of human cadavers during med-
transplant waiting list because the organ they needed ical education became a routine practice. The issue of
was not donated in time.2 The situation is no better biotechnological achievements and their social assimi-
in Europe. In certain countries of the Eurotransplant lation contemplates a much deeper dialogue than is
area—Austria, Belgium, Luxembourg, Germany, the being conducted in contemporary ethical discussions.
Netherlands, and Slovenia—as at 1 July 2004, there This paper is an attempt in that direction.
were 15,585 people on the waiting list.
In the year 2002, 12,644 patients were on the wait- Essential Questions and Values
ing list for kidneys but only 3,043 could get an organ.
In the year 2003, in the Netherlands and Germany, The basic ethical principle involved in organ trans-
1,182 and 9,479 patients, respectively were on the plantation is whether a person has a right to enjoy
waiting list for kidney transplantation but only 406 and life on the basis of organs belonging to others. Once
2,111 transplantations could be conducted.3 In India we choose to answer this question in the affirmative we
there is no database and reliable studies are yet to be concede that we are prepared to inflict harm on others
conducted but in view of the country's large population in order to improve our health or to prolong our life.
of nearly 1.2 billion and the rising incidence of end Thus we sacrifice the long cherished principle of non-
stage renal disease (ESRD), the requirement for kidney maleficence in medicine. Whether the organs come
transplantation alone is expected to be around 80,000 through donation, gift, or sale is a matter of individual
per year but not even 5,000 transplants are conducted. choice and circumstances. Even if a person gives his
Strategies such as liberalization of the concept of brain organ willingly and without any thought as to recom-
stem death, introduction of presumed consent, routine pense he suffers harm to his body.
harvesting, required request, mandated choice, raising It is a fact that in every gift or donation some kind
the donor's upper age limit, relaxation of restrictions of expectation is involved, though it may not be a ma-
imposed on donations among family members, and terial consideration. In the case of live organ donors
Source: From Journal of Medical Ethics: The Journal of the Institute of Medical Ethics 31, 6 (2005), 362-5. Copyright © 2005 BM J
Publishing Group Ltd. Reprinted with permission.
II AI IIIIJJ
the organ is donated to a particular person who, in fact, sells his organ does so because he knows that his organ
may not be the neediest or the most deserving bearing is going to save the life of a fellow human being and
in mind the seriousness of illness, period of waiting, as such he is convinced that he is not doing anything
age, family circumstances, capability to afford post- immoral or inhuman. Had he known that his act would
transplantation therapy, and other criteria. This means lead to loss of life, property, or inflict any other kind of
that the act of donation is tainted with considerations of injury on the buyer he would not have sold his organ.
personal relationship, choice, and preference. In other A criminal has no such moral conviction or justification
words the donation is not a candid act of altruism or and commits the act solely for his personal gain, with-
human solidarity, but rather :.s motivated by the desire out caring for the loss or injury suffered by the victim.
to save the life of a near and dear one, which may, at As such, it seems that the legislative strategies in the
times, be to secure one's own comfort and future. Such area of organ transplantation have not been realistic.
urges and motives also consnute considerations other
than altruism since they are aimed at pleasure and Arguments against Financial
fulfillment. Even a donation made to a stranger is not Incentive and Sale
without considerations of possible benefit. Such cases
may be motivated by the desire to discharge a religious Policies on organ transplantation reflect a unique social
duty, to correct a wrong done in the past, to gain mental paternalism. Objections against the sale of organs such
or moral satisfaction, or to be seen as a good Samaritan. as "(1) the dilution of altruism in society, (2) the risk
Once the practice of organ donation by the genetic- that the quality of donated organs would decrease; (3)
ally related and also by strangers, based on altruism, doubts about the voluntariness of those who accept
has been accepted as ethically sound, the following financial incentives for donation, and (4) the treatment
components of organ removal stand morally vindicated: of human beings and their parts as commodities'
do not reflect an objective approach. Recent critics of
a. a person's expectation to enjoy life with the help of markets in organs give two main reasons to support
organs belonging to others is valid, and their opposition: (1) "the integrity of the human body
b. the breach of a donor's bodily integrity and the should never be subject to trade," and (2) a system is
consequent harms are permissible. unethical "when it penalises the weakest people and
exacerbates discrimination based on census" and gen-
Judged on these values a person's act of severing erates "the risk of exploitation of vulnerable donors.""
his/her organ in order to liberate a fellow being from a Some are more skeptical and feel that the "poor of the
terminal illness or to save his/her life cannot be dubbed developing world could become a vast reservoir" of
as immoral simply because the act is accompanied by organs for the developed worldu and that the poor in
a reasonable material consideration. "When a person a "starved country" can never be "fully informed and
sells an organ he or she acts both selfishly, in advanta- autonomous donors.' Others feel that it amounts to
ging him or herself, and altruistically, in contributing "exploitation of potential donors."H
to a public good."6 The presence of considerations is Arguments against organ sale are thus grounded in
not a sufficient reason to transform a simple act into a two broad considerations: (1) sale is contrary to human
sin. Otherwise, selling water to the thirsty would be an dignity, and (2) sale violates equity. Let me examine
equally big sin—in fact, rather a bigger one. these one by one.
To equate an organ vendor with a criminal com-
mitting a heinous crime, as is reflected in the legisla- Is Sale Contrary to Human Dignity?
tive strategies of many jurisdictions, is misconceived. In contemporary ethical deliberations human dignity has
Several jurisdictions provide stringent punishment for become a very handy tool to measure the ethical content of
organ sale in utter disregard of the circumstances that biotechnological applications, at times, without appreciat-
compel a person to sell her/his organ. The punishment ing its true nature, ambit, and implications. It is not within
ranges from three months' imprisonment and/or a fine, the scope of this paper to deal with human dignity in its
as in the UK,7 to eight years' imprisonment and/or a entirety but it may, however, be worthwhile to know what
fine, as in Venezuela.8 In India the punishment may it means in essence. Essentially speaking, human dignity
extend up to seven years' imprisonment and/or a fine is an expression of the human content of Homo sapiens. It
up to Rs20,000.9 Such punishments are prescribed is an expression of the properties or virtues due to which a
only for serious offences, and it is thus clear that organ human creature is known as a human being. These are the
sale is treated as a serious offence, worldwide. Such an characteristic or attributes that are unique to the human
approach does not seem to be correct. A person who race and not possessed by any other living form. What
are these virtues? These virtues, known in Vedic thought themselves. This means that the whole exercise of organ
as dharma, are ten in number—namely, love, trust, right- transplantation is inconsistent with the principle of free
eousness, compassion, tolerance, fairness, forgiveness, and informed consent and therefore is unethical. Why
beneficence, sacrifice, and concern for the weak With then should only the vendor be dedared an offender?
these human virtues in mind, any act done to save the life Arguments linking a person's autonomy to bodily
of a human being or to liberate him from suffering cannot "integrity" or "fullness" and on this basis declaring
be construed as contrary to human dignity. The presence organ sale as "misuse" of "our autonomy"15 appear mis-
of a consideration does not alter the basic content of an conceived because if that is the case "integrity" or "full-
act such as an organ sale, which is grounded in the need ness" is also breached in the case of donation, which is
to save at least two human lives, one from terminal illness not considered to be unethical.
(the recipient), the other from hunger (the donor). The In order to justify the failure to provide organs to the
concept of human dignity does not demand that people needy, various considerations, such as old age, associated
should be forced to die a premature death where an illness diseases, poor prognosis, and irresponsible behaviour,
can be cured nor that people who donate organs should are brought up in order to exclude a number of potential
die of hunger and their families be left to starve. Rather, it organ recipients, knowing fully that in such cases "the al-
will be contrary to human dignity to promote such an act. ternative to transplantation is death."16 Organ scarcity has
The matter of payment is a logistical dimension, not the polluted the moral concepts involved in transplantation
substantial aspect of the transaction. and some feel "alcoholics should be given lower priority
Retrieval of organs from the dead by presum- for a liver because of their moral vice of heavy drinking"17
ing consent on their behalf or the act of declaring a and "lung transplantations" should not be offered "to
person brain stem dead in order to remove organs people who smoke or have other substance abuse in the
from his body are devices designed solely to augment last six months."18 Despite the fact that brain stem death
organ supply but they are not regarded as contrary to and human death are not the same, the definition of
human dignity. Prohibition on sale of organs makes death has been liberalized in order to give an "incessant
matters worse by restricting transparency, fairness, and push to expand the pool of potential organ donors."19
choice and by generating arbitrariness, fear, and bribes. Thus, the prohibition on sale is not without heavy costs,
Vendor and buyer are rendered vulnerable because of which include untimely death, poor quality of life, higher
the introduction of an unwarranted legal component disease burden, moral bias, and premature certification
that brings in many players such as police, lawyers, of death. Is this consistent with human dignity?
adjudicators, and social activists, each with their own It is significant that the concept of human dignity
philosophy and interest, thereby transforming a simple is being selectively applied in the case of certain tissues
activity into a highly complex exercise. only. Blood, bone marrow, sperm, and eggs are being
The argument that there cannot be genuine and openly sold and a woman can "command $50,000 for
free consent to the sale of organs is not well founded. her donated eggs" (Kahn, 14: p. 1) but their sale, it
In fact such an argument is an antithesis of the con- seems, does not attract notion of human dignity.
cept of autonomy. The decision to sell an organ, taken
by a person after considering all circumstances, con- Is Sale Violative of Equity?
sequences, options, and possibilities, cannot be dis- (1) The apprehension that organs will become costly,
regarded by others on the ground that it has been taken going beyond the reach of the common man is un-
under undue influence or inducement. The individual founded. Costs can always be controlled by the state
is the best judge as to what is best for him in a given through the use of regulatory mechanisms, as is done
situation and so long as his decision does not affect in the case of other goods and services. Furthermore,
others he cannot be stopped from acting upon his deci- socioeconomic inequalities are present in all walks
sions. If the vendor is not able to give free and informed of life, not only in organ transplantation. The whole
consent because of the pressures of poverty and the lure healthcare system is subject to market forces. Many
of money, the buyer is also not able to give such consent drugs, many pieces of equipment, appliances, proced-
because of the pressures of illness and the urgent need ures, and services are prohibitively costly and are not
to save his life. The donor too is unable to give free accessible to all those who need them. In many jurisdic-
and informed consent because of the fear of losing a tions, they have to be imported from the developed and
near relative and possibly the consequent loss of sup- industrialized countries, which, at times, monopolize
port and security. Thus none of the parties involved is their trade. Despite strong objections from the public
capable of giving free and informed consent because who desire health care to be available as a welfare meas-
of the compelling circumstances in which they find ure there has been an increasing commercialization of
II J11 1111111
healthcare services. Organ transplantation is also a part their existence. If, finally, selling their organs is the only
of this overall milieu. way to get the money they need to prolong their exist-
(2) The purchase of organs is likely to have only a ence, even if only temporarily, how can society stop
marginal impact on the cost of -xansplantation procedures. them from doing so when society itself is unwilling to
In many countries, including India, where there have been provide them with adequate means to survive.
reports of organ trafficking, kidneys are sold for as little as Why this sudden concern for the poor? A society
US$400-500, while reports on the total cost of a kidney passive to their problems for ages has no authority to
transplantation vary widely, ranging from US$1,0002° to interfere with the arrangement evolved by them to safe-
US$8,000.21 Dr Raymond Crockett, debarred from prac- guard their survival. If, however, society truly feels they
tising in Britain in 1990, for professional misconduct, should not sell their organs, then their genuine needs
arranged for kidneys to be bought from Turkish people should be addressed so that they are not forced to sell
for £2,500-3,500, but charged each patient £66,000 for their organs. What kind of morality is it which snatches
the transplantation.22 This shows that the cost of organs is from the poor the only asset offered to them by nature?
just a fraction of the total transplantation cost, which, in The prohibition on the sale of organs has worsened the
fact, is much higher if the pos:-transplantation immuno- lot of the poor. Buyers quite often refuse to pay or do not
suppressive therapy and the other follow-up care is taken pay the agreed price. The vendor cannot assert his claim
into account. The apprehension in some quarters that because of the fear of being prosecuted. Thus the strategy
organ sales will create a market mechanism that will greatly that was evolved for protecting the poor has been causing
increase transplantation costs is therefore misconceived. just the opposite effect. There is one more aspect to the
Rather, the free availability of organs will reduce the costs of above issue. If a person who is not poor and in whose case
transplantation by curbing the expenses incurred in clan- there is no possibility of being exploited chooses to sell his
destine operations and the middle men who are invariably organ will he be allowed to do so? If not, this means that
associated with the organ trade, as has been made clear by the reasons for prohibiting organ sale are grounded not in
the recent exposures in London23 and the Punjab.24 Organ the concern for the poor, but in some other considerations.
transplantation is a costly medical intervention, mainly be- (4) The argument that permitting organ sale is not
cause of the high fees of the surgeons and others involved an equitable proposition since it restricts availability
in the process. Organ sale is unlikely, therefore, to increase of organs only to affluent sections of the population is
the transplantation costs substantially. Moreover, if organ misconceived. How can it be fair to deny health care to
sale is legitimized the cost can always be regulated by the those who wish to buy it using their own money, earned
state, as is being done in the case of other commodities. by honest means, on the ground that it is not available
A better option to reduce transplantation costs is to regu- to others because of their inability to pay? If that is fair,
late surgeons' fees, nursing home charges, and the price of why have we chosen a system of pricing for health care
equipment, appliances, and drugs. when many cannot pay even for the basic necessities,
(3) With regard to the concern that once organ such as food and drinking water? Ours is an essentially
sale is legitimised organs will mainly be sold by "those heterogeneous society and equity has to be defined in
who cannot afford to keep their organs,"25 the matter realistic and pragmatic terms.
requires deeper examination. The evidence thus far (5) Prohibition on organ sale generates inequity by
shows that in almost all cases, organs have been sold exerting undue pressure on the near relative who may
by persons in a state of abject poverty. In one case, in "feel compelled to overlook the risks of organ dona-
India, when I asked an organ seller why he had sold tion when their loved one stands to receive so great a
his kidney his reply was devastating "I had nothing benefit."26 This pressure amounts to coercion, which
else to sell!" People surrounded by such brutal poverty is as bad as that exerted by the poverty and as such
and social deprivation do not have many options. Even "should equally rule out donation."' The sale provides
when their organs are intact their lot is miserable be- a wider choice, the vendor may be much healthier than
cause they suffer from hunger, diseases, and scorn. So- the donor, and his or her organ may be much more
ciety has so far done nothing to alleviate their suffering, compatible with the recipient. Furthermore,
The selling of an organ may provide them with some
additional means and prolong their existence. If the [during] removal of a vital part like a kidney the donor
sale of organs amounts to exploitation of the poor it is subjected to a major surgical intervention which is
is no more than a continuum of the long drawn-out not without risk. The recipient who is already sick is
process of their exploitation, which has been watched also exposed to a major surgical procedure with pos-
by society for centuries. The poor have been selling sibilities of complications. This means that two mem-
all that they possess for centuries in order to continue bers of the same family are placed in a vulnerable
situation thereby affecting the fate of other members physical or mental disability or disease. It cannot be
of the family also. In case of any complications or equated with justice, art, or spirituality. It is also worth re-
untoward outcome the said family may be the victim membering that the right to relief from pain and suffering
of serious adversities. As such it seems safer to accept is intrinsic-that is, it is non-divestible. Any social policy
donation from a member of a different family? leading to infringement of such a valuable right has to
be founded on equally vital considerations. Donor and
(6) There is also a concern that there may be trans- recipient are the major stakeholders in policies relating
national movement of organs. The affluent countries, to organ transplantation and as such any policy that does
with the power of their money, may drain organs from not take into account their perspectives and views is not
the poorer countries thereby making the populations of valid. The sale of organs is essentially rooted in the urge
the poorer countries even more vulnerable. This again to survive. The recipient wants to survive the threat of a
is a problem of regulation, which can always easily be terminal illness, the seller wants to survive the threat of
handled by banning exports of human organs or by poverty. Ethicists continue to debate the moral content
adopting other suitable strategies. of biotechnological promises, subordinating therapeutic
(7) The purpose of allowing the sale of organs is not advantages to "higher" goals but I am forced to draw the
to improve the health status of the sellers or to award same conclusion that I drew in 1995-that is:
them "a long term economic benefit," as conceived
by some.29 Such economic or health benefit does not Neither the diseased persons nor the genetic rela-
occur in altruistic donations either, but they are permis- tions provide an answer to trading in human body
sible. The reasons for permitting the sale of organs are parts. The live human body constitutes a vital source
grounded in the concern to save the lives of terminally of supply of organs and tissues and the possibilities
ill patients with the help of available medical knowledge of its optimum utilisation should be explored. There
and technology by curbing "a contrived shortage created is no scope for dogmatic postures and open minded-
by existing organ procurement policies."30 As regards to ness should be the approach while dealing with the
poverty and ill health, they are rooted in factors other issue of organ transplantation. Society owes a duty to
than organ sale and need to be addressed accordingly. save the life of a dying man and in the event of failure
to do so, it is absolutely immoral to interfere with his
own arrangements by making unrealistic laws. The
Conclusion
scarcity needs to be urgently overcome otherwise
Those who were earlier destined to die carry a hope unwarranted trade and crime are liable to thrive.
to survive, provided the biotechnology is allowed to Commercialisation should be curbed by making the
unfold itself. It may be worth appreciating that medicine enforcement agencies efficient and not by depriving
is always need-based-that is, it is an aid to overcome a needy person of his genuine requirements.28
Notes
1. Rohter, L. 2004. 'Tracking the Sale of a Kidney on 9. Transplantation of Human Organs Act 1994. Gazette of India
a Path of Poverty and Hope," in New York Times 1995 Feb 4: Part 2, section 3, sub-section (i).
(23 May): 1. 10. American Medical Association: Council on Ethical and Judicial
2. The United Network for Organ Sharing (uNos), [online], Affairs. 1995. "Financial Incentives for Organ Procurement
accessed 10 July 2004 at wwwunos.org. Ethical Aspects of Future Contracts for Cadaveric Donors," in
3. Eurotransplant Foundation, [online], accessed 1 July 2004 Archives of Internal Medicine b 5.581-9.
at wwwtransplant.org. 11. Marino, I.R., Cirillo, C., and Cattoi, A. 2002. "Market of
4. Siegel-Itzkovich, J. 2003. "Israel Considers Paying People Organs Is Unethical under Any Circumstances," in stij
for Donating a Kidney" in stij 326: 126-7. 325: 835.
5. Tuffs Heidelberg, A. 2003. "German Surgeon 12. Wigmore, S.J., Lumsdaine, J.A., and Forsythe, J.L.R. 2002.
under Investigation over Organ Trading," in BMJ "Defending the Indefensible," in BMJ 325: 835-6.
326: 568-9. 13. Moslmann, E 2002. "The Right to Buy or Sell a Kidney," in
6. Harris, J., and Erin, G. 2002. "An Ethically Defensible Lancet 360: 948.
Market in Organs," in BMJ 325: 114-15. 14. Kahn, J. 1999. 'Wanted: Tall, Smart and Fertile," in
7. Human Transplant Act 1989 s 1(5). 1989. Hmso: London. Bioethics Examiner 3: 1-4, at 4.
8. Law of 19 July 1972 of Venezuela on Organ 15. Cohen, C.B. 1999. "Selling Bits and Pieces of Human to
Transplantation. 1972. International Digest of Health Make Babies: The Gift of the Magi Revisited," in Journal of
Legislation 23: 636. Medical Philosophy 24: 288-306, at 295.
II UM 11141.1
16. Schmidt, V.H. 1998. "Selection of Recipients for Donor 23. Dyer, 0.2002. "GP Struck off after Offering to 'Fix' Kidney
Organs in Transplant Medicine," in Journal of Medical Sale," in BMJ 325: 510.
Philosophy 23: 50-74, at 52. 24. Kumar, S. 2003. "Police Uncover Large Scale Organ
17. Glannon, W 1998. "Responsibility, Alcoholism, and Liver Trafficking in Punjab," in BMJ 326: 180.
Transplantation," in Journal of Medical Philosophy 23: 25. Kluge, E.-H. 2000. "Improving Organ Retrieval Rates:
31-49, at 31. Various Proposals and their Validity," in Health Care
18. Anon. 2001. "No Transplants for Smokers," in The Analysis 8: 279-95, at 283.
Hindustan Times (9 Feb.): 12. 26. Kahn, J. 2003. "Dying to Donate," in Bioethics Examiner 7:
19. Youngner, S., and Arnold, RM. 2001. "Philosophical 1-4, at 4.
Debates about the Definition of Death: Who Cares?," in 27. Radcliffe Richards, J. 1996. "Nefarious Goings On," in
Journal of Medical Philosophy 26: 527-37, at 527. Journal of Medical Philosophy 21: 375-416, at 377.
20. Swami, P 2003. "Punjab'; Kidney Industry," in Frontline 28. Kishore, A.R. 1996. "Organ Transplantation: Consanguinity
(14 Feb.): 115-17. or Universality?," in Medical Law 15: 93-104.
21. Anon. 2000. "Kidney Transplant Racket Busted. 29. Goyal, M., Mehta, R.L., and Lawrence, J.S., et al. 2002.
Delhi Surgeon, Five Donors from Andhara Pradesh "Economic and Health Consequences of Selling a Kidney
among Nine Arrested," in The Hindustan Times in India," in fAMA 288: 1589-93.
(7 Dec.): 1. 30. Barnet, A.H., Blair, R.C., and Kaserman, D.L. 1992.
22. Anon. 2000. "Doctor Appeals against Ban on Kidney "Improving Organ Donation: Compensation versus
Sales," in The Times (23 Sept.): 11. Markets," in Inquiry 29: 372-8.
7.6 Cases
Case 1 inhibiting the doctors in their judgment of what

Optimal Care versus Cost should be done for him. That is to be deplored.
I understand that there are budgetary problems
Containment: What Is a Doctor to Do? confronting the health care system. I raise it in
passing to point out that there were a number of
In a BC Supreme Court case, :he plaintiff sued doctors references to the effect of financial restraint on the
for negligence claiming that her husband died because treatment of this patient. I respectfully say it is
the doctors had not taken a CT scan and thus failed something to be carefully considered by those who
to provide timely, appropriate, and skillful emergency are responsible for the provision of medical care
care. In the course of the judgment against the doctors, and those who are responsible for financing it. I
Mr Justice Spencer of the British Columbia Supreme also say that if it comes to a choice between a phys-
Court commented upon the relationship between ician's responsibility to his or her individual pa-
budget reductions in health care and physicians' re- tient and his or her responsibility to the medicare
sponsibilities toward their patients: system overall, the former must take precedence
in a case such as this. The severity of the harm that
I must observe that throughout this case there were may occur to the patient who is permitted to go
a number of times when doctors testified that they undiagnosed is far greater than the financial harm
feel constrained by the British Columbia Medical that will occur to the medicare system if one more
Association standards to restrict their requests for CT scan procedure only shows the patient is not
CT scans as diagnostic tools. No doubt such so- suffering from a serious medical condition.'
phisticated equipment is limited and costly to use.
No doubt there are budgetary restraints on them. Judge Spencer's reasoning can be usefully com-
But this is a case where, in my opinion, those pared to the discussions of physicians' roles in rationing
constraints worked against the patient's interest by health care resources, discussed also in Chapter 4.
Note
1. Law Estate v Simice (1994), 21 C.C.L.T. (2d) 228 (B.C.S.C.).

Cases 1 413
Case 2 gets an idea: he could hold an estate sale of his organs

Having an Estate Sale of One's Organs after his death. The highest bidder would get the organ,
while his estate would get the highest price. "I think
it's a great idea," he tells his wife. "I don't know," she
Fletcher knows that his death is imminent, and he's responds hesitantly, "I think that sort of thing is illegal
been putting his affairs in order. He wishes he had in Canada." "Why should it be?" Fletcher demands in-
more money to leave his family, especially his grand- dignantly. "It's my body, my organs. I can auction off
daughter, who would make a great doctor if she only my car, my piano, and my mother's wedding jewelry,
had medical school tuition, and his eldest grandson, so why cant I auction off my organs? It's not like I'd be
who would love to go to New York to drama school. hurting anyone, in fact I'd be making sure some people
While watching a news report about transplant tour- who might not get an organ get one sooner. I think it's
ism and the extreme lack of organs in Canada, Fletcher a win-win situation."
Case 3 the year-long recovery period. He is reluctant to have

Rationing Services to an Elder Who Is the surgery, stating that it's his life, his choice, and that
he may choose not to have the surgery and "let what-
Responsible for His Medical Condition ever happens, happen." His wife is uncomfortable with
that choice because his death would leave her without
Frank J. is 63 years old and semi-retired, work- income or a caregiver. Frank reluctantly agrees to see
ing part-time at a non-strenuous job to support his the surgeon to discuss his options.
invalid wife. The couple have no children or other The surgeon tells Frank that in order to receive
family members. Frank is overweight but not obese; the surgery, he must give up smoking and drinking
he smokes one-and-a-half to two packages of ciga- and must agree to certain lifestyle changes, including
rettes a day, and he has been an alcoholic for the last a healthier diet and more exercise. This ultimatum
30 years, drinking whisky slowly but steadily from enrages Frank, who states that smoking and drinking
morning to night. He has had good health all his life are his only enjoyments in life and if those are gone,
in spite of his sedentary lifestyle. He has never been he has nothing left to live for. However, now that he
hospitalized and in fact has never had an illness more knows that the surgery is no longer entirely his choice,
serious than the flu. His mother's death by medical he demands the surgery without limitations of any kind.
mistake and his wife's chronic poor health have given He tells the surgeon that he's contributed to "the [health
him a distrust of medical professionals and an intense care] system" all of his life and taken nothing from it,
dislike of hospitals. so it's wrong of the ficPs to deny him treatment the first
Frank has come to his doctor with shortness of time he needs it. The surgeon explains that the very in-
breath and intermittent chest pains. Testing indicates vasive surgery will not be as successful if he continues to
that Frank is suffering from congestive heart disease smoke, drink, and live a sedentary lifestyle afterwards.
and that he will need a triple-bypass surgery. Frank is Frank says he doesn't care—the system "owes" him and
frightened, both of the prospect of heart surgery and of cannot ref use his need.
Case 4 receive a kidney from an Indian woman. Usually pa-

Buying a Kidney in India but tients never meet the donors, but a nurse at the hospital
reveals to Mr B. that the donor is a healthy widow with
Requesting Canadian After-Care five children who is using the money to help pay her
bills.
Forty-four-year-old Harjumar B. has been suffering The surgery goes well, but upon his return to
from kidney disease for years. He is fed up with having Canada Mr B. quickly becomes very sick. He goes to
dialysis every other day while he waits up to six years his local hospital where it becomes apparent that he
for a donor kidney, he wants a more immediate solu- has developed a serious infection that will require a
tion. After liquidating some of his RSP savings, Mr B. lot of medication and time in the hospital to treat. As
travels to India where a broker has arranged for him to Mr B.'s daughter is passing the nursing station one day
II ,JII 1111.1.1
on the way to visit her father. she overhears the nurses "I don't think it's fair at all. Our medical care wasn't
talking about her fathen case. "Why should he be good enough for him when he decided he wanted a
able to jump the line and go out of the country to buy kidney—why is it good enough now when he's got
his kidney, when others can't afford to do it?" won- an infection? He wouldn't even have the infection if
ders one nurse. "I don't know," says the other one, "I'd he hadn't had the surgery outside the country. Why
probably do it if I was dying and could afford it. After should Canadian taxpayers pay for something he's
all, it's your life, you know" The first nurse replies: done to himself?"
Case 5 Security and the Guaranteed Income Supplement—

Does Clifford Olson Deserve which together totaled $1,169.47 monthly after he had
reached the age of 65. On June 1, the federal govern-
Medical Treatment? ment terminated Olsons payments. Human Resources
Minister Diane Finley called the fact that he had been
In 1980-1, Clifford Olson strangled, bludgeoned, and receiving them "outrageous" and "offensive," stating
stabbed 11 British Columbia children and youths be- that "It's wrong and obviously unfair that prisoners
tween the ages of 9 and 18 to death. After his arrest in who break the rules receive the same entitlements [as
August 1981, Olson made a cash-for-bodies deal that everyone else] ."2
paid $10,000 per body located into a trust for Olson's In September 2011, it was revealed by the com-
wife and infant son; Joan Hale was paid $100,000. missioner for corrections that Olson was suffering from
Olson pled guilty to 11 counts of murder and was metastatic terminal cancer and had been transferred to
given 11 concurrent life sentences and "dangerous a hospital in Laval, Quebec, to undergo treatment. He
offender" status, meaning that he would likely never died September 30 at the age of 71. This information
have been released from prison. By his own admis- was revealed to the victim's families under their right
sion, this self-proclaimed "beast of British Columbia" to know Olsons movements; however, out of respect
was guilty of between 80 and 200 murders and would for Olson's privacy, the commissioner's office would not
kill again if ever released. While in prison he revelled release information to the media. The idea that Olson's
in his multiple serial murderer status, including writ- privacy was worthy of respect and that he was receiving
ing book manuscripts and making a series of video medical care—whether treatment for cancer previously
tapes describing the acts of torture and sexual assault or palliative care currently—enraged many Canadians.
he perpetrated on his victims, which included driv- Some argued that Olson didn't deserve any treatment at
ing nails into their skulls, then asking them how it all, that he should suffer and die like the beast he was.
felt. Five years to the day after they found the body Some argued that if we live according to a social con-
of Daryn Johnsrude, Olson continued his torment by tract, then Olson broke that contract and is not entitled
sending a letter to the victim's• mother and stepfather to the same treatment as those who keep it. On the
describing in exact detail what he did to their 16-year- other hand, some argued that Olson deserved the best
old son.' medical treatment Canadian health care could offer, the
In March 2010, it was revealed that Olson had same as any other patient.3 No consideration of his past
been receiving two federal government benefit cheques actions should be included in decisions about current
per month while in prison—the Canadian Old Age health care treatment.
Notes
1. CBC News, "In Depth: Clifford Olson, The Beast of harper-cuts-clifford-olsons-government-pension-payments/
British Columbia," 19 July 2006, http://www.cbc.ca/news/ article1587995.
background/olson-cliff ord 3. Bill Kaufmann, "Olson deserves 'best medical
2. Jane Taber, "Harper Cuts Clifford Olsons Government treatment': Lawyer," The Toronto Sun 22 September
Pension Payments," The Globe and Mail 1 June 2010, http•1/ 2011, http://www.torontosun.com/2011/09/22/
www.theglobeandmail.com/news/politics/ottawa-notebook/ olson-deserves-best-medical-treatment-lawyer.
Study Questions
7.7 Study Questions
1. What is the egalitarian system of providing health care? What is the libertarian? What
is the liberal? What are the strengths and weaknesses of each? Which do you prefer,
and why?
2. Which areas of the Canadian health care system would you like to see improved (and
how) and which left alone?
3. How are elements of egalitarian, liberal, and libertarian approaches to macroallocation
reflected in Ronald Dworkin's approach to macroallocation of health resources? What
do you think Dworkin's criticism of the libertarian approach to allocation of health
resources would be?
4. What are the problems with using "the rescue principle" (also called "the rule of rescue")
as a means of allocating scarce resources?
5. What are the different metrics for measuring the health benefits of spending scarce
medical resources, and what are their advantages and disadvantages?
6. What is the complex criteria method of microallocation? What is the random selection
method? Do you agree with either, or would you prefer a system that also takes into
account age and responsibility for one's condition (or anything else)?
7. What is age-based rationing, and what are the arguments for and against it? Which
side do you take, and why?
8. Do you think that persons should be offered reduced health care for conditions
brought on by such things as smoking, drinking, overeating, eating fatty foods, under-
exercising, engaging in risky activities, or working too hard?
Health Care Delivery Systems

Armstrong, Pat. 1997. "Managing Care the Canadian Way" Humane Health Care Inter-
national 13 (1): 13-14.
DeGrazia, David. 1996. "Why the United States Should Adopt a Single-Payer System of
Health Finance." Kennedy Institute of Ethics Journal 6 (June): 145-60.
Freedman, Benjamin, and Francoise Baylis. 1997. "Purpose and Function in Government-
Funded Health Coverage."Journal of Health Politics, Policy and Law 12 (1): 97-122.
Neilsen, Kai. 1989. "Autonomy, Equality, and a Just Health Care System." The International
Journal of Applied Philosophy 4: 39-44.
Woolhandler, Steffie, and David U. Hiimmelstein. 1989. "A National Health Program:
Northern Light at the End of the Tunnel."JAMA 262: 2136-7.
Macroallocation
Daniels, Norman. 2007. Just Health: Meeting Health Needs Fairly. Cambridge: Cambridge
University Press.
Englehart, Tristram. 1986. The Foundations of Bioethics, ch. 8. New York: Oxford University
Press.
Grogan, Collen M. 1992. "Deciding on Access and Levels of Care: A Comparison of Canada,
Britain, Germany and the United States." Journal of Health, Politics, Policy, and Law 17
(2): 213-32.
Gutman, Amy. 1981. "For and against Equal Access to Health Care." Millbanh Quarterly
59: 542-60.
Nozick, Robert. 1971. Anarchy, State, and Utopia. New York: Basic Books.
II JZI 11123.1
Rawls, John. 1971. A Theory of Justice. Cambridge, MA: Harvard University Press.
Sting], Michael. 1996. "Equality and Efficiency as Basic Social Values." In M. Stingl and D.
Wilson, eds, Efficiency vs. Equality: Health Reform in Canada, 7-19. Halifax: Fernwood.
Wiklar, Daniel, and Sarah Marchand. 2009. "Society's Allocation of Resources for Health."
In Helga Kuhse and Peter Singer, eds, A Companion to Bioethics, 2nd edn. Oxford: Wiley
Blackwell.
Mesoallocation/Rationing
Browne, Alister. 2010. "The Ethics of Aggressive Discharge Planning." Cambridge Quarterly
of Healthcare Ethics 19 (1): 75-85.
- --. 2008. "Violating Rules of Rationing." Canadian Journal of General Internal Medicine
3 (3).
- --T. Anderson, D. Brown, C. Cooledge, B. Leal, D. McDonald, and M. Saxe-Braithwaite.
2005. "How to Make Allocation Decisions: A Theory and Test Questions." Healthcare
Management Forum 18 (1): 32-3.
Daniels, Norman. 1994. "Four Unsolved Rationing Problems: A Challenge." Published
with replies to the challenges by Frances M. Kamm, "To Whom?"; Eric Rakowski, "The
Aggregation Problem"; John Broome, "Fairness versus Doing the Most Good"; Mary
Ann Baily, "The Democracy Problem." Hastings Center Report 24 (4): 27-42.
- --. 1991. Is the Oregon Rationing Plan Fair? JAMA 265(17): 2232-5.
Klevit, H.D., A.C. Bates, T. Castanares, E.P. Kirk, R. Sipes-Metzler, and R. Wopat.
"Prioritization of Health Care Services: A Progress Report by the Oregon Health Services
Commission." Archives of Internal Medicine 151: 912-16.
Martin, Doug, and Peter Singer. 2003. "A Strategy to Improve Priority Setting in Health
Care Institutions." Health Care Analysis 11 (1): 59-68.
Menzel, Paul T. 1990. Strong Medicine: The Ethical Rationing of Medical Care. New York:
Oxford University Press.
Nelson, James Lindmann. 1996. "Measured Fairness, Situated Justice: Feminist Reflections
on :-.1ealth Care Rationing." Kennedy Institute of Ethics Journal 6: 53-68.
Wikler. Daniel. 1992. "Ethics and Rationing `Whether,' How,' or 'How Much'?" Journal of
the American Geriatrics Society 40 (4): 398-403.
Microallocation
Broome, John. 1990. "Fairness." Proceedings of the Aristotelian Society, New Series 91:
87-101.
Browne, A. 2013. "Blood in a Time of Scarcity" Cambridge Quarterly of Healthcare Ethics 22
(2): 159-69.
Childress, James E 1970. "Who Shall Live When Not All Can Live?" Soundings 53: 339-55.
Shaw, George Bernard. 1946 (originally published 1911). The Doctor's Dilemma. Harmond-
sworth, Middlesex: Penguin Books.
White, D.B., M.H. Katz, J.M. Luce, and B. Lo. 2009. "Who Should Receive Life Support
during a Public Health Emergency? Using Ethical Principles to Improve Allocation
Decisions." Annals of Internal Medicine 150 (2): 132-8.
Age-Based Rationing
Brody, Baruch. 1988. "The Macro-allocation of Health Care Resources." In H.M. Sass and
R. Massey, eds, Health Care Systems: Moral Conflicts in European and American Public
Policy, 213-36. Dordrecht: Kluwer Academic Publishers.
Callahan, Daniel. 1995. Setting Limits: Medical Goals in an Aging Society (with "A Response
to my Critics"). Washington, DC: Georgetown University Press.
Childress, James F. 1984. "Ensuring Care, Respect, and Fairness for the Elderly" in Hastings
Center Report 14: 27-31.
Daniels, Norman. 1988. Am I My Parent's Keeper? An Essay on Justice between the Young and
the Old. New York: Oxford University Press.
Fleck, Leonard M. 2009. Just Caring: Health Care Rationing and Democratic Deliberation.
New York: Oxford University Press.
Harris, John. 1985. The Value of life. London: Routledge and Kegan Paul.
Jecker, Nancy S. 1991. "Age-Based Rationing and Women." Journal of the American Medical
Association 266: 3012.
Williams, Alan. 1997. "Intergenerational Equity: An Exploration of the 'Fair Innings' Argu-
ment." Health Economics 6: 117-32.
Responsibility for Condition/Luck Egalitarianism

Alvin H. Moss and Mark Siegler. 1991. "Should Alcoholics Compete Equally for Liver
Transplantation?" Journal of the American Medical Association 265 (10): 1295-8.
Anderson, Elizabeth. 1999. "What Is the Point of Equality?" Ethics 109: 287-337.
Arneson, Richard.1989. "Equality and Equal Opportunity for Welfare." Philosophical Studies
56: 77-93,
Atterbury, C.E. 1986. "The Alcoholic in the Lifeboat: Should Drinkers be Candidates for
Liver Transplantation?" Journal of Clinical Gastroenterology 8 (1): 1-4.
Katharine Browne. 2016. "VoluntarySterilisationandAccesstoIFVinQuebec ." http://jme.bmj
.com/content/early/2017/08/05/medethics-2016-103726
Cohen, G.A. 1989. "The Currency of Egalitarian Justice." Ethics 99: 906-44.
Dworkin, Ronald. 2003. "Equality, Luck and Hierarchy." Philosophy & Public Affairs 31:
190-8.
Nielsen, Morten Ebbe Juul, and Martin Marchman Andersen. 2014. "Should We Hold the
Obese Responsible? Some Key Issues." Cambridge Quarterly of Healthcare Ethics 23 (4):
443-51.
Scheffler, Samuel. 2003. "What Is Egalitarianism?" Philosophy & Public Affairs 31: 5-39.
Pubk i_r ean
8.1 Introduction
Public health is the part o f health care dealing with anything affecting the health o f popula-
tions that can be addressed through collective or government action. It is often contrasted
with clinical medicine or health care, segments of health care considered in many of the
other chapters in this text. The definition of public health has two important components.
One is that the patient in public health is not an identified individual like someone in a
hospital bed. Instead, the "patient" is a large population or the community as a whole. The
other component is that the decision-makers are not the individual patient or his or her
family in consultation with a medical professional deciding on a treatment that will affect
only that person. Instead, the decision-maker is usually a department of the government—
federal, provincial, or municipal—deciding on a policy that will affect many people.
What kinds of moral issues are there in public health? Many of them concern the
conflict between an individual's rights or freedoms and the well-being of other members
of the community or population. Others are about how to balance harms and benefits
to determine which programs will be for the good of the greatest number or which are
most fair and just. These problems are illustrated in the four topics we've selected for this
chapter: collective action problems in health care, medical screening, harm reduction pro-
grams, and inequalities in health prospects among different groups of people in Canada.
Collective Action Problems
Types of Goods
rival (or rivalrous) good Economists sometimes divide goods—things we might want or need—into four different
a good where one person's kinds.' The division is based on two features. A good can be rival or not in the benefits it
benefit reduces or eliminates a
benefit to others. h a non-rival
provides to users or consumers. An organ transplant is an example of a health care good
good, one person's benefit or that is rival. If one patient is transplanted with a particular kidney (or "consumes it" in
consumption does not affect the words of an economist), then another cannot have that transplanted kidney. But there
another person's. are health care goods that are not rival. The knowledge gained from medical research
can benefit (or be consumed by) one patient without subtracting from any other patient's
excludable good a good benefit. The other feature dividing types of goods is that getting benefits from the good
that people can be prevented may be excludable or not. Suppose medical researchers have found a new drug to treat
from benefiting from without
multiple sclerosis. Access to the new drug is excludable; it can easily be denied to anyone
great cost. A non-excludable
good is one where people can
not willing or able to pay, even when it would be medically beneficial. But access to the
be stopped from consuming it informat:.on about the drug's effectiveness is not excludable. Once the information about its
only at great cost. effectiveness is available to some members of the general public, there is no realistic way to
Collective Action Problems 419
Table 8.1 Types of Goods

Excludable Not Excludable
Rivalrous Private Goods Common-Pool Goods common-pool resource
hospital beds; tangible products of blood collected by Canadian Blood good a good that is rival
medical research (e.g., MRI machines) Services for transfusion; organs for (in consumption) but is not
transplant; access to family doctor
excludable.
appointments if eligible
Not Rivalrous Club Goods Public Goods
electronically stored medical records; knowledge from medical research; "herd club good a good that is not
access to electronically stored medical immunity"; food and drug safety rival (in consumption) but is
journals excludable.
prevent others from also finding this out (or "consuming" this knowledge). The combina-
tion of these two features (rival/non-rival and excludable/non-excludable) gives us the four
categories of goods in Table 8.1.
Collective Action Problems in Health Care

Collective action problems appear for goods in the second column. Because people collective action problem
cannot be excluded from consuming or benefitting from the good once it is available, any failure of a group of
individuals to achieve a result
there can be free riders and there can be overconsumption or overuse. Free riders use or
(for example, the production
consume the good without paying or contributing sufficiently to its production. Overusers of a good) that everyone
consume or use it in such great amounts that production or supply of the good cannot values at least as much as the
keep up. In Chapter 4, Milton Weinstein in "Should Physicians Be Gatekeepers of Medical result of everyone acting in
Resources?" (see page 181) considers overconsumption of the medical services provided their own self-interest.
by, for example, a Canadian provincial or territorial health plan. In this chapter, Katharine
Browne in "The Measles and Free Riders" (see page 427) and Walter Glannon in "Free free rider a person who
Riding and Organ Donation" (see page 431) consider free rider examples. benefits from or consumes a
good without contributing
It is easy to understand how someone can free ride when it comes to an organ for trans-
sufficiently to its production or
plant or blood for transfusion. Others donate their organs or their blood. These goods are now continued existence.
available to anyone in medical need. No one in Canada is excluded from the benefit because
they do not have a signed organ donor card or have not recently given blood. Anyone who
has an organ transplant or a blood transfusion without contributing enough in some way to
maintain the pool of organs or a blood supply is a free rider. But how can there be free riders
when it comes to vaccinations against an infectious disease like measles, mumps, or polio? Vac-
cinations produce two benefits. The person vaccinated has their own risk of being infected re-
duced. This private benefit is consumed by the person. But the person's risk of infecting others
is also reduced. This secondary benefit goes to others. These others can't be excluded from
getting or consuming it. When enough individuals are vaccinated these secondary benefits add
up to a collective or "herd" immunity, as Browne explains. Someone who gets this secondary
benefit from others being vaccinated without themselves being vaccinated is a free rider.
Assuming that it is wrong to free ride in these cases, both Browne and Glannon con-
sider what is to be done to discourage free riding. In the case of vaccinations and organ
donation, it is often suggested that medical professionals could encourage people to be
vaccinated or to sign a card saying they agree to donate their organs after they die. Some
are not vaccinated because they believe the risk of bad side effects from certain vaccines is
so great that the benefits to themselves or their children, and perhaps even the secondary
benefits to others, do not outweigh the costs to them and their families of being vaccinated.
Although some people have religious or other convictions that forbid them from donating
an organ upon death, in many cases not signing a donor card and free riding is not the result
of considered thought. It turns out, unfortunately, that medical professionals are mostly
unsuccessful at persuading these individuals to change their minds.2 Given this failure, an
II MI 111Mial
8 Public Health
idea that Glannon favours to increase the number of organs donated after death is to adopt
what is called an "opt-out" system. The system now in place in Canada is an "opt-in" one.
Under an opt-in arrangement, people have to indicate—by signing an organ donor card—a
willingness to donate their organs after death. Under an opt-out scheme, they would have
to indicate—perhaps by going somewhere and signing up—an unwillingness to donate. A
far greater percentage of people than are currently registered to donate their organs say that
they like the idea of donating their organs after death.3 An opt-out system would make them
walk their talk Another possibility is to have the government provide incentives for people
to contribute to the production of the good, either the existence of herd immunity in the
case of vaccination, or a pool of solid organs for transplant in the case of organ donation.
This is a solution Browne investigates for vaccinations in her article. In some jurisdictions,
children who are not vaccinated are not allowed to attend school.4 The legal incentives for
organ donation can be as strong. In Israel, even if one patient has spent less time on the
waiting list for a transplant than another, or is not as good a match for an available organ,
priority can go to the one who has indicated a willingness to donate organs after death.5
AT there differences between organ or blood donation on the one hand and vaccina-
tion on the other that will alter what should be done about free riding? Most of us think
that donating an organ or blood is a nice thing to do but nothing we are morally required
to do. We do not harm others unjustly by not donating, Not being vaccinated, however,
is different. Anyone not vaccinated risks infecting others. They risk wrongly infringing the
rights others have not to be harmed. This is the kind of circumstance where John Stuart Mill
(as Chapter 1 explains) would say that the government should intervene. But this difference
between causing harm and not bringing or doing good may not be so great. Imagine, says
James Rachels in "Active and Passive Euthanasia" in Chapter 6 (see page 301), two very
similar but slightly different cases. In one, person A opens the bathroom door and pushes
a baby into a bathtub where it drowns. In the other, person B opens the bathroom door,
finds that a baby is drowning in a bathtub, and does nothing to save it. Person A is surely
violating the baby's right not to be harmed. A non-vaccinating individual is like person A.
Person B simply does not help the baby. This is like the non-organ-donating person But is
person A really acting in any morally worse way than person B in Rachels's examples? If so,
is anyone who is not vaccinated acting in a morally worse way than someone who does not
donate organs after death or blood before death?
There are many more examples of collective action problems in health care. Another
example is related to the topic of Chapter 9—research with humans as subjects. This ex-
ample is mentioned by Glannon in his article on organ donation. As Table 8.1 states, the
knowledge gained from medical research is a benefit often available to everyone. (Some
people, it is true, will benefit more than others.) If the right way for most people to con-
tribute to medical research is to be a subject, then not being a subject is to be a free rider.
Being a subject in medical research is usually thought of as something wonderful to do but
not wrong not to do. But perhaps free riding in this case is as wrong as it is in the case of
not signing an organ donation card.°
Screening
In medical screening, people who have no symptoms of an illness are tested for predictors,
screening test a medical precursors, or even the early presence of the illness. For example, screening tests for many
test that looks for predictors medical problems are offered to all newborns in all provinces across Canada as part of provincial
or the early stages of illness
health care programs.' Very rarely are these tests mandatory, parents can decline to have them
when a person shows no
symptoms.
done if they wish. Most states in the United States, however, require that some of these tests be
done.8 Another example of a screening test is the one for type II diabetes. This is recommended
by Diabetes Canada for certain individuals who are 40 years of age or older or who are at high
risk of getting this illness.9 Provincial health plans cover the cost of this screening test.
Screening
Screening programs certainly seem to be a good thing. What parents would not want
to have their baby tested for a treatable illness like phenylketonuria (an inherited disorder
that causes the build-up of the amino acid phenylalanine, which in turn leads to serious
health problems)? This disease is part of newborn screening in every province. But even if
this particular screening program is beneficial, this does not tell us that every screening test overdiagnosis a correct
is worth encouraging people to have or worth spending public funds on. It also certainly diagnosis of illness (for
does not settle whether being screened should be a voluntary choice, left up to an individ- example, cancer) where the
ual to decide on, or something that is mandatory. Some of these topics are taken up in the illness will do no harm to the
person during that person's
articles included in this chapter.
lifetime.
Let's focus on whether publicly funded voluntary programs should be offered—that is,
programs in which no one is compelled or required to be screened. What are the benefits
overtreatment medical
of these programs? Sometimes illnesses that are not life-threatening can be prevented or
treatment for an illness given
at least have their damage reduced through early screening. This happens in the case of when the illness will do no
glaucoma (a condition that causes damage to the optic nerve). Sometimes screening saves harm to the person during
lives through the early detection of an illness, such as cancer. Sometimes screening saves that person's lifetime.
the lives of many by detecting a communicable infection like HIV or TB. But do these
programs always save lives? Sometimes the way numbers about survival and success rates false positive a test result
are presented for these programs can be misleading regarding lives saved and can lead an that is positive but incorrect.
unwarranted enthusiasm for a screening program. This is what Darren Shickle and Ruth A test result is positive when
thetestsays a patient has the
Chadwick call "screeningitis" in "The Ethics of Screening" (see page 436).
illness being tested for or a
There are also harms from mass screening programs. One type of harm is overdiagnosis predictor for it.
and overtreatment, as they are sometimes called. This kind of problem is considered by the
Shickle and Chadwick article and by Timothy Caulfield in "Direct-to-Consumer Genetics
false negative a test result
and Health Policy" (see page 434). Caulfield concentrates on genetic screening tests like that is negative but incorrect.
those provided by 23andme and other companies here in Canada and around the world, A test result is negative when
but the problem also occurs for other screening tests. Another kind of harm that can result the test says a patient does
from mass screening programs comes from the mistakes that screening tests make. Four not have the illness being
tested for or a predictor for it.
different outcomes or results are possible with any screening test. The result of a screening
test for a disease may be positive (the disease is present according to the test) or negative
(it is not present according to the test). Each of these results may be correct or true on the positive right a right to be
provided with something (a
one hand or incorrect or false on the other hand. A correct or true result means that what
good or benefit) by some
the test says is right, but most medical screening tests get it wrong sometimes, producing institution or person or group.
a false positive or false negative result. Of course, no one knows when using the test It corresponds to a positive
whether a particular result is a true or false one. All that is known is what the screening test duty on the part of that
says and, from research done on the test, the rate at which it gets it correct or not. (If other institution or person or group
to provide the good or benefit
information from a patient's medical record is also available, it can help to resolve some of
to those with the right.
the uncertainty.10) This issue is discussed by Shickle and Chadwick in a general way. Case 3
in this chapter considers what happens in a specific example (breast cancer screening).
negative right a right
Aside from considerations about these harms and benefits for publicly funded screening
not to be stopped by some
programs, what other moral issues might there be? One comes up in the chapter on medical institution or person or group
information (Chapter 3). Suppose people—for example, the parents of babies in Canada— from getting something
have a positive right to information about their own health and that of their small children. (a good or benefit). It
This would seem to mean that they have a right that public health authorities, someone, or corresponds to a negative
duty on the part of that
some group provide them with effective screening tests. On the other hand, if these people
institution or person or group
have only a negative right, then all that public health authorities are morally required to do not to interfere with those
is avoid getting in the way of people paying privately for their own screening tests. This is the with the right from getting
main issue considered by Caulfield in "Direct-to-Consumer Genetic Testing." thegood or benefit.
Harm Reduction Programs
Harm reduction or minimization programs are public health measures aimed at reducing
the harmful effects resulting from people's unhealthy or risky behaviours. These types of
II j11 1111.1,1
422 8 Public Health
programs are not to be confused with ones that have the goal of eliminating unhealthy be-
haviours, which are called "cessation" or "abstinence" or perhaps "zero-tolerance" or "war
on" (as in "war on drugs") programs. An example of a cessation program would be to forbid
bike riding on city roads because people can get hurt riding their bikes. An example of a
harm minimization program, however, would be to allow bike riding while adding special
bike lanes, encouraging riders to wear helmets, and requiring that bikes have appropriate
lights and reflectors for night-time riding.
Harm reduction programs, at least of the kind we are concerned with here, offer medical
and health care services to people who behave in risky or unhealthy ways. Most of us will
think of methadone maintenance programs for those addicted to heroin or needle-exchange
programs and supervised injection sites for intravenous drug users when it comes to these
kinds of programs. Other kinds of harm reduction programs in which medical profession-
als participate include measures and interventions to help sex trade workers avoid sex-
ually transmitted diseases, college students to overcome the after-effects of binge drinking
sessions, and competitive athletes to deal with side effects from performance-enhancing
drug use. None of these programs are attempts to stop the activities; they just deal with the
harmful results.
There are also less obvious examples of harm reduction programs. Suppose you are
diabetic, have a heart condition, or have high blood pressure partly as a consequence
of an unhealthy diet. Or suppose you will need a liver transplant if you continue to
drink alcohol in large amounts. In these scenarios, any medical services provided to
you to reduce or help combat the harmful effects of your behaviour—so long as they
aren't aimed primarily at getting you to stop the behaviour—will be an example of harm
reduction or minimization.
There are many arguments against harm reduction programs such as safe injection
sites and against medical professionals participating in them. One is that health care profes-
sionals will have "dirty hands" if they provide services in these kinds of harm reduction
programs.'2 They are participants or helpers in something that is morally bad, and so they
are dirtying their hands by being involved. For this argument, the real question is whether
the unhealthy behaviour is immoral. In some cases it is certainly illegal. This is why safe
injection sites and some other kinds of harm reduction programs need special legal permis-
sions, as noted by Dan Small in "Canada's Highest Court Unchains Injection Drug Users"
(see page 443). But are the behaviours—using addictive drugs, for example—immoral?
One possibility is that people who use these drugs can't control their behaviour. If these
individuals are not able to control their actions voluntarily, they are not acting immorally.
After all, as the philosopher Immanuel Kant famously said, "Ought implies can."12 If the
behaviours are not immoral, then helping people to avoid the harmful effects of the behav-
iour leaves one's hands clean. This line of thinking is considered by Dan Small in his article
(see page 443) in this chapter.
A different argument is considered in "Virtue Ethics as an Alternative to Deontological
and Consequential Reasoning in the Harm Reduction Debate" by Timothy Christie, Louis
Groarke, and William Sweet (see page 451). This is an argument concerned with the conse-
quences of having these harm reduction programs. The idea is that by removing or helping
to minimize some of the bad effects of unhealthy behaviour like intravenous drug use,
health care professionals and these programs encourage the behaviour and maybe even
increase it. This idea is not completely silly. Going back to the example of riding a bike on
city roads, researchers have found that some bike riders compensate for the safety benefits
wearing a helmet should bring them by riding in riskier ways. In other words, harm re-
duction encourages unhealthy, dangerous behaviour by reducing or removing effects that
would otherwise discourage the behaviour.13 Does the same thing happen with drug users
and sex trade workers? If it does, then harm reduction programs are not so beneficial
after all. Fortunately, the factual evidence we have shows that methadone maintenance or
Harm Reduction Programs
needle-exchange programs and supervised injection sites do not increase drug use. Some
of this evidence is provided in the article by Small.
Some other arguments say there is nothing morally wrong with harm reduction pro-
grams, or with medical professionals offering their services in these programs, and that
these programs should exist. One of these arguments is found in Small's article. It relies on
the reasoning in a judgment in a Supreme Court of Canada case. Small argues—as noted a
few paragraphs ago—that users of addictive drugs like heroin do not freely participate in
their unhealthy behaviour; it is outside their control So, like others whose medical needs
are the result of factors outside their control, addictive drug users should receive as large
a share as any other medical need when the budget for scarce medical resources is being
allocated. Furthermore, their share should be provided in a way that is most medically
beneficial and efficient. That way is harm reduction efforts like methadone maintenance
programs for heroin addicts and clean needle-exchange programs and supervised injection
sites for intravenous drug users.
Equality in Health
As long ago as 1986, Health Canada recognized that certain population groups in Canada
have significantly lower life expectancy, poorer health, and a higher prevalence of disability
than the average Canadian.14 The map in Figure 8.1 has more recent data from Statistics
Canada that shows some of what Health Canada was talking about. In this case, the popu-
lation groups are people living in different health regions across the country." health region a geographical
The map (Figure 8.1) shows that there are differences in expected life spans—an im- area defined for administrative
purposes by a provincial or
portant measure of general health—in different health regions across Canada. Some areas
territorial ministry of health.
are much better than the Canadian average and some much worse. These differences are
Compared to the national

average of 80.7 years
Significantly higher
M Not significantly different
=I Significantly lower
Sparsely populated
titer*
. • .47
Regina Winnipeg 11.011)..
AA.Charlottetown
Fredericton 1/7 Halifax
Ottawa
Figure 8.1 Expected Life Span Differences across Canada

Source: Statistics Canada. Vital Statistics. cansinATable 102-4307. Life expectancy. Produced by the Geography Division for Health Statistics Division, Statistics
Canada, 2011.
II .111 1111:41
8 Public Health
greater than just accidental chance would allow. This phenomenon is repeated on a smaller
scale too—that is, inside what Statistics Canada calls health regions. In Vancouver, where
the editors of this textbook live, the Vancouver Coastal Health regional health authority
says that those living in Vancouver—Westside are expected to live 11.5 per cent longer than
those in Vancouver—Downtown-Eastside.i°
Are these differences or inequalities in health unjust or unfair? Differences, even ones
as dramatic as these, do not by themselves mean that there is injustice or unfairness. As
Norman Daniels tells us in "When Are Health Inequalities Unjust?" (see page 458), they
are unjust only if what causes them is something that could have been changed or done
differently. So what is responsible for these differences in health status in various regions of
the country? Are they things that could have been done differently? We know, for example,
that there are great differences in the distribution of medical care, differences considered in
Chapter 7. Urban areas usually have much better health services—larger, easily accessible
hospitals, among other things—than rural or sparsely populated areas. This situation is
certainly part of the story, and it is of course something that could be changed by govern-
ment action. But according to the Canadian Medical Association (CMA) and many social
scientists, the distribution of medical care is actually only a small part of the explanation
for the dramatic differences among health regions.'7 Figure 8.2 shows what the CMA tells
us is responsible for the health of Canadians and for the great population-level differences
in average health.
As Figure 8.2 notes, a substantial proportion of what contributes to individual
and population-level differences in health are factors like income, education, housing
WHAT MAKES
CANADIANS SICK?
INCOME
EARLY CHILDHOOD DEVELOPMENT
DISABILITY
EDUCATION
SOCIAL EXCLUSION
SOCIAL SAFETY NET
50% > YOUR LIFE GENDER

EMPLOYMENT/WORKING CONDITIONS
RACE
ABORIGINAL STATUS
SAFE AND NUTRITIOUS FOOD
HOUSING/HOMELESSNESS
COMMUNITY BELONGING
ACCESS TO HEALTH CARE

25% > YOUR HEALTH CARE HEALTH CARE SYSTEM
WAIT TIMES
15% ~•> YOUR BIOLOGY

BIOLOGY
GENETICS
10% YOUR ENVIRONMENT

AIR QUALITY
CIVIC INFRASTRUCTURE
Figure 8.2 What Makes Canadians Sick

Source: Canadian Medical Association. 2013 ."Health Equity a n d the Social Determinants of Health:' https://www.cma.ca/En/
Pages/health-equity.aspx.
Equality in Health 425
conditions, and other aspects of a person's daily life. Some of these are what social scien-
tists call the social determinants of health.'7 But Statistics Canada tells us that there are social determinants of
disadvantages in these social determinants in the health regions where people have shorter health the social and
economic conditions in
life expectancies and corresponding advantages in the health regions where people have
which people are born, grow,
longer life expectancies.'8 The deprived regions have in common lower incomes, high live, work, and age that have
long-term unemployment rates, lower levels of education, small immigrant populations, effects on their health.
and relatively large Aboriginal populations. So a substantial proportion of what contributes
to the population-level health differences shown in the map in Figure 8.1 are factors like
these advantages or disadvantages in social determinants. Can these factors, which are
largely not the result of the behaviour or habits of those living in these regions, be changed
and improved? Surely some can be, as Daniels tells us. But Statistics Canada also notes that
these disadvantaged regions also tend to be ones with some of the highest rates of smok-
ing, obesity, and heavy drinking rates in Canada.18 Thus, some of the factors that produce
the health problems in these locations are the result of behaviour. Can these behaviours
be changed? Perhaps they can be—as Daniels suggests—and largely by improvements in
social and economic circumstances.
Clearly, some things could be changed to eliminate the differences in health among
people living in these regions. However, that fact is not enough to prove that the inequal-
ities are unjust inequities needing to be changed. What more is needed to prove that
point? Two quite different answers to this question can be found at various points in the
article by Daniels. One of the answers is backwards-looking. It argues that population-
level differences in health are unfair and should be corrected if these differences resulted
from controllable and unjust past disadvantages for the populations with worse health—
for example, if they were produced by laws or practices that discriminate unfairly on
the basis of race, ethnicity, social class, or gender. The other answer is forwards-looking:
population-level differences in health are unfair and should be corrected if these differ-
ences produce future controllable and unjust disadvantages for the populations with
worse health. For example, the inequalities are unjust in a forwards-looking way if they
put people at a disadvantage when it comes to competing on equal terms with others
for economic and social goods like a spot in medical school or a good job after gradua-
tion. These are not the only possible answers. Some authors (not included here) defend
more radical ideas.'9 For example, according to one radical backwards-looking answer,
population-level differences in health are unfair and should be corrected if these differ-
ences resulted from any past disadvantages for the populations with worse health. On
this view, it doesn't matter whether the disadvantages were unfair or unjust. (They do on
one of the answers Daniels provides.) They could, for instance, be the result of bad luck
rather than factors like discrimination.
One question that can be asked about all of these views—backwards-looking or
forwards-looking, radical or not so radical—is whether what is important is absolute or
comparative health deprivation. Here is what this means. The average life span for Can-
adians living in disadvantaged health regions is just over 71 years, whereas the expected
life span is nearly 85 years in the most advantaged health regions.18 A moral approach
that focuses on comparative health deprivation sees this inequality or difference as the key
focus. If the expected life span in every health region were pretty much the same—say,
around 74 years—there would be nothing to complain about morally. There would be
no unfair difference because there would be no difference. Of course, it might be better
if the average life span were longer, perhaps even nearing the 84 or 85 years of the most
advantaged health regions today, but in terms of comparative inequality in health, a longer
life span would not be a moral improvement or anything required by justice. On the other
approach, absolute health deprivation, the key focus is on the fact that the health of some
groups—their expected life spans, rates of disability, and other aspects of average general
health status—is much worse than it could be. The better health status of those in advan-
taged regions shows how much better the health status of the disadvantaged groups could
II .1II IIILLI
8 Public Health
be if something in their lives or situations were different and improved. Therefore, justice
requires not so much erasing the differences between those health regions as bringing all
regions up to the level that those regions with the best of health show is possible.
Notes
1. Gregory Mankiw, Microeconomics, 5th edn, ch. 11 (Mason, OH: South-Western Cengage
Learning, 2009).
2. Brendan Nyhan and John Reifler, "Does Correcting Myths about the Flu Vaccine Work? An
Experimental Evaluation of the Effects of Corrective Information," Vaccine 33 (2015), 459-64; Eve
Dube et al., "Understanding Vaccine Hesitancy in Canada: Results of a Consultation Study by the
Canadian Immunization Research Network," PlosOne 11 (2016).
3. Richard H Thaler, "Opting in vs Opting out," The New York Times (26 Sep 2009). Ipsos Reid,
"Canadians willing to donate their organs in the case of their death" (11 Jul 2006). Available at
http://www.marketwired.com/press-release/canadians-willing-to-donate-their-organs-in-the-case-of-
their-death-603162.htm.
4. Heather MacDougall, Laurence Monnais, and Monica Brown, "The Big Picture: Historical, Social
and Cultural Features of Vaccine Hesitancy," presentation to the Canadian Public Health Association
conference, Toronto, June 2016.
5. Jacob Lavee, Tamar Ashkenazi, Gabriel Gurman, and David Steinberg, "A new law for allocation of
donor organs in Isreal," The Lancet vol 375 (2010), 1131-1133.
6. Fr tz Allhoff, "Free Riding and Research Ethics," American Journal of Bioethics 5 (2005), 50-1;
G. Owen Schaefer, Ezekiel Emanuel, and Alan Wertheimer, "The Obligation t o Participate in
Biomedical Research," Journal of the American Medical Association 302 (2009), 67-72.
7. Lift Grainger, "Newborn Screening in Canada: What to Expect," Today's Parent 20 October 2014;
Canadian Organization for Rare Disorders, "Newborn Screening in Canada Status Report," Canadian
PKJ and Allied Disorders, September 2015.
8. National Newborn Screening and Global Resource Center, "US Newborn Screening Programs,"
http://genes-r-us.uthscsa.edu/resources/consumer/statemap.htm.
9. Canadian Diabetes Association Clinical Practice Guidelines Expert Committee, "Screening for Type 1
and Type 2 Diabetes," http://guidelines.diabetes.ca/browse/chapter4.
10. This uses information from the medical record to establish a prior probability for having or not
having the condition being screened for.
11. John Kleinig, 'The Ethics of Harm Reduction," Substance Use and Misuse 43 (2008), 1-16.
12. Immanuel Kant, 1998, Critique of Pure Reason, p. A548/13576 (Cambridge and New York: Cambridge
University Press).
13. Jake Olivier and Prudence Creighton, "Bicycle Injuries and Helmet Use: A Systematic Review and
Meta-Analysis," International Journal of Epidemiology 46 (2017), 278-92; Tim Gamble and Ian
Walker, "Wearing a Bicycle Helmet Can Increase Risk Taking and Sensation Seeking in Adults,"
Psychological Science 27 (February 2016), 289-94; Ross Phillips, Aslak Fybri, and Fridulv Sagberg,
"Risk Compensation and Bicycle Helmets, Risk Analysis 31 (2011), 1187-95.
14. J. Epp, Achieving Health for All: A Framework for Health Promotion (Ottawa: Health and Welfare
Canada, 1986); Dennis Raphael, "Health Inequalities in Canada: Current Discourses and
Implications for Public Health Action," Critical Public Health 10 (2000), 193-216.
15. Other population-level differences exist in Canada. See Juha Mikkonen and Dennis Raphael, Social
Dererminants of Health: The Canadian Facts (Toronto: York University School of Health Policy and
Management, 2010); Margo Greenwood, Saraj de Leeuw, Nicole Lindsay, and Charlotte Reading,
ed5.., Determinants of Indigenous Peoples' Health in Canada, (Toronto: Canadian Scholars' Press, 2015);
Pat Armstrong and Ann Pederson, eds, Women's Health, 2nd edn. (Toronto: Women's Press., 2015).
16. Ken MacQueen, "Vancouver's Downtown Eastside Gets New Lease on Life," Maclean's 15
October 2012.
17. Dennis Raphael, ed., Social Determinants of Health: Canadian Perspectives, 3rd edn., (Toronto:
Canadian Scholars' Press, 2016).
18. Lawson Greenberg and Claude Normandin, "Disparities in Life Expectancy at Birth" (Ottawa:
Statistics Canada, 2015), http://www.statcan.gc.ca/pub/82-624-x/2011001/article/11427-eng.htm.
19. Shlomi Segall, "Affirmative Action in Health," in Shlomi Segall, Equality and Opportunity, (New York:
Oxford University Press, 2013), ch. 3.
Browne: The Measles and Free Riders: California's Mandatory Vaccination Law
8.2 Collective Action Problems
The Measles and Free Riders: of further spread of the disease is minimized, because
there are fewer transmission points (i.e., nonimmune
California's Mandatory individuals) by which the virus can spread? Achieving
Vaccination Law herd immunity helps to protect vulnerable members
of the population who cannot be vaccinated, such as
Katharine Browne
infants and individuals with certain health issues.
The percentage of the population that needs to be
Introduction vaccinated for herd immunity to obtain varies, depending
on the illness. In most cases, herd immunity obtains
In response to the measles outbreak—famously when at least 90% of the population is vaccinated against
linked to Disneyland—and to the growing number an illness. In order to achieve herd immunity from the
of unvaccinated children, California recently passed measles, which is highly contagious, 96%-99%3 of the
legislation to eliminate all nonmedical exemptions population needs to be vaccinated. The World Health
for school-mandated vaccines.' According to the Organization (WHO) estimates that the United States
new legislation, children who do not receive school- national average for vaccination against the measles is
mandated vaccines and who do not have a medical 91%.4 In 2013-2014, 92.3% of kindergarten children
excuse will be prohibited from attending public and in California were vaccinated with the vaccine against
private schools. This makes California the largest state measles—the MMR (measles-mumps-rubella) vaccine.5
to restrict vaccine exemptions to those that are medic- A recent analysis of the data from the Disneyland-linked
ally necessary and could set a precedent that would measles outbreak revealed an estimated vaccination rate
pave the way for other states to follow suit. However, of 50% to 86% among those exposed to the illness.6 It is
the legislation has been met with significant resistance. thus clear that vaccination levels are lower than would
Some opponents to it have claimed that implementing be desirable to guarantee herd immunity against the
mandatory vaccinations violates medical freedom and measles. What is not clear is what should be done about
parental autonomy, emphasizing that wherever there this. The debate over mandatory vaccination is a heated
are risks associated with vaccination, there should be one, often characterized as one between science and its
choice regarding whether to vaccinate. opponents. Antivaccinators are often characterized as
In this article I take up a game-theoretic perspective selfish, ill-informed, and irrational. And, although many
on the problem of vaccination compliance. Such an ap- of the claims on which the antivaccine movement rests
proach characterizes parental decisions to vaccinate their are arguably without scientific credibility,' it is not true
children as a collective action problem. Collective action that vaccines carry with them no risk, and there remains
problems reveal the presence of an incentive to free ride— a legitimate question about the extent to which the state
to benefit from the efforts of others who vaccinate their can interfere with people's choices to vaccinate. On the
children without carrying any of the costs of vaccinating one hand, if fewer people are vaccinated, herd immun-
one's own. I argue that California's legislation represents ity drops, and this puts everyone at risk. On the other
a reasonable means to overcome the free rider incentive hand, it is desirable for coercive measures to be kept to
and improve compliance with vaccination requirements. a minimum whenever possible. Thus, it would be best
Although the focus of this article is on California's legisla- if compliance with vaccination requirements could be
tion, this conclusion has implications for other jurisdic- secured voluntarily.
tions where vaccination compliance is a problem.
Securing Voluntary Compliance
Herd Immunity, Vaccination, and the Law
To see if and how voluntary vaccine compliance might
The Holy Grail in the vaccine world is to achieve what is be achieved, consider the following. Parent A is decid-
known as "herd immunity" This describes the point at ing whether to have her children vaccinated against the
which the majority of a population is immune to a par- measles. Factors relevant to her decision may include
ticular illness. Once herd immunity is achieved, the risk the following: (1) the risk associated with the vaccine,
Source: From Cambridge Quarterly of Healthcare Ethics 25, 3 (2016), 472-8. Reprinted with permission.
II 'Al HILL!
8 Public Health
(2) the effectiveness of the vaccine in protecting against those choices.'2 Game theory is the study of strategic
the measles, (3) the risk of contracting the measles interaction among rational individuals. The players in
without the vaccine, and (4) the risk of complications the game are taken to be ideally rational, where this is
arising from the measles. usually understood to mean they are expected utility
The antivaccine movement tends to latch on to maximizers. In a strategic choice situation involving
the first factor concerning vaccine safety and has been two or more rational players, an individual will choose
fueled largely by the (discrechted8) link drawn between on the basis of what she expects her opponent(s) to do:
the MMR vaccine and autism. Research concerning par- her action will constitute a best reply to the action she
ental attitudes toward vaccination reveals that parents anticipates the other player(s) will choose.
may also deem vaccination unnecessary, either because Consider Parent As decision whether to vaccinate
they assign a low probability to the possibility of their her children against the measles with the MMR vaccine.
children contracting the disease or because they believe Assume, for the sake of the argument, that the MMR vac-
they can control whether their children are exposed to cine is 100% effective" at guarding against the measles;
the disease.9 The legacy of successful vaccination pro- that when more than 96% of the population vaccinates
grams is that the dangers of illnesses like the measles their children, herd immunity is achieved; that wher-
are unfamiliar to parents and risks of exposure and ever there is herd immunity there is a 0°k chance of
complications may be perceived as low. contracting the measles; that when less than 96% of the
Insofar as these factors impede achieving herd population vaccinates their children, there is no herd
immunity, one strategy to boost voluntary vaccination immunity, and that where there is no herd immunity
compliance would be through public health efforts de- the chance of contracting the measles is 100%.
signed to educate parents about the risks and benefits of The matrix in Table 1 represents Parent As choice
vaccination and the dangers of contracting the measles. situation, where "benefit" stands for the benefit of im-
However, it is not clear that such education programs munity from the measles, "cost" for the myriad costs
will be sufficient, for two reasons. First, there is little em- associated with vaccination, and "measles" represents
pirical evidence that parental attitudes can be changed harms associated with contracting the measles.
through public education efforts. Although there is a Parent A can choose between two strategies: "vac-
growing literature concerning reasons for parental re- cinate" or "don't vaccinate." The first column of the
fusal of vaccination, there remains surprisingly little data matrix represents Parent As choice situation when there
on which interventions are most effective in reversing is herd immunity. In such a case, the option to vaccinate
vaccination ref usal.'° Furthennore, the data that do exist carries with it both the benefit of immunity from vac-
suggest that parental attitudes concerning vaccination cination and the costs of vaccination. This cost can range
are not readily changeable. A recent American study of from the inconvenience of going to the doctor to obtain
vaccine promotion revealed that education about vac- the vaccine or more serious side effects associated with
cination tends not to influence parents to choose to vac- the vaccine. Importantly, however, when there is herd
cinate and, in fact, often has the opposite effect.il immunity the individual benefit of immunity obtains
Second, educational efforts may not address all rel- regardless of whether Parent A vaccinates, because pro-
evant factors involved in parents' decisions to vaccin- tection against the measles is already achieved by the
ate their children. Indeed, it is the third factor named efforts of the 96% of the population who vaccinate their
previously—the risk of an unvaccinated individual children. Because, in this scenario, "don't vaccinate" im-
contracting the measles—that is especially problem- poses no cost to Parent A but permits enjoyment of the
atic. More specifically, the facts that (1) the risk of con- benefit of herd immunity, Parent A should choose "don't
tracting the measles is dependent on herd immunity vaccinate." In so doing, Parent A will be free riding off
and that (2) herd immunity is dependent, largely, on of the efforts of the 96%—enjoying the benefits of herd
whether other parents choose to vaccinate their chil- immunity from vaccination without bearing any of the
dren pose a serious challenge to securing voluntary costs associated with vaccinating.
vaccination compliance.
Table 1 Parent A's Choice Situation
Game Theory and Vaccination > 96% vaccinate < 96% vaccinate
(herd immunity) (no herd immunity)
The interdependence of parental choices to vaccinate
(that is, the fact that one parent's choice to vaccinate Vaccinate Benefit + cost Benefit + cost
will be influenced by the choice of another, and vice Don't vaccinate Benefit + no cost No cost + no benefit +
measles
versa) sets the stage for a game-theoretic analysis of
The second column in the matrix represents occasional measles outbreaks like the one that started
Parent A's choice where there is no herd immun- in California. But this still leaves open the possibil-
ity. Here "vaccinate" carries with it the same costs as ity of free riding, and there are good reasons to close
before but confers the benefit of immunization that that opportunity. First, eliminating free riding ensures
would otherwise be unavailable. If a parent chooses the that no one takes unfair advantage of the collective
option "don't vaccinate," she and her children avoid efforts of others and everyone shares in the burden.
the costs associated with vaccinating, but the children Second, permitting free riding increases the risk of
face the potential harms associated with contracting the infection in the community and thus puts vulnerable
measles. Her choice to vaccinate in this situation then members of the population—those who, for medical
depends on whether the potential harms associated reasons, cannot be immunized—at risk of contract-
with the measles are greater than the costs associated ing the illness. Finally, the risks of measles for anyone
with vaccination. should not be minimized. Although there have been
In this matrix, when 96% of the population vac- no measles-related deaths reported to date in 2015 in
cinates, Parent A, insofar as she is rational, will reason the United States, from a global perspective there has
to the conclusion that it is to her advantage to refrain been significant mortality over the years. For example,
from vaccinating. That way she could enjoy the bene- in 2013, there were 145,700 measles-related deaths
fits of herd immunity without bearing any of the costs worldwide.14 Meanwhile, the WHO has estimated that
associated with receiving the vaccine. This is a free vaccination prevented 15.6 million measles-related
rider temptation. Thus, the basic logic in the case of deaths between 2000 and 2013.15 There are also (often
vaccines is this. If a sufficient number of others get vac- serious) complications that accompany the measles,
cinated, then everyone will be better off. But it would such as hearing loss, respiratory infections such as
be best for any particular individual if all others got pneumonia, and encephalitis.16
vaccinated and he or she did not. The trouble is that An alternative, or possible supplement, to educa-
if everyone thought that way, no one would become tional efforts would be to appeal to the common good in
vaccinated. There would be no herd immunity, and order to encourage vaccination. The free rider problem
everyone would be at risk of falling ill. emerges because of a tension between individual self-
In this picture of vaccination as a collective action interest and the good of the collective. If that tension
problem, it emerges that public education about the could be resolved, then the problem would disappear.
safety and efficacy of vaccination is of limited value in So if people can be encouraged to care about the wel-
encouraging parents to vaccinate their children. Until fare of the collective rather than just their own (or that
the threshold for herd immunity is met, all unvaccin- of their children), then vaccination may be achievable
ated individuals are at risk of contracting the measles, without the use of force. But to the extent that parents
and education can show parents that on a simple cost- are concerned with their own children's welfare and
benefit analysis they should vaccinate their children. not necessarily the common good (or the welfare of
However, once herd immunity is achieved, it is rational other people's children), and to the extent that such
for parents to refrain from vaccinating their children, concerns are not always readily changeable, appealing
and no amount of education about the risks and bene- to those elements will prove ineffective. Add to this
fits of vaccination can show otherwise. equation a risk of serious side effects from vaccinating
one's own children, and appeals to the collective good
Policy and Free Riders will quickly lose force.
This suggests that heavier-handed interventions
Ensuring that populations are adequately protected may be needed to ensure vaccination compliance. Such
against vaccine-preventable illnesses like the measles measures are common in other policy fields. Take, for
requires a means to both achieve and maintain herd im- example, littering. Although it is best for everyone if
munity. One policy option is to continue with efforts people refrain from littering in the streets, an individual
at education about the risks and potential benefits of person's small contribution of litter makes minimal dif-
vaccination but allow the exemption of conscience ference to the cleanliness of her city's streets. There is
(i.e., for philosophical and religious reasons) and a temptation to free ride. But if everyone reasons this
otherwise take no measures that go beyond the volun- way, no one will use a trashcan, and the city's streets will
tary. That would preserve individual freedom (keep- become full of litter. Because self-interest will not guar-
ing antivaccinators happy) and may help to achieve antee clean streets, and because appeals to the common
herd immunity but would do nothing to maintain good of keeping the city's streets clean sometimes prove
it once it is achieved. The likely result would be ineffective, fines are implemented for littering.
II J11 111111
8 Public Health
Following this logic, we might say that similarly incentives to generate the desired result—is that it priv-
heavy-handed interventions may be needed in the case ileges those who can afford to forgo the state-sponsored
of vaccines. At the extreme end of the spectrum of such financial benefit. Consequently, the burdens of securing
measures would be forced vaccination—administering the public good of herd immunity may not be equally
vaccination against the will of parents or children. shared across the population because of disparities in
But to the extent that public policy should aim at the wealth. The rich can free ride off the efforts of the poor.
mildest measure possible to produce the desired result,
such an intervention should be a last resort, and milder Conclusion
measures should be tried first.
California's proposed legislation is one such meas- A full comparative analysis o f the myriad mechan-
ure. To be sure, it does have disadvantages. It runs the isms by which governments can and should encour-
risk that parents who remain opposed to vaccination age measles vaccination lies beyond the scope of this
will remove their children from the school system article. Having said that, a game-theoretic approach
instead of complying with mandates. The alternative demonstrates that legislative measures similar to those
to public or private schooling is home schooling. If implemented in California, although perhaps im-
unvaccinated children are homeschooled together, this perfect, are defensible on both pragmatic and moral
may jeopardize herd immunity, especially given clus- grounds. Pragmatically, they provide a good chance of
tering effects, whereby unvaccinated individuals tend achieving herd immunity in the presence of free rider
to pool together, which increases the probability of a temptations. Morally, closing the loopholes that permit
disease outbreak.17 Furthermore, this alternative runs free riding ensures that everyone shares in the burden
the danger of disadvantaging the unvaccinated children equally. Furthermore, in the context of the United
by depriving them of an education to which they are States, Californiab legislation has the advantage over
entitled. Children who are homeschooled may also be Australia's policy in being an extension of the famil-
deprived of socialization tha: they would receive in the iar and generally accepted practice of requiring those
school system. who attend school to be vaccinated, altering only the
There are alternatives to California's proposed grounds on which one can object to vaccination.
legislation. Australia, for example, has recently an- To be sure, voluntary compliance with vac-
nounced its "no jab, no pay" policy, which prevents cination requirements would be most desirable. By
parents who refuse vaccination for philosophical or re- removing all nonmedical vaccination exemptions, the
ligious reasons from receiving annual tax and childcare proposed legislation infringes on the conscience of
benefits.' The benefits forgone by parents who refuse parents (as any policy that toughens up on exemptions
immunization may amount to as much as $11,000 would). However, insofar as the alternative to closing
per child.19 Whether this intervention will be effective those exemptions is periodic outbreaks—and that is
in boosting vaccination compliance is yet to be seen. unacceptable—compromises to consciences are un-
However, independently of its effectiveness, the main avoidable. California's legislation is arguably the best of
drawback to Australia's policy—or, for that matter, imperfect alternatives and represents a defensible first
to any kind of intervention that relies on economic attempt to achieve and maintain herd immunity
Notes
1. S.B. 277,2015-16 Leg., Reg. Sess. (Cal. 2015); available 5. Seither R, Masalovich S, Knighton CL, Mellerson J,
at http://www.leginfo.ca.gowpub/15-16/ bil]/sen/sb_0251- Singleton JA, Greby SM. Vaccination coverage among
0300/sb_277_bill_20150219_introduced.pdf (last children in kindergarten—United States, 2013-14 school
accessed 1 Sept 2015). year. Morbidity and Mortality Weekly Report 2014 Oct
2. Fine P, Eames K, Heymann DL "Herd immunity": A rough 17;63(4):913-20; available at http://wwwcdc.gov/mmwr/
guide. Clinical Infectious Diseases 2011;52(7):911-16. preview/mmwrhtml/mm6341a1.htm (last accessed 1
3. Plans-Rubi6 P Evaluation of the establishment of herd Sept 2015).
immunity in the population by means of serological 6. Majumder MS, Cohn EL, Mekaru SR, Huston JE,
surveys and vaccination coverage. Human Vaccines & Brownstein JS. Substandard vaccination compliance and
Immunotherapeutics 2012;8(2):184-8. the 2015 measles outbreak. Journal of the American Medical
4. World Health Organization. Global Health Observatory Association Pediatrics 2015;169(5):494-5.
repository: Measles (mcv) data by country; available 7. Wakefield AJ, Murch SH, Anthony A, Linnell J, Casson
at http://apps.who.int/gho/data/node.main.A826 (last DM, Malik M, et al. Ileal-lymphoid-nodular hyperplasia,
accessed 1 Sept 2015). non-specific colitis, and pervasive developmental disorder
in children. Lancet 1998;351(9103):637-41. Retraction in: 13. According to the Centers for Disease Control and
Lancet 2010;375(9713):445. Prevention, one dose of the MMR vaccine is 93% effective
8. E.g., Taylor B, Miller E, Farrington C F, Petropoulos MC, and two doses are 97% effective. Centers for Disease
Favot-Mayaud I, Li], et al. Autism and measles, mumps, Control and Prevention. Measles Vaccination; 2015 Apr 7;
and rubella vaccine: No epidemiological evidence for a available at http://www.cdc.gov/measles/vaccination.html
causal association. Lancet 1999;353(9169):2026-9; Jain A, (last accessed 1 June 2015).
Marshall], Buikema A, Bancroft T, Kelly JP, Newschaffer 14. World Health Organization. Measles fact sheet; 2015 Feb;
CJ. Autism occurrence by MMR vaccine status among US available at http://www.who.int/ mediacentre/factsheets/
children with older siblings with and without autism. fs286/englast accessed 1 Sept 2015).
JAMA 2015;313(15):1534-40. 15. See note 14, World Health Organization 2015.
9. Smailbegovic MS, Laing GL, Bedford H. Why do parents 16. See note 14, World Health Organization 2015.
decide against immunization? The effect of health 17. Salathe M, Bonhoeffer S. The effect of opinion clustering
beliefs and health professionals. Child: Care, Health and on disease outbreaks. Journal of the Royal Society Interface
Development 2003;29(4):303-11. 2008;5(29):1505-8.
10. Sadaf A, Richards JL, Glanz J, Salmon DA, Omer SB. 18. Ireland J. Abbott government closes religious
A systematic review of interventions for reducing loophole for children's vaccinations. Sydney
parental vaccine refusal and vaccine hesitancy. Vaccine Morning Herald 2015 Apr 19; available at httpf/
2013;31(40):4293-304. www.smh.com.aufiederal-politics/political-news/
11. Nyhan B, Reiner J, Richey S, Freed GL. Effective messages abbottgovernment- closes-religious-loophole-for-
in vaccine promotion: A randomized trial. Pediatrics childrens-vaccinations-20150419-1mo4rv.html?utm_
2014;133(4):e835-42. content=bufferd74d36zutm_medium=social&utm_
12. For other game-theoretic treatments of vaccination, source=twitter.com&utm_campaign=buffer (last accessed
see Bauch CT, Earn DJD. Vaccination and the theory of 1 Sept 2015).
games. Proceedings of the National Academy of Sciences 19. Izadi E. Religious vaccination exemptions will
of the United States of America 2004;101(36):13391-4; completely end in Australia. Washington Post 2015 Apr
Shim E, Grefenstette]] , Albert SM, Cakouros BE, Burke 20; available at httpi/www.washingtonpost.com/news/
DS. A game dynamic model for vaccine skeptics and acts-of-faith/wp/2015/04/20/ religious-vaccination-
vaccine believers: Measles as an example. Journal of exemptions-will-completely-end-in-australia/ (last
Theoretical Biology 2012;295:194-203. accessed 1 Sept 2015).
Free Riding and Organ Donation Childress, claim that an opt-out policy "would not
likely be adopted in the United States, and if it were
Walter Glannon adopted, it probably would not increase the number
of organs for transplantation because so many citizens
With the gap between the number of transplantable would opt out."4
organs and the number of people needing transplants Offering financial incentives such as paying fu-
widening, many have argued for moving from an neral expenses for a deceased donor's family is one way
opt-in to an opt-out system of deceased organ dona- of increasing the number of organs. A regulated market
tion. In the first system, individuals must register their in organs is another.' In the UK at least, there is broader
willingness to become donors after they die. In the support for presumed consent. Among the alternatives
second system, it is assumed that individuals wish to to opting in, this system has the best prospect of even-
become donors unless they have registered an objec- tually being adopted. Presumed consent would not
tion to donation. Opting out has also been described obviate but retain the basic idea of consent, because
as presumed consent. Spain has had the most success- individuals would still have the right to opt out and
ful presumed consent policy, resulting in a substantial choose not to donate.6
increase in the donation rate.' Despite support for One argument for presumed consent is that it
an opt-out system from the British Medical Associa- would reduce the incidence of free riding. For a free
tion and other groups since 1999, the report of a task rider, it is not rational to contribute to the organ pool
force delivered in November 2008 recommended that because one's donation has a negligible effect on the
the current opt-in system in the UK be retained.2 The number of transplantable organs, the personal cost of
USA has also retained an opt-in system.3 Two promin- donating is too high, and one could refuse to donate
ent American bioethicists, Tom Beauchamp and James but have equal access to an organ transplant if one
Source From J Med Ethics (2009), 35, 10: 590-591. Reprinted with permission.
II au 111111
8 Public Health
developed organ failure.' The third reason is espe- they will not. In either case, not donating seems to
cially significant because in the current opt-in system be a rational choice for an individual. Dead donors
access to a transplant is not conditional on donation. cannot benefit from an increase in the supply of
If it were, then it would be rational to donate. In the organs, because by the time they give up their organs
absence of such a condition, and given differences in they have ceased to exist. Mere willingness to donate
people's beliefs, it may be too strong to claim that one does not make a difference to the organ supply; only
is rationally required not to donate. A weaker and more actual donation does. However, this assumption is
plausible claim is that not donating is rationally permit- mistaken because the effect of donating or failing to
ted. Many refuse to donate organs after death because donate is not limited to the time when one dies. If
of ungrounded fears about the transplant system. On a person is in organ failure and needs a transplant,
rational grounds, though, some may find the current then his or her access to a transplant now is possible
opt-in system too costly in terms of the time and effort because of the decision of another person to donate
required to register as an organ donor. For those with before the need for the transplant arose. Stating on a
strong religious objections to deceased organ donation, donor registry or driver's licence that one would like
refusing to donate is rational because this decision is to give up one's organs for transplantation does make
consistent with their belief in the sanctity and integ- a difference. Many donors die unexpectedly, and stat-
rity of the body after death. There are also divergent ing at an earlier time that one wishes to donate one's
views on donation among families. Whereas some organs after death enables organ procurement. The
families may find meaning in donating a deceased result of more individuals stating that they wish to be
child's or spouse's organs, others may be distressed by deceased donors is more transplantable organs. The
the thought that donating would violate their belief in effect of an individual refusing to donate might seem
the bodily integrity of a loved one. This psychological negligible with respect to the total number of organs,
cost would not be outweighed by any benefit to trans- but many people opting out would create a collective
plant recipients or any psychological benefit to families action problem, the result of which would be fewer
from knowing that the transplanted organs saved other organs than the already limited number currently
people's lives. For those who object to donation, pre- available.8 Fewer organs would mean that a person
sumed consent would not threaten individual auton- developing organ failure and needing a transplant
omy and the right to make decisions regarding one's would have to wait longer to receive one. One's con-
body. They would have to make the time and effort to dition could deteriorate while waiting to the point
express an objection to having their organs procured at which one would no longer be a candidate for a
for transplantation. transplant and would die. In light of the increasing
In all of these cases, Tnrsonal cost rather than probability of organ failure as one ages and the col-
risk is the core issue, as there is no risk for an indi- lective effect of many individuals refusing to donate,
vidual once he or she has been declared dead. This free riding by opting out of organ donation may not
distinguishes deceased organ donation from live organ be a rational choice.
donation. The risk of short and long-term morbidity Hypertension and type 2 diabetes have been in-
and mortality in live kidney donation, for example, is creasing among adults in developing and developed
low; but it is not zero. Rationality alone is sufficient to countries, and these conditions are the main causes of
explain why so many people do not donate. It is not kidney failure. In many cases, these conditions are not
because people fail to act on a moral duty but because predictable and can develop in a short period of time.
donation is not in their own best interests. The main Hepatitis C and other viral infections have increased
problem with free riding is rot that any benefit from the incidence of liver failure, although this is not as
contributing to the organ pool is not worth the cost. high as the incidence of kidney failure. One might think
Rather, free riders know that they can benefit from that it is irrational to contribute to the total number of
others donating without having to incur any costs of organs because it is unlikely that one would develop a
their own. This may be unfair to those who donate, condition requiring a transplant. Healthy young adults
but it seems rational for one not to contribute to a col- may reason in this way. Yet as we age the probability of
lective good to which one will have access if one needs diseases such as hypertension and diabetes increases
a transplant. among the general population, and the probability of
This idea rests on the crucial assumption that an individual developing one of these diseases and
one's willingness to donate makes no difference to kidney failure increases as that individual ages. Re-
one's access to organs. Eitl- er a sufficient number fusing to donate may be rational the younger one is
of other people will contribute to the organ pool or because it is unlikely that one would develop kidney
Glannon: Free Riding and Organ Donation 433
failure at an earlier stage of life. Yet given the increas- optout system would be consistent with this sense
ing probability of kidney failure as we age, refusing to of obligation, but many will fail to discharge any
donate may not be so rational at a later stage of one's presumed obligation to donate. Moral obligations
life. This probability may be low in absolute terms. Yet are not socially or legally enforceable and are not
in light of the collective effect of many people failing subject to sanctions if they are not discharged. Be-
to donate over time, the cost of donating relative to cause no one can be forced to give up their organs,
the cost of not receiving a transplant makes donation a and because access to a transplant is not conditional
rational choice. on willingness to donate, those who choose not to
The effect of many people opting out would fur- donate would not be denied a transplant for making
ther reduce the availability of organs and limit access this choice.
to a transplant if one developed kidney or other Still, the question is not whether there is a moral
types of organ failure. Each person may appear to be obligation to donate, but whether it is rational for self-
better off by acting alone; but all people are worse interested individuals not to donate. Regarding rational-
off by acting collectively in the same way. Clearly, the ity, there are two differences between participation in
cost of not receiving an organ transplant and dying medical research and organ donation. It may be rational
prematurely is greater than any cost associated with for one not to participate in research because, even if the
having one's organs taken from one's body after death. contribution is not negligible, any risk to which one may
The organ shortage is due largely to the failure of expose oneself will probably not be outweighed by any
people to donate in the current opt-in systems in the benefit. Although a subject in the active arm of a ran-
UK and USA. If Beauchamp and Childress4 are cor- domized controlled phase II or III drug trial may benefit
rect, the situation would not improve in an opt-out if the drug is more effective than standard therapy, gen-
system, because many people would still choose not erally any benefit from being a subject would probably
to donate. not accrue to that subject but to future patients. In con-
The collective action problem of opting out of trast, there is no risk in deceased organ donation and
organ donation is analogous in many respects to the thus no risk-benefit calculation for an individual. Yet
collective action problem of opting out of medical one who refused to donate organs could be adversely
research. If asked to participate in a clinical trial affected during one's life if one developed organ failure
comparing a standard therapy with an experimental and did not receive an organ transplant. This would be a
one, an individual could free ride by refusing to par- consequence of a shortage of organs, the collective effect
ticipate in the trial. One might believe that being a of many individuals refusing to donate.
research subject would have a negligible effect on Those who already have chronic conditions
the scientific knowledge gained from the research such as hypertension or diabetes would be excluded
and would not be worth the cost of participating. from deceased organ donation and thus would not
One might also believe that one could benefit subse- be part of this collective action problem. The critical
quently from treatments deriving from the collective age group would be those roughly between 30 and
acts of other research subjects without having to 55 years of age who are generally healthy and have
be a subject. In both deceased organ donation and none of the obvious risk factors for organ failure,
medical research, the benefits of an individual's par- because this is the period when most people decide
ticipation accrue to other people. On these grounds, to donate. It may seem rational for each individual
it seems at least rationally permissible not to par- in this group to free ride by ref using to donate. Yet
ticipate. Yet scientific knowledge and new therapies individual actions cannot be separated from the col-
based on that knowledge would be limited if many lective actions of others. As the collective effect of
people refused to participate in research. Some have many people opting out of donation would be fewer
argued that we have an obligation to participate in organs for transplantation, an individual who may
medical research because participants in the past eventually need a transplant may not receive one
put themselves at some risk and incurred costs for and thus may not be acting rationally by opting out.
scientific knowledge from which we have benefited.9 As noted, it may be rational for those with certain
We should do the same now to benefit people in the religious beliefs not to donate. For others, though,
future.1° One could present a similar argument that the absence of any risk, the relatively low personal
we have an obligation to donate our organs after cost, the limited supply of organs, and the increasing
death because organ donation and transplantation probability of developing organ failure and needing a
are two components of health care from which most transplant as we age make deceased organ donation
of us have benefited over the course of our lives. An a rational choice.
II ail 111111
8 Public Health
Notes
1. Quigley M, Brazier M, Chadwick R, et al. The organs 5. Erin CA, Harris J. An ethical market in human organs.
crisis and the Spanish model: theoretical versus pragmatic J Med Ethics 2003;29:137-8.
considerations. J Med Ethics 2008;34:223-4. 6. Harris J. Organ procurement: dead interests, living needs.
2. Organ Donation Task Force. The potential impact of an J Med Ethics 2003;29:130-4.
opt-out system for organ donat on in the UK: an independent 7. Tuck R. Free riding. Cambridge, MA: Harvard University
report from the Organ Donation Task Force. London: ODTE Press, 2008.
17 November 2008. 8. Olson M. The logic of collective action. Cambridge, MA:
3. Childress], Liverman C. Organ donation: opportunities for Harvard University Press, 1965.
action. Institute of Medicine. Washington, DC National 9. Harris]. Scientific research is a moral duty. J Med Ethics
Academies Press, 2006,205-28. 2005;31:242-8.
4. Beauchamp T, Childress]. Principles of biomedical ethics, 10. Schaefer GO, Emanuel EJ, Wertheimer A. The obligation to
6th edn. New York Oxford University Press. 2009:110. participate in biomedical research. JAMA 2009;302 67-72.
8.3 Screening for Disease
Direct-to-Consumer Genetics and highlights the degree to which this is so. This study
found that an astounding 78% of those who would
Health Policy: A Worst-Case Scenario? pursue testing would ask their physician for assistance
Timothy Caulfield with interpreting the data Further, 61% of respondents
felt that physicians have a professional obligation to
There is currently little evidence that the information help with this interpretive process.
provided by personal genomics companies—such as From a health policy perspective, this data sug-
23andMe and Navigenics—on a direct-to-consumer gests the potential for a worrisome trend: the driving
(DTC) basis, has any real health value. To be fair, re- up of healthcare utilization costs with little actual
search directly on-point is lacking. However, emerging health gain. The trend could play out as follows. En-
analysis suggests that the personal genomic informa- ticed by marketing strategies and the high profile of
tion that can be obtained from these companies is genetics in popular culture, individuals access the ser-
simply not of a nature that can support useful health vices of personal genome companies, hoping to obtain
applications (Kraft and Hunter 2009; Jakobsdottir et al. useful health information. They take this information
2009; Magnus et al. 2009). to a healthcare professional (e.g., family physician) for
Despite this skepticism, reflected at times in help interpreting the data. The healthcare professional
both the popular press and government reports (e.g., likely does not currently have the skills or knowledge
Federal Trade Commission [ETC] 2006), DTC personal to assist (Baars et al. 2005). The patient is left to inter-
genomics receives significant attention and was even pret the data on his/her own, or with a varying degree
selected as the 2008 Invention of the Year by Time of help from the company that sold the service. Since
Magazine (Hamilton 2008). II also seems to engender a the data is of only marginal health value, little or no
reasonable amount of interest in the general public, as health benefit is achieved. But the interaction with the
highlighted by McGuire and colleagues' survey (2009). physician costs the system both time and money.
Significantly, this survey revealed that the public ap- One could argue that despite the lack of a scien-
pears to be interested in the testing because they view tifically verified value to personal genomic services, the
it as valuable health information. results may still motivate healthy lifestyle behaviors by
Of course, this finding makes perfect sense. suggesting areas where change would be advantageous
Given that people hear in the media that genes influ- (e.g., "your genes say you should eat more broccoli,
ence popularity, ability to empathize, and likelihood exercise and sleep more!"). While this result is possible
of cheating while in a relationship, it is unsurprising (indeed, some people appear to be motivated by horo-
that they expect genetic data to provide significant scopes), the evidence on behavior change is not encour-
health information. The McGuire and colleagues' study aging. Bottom line It is tough to get people to change
Source: From The American Journal of Bioethics 9 (2009), 6-7,48-50.

Caulfield: Direct-to-Consumer Genetics and Health Policy: A Worst-Case Scenario?
unhealthy behaviours, even in the face of health infor- providers and then accessing the publicly financed
mation that tells them change is a good idea (Caulfield system for interpretive help.
et al. 2009). There is some evidence that genetic risk What is needed to respond to and moderate this
information will motivate non-health behavior change, issue? It seems inevitable that healthcare providers
such as the purchasing of insurance (Zick et al. 2005), will be asked by patients to assist in the analysis of
but a meaningful and sustained change in diet and the information generated by DTC companies. Ac-
exercise likely requires more than receiving a test result cordingly, it is axiomatic that physicians should be
(Story et al. 2008). This seems particularly so because the provided with the resources necessary to deal with pa-
risk information provided by DTC companies generally tients' questions (Guttmacher et al. 2007). While this
has minimal predictive power (Magnus et al. 2009). For measure will not erase the cost of a physician visit,
example, would an increased risk of 0.5% over a back- it will help make the visit more useful and, perhaps,
ground risk of 1% motivate long-term behaviour change? lesson the likelihood of future costs. Even more pref-
It is also worth considering whether test results erably, as noted by others, individuals should visit a
might have the opposite impact, providing individuals knowledgeable healthcare provider before ordering a
with a justification for fatalistic inaction (Claassen et test, to discuss the potential health implications and
al. 2009) or a false sense of security via the presence limits of DTC testing (FTC 2006). The need to provide
of a "health protective" gene. In other words, could resources and education to physicians is a common
genomic tests lead to less healthy behaviour? A specu- recommendation (Burke et al. 2002; and Burke et al.
lative concern, certainly, but humans do have a bit of a 2009) and is supported by the McGuire data (McGuire
tendency toward complacency and the adoption of un- et al. 2009). In addition, companies should be encour-
healthy habits. It may not take much—even a genetic aged to offer comprehensive and accurate information
test of marginal predictive value may suffice—to nudge about the limitations of their services. Finally, there
us further toward a sloth-like lifestyle. is a need for publicly available and widely dissemin-
To classify this situation as a "worst-case scenario," ated independent information about genetics and DTC
as I do in my title, is likely a bit alarmist. However, there testing services, something that is already starting to
seems little doubt that a health policy issue is germinat- emerge (FTC 2006). If framed appropriately, these sug-
ing, if not already starting to bloom. The direct costs gestions may help to lower the health expectations of
to the healthcare system, such as the cost of a visit to a future consumers.
family physician or, perhaps, anxiety-driven follow-up This brief comment ends with the traditional aca-
testing, may not be substantial. Yet, in this time of fiscal demic plea for more research. So many of the concerns
restraint, can any cost be ignored? In countries with a associated with DTC genetic testing remain largely
public healthcare system, such as my home country of conjecture. As it appears here to stay, we need more
Canada, this situation is particularly worrisome. Con- research on its potential impact on the healthcare
sumers would be paying for information from private system (Mouchawar et al. 2005) and on individuals.
References
Baars, M.J., Henneman, L., and Ten Kate, L.P. 2005. Deficiency different types of genetic risk information: Exploring the
of knowledge of genetics and genetic tests among general role of self-malleability. Psychology & Health (January 16,
practitioners, gynecologists, and pediatricians: A global 2009). Available at: http://www.tandfonlinecom/doi/
problem. Genetics in Medicine 7 (9): 605-160. abs/10.1080/08870440802460434 (accessed March 7, 2009).
Burke, W, Acheson, L., Botkin, J., et al. 2002. Genetics in Federal Trade Commission (FTC). 2006. At-Home Genetic
primary care: A USA faculty development initiative. Public Tests: A Healthy Dose of Skepticism May Be the
Health Genomics 5 (2): 138-146. Best Prescription. Available at: httpi/www.ftc.gov/
Burke, S., Martyn, M., Stone, A., Bennett, C., Thomas, H., and bcp/edu/pubs/consumer/health/hea02.pdf (accessed
Farndon, P 2009. Developing a curriculum statement March 7, 2009).
based on clinical practice Genetics in primary care. British Guttmacher, A.E., Porteous, M.E., and McInerney, J.D. 2007.
Journal of General Practice 59 (559): 99-103. Educating health-care professionals about genetics and
Caulfield, T., Shelley, J., Alfonso, V., and Bubela, T. 2008. genomics. Nature Reviews Genetics 8: 151-157.
Nutrigenomics and the promise of prevention: Hamilton, A. October 30, 2008. TIME's Best Inventions of 2008:
Representations and realities. Health LawJournal special Invention of the Year—The retail DNA test. Time Magazine.
edition: 41-66. Available at: http://www.time.com/time/specials/packages/
Claascen, L., Henneman, L De Vet, R, Knol, D., Marteau, article/0,28804,1852747_1854493,00.html (accessed
T., and Timmerman, D. 2009. Fatalistic responses to March 8, 2009).
II JII 111111
8 Public Health
Jakobsdottir, J., Gorin, M.B., Conley, Y.P., Ferrell, R.E. and Weeks, Mouchawar, J., Hensley-Alford, S., Laurion, S., et al. 2005.
D.E. 2009. Interpretation of genetic association studies: Impact of direct-to-consumer advertising for hereditary
Markers with replicated highly significant odds ratios may breast cancer testing on genetic services at a managed care
be poor classifiers. PLoS Genetics 5(2): e1000337. organization: A naturally-occurring experiment. Genetics in
Kraft, P., and Hunter, D. 2009. Genetic risk prediction—Are Medicine 7 (3): 191-197.
we there yet? New England Journal of Medicine 360: Story, M., Kaphingst, K.M, Robinson-O'Brien, R., and Glanz,
1701-1703. K. 2008. Creating healthy food and eating environments:
Magnus, D., Cho, M. K., and Cook-Deegan, R. 2009. Direct- Policy and environmental approaches. Annual Review of
to-consumer genetic tests: Beyond medical regulation? Public Health 29: 253-272.
Genome Medicine 1: 17. Zick, C.D., Mathews, C.J., Roberts, J.S., Cook-Deegan,
McGuire, A., Diaz, C., Wang, T., and Hilsenbeck, S. R., Pokorski, R.J., and Green, R.C. 2005. Genetic
2009. Social networkers' attitudes toward direct-to- testing for Alzheimer's disease and its impact on
consumer personal genome testing. American Journal of insurance purchasing behavior. Health Affairs
Bioethics 9 (6-7): 3-10. (Millwood) 24 (2): 483-490.
The Ethics of Screening: Is

comes back "negative." This anxiety-state may be de-
"Screeningitis" an Incurable Disease? scribed as that of the "worried well" (figure 1).
Darren Shickle and Ruth Chadwick Consider a population, some of whom have a disease,
while others do not. The diagnosis of "disease" may re-
Most Western societies are concerned about health care quire extensive investigation (or may only be made at post-
expenditure in view of an ageing population and a growth mortem). Such a "gold standard" test is important prior to
in the number of effective health care interventions. Scar- embarking on treatment. Screening should be a simpler,
city of resources means that health care systems are ex- cheaper, and quicker test to select for further investigation.
pected to produce more health gain with relatively fewer However, cases with a positive screening-test finding may
resources. It is assumed that f "prevention is better than not have the disease (falsely labelled as "positive"), and
cure," then it will also be "cheaper." Cost benefit analy- similarly, disease may be present in individuals who have
sis tends to consider tangible costs and benefits which a negative screening result, i.e., "false negatives" (figure 2).
accrue directly to the health service since these are easier
to measure and are the only ones that will appear on
balance sheets at the end of the financial year. However,
there are ethical implications that arise from ignoring the STOP!
intangible and indirect aspects of this equation. Are YOU Healthy?

Screening may be defined as a selection procedure
for further investigation, applied to a population of ARE YOU SURE??
asymptomatic individuals, with no personal or family Did you know that there is a
history to suggest that they are at a higher risk of the
disease than the rest of the population. It may be a form malignant disease called Screeninoitis ?
of "secondary prevention," which aims to detect dis- DON'T PANIC

ease in presymptomatic individuals in order to provide
There is a screening test available ...
more effective treatment in the early stages of disease,
or "primary prevention," which aims to identify risk ... and it is accurate...
factors or carrier states. Individuals offered screening
would usually consider themselves to be "healthy" ... MOST of the time ...
(unless they have coincidental illness). We tend not to ... BUT the treatment is PAINFUL!! ...
consider our health until either we or someone close
to us loses it. The offer of screening implies that we AND HAS A LOW SUCCESS RATE!!!
may not be healthy after all. This may result in a rise

in anxiety, even if small and short-lived, until the result Figure 1 Screening and the "Worried Well"
Source: From Journal of Medical Ethics 20,1 (1994), 12-18. Reprinted with permission.
Shickle/Chadwick: The Ethics of Screening: Is "Screeningitis" an Incurable Disease?
would raise specificity and hence target resources at

Disease
those with the highest risk of morbidity and potentially
the most to gain from treatment. In contrast, a test with
Present Absent high sensitivity would tend to maximise identification
true false
of diseased people in the screened population, i.e.,
positive positive positive A +B there would be relatively few false negatives, but there
Finding A B would also be unnecessary investigations or treatment
for others. Thus an increase in sensitivity will be at the
false true
negative negative negative C+D expense of specificity and vice versa. There are costs
C D and benefits associated with each of the four permuta-
tions of true/false positive/negative (figure 3).
A+C B+D
When a screening programme is evaluated, the
Sensitivity = Proportion of patients with the
benefits of being a true positive tend to be considered
disease in whom the finding is and sometimes the reassurance associated with a
positive negative result. However, associated costs tend to be
A ignored. If costs outweigh benefits, then either the bal-
A+C ance between sensitivity and specificity should be al-
Specificity Proportion of those without the tered or else the screening test should not be used at all.
disease in whom the finding The preference for specificity or sensitivity should
is negative depend on the consequences of making the diagnosis
D or not. High specificity should be desirable if:
B+D
1. the diagnosis is associated with anxiety or stigma;
2. further investigations are time-consuming, pain-
Figure 2 Possible Outcomes of Screening Test
ful, and/or expensive;
3. cases are likely to be detected by other means
before it is "too late" for effective treatment; and
It is not always possible to make an absolute dis- 4. treatment, especially if painful or expensive, is to
tinction between those individuals in whom the disease be offered without further investigations.
is present or absent even with a "gold standard." For
some conditions there will be a continuous distribu- In contrast, high sensitivity would be desirable if:
tion of variables, such that at one extreme, individuals
could be considered "diseased," while those at the 1. adverse consequences of missed diagnosis for the
other are "healthy," for example, in the context of hear- individual, for example, late treatment, might be
ing or visual acuity. For other biological variables, for significantly worse than early,
example, blood pressure or haemoglobin, the extremes 2. adverse consequences of missed diagnosis for so-
at both ends of the distribution would be "unhealthy," ciety, for example, with a serious communicable
while those in the middle would be "normal." disease; and
For the purposes of screening there should be a 3. diagnosis is to be confirmed by other investigations
threshold to trigger further investigations or treatment. so that the period of anxiety is short, or the correct
The position of this cut-off in the distribution should diagnosis is given before treatment is started.
be based on the associated risk of morbidity or mor-
tality that warrants further intervention. The choice of For some screening tests the choice between sensi-
threshold may therefore be arbitrary or depend on the tivity and specificity is difficult. Consider HIV (Human
resources available. For example, a blood cholesterol Immunodeficiency Virus) screening. The diagnosis is
of 6.5 mmo1/1 or above is considered to be associated associated with high levels of anxiety and stigma, there
with a significantly high risk of coronary heart disease.' may be no other investigations to refute the diagnosis,
However, between 25 and 36 per cent of the UK popula- treatment is expensive, with many side-effects, and
tion' have a blood cholesterol above this threshold, and may be ineffective. However, there are important con-
it would be unrealistic to treat this number of people. sequences of a missed diagnosis for society.
A screening test with high specificity would tend The ethicist has a role in resolving dilemmas
to detect only people with the disease, i.e., there would arising from the conflict of sensitivity and specificity.
be relatively few false positives. Raising the threshold Moral conflict can arise in circumstances where an
AI II .III IJILLI
8 Public Health
intervention is associated with both good and harm for but I will never use it to injure or wrong them." How-
an individual. Utilitarianism is an attractive theory in ever, in circumstances where both good and harm are
such circumstances, since the correct solution would likely to result from an intervention, an objective rule
depend on whether the utility arising from the sum of may be required to resolve such an ethical dilemma.
the benefits outweighs the disutility associated with There are frequent appeals for funds to allow a child
harm. The person usually considered best placed to to receive a "life-saving" treatment that is not available
assess utility is the person him/herself; however, does from the UK National Health Service. It is difficult for a
the effect of illness cloud the person's judgement? In health planner to withdraw funding from a programme
any case, should the autonomy of an individual with that can save life because of the consequent publicity that
a communicable disease have primacy over the desire would be generated through "shroud waving." Vested in-
of society to place that person in quarantine to prevent terest groups would argue that the benefits of screening
other individuals from becoming infected? Similarly, to detect one individual who may otherwise have died
should a couple be allowed to have a handicapped without early treatment would justify any harm caused.
child, when prenatal screening may have detected an The principle of double effect permits harm to others
abnormal fetus which could have been terminated? provided the harm is indirect and was not intended,
The professional duty to both non-maleficence although it may have been foreseen. However, this prin-
("do not harm") and beneficence ("do good") is en- ciple has been widely criticised3" and is unhelpful in
shrined in the Hippocratic oath2: "I will use treatment the resolution of the dilemma of whether to screen or
to help the sick according to my ability and judgement, not. It can apparently excuse the harm caused both by
True Positive False Positive

Costs Benefits Costs Benefits
May be no increase in life Earlier treatment (cheaper; Unnecessary anxiety and Provides an opportunity
expectancy if treatment is of more pleasant; better stigma until correct diagnosis to counsel on unhealthy
no effect but patient may have prognosis) is made lifestyles:"next time could
to live longer with diagnosis be f or real"
because it is made earlier (lead Benefits of"sick role" (excused Unnecessary investigations
time bias) social responsibilities and treatment (time-
provided seek and comply consuming; expensive; pain;
Anxiety:"Worried ill" with treatment) side-effects; complications)
Stigma
Explanation of symptoms: Lingering doubts: patients
Reduced quality of life, for knowing the cause of tend to believe doctors! May
example, due to side-effects or symptoms so that they can be be difficult to reassure patient
complications of treatment "treated" can relieve anxiety that initial diagnosis was
incorrect
False Negative True Negative

Costs Benefits Costs Benefits
False reassurance ("Unworried Spared anxiety if treatment of Anxiety while awaiting results Reassurance
ill") no benefit of test ("Worried weir)
May legitimizeun healthy May legitim ise"un healthy

lifestyle" lifestyle:for example, If my
chest x-ray is normal, then
Later treatment (unpleasant; smoking cannot be doing me
more expensive; worse any harm!"
prognosis)
Increased cost per desired

outcome
Figure 3 Costs and Benefits Associated with Being a True Positive, False Positive, False Negative or
True Negative
providing and withholding screening. For example, it For it is not possible to know which members of a
is accepted that all screening procedures are associated population will be "true positives" before screening,
with false positives and negatives. This harm, however, and whether these individuals will benefit from the
although foreseen, can be described as an unintended treatment offered. Glover has shown that this con-
side-effect. On the other hand, if screening is withheld, sideration is irrelevant, morally speaking: whether
individuals who could have benefited from screening, the harmed individual can be identified prospect-
via earlier treatment of their disease, will be denied these ively or retrospectively makes no difference to the
advantages. Thus the decision to withhold arguably wrong of harming them.9 Harris maintains that an
causes harm which is foreseen but not intended. individual remains "negatively" responsible even if
The principle of acts and omissions might be in- he was not aware that he could have prevented harm,
voked to escape responsibility for the harm that results although he recognises that the individual should
from a decision to withhold screening. According to not be blamed.
this principle, failing to prevent some harm is less bad, Another attempt to limit the moral demands upon
morally speaking, than actively bringing it about,6 and us is to try to establish priorities between different
the withholder has not, it might be claimed, actively kinds of moral requirements. Frankena, for example,
caused the harm. Like the principle of double effect, this proposed that beneficence includes four elements'°:
principle has been heavily criticised.' In the light of the
evidence that haemophiliacs can be infected with the 1. One ought not to inflict evil or harm (what is bad).
HIV virus as a result of transfusions of infected blood, 2. One ought to prevent evil or harm.
for example, it is arguably just as bad to fail to screen 3. One ought to remove evil or harm.
the blood, with the result that, say, 10 per cent will 4. One ought to do or promote good.
be infected, as it would be to inject those 10 per cent
with the virus deliberately. If we fail to prevent some- Frankena ranked the four elements such that in cases
thing bad from happening where we are in a position of conflict, non-maleficence took moral precedence
to do so, we are responsible. over beneficence. Foot's proposed priority of what she
Harris8 identifies "positive" moral responsibility calls "negative" duties to refrain from injuring someone
where something happens because someone did some- and "positive" duties to help them is similar.'
thing and "negative" responsibility because someone There are at least two problems in applying these
allowed or permitted or let it happen. He concludes distinctions to the decision concerning whether or
that there is equal responsibility for what happens not to carry out a screening programme. First, the
because of our actions and what happens because of primacy of non-maleficence is not absolute. For
our inaction or non-action. Thus, it would be an in- example, a surgical incision will cause injury, but
adequate defence to argue against a screening proced- the consequence of the operation may be that the
ure on the grounds of the need for a needle-prick or patient's life is saved. Thus, the injury caused by
anxiety that may be caused to the "worried well," since taking a blood sample as part of a screening pro-
there is "negative" responsibility for the "true positives" gramme may be considered trivial in comparison to
who would otherwise have been identified. the potential benefits. Second, it is not always clear
whether a decision not to do something, such as
Unlimited Moral Demands carry out a screening procedure, should be described
as a maleficent infliction of injury or a failure to do
The problem that arises from the abandonment of the good. It is easy to imagine contexts in which the fail-
principle of acts and omissions is that it seems to place ure to screen blood before transfusion to haemophil-
unlimited moral demands upon us. It might indicate iacs could accurately be described as a maleficent
a moral requirement to carry out any screening pro- infliction of injury.
gramme that could conceivably prevent harm. This has It is difficult, therefore, to accept any of these pro-
led to the search for ways of placing limits on what we posals for drawing moral distinctions. There must be
are required to do by attempting to draw moral distinc- room for a trade-off between the harm brought about
tions between different kinds of cases. by screening and the harm brought about by failure to
In the screening context it might be argued that screen. This suggests that cases should be considered
it is morally significant that those who are harmed on their merits. Even if we accept, as we should, that
through screening are identifiable, whereas those there is negative responsibility for the harm we fail to
that are harmed through failure to screen are not. prevent by withholding screening, this does not mean
II ill 111111
8 Public Health
that we should carry out every screening programme Counselling Should Not Be Prescriptive
that could conceivably prevent harm. A useful test is that
proposed by Singer: "If we :an prevent something bad There are further problems in placing the responsibility
without sacrificing anything of comparable moral sig- of whether benefits outweigh costs upon the individ-
nificance, we ought to do it."12 ual. Counselling should, as far as possible, facilitate
There is a danger that advances in medical tech- the autonomous decisions of the individual, for ex-
nology will be translated into screening services and ample, allowing informed consent to be obtained for
offered by "enthusiastic" scientists to an "unsuspecting" screening. It is unlikely however, that any individual
public, without consideration of the consequences. could absorb all the necessary information during a
Even if the society is consulted, there is a general ex- few counselling sessions, especially if this is during a
pectation among the public that medical advance is time of anxiety. While counselling should not be pre-
generally "a good thing." scriptive, the fact that a provider of health care places
The harm consequent from a screening test for a positive value on screening such that the service is
any individual will usually be trivial in comparison offered may give the counsellee the impression that the
with the potential harm from not offering screening "correct choice" of action would be to accept screen-
and saving a life by providing effective treatment. ing. Thus arguably, even if a counsellor attempts to give
However, the way in which individuals perceive risk information in a neutral fashion, counselling cannot be
is complex: some are "risk lovers" while others are truly "non-directive."
"risk averse" or "risk neutral."13 For example, the risk We would not wish to argue that screening
of an HIV-positive doctor infecting one of his/her pa- should not be offered because information about
tients is very small, yet there has been demand within risk, utility, opportunity costs, etc. is too complex,
the media for regular, universal HIV testing of health- nor that we should not encourage individuals to in-
care professionals. It has been argued that patients are clude such considerations in their decision-making.
more concerned with the nature of risk than with its Unless specified in law, we believe that screening
precise numerical value: "let us not get so preoccu- should not be overtly directive, although some may
pied with the risks that we lose sight of the stakes."14 disagree if the consequences of the condition in ques-
However, individuals may place differing amounts of tion are sufficiently serious. However, it should be
disutility on these states. For example, some patients recognised that the obtaining of apparently informed
may be happy to receive treatment for mild hyperten- consent does not remove any responsibility for harm
sion, even though there is only a small decrease in from those who offer screening. There is a duty on
risk of a "stroke" or "heart attack" following treatment; health-service planners, however, to offer screening
other individuals may be less willing to accept the tests that the majority of the target population would
label of "ill-health." wish to accept.
An individual may choose screening even though The principles proposed by Wilson and Jungers'
there is a high risk of relatively trivial harm and only a in the mid-1960s have attracted much criticism. How-
small chance that he/she will benefit. Using utilitarian ever, these criteria may be useful guidelines for the
principles, the net utility or disutility for society aris- development of screening programmes:
ing from an intervention is equivalent to the sum of the
change in utility to its individual members. However, 1. The condition sought should be an important
the sum of trivial disutility for the many may exceed problem.
the large utility gain for the few. Others have raised 2. There should be an acceptable treatment for pa-
similar concerns about the potential net disutility asso- tients with recognised disease.
ciated with health promotion. For example, Skrabanek 3. Facilities for diagnosis and treatment should be
pointed out that "the proverb, 'a stitch in time saves available.
nine,' may be sound advice for mending socks but it 4. There should be a recognised latent or early symp-
makes little sense if a thousand people need one stitch tomatic stage.
(in its medical equivalent) :o save one person from 5. The natural history of the condition, including
nine stitches." It is, however, unreasonable to expect its development from latent to declared disease,
the individual to include "opportunity costs" in his/her should be adequately understood.
consideration of providing consent, i.e., resources can 6. There should be a suitable test or examination.
only be spent once, and hence benefits may have to be 7. The test or examination should be acceptable to
foregone by others. the population.
8. There should be agreed policy on whom to treat as consequences for the individual or society as a whole.
patients. Screening for a condition which is not "important" is
9. The cost of case-finding (including diagnosis unlikely to result in net utility or accrue more utility
and subsequent treatment of patients) should be than other interventions competing for scarce resour-
economically balanced in relation to the possible ces. However, it may be worth considering screening
expenditure as a whole. for relatively trivial conditions, providing screening is
10. Case finding should be a continuous process and cheap, does not cause anxiety, and there is an effective
not a "once and for all" project. treatment with few side-effects.
Wilson and Junger do not, however, specify
Wilson and Junger used the term "principles" for "ease whether patients, a third party, or society as a whole,
of description rather than from dogma." It is unlikely should prioritise importance. John Stuart Mill, in his
that any screening programme will be able to fulfil all essay On Liberty, argues that "there is no one so fit to
of these criteria to everyone's satisfaction. The question conduct business, or to determine how or by whom
therefore arises as to whether each criterion has equal it shall be conducted, as those who are personally
merit or whether there is a hierarchy of importance. interested in it."19 We have reservations about apply-
Wilson and Junger felt that "of all the criteria that a ing this principle of utilitarianism in the context of
screening test should fulfil, the ability to treat the health-care resource allocation. It is difficult for an
condition adequately when discovered, is perhaps the individual patient to be objective as to whether his/her
most important." However, it is important not to be health problem is more important than that of another
too restrictive in our thinking about what represents patient or whether he/she deserves scarce resources in
an "adequate" treatment. For example, with some cases preference to others.
of prenatal diagnosis, couples may benefit from genetic Attempts have been made to involve society in the
counselling such that they may "prepare themselves" prioritisation process, for example, in Oregon, USA.2°
for the birth of an affected child. In this circumstance, However, such consultation was expensive and time-
the counselling could be considered to be a form of consuming, and the complexity of the exercise meant
treatment. An analogous question is whether termina- that there were inconsistencies in the initial priority-list.
tion of pregnancy can be considered a treatment and if For example, treatment for crooked teeth was ranked
so for whom. However, the debate on that question has higher than treatment for Hodgkin's lymphoma. In
been explored elsewhere.'8 view of these difficulties, we see little alternative to the
The requirement for an acceptable treatment existing decision-making process, provided that this is
is compatible with our argument for a proper con- in consultation with experts and lay organisations. We
sideration of all costs and benefits. The advocates recognise that this will be biased by the social values of
of screening have a duty to demonstrate potential those involved.
benefit prior to the introduction of screening. Im- The criteria for suitability and acceptability of
proved prognosis following the use of an "effective" a test for screening will vary from one condition to
treatment will usually be the most important of these another. While the test should have high validity and
benefits. repeatability, a compromise between sensitivity and
The importance of availability of facilities for diag- specificity may be required. A test associated with high
nosis and treatment is self-evident. Informing patients costs (both tangible and intangible) may be acceptable
that they have a treatable disease but withholding any to an individual or society as a whole if the condition
intervention is likely to cause more harm than not per- is severe and an effective treatment is available. For
forming screening. example, most women will endure the indignity and
embarrassment associated with taking a cervical cytol-
Helpful Pointers ogy smear if it means that carcinoma of the cervix may
be prevented.
The remaining "principles" are, in our view, helpful The requirement for an understanding of a con-
pointers as to which screening programmes are desir- dition's natural history and the recognition of a latent
able and, if fulfilled, will describe situations where net stage are important in the development of techniques
utility is likely for diagnosis and treatment but in themselves are not
Wilson and Junger recognise the difficulty in essential. Screening for HIV is performed, yet the nat-
defining "importance" and include conditions with ural history of acquired immunodeficiency syndrome
high prevalence and those which have very serious (AIDS) is still not "adequately understood."
III II I NJ
442 8 Public Health
The balancing of economic costs resulting from Wilson and Junger Criteria Endorsed
case-finding with those due to possible expenditure
as a whole is compatible with our modified utili- An assessment of net utility will assist in choosing
tarian approach and with the principle of justice. service options which will result in the "most hap-
If screening is available for one individual, then piness" for the population. We recognise that as-
justice would expect screening to be offered for signing utility values within the cost/benefit equation
another individual who has a similar risk of having is difficult. Obtaining a consensus as to the objective
the disease and a similar potential to benefit from of a screening programme is an essential first step.
treatment. If a screening programme is considered For example, what is the purpose of genetic-carrier
to be desirable, why then should future patients not screening-empowerment of the individual in making
benefit from screening? Wilson and Junger argue that reproductive choice or reducing the number of births
screening should be a continuous process, since the of handicapped individuals?
"start-up" costs associated with screening (providing We endorse the Wilson and Junger criteria, since
accommodation, purchas:.ng equipment, training if the majority are fulfilled (especially those relating to
staff, etc.) are usually large in comparison with the effectiveness of treatment), then the conclusions will
marginal costs of each test performed. While these approximate to those of a utilitarian assessment.
points are true, screening for prevalent (existing) The principle of justice might suggest that
cases requires different considerations from screen- access to health care is a moral right for all. How-
ing for incident (new) cases. Thus, the case-yield ever, scarcity of resources means that there must
and the consequent utility may be high in the first be prioritisation. There is no reason why philoso-
wave of screening if the natural history of a condi- phers should not be interested in the resolution
tion is slow, and hence there may be a large number of moral dilemmas in order to maximise utility
of undetected cases in a population. However, the from limited health care resources. Such choices
number of new cases that develop during a screening have to be made, and it is better that such deci-
cycle may be small, and hence the consequent utility sions are made with an explicit, ethical mechan-
for these few cases may not outweigh disutility or ism rather than them depending on the values of a
opportunity costs. few health service planners.
References
1. Standing Medical Advisory Committee. Blood chol- 12. Singer P Practical ethics. Cambridge: Cambridge University
esterol testing: the cost effectiveness of opportunistic Press, 1979: 169.
cholesterol testing. London: HMSO, 1990. 13. O'Brien B. "What are my chances doctor?"-a
2. Quoted in: Beauchamp T L, Childress J E Principles of review of clinical risks. London: Office of Health
biomedical ethics [3rd ed). New York Oxford Univer- Economics, 1986.
sity Press, 1989: 120. 14. Fraser F C Evolution of a palatable multifactorial
3. See reference 2: 127-134. threshold model. American journal of human genetics
4. Glover] C B. Causing death and saving lives. Harmond- 1980; 32: 796-813.
sworth: Penguin, 1977: 86-91. 15. Skrabanek P Why is preventive medicine exempted
5. Foot P The problem o f abcrtion and the doctrine o f the from ethical constraints? Journal of medical ethics 1990;
double effect. In: Rachels J, ed. Moral problems [2nd 16: 187-190.
ed). New York: Harper & Row, 1975. 16. Clarke A. Is non-directive genetic counselling possible?
6. See reference 4: 92-116. Lancet 1991; 338: 998-1001.
7. See reference 4 and also Harris J. Violence and re- 17. Wilson J M G, Junger G. The principles and practice of
sponsibility. London: Routledge & Kegan Paul, 1980 screening for disease. Public Health Papers, 34. Geneva:
passim. World Health Organization, 1968: 26-39.
8. See reference 7: Harris and ilso Harris J. The value of life: 18. Clarke A. Response to: What counts as success in
an introduction to medical ethics. London: Routledge genetic counselling? Journal of medical ethics 1993;
and Kegan Paul, 1985: 28-33. 19: 47-49.
9. See reference 4: 100-102. 19. Mill J S. On liberty. In: Gray J. On liberty and other
10. Frankena W. Ethics [2nd ed). Englewood Cliffs, NJ: essays. Oxford: Oxford University Press, 1991: 121.
Prentice-Hall, 1973: 47. 20. Dixon J, Welch H G. Priority setting: lessons from
11. See reference 5: 59-70. Oregon. Lancet 1991; 337: 891-894.
Small: Canada's Highest Court Unchains Injection Drug Users
8.4 Harm Reduction Programs
Canada's Highest Court Unchains The journey for Insite has been wrought with
challenges because it confronts our inner web of belief
Injection Drug Users; Implications about how to best approach addiction. Supervised in-
for Harm Reduction as Standard of j ection exists in a moral minefield at the very heart of
Healthcare our culture. By culture, I am speaking about what we
believe to be right and wrong, the implicit and explicit
Dan Small values that are the building blocks of our understand-
ing, practice, and societal approaches to people with
addictions.3 As such, it is my belief that this legal vic-
The philosophers have only interpreted the world tory is about something much more fundamental than
in various ways; the point, however, is to change it.' the legal, medical, or scientific issues that arise from
it. It points the way towards a shift in our cultural
After a long legal and cultural battle, North America's orientation that allows for addiction to be constructed
only supervised injection facility, Insite, is finally safe as a social issue best addressed, metaphorically, by the
from arbitrary political interference. This was a case Chief of Medicine rather than the Chief of Police.
where personal experience, activism, advocacy medi- This cultural change is best illustrated by a story
cine, and science stood side-by-side to protect the from only a decade ago when harm reduction was not
rights of even the most marginalized members of the widely accepted or understood in Canada. Harm reduc-
community to life-saving healthcare. The case began at tion innovations involve attempting to reduce harms
the Supreme Court of British Columbia and eventually associated with drugs, such as fatal overdoses, but to
made its way to the Supreme Court of Canada (ScC). At do so without necessary eliminating the use of drugs
the centre of the case were the personal stories of people (abstinence). The Harm Reduction Journal provides the
who relied on supervised injection to stay alive together following definition:
with testimony from scientists, physicians, healthcare
officials, and the operators of Insite. On 30 September We define "harm reduction" as "policies and
2011, Supreme Court of Canada drew a legal line in programs which aim to reduce the health, social,
the sand that highlights the rights of people with addic- and economic costs of legal and illegal psycho-
tions to the security of their person under section 7 of active drug use without necessarily reducing drug
the Charter of Rights and Freedoms (Charter of Rights)? consumption."4
The Charter enshrines the values of Canadian culture
regarding the rights of individuals with respect to the In the year preceding the opening of Insite in
provincial, federal, and territorial governments. The 2003, I was collaborating on a draft press release in
judges of Canada's highest court are appointed from a response to local opposition to harm reduction. The
wide variety of political backgrounds. The decision was press release simply stated that addiction is, primarily,
unanimous and reinforced the foundation of our under- a healthcare matter. At that time, the notion of pub-
standing of addiction as a healthcare matter in Canada. licly stating that addiction is a healthcare rather than
As a caveat, I am not a distant academic examin- criminal justice issue was so controversial that I could
ing supervised injection from the point of view of a not convince anyone in the establishment to lend their
removed observer. I am part of the senior management name to the press release. In the end, two commun-
of the non-profit organization that founded and oper- ity activists agreed to sign their names to what was,
ates Insite, the PHS Community Services Society (PHS), at the time, a provocative public statement. Now, ten
and, as such, I have been intricately involved in the years later; this has become an established legal fact in
development, set-up, management, and advocacy for Canada. Addiction is a healthcare matter.
Insite. I am a participant observer, and so this com- The PHS initiated the case at the Supreme Court of
mentary is written from the point of view of my per- British Columbia in 2008 to protect the program from
sonal experience. closure by a conservative federal government. We did
Source From Harm Reduction Journal 9, 34 (2012). Reprinted with permission.

II all III Bad
8 Public Health
so at a time when there was no formal institutional sup- The second fact that was established was that
port for a struggle to protect Insite in the courts. The drugs, in and of themselves, do not necessarily cause
only intervener was the British Columbia Civil Liberties serious interruptions in health. Rather, it is the method,
Association. Despite their role in co-managing Insite, mechanism, and context within which drugs are in-
Vancouver Coastal Health (vcx) counseled our organ- gested that brings about danger:
ization not to turn to the courts to protect the program
and refused to provide any formal assistance for the Controlled substances such as heroin or cocaine
PHS to legally defend it. In the initial case, neither [the] that are introduced into the bloodstream by injec-
Vancouver [Coastal] Health Authority (vcx) nor the tion do not cause Hepatitis C or HIV/AIDS. Rather,
Attorney General of British Columbia sought inter- the use of unsanitary equipment, techniques and
vener status. Fortunately, the PHS was able to obtain procedures for injection permits the transmission
pro bono representation by three lawyers: Monique of those infections, illnesses or diseases from one
Pongrecic-Speier, E Andrew (Drew) Schroeder, and individual to another" (p. 33, para. 87).
Joseph Arvay, all of whom nobly took on the case when
it felt like Insite had been backed against the wall. A This is the foundation of supervised injection as an
related and important case, entered by the Vancouver intervention. The point of intervention focuses on
Area Network of Drug Users (Respondent/Appellant on reducing the harms associated with drug use with-
cross appeal) was heard jointly. out forcing abstinence as a precondition for receiving
The PHS case focused on two overarching themes. healthcare.
The first theme related to the division of powers be- The idea that drugs, as substances, are not auto-
tween the federal and provincial governments and matically intrinsically evil or dangerous has been a
essentially argued that the operation of Insite falls culturally controversial notion. There are, broadly
under the jurisdiction of the Province of British Col- speaking, two competing overarching narratives about
umbia. The second argument pertained to the Char- addiction.3 The first narrative focuses, essentially, on
ter. The Charter is the first part of the Constitution of drug use and the drugs themselves as intrinsically dan-
Canada and contains a passage of particular relevance gerous. The second does not centre on the drugs them-
to Insite.2 This portion of the Charter, section 7, selves but instead concentrates on the way drugs are
i s of central importance to Insite and states that: administered (e.g., clean versus unclean syringes) and
"Everyone has the right to life, liberty and security of the psycho-socio-cultural context of their use (such as
the person and the right not to be deprived thereof criminalization, poverty, and mental pain).
except in accordance with :he principles of funda- This distinction can be traced to the work of
mental justice"' (p. 4). The PHS argued the federal psychologist Bruce K. Alexander who first discussed
Health Minister's withholding of an exemption from two orientations for constructing addiction 30 years
the Controlled Drugs and Substances Act (cDSA) that ago.' The first orientation essentially constructs ad-
was required at that time for Insite's operation was diction as follows: first, a person takes a drug, then,
prejudicial and arbitrary. More so, it jeopardized the eventually, the drug takes the person as a result of
life chances of people who need the facility to access repeated exposure. Understanding addiction in terms
life-saving healthcare. of an exposure to drug concentrates on the eradica-
The case at the BC Supreme Court laid the ground- tion of drugs as the point of intervention. Following
work for the facts that would later form the founda- on from this reasoning, it is the reduction in drug
tion for the landmark ruling at the Supreme Court of exposures that will ultimately reduce or eliminate
Canada. Given the case's importance, it is worth exam- addiction. The rival narrative about addiction, de-
ining at a high level. There are, in my view, four key scribed in terms of coping or adapting by Alexander
findings of fact in this first case. and his colleagues, constructs the problem altogether
The first key fact pertains to the notion that addic- differently. According to this alternative explana-
tion is a healthcare matter. The Government of Canada tion, people take drugs; drugs do not take people.
conceded this as an indisputable fact. This perspective maintains that people misuse drugs
The presiding judge at the BC Supreme Court, due to impoverished conditions and psychosocial
Justice Ian Pitfield, highlighted this absolutely critical pain that require extraordinary coping strategies. In
cultural admission in his Reasons for Judgment when turning focus away from the dangers of the drugs
he stated: "drug addiction is an illness' (p. 20). The themselves to the ways in which they are being used,
declaration of addiction as an illness allows for the de- the original court decision was aligned with this
votion of healthcare resources to addressing it. latter orientation.
The third legal fact pertained to the notion that injection is sacrosanct. Without supervised injection,
effective medical interventions are available to meas- people might perish from fatal overdoses before real-
urably reduce the harms of addiction. The primary izing the opportunity to one day pursue detox, treat-
healthcare intervention in Insite is the provision ment, and abstinence. Instead, all that would be left
of sterile injection equipment and the supervision would be a mortality statistic: a faint reminder that
of injection: they ever lived.
In summary, there were four legal facts that go
The risk of morbidity and mortality associated with to the heart of a particular cultural understanding of
addiction and injection is ameliorated by injection addiction:
in the presence of qualified health professionals°
(p. 33, para 87). 1. Addiction is a healthcare matter.
2. Drugs do not cause deadly HIV, HCV, and fatal
The finding of this fact, based on the scientific and overdoses: unclean needles and unsupervised in-
medical evidence before the court, established that a j ection do.
supervised injection facility helps to prevent disease 3. Supervised injection is effective at preventing
and death. Contrary to its popular characterization morbidity and mortality. Harm reduction opens
as an isolated program, Insite also offers detox and the door to a range of healthcare (e.g., detox,
treatment on site. treatment).
The final fact is perhaps the most culturally con- 4. Abstinence, though laudable, is not compulsory for
troversial because it foregrounds the fact that effective effective healthcare interventions, with measurable
healthcare interventions exist for addiction that do not outcomes (e.g., such as saving lives by interven-
demand abstinence. Justice Pitfield understood that ing in otherwise fatal overdoses or preventing fifv)
while Insite is not traditional treatment, it is, nonethe- for addiction. The idea of condoning or enabling
less, important healthcare: addiction with supervised injection takes second
place to keeping people alive.
While users do not use Insite directly to treat
addiction, they receive services and assistance at The establishment of these four key facts, in my view,
Insite which reduce the risk of overdose that is a laid the groundwork for both a legal and a cultural vic-
feature of their illness, they avoid risk of being in- tory with respect to the notion of supervised injection.
fected or of infecting others by injection and they As the case advanced, it gathered cultural mo-
gain access to counselling and consultation that mentum as part of a growing acceptance of a particular
may lead to abstinence and rehabilitation. All of understanding about addiction as described above. At
this is healthcare° (p. 51, para. 136). the final stage, 14 interveners had joined the proceed-
ings including the Vancouver Coastal Health Authority,
This recognition of this fact feeds into the cultural anx- Canadian Nurses Association, British Columbia
ieties about somehow enabling or encouraging addic- Nurses' Union, Registered Nurses' Association of
tion by not outlawing it with vehemence. The ruling Ontario, Association of Registered Nurses of British
also addressed this culturally notorious notion of over- Columbia, Canadian Medical Association, Canadian
looking addiction: Civil Liberties Association, Canadian HIV/AIDS Legal
Network, International Harm Reduction Associa-
Society cannot condone addiction, but in the face tion, CACTUS Montreal (a non-profit organization
of its presence it cannot fail to manage it, hope- dedicated to providing non judgmental assistance
fully with ultimate success reflected in the cure of and risk reduction for at risk individuals including
the addicted individual and abstinence" (p. 54, those who use illicit drugs, street involved youth, sex
para. 144). trade workers as well as transvestite and transsexual
persons), Canadian Public Health Association, British
The rival perspective, abstinence at all cost, would Columbia Civil Liberties Association, Attorney Gen-
presumably withhold supervised injection as a health- eral of Quebec and Dr. Peter AIDS Foundation. The
care intervention even if it resulted in preventable fatal Attorney General of British Columbia was a respondent
overdoses. This was the very reason the PHS entered with regard to the doctrine of inter jurisdictional
the courts: we believe that harm reduction is a door- immunity that was won at the Appeal Court level and
way into treatment, detox, and abstinence and that argued for localized control over Insite as a provincial
the safeguarding of human life offered by supervised initiative.
MI Ali II .1 II .al IIILA
8 Public Health
Sadly, there has always been a psychosocial phe- Pussin outlawed the employment of chains to imprison
nomenon that is culturally expunged or relegated to the the mentally ill in 1797 while serving as the governor
shadows due to the moral anxieties that it creates in of Bicetre.12 Pussin is an important forerunner in the
the wider community. This has been true in the case history of humanistic mental health services.
in the present and past with issues or experiences that In popular culture, physician Phillippe Pinel is
make people uncomfortable such as death, sexuality, widely thought of as being the first individual to liber-
mental illness, or addiction, which are sequestered in ate the mentally ill from chains. The renowned painting
social life and institutional settings.8 This process of Madwomen at the Salpetriere, painted by Tony-Robert
sequestering, or hiding away of the social phenomena Fleury, has helped to immortalize this legend. The
that alarm or anger us, also takes its shape in the form painting shows Pussin removing the chains of psychi-
of cultural erasures and silences, things that are unsaid atric patients while Pinel looks on and symbolizes a
which can, in actuality, be more influential than what is transition towards more humanistic approaches to
said.9 Addiction is just such a phenomenon; people with the mentally ill in the 18th century. Today the paint-
addictions have been sequestered, silenced, and erased ing hangs in the entrance to the Library Charcot in the
from positive social life. Their personhood has been so Salpetriere hospita1.14
dramatically undermined that their identities are some- In fact, it was Pussin who inspired Pinel to ban the
times socially spoiled,1° leaving them metaphorically use of chains for detaining people living with mental
chained. Throughout the education campaign to save illnesses.12,15 After having worked at Bicetre between
Insite, we attempted to combat this identity erosion by 1793 and 1795, Philippe Pinel was so inspired by the
highlighting the personhood of people living with ad- work of Pussin that he credited him with the emancipa-
dictions by echoing the idea that everyone living with tion of the mentally ill and the first actual application of
addictions was someone's son or daughter. humanistic psychiatric treatment. When Pinel moved
The field of mental health provides an interest- to Salpetriere, the largest hospital in Paris, he estab-
ing analogy with respect to the process of liberation lished a post for Pussin, who took up the position there
of people with pain from societal shackles of stigma from 1801 until his death in 1811.14 At Salpetriere
and incarceration. Psychiat.:y's approach to people in Paris, Pussin and Pinel worked together to apply
with mental illnesses transitioned towards humanistic humanistic approaches to psychiatric treatment.
treatment in the 1780s under the leadership of more There is some humanistic truth at the heart of
humanistic psychiatrists and reforms in the mental legendary characterization of Pinel as the person that
health field. Physician Vincenzo Chiarugi (1759-1820) liberated the mentally ill from their chains. He did, in
led a movement espousing the humane treatment of the fact, show significant leadership by moving away from
mentally ill, which took its first applied step, in prac- abandonment and brutal imprisonment to a therapeutic
tice, in 1788 with the opening of Hospital Bonifacio in approach based on medical science and compassion.12
Florence, Italy, where he served as the physician direc- Similarly, the Insite legal case helps to make a similar
tor.'' Chiarugi's approach was in keeping with the goals transition from the cruelty of criminalization in addic-
of Grand Duke Pietro Leopold of Tuscany, a socially tion to a healthcare model where people with addiction
conscious aristocrat who ordered the establishment have fundamental rights to life-saving healthcare.
of Bonifazio, and was predicated on respectful and The mental health field transitioned from a model
humane treatment.12 He is a significant figure in the based on incarceration and neglect of the mentally ill
history of humanistic medicine and one of the fathers to an approach based on compassion, science, and
of compassionate psychiatry. Chiarugi is thought to be medical treatment in the 18th century. In the addic-
the first figure to forbid the use of chains to restrain tion field, this transition has taken another two cen-
the mentally ill (a policy which he established during turies (21st century). The metaphor of chaining of the
his role as physician director of Santa Doretea hospital addicted goes beyond the symbolic. In 2009, the US
before 1793).12 incarcerated more than 400,000 individuals for non-
Similarly, Jean-Baptiste Pussin (1746-1811) and violent drug offenses (a greater number than those
Madame Marguerite Pussin (1754—?) helped to infuse incarcerated for all offenses in the 27 nations of the
compassion into the practice of mental health.12,13 After European Union combined).16
having been a patient at Bicare hospital in the sub- The incarceration of the addicted has been s o dra-
urbs of Paris, Jean-Baptiste Pussin went on to become matic that, when seen through the lens of epidemiol-
the director of a psychiatric ward from 1785 to 1802 ogy, it can be considered as a catastrophic event that
during which time he implemented a series of compas- has resulted in tremendous suffering and death. The
sionate improvements in the treatment of mentally ill.14 epidemiological tool of years of life lost (YLL) is useful
for examining the impact of a criminal justice ap- of a localized response to a healthcare emergency.
proach to addiction. Drucker defines years of life lost Supervised injection, in this case, began from the
as "the number of years between the victim's age at ground up rather from a top-down policy. The goals
death and the age that his or her usual life expectancy of Insite originated out of local need, inspired by the
would predict. Thus, for the average American with idea that people with addictions deserved something
a life expectancy of 75 years, a child's death at age better than death from overdose. The goals and out-
ten implies a loss of sixty-five potential years of life"16 come measures were simple. The program aimed to
(p. 69). Building on this logic, Druckernotes that 1,513 provide a doorway to life, supported housing, phys-
people died representing an estimated 47,000 YLL icians, healthcare services and supports. The intent
in the Titanic disaster, 2,819 deaths representing of the program was to provide its interventions (e.g.,
an estimated 104,303 YLL in the World Trade Center clean syringes, supervised injection) in an accessible
tragedy and reasons that over the past 35 years (since way without barriers (such as abstinence or onerous
the introduction of the Rockefeller drug laws in intake procedures).
New York), more than 7 million people have been in- One thing that is clearly demonstrated by the
carcerated. This translates into an estimated 14 million case of Insite is that science is not enough, on its
YLL representing 350,000 deaths in a group of the own, to change public policy, especially in stigma-
same age.16 tized areas like addiction. The simple existence of a
The Insite victory is emblematic of a different scientific evidence base does not automatically lead
cultural understanding of addiction that is supplant- to changes in policies or practice. Policy-makers
ing a traditional one. This newer approach assumes and elected officials need to pay attention to the
people living with addictions are in need of health- evidence base. In some disappointing instances, as
care rather than punishment through the criminal the case of Insite has shown, policy-makers need to
justice system. be forced to pay attention to the established facts.
By focusing on the federal Health Minister's re- Conversely, scientists need to take an active role in
fusal to provide an exemption for Insite under the affecting public policy when the evidence indicates
existing regulatory framework, the scc did not have the need for change.'' The SIF, for instance, had more
to make any alterations to existing provincial and than three dozen peer-reviewed papers associated
federal jurisdictions over the program. Canada's fed- with its evaluation.18-47 Despite the fact that much
eral government attempted to argue that the federal of the evaluation was paid for by the Government of
minister of health had, technically, never "not given" Canada, they chose to ignore the scientific findings.
a permit for Vancouver's supervised injection site and The evidence base generated by Canada's supervised
therefore never formally jeopardized its operation. injection trial should have earned it a medical exemp-
However, the court ruled that it was self-evident that tion (the next stage of operational permit after the
the federal Health Minister had every intention to scientific authorization originally granted to Insite),
close Insite: but it was not provided. Uncompromising advocacy,
including public protest and legal challenge, was re-
The Minister of Health must be regarded as having quired to obtain the permit.
made a decision whether to grant an exemption, In this way, science in healthcare and applied re-
since he considered the application before him search are not the same.' While science in healthcare
and decided not to grant it. The Minister's deci- is portrayed as "objective," applied research is seen
sion, but for the trial judge's interim order, would as rooted in the context of the community needs.
have prevented injection drug users from access- From my perspective, science in healthcare needs
ing the health services offered by Insite, threat- to move more towards clinical application. Yes, it
ening their health and indeed their lives [emphasis needs to be sound and rigorous, but its main purpose
added]2 (p. 9). should be to serve the patients, families, and the com-
munity. Scientific evaluation of the Insite may have
The Scc concluded that the Minister's intention to shut been a necessary condition, but it was certainly not
Insite would have threatened the lives of the people a sufficient condition to bring about a permanent
who rely on the program. It also noted the program change in public policy or a sustainable supervised
would not have remained open had it not been for pro- injection facility.
tection provided by the Supreme Court of BC. Sometimes scientists and bureaucrats spend too
National public policy with respect to supervised much time worrying about protecting objectivity at
injection in Canada began with the establishment the expense of advocacy. Yet, there are, time and time
8 Public Health
again, instances where advocacy needs to be under- Correspondingly, one shouldn't be required in the
taken and undertaken vigorously. The supervised case of harm reduction programs. If the evidence base
injection facility was one of those times. It demanded is there to support an intervention, then we should
public advocacy. Despite the victory, on the day of move past consensus-building when it comes to life-
court announcement, I felt relief more than elation. I saving healthcare. The SCC has ruled that the morality
couldn't help but wonder what would have happened of an activity, such as drug addiction, isn't enough
to Insite if things had been different? If we had lost to ignore someone's rights to security of their person
the court case, would the various stakeholders have under section 7 of Canada's Charter of Rights and
chosen the ethically sound course of action by con- Freedoms (the Charter):
tinuing to operate in spite of an unjust law prohib-
iting the facility's operation? There would have been Additionally, the morality of the activity the law
a strong ethical case for breaking the law and keep- regulates is irrelevant at the initial stage of de-
ing Insite open. There was no scientific uncertainty termining whether the law engages a s. 7 right.
about the effectiveness of Insite at engaging a hard to Finally, the issue of illegal drug use and addic-
reach population in healthcare and saving lives. The tion is a complex one which attracts a variety
only equipoise was political. Had we lost, would we of social, political, scientific and moral reac-
have been forced to live through a medical, legal and tions. While it is for the relevant governments to
ethical disaster while people died of preventable over- make criminal and health policy, when a policy
doses because it was against the law? These are dark is translated into law or state action, those laws
questions that haunted Insite right up until the very and actions are subject to scrutiny under the
moment that the final decision was rendered. Thank- Charter"2 (p. 9).
fully, we never had to publically confront what might
have been had we lost. The suitability of supervised injection shouldn't be
In my view, there are three important cultural debated any longer as though it were on par with a
ramifications of this case. Firstly, this legal decision discussion on a sports show about which sports team
says a lot about what it is to be a person, to have per- will win the championship. Supervised injection is
sonhood, in Canada. Personhood symbolizes our con- healthcare, and whether it is required in a jurisdiction
nection to the wider human family.49 What it is to be needs to be determined by evidence and not arbitrary
a person exists in the borderland of human relations opinions or fickle political stances in search of votes. If
where personal agency and meaning are psychosocially science and medicine have established the best course
constructed as part of an inner and outer conversation. of action, then we shouldn't turn to opinion polls to
Socially devalued features of our selves such as addic- determine the best healthcare.
tion reduce our opportunities to be on the threshold Finally, it is my opinion that this ruling provides
of a successful life.10 The personhood of people with further affirmation that many healthcare providers
addiction is often undermined or threatened by poli- know. harm reduction needs to be an explicit part of
cies, programs, and implicit or explicit exclusion (e.g., the standard of care now: Every single health author-
drug courts and therapeutic communities are typically ity and region in Canada should have a proactive policy
founded on the principle that addicts must abdicate a detailing best practices in harm reduction when it is
portion of their self-determination). The personhood epidemiologically indicated. It is simply not acceptable
of people with addictions in Canada has been empha- to pretend harm reduction doesn't exist or to let moral
sized, and their constitutiona_ rights feature promin- opposition rather than evidence-based analysis guide
ently in this legal ruling. decisions in this area.
The second consequence of this case is that other Any jurisdiction that doesn't have a positive
jurisdictions may, and should, establish supervised policy on harm reduction is misguided. An example is
injection if the epidemiological variables demand provided by the City of Abbotsford in British Colum-
it. On this point, I disagree with the overemphasis bia. In June 2005, Abbotsford amended their zoning
on obtaining consensus from an overly broad col- (bylaw no. 1378-2004) in order to overtly exclude
lection of stakeholders (e.g., the municipal govern- harm reduction:
ment, the local police) who are typically consulted
in order to obtain their blessing so that lives can be Prohibited uses would include safe injection
saved by supervised injectio a. A letter of support sites, needle exchanges, mobile dispensing vans,
from the Chief of Police or Mayor of a city would not methadone treatment facilities and other types or
be required to establish a cancer treatment program. similar uses.
The experience of Insite should be an important fact that the program has survived is itself an indica-
lesson for jurisdictions that ignore, or in the case of tor of social change, and I believe that this ruling
Abbotsford, outlaw, harm reduction as part of health- signals that we have reached an important cultural
care. They do so at their own peril legally, medically milestone. We've gone so much further than hoping
and ethically. that addiction will one day be understood as a
This commentary is not meant to be a distant healthcare matter. The very fact that Insite can exist,
scientific paper but, instead, an experience-based with the permanence of a Supreme Court decision,
and socially positioned interpretation of events that supports harm reduction as part of the standard of
I have lived. As one of the creators of Insite, I had care, with the sustainability it deserves. With this
always imagined that establishing an injection site ruling, we've moved beyond hopefulness to a point
would be the most difficult task to accomplish. In in our history where people with addictions have
fact, it seems to me that the protection of it, once been unchained. Canada's highest court has spoken.
established, has been an even larger challenge. The It's the law.
struggle to protect Insite has distracted us from many
other important prevention, treatment, enforcement, If you have built castles in the air, your work need
and harm reduction initiatives that need to be es- not be lost; that is where they should be. Now put
tablished to comprehensively address addiction. The the foundations under them" (p. 255).
References
1. Marx K: Selected Works. In Theses on Feuerbach. 1845 12. Gerard DL: Chiarugi and Pinel considered: soul's brain/
edition. Edited by Marx K, Engels F Moscow: Progress person's mind. J Hist Behav Sci 1997, 33(4):381-403.
Publishers; 1989:13-15. 13. Schuster J-P, Hoertel N, Limsin F: The man behind
2. Supreme Court of Canada: Canada (Attorney General) Philippe Pinel: Jean-Baptiste Pussin (1746-1811).
v. PHS Community Services Society. In SCC 44 33556. Br J Psychiatry 2011, 198:241.
Edited by Canada SC. Ottawa: Supreme Court of 14. Schuster J-P, HoertelN , LimosinF :The man behind Philipe
Canada; 2011. Pinel: Jean-Baptiste Pussin (1746-1811)-psychiatry in
3. Small D: Mental illness, addiction and the supervised pictures. Br J Psychiatry 2011, 198:241.
injection facility. Visions: BC's Mental Health and Addic- 15. Weiner DB: Philippe Pinel "Memoir of Madness" of
tions Journal 2004, 2(1):37-39. December 11, 1794: a fundamental text of modern
4. Harm Reduction Journal. Definition of Harm Reduction; psychiatry. Am J Psychiatry 1992, 149(6):725-732.
2012. [cited 2012 24 April]; available from: http://www. 16. Drucker E: A Plague of Prisons: The Epidemiology of Mass
harmreductionjournal.com. Incarceration in America. New York: The New Press; 2011.
5. Solomon R, Green M: The First Century. The history of 17. Small D, Drucker E : Policy makers ignoring science and
nonmedical opiate use and control policies in Canada, scientists ignoring policy: the medical ethical challen-
1870-1970. University of Western Ontario Law Review ges of heroin treatment. Harm Reduction Journal 2006,
1982, 20(2):307-337. 3 (16): 1-14 .
6. Pitfield, Justice I: PHS Community Services Society v. At- 18. DeBeck K, Wood E, Zhang R, Tyndall M, Montaner J,
torney General of Canada: BCSC 661: The Supreme Court Kerr T: Police and public health partnerships: evidence
of British Columbia; 2008. from the evaluation of Vancouver's supervised injection
7. Alexander BK, Hadaway PF: Opiate addiction: the facility. Substance Abuse Treatment, Prevention and Policy
case for an adaptive orientation. Psychol Bull 1982, 2008, 3(1):1-5.
92(2):367-381. 19. Fairbairn N, Small W, Shannon K, Wood E, Kerr T:
8. Giddens A: Modernity and Self-Identity: Self and Society Women's experiences in North America's first medic-
in the Late Modern Age. Stanford: Stanford University ally supervised safer injection facility. Soc Sci Med
Press. Cambridge: Polity Press; 1997. First pub. 1991, 67(8):817-823.
Cambridge: Polity Press. 20. Kerr T, Kimber J, DeBeck K, Wood E: The role of
9. Saris AJ: Telling stories: life histories, illness narratives safer injection facilities in the response to HIV/AIDS
and institutional landscapes. Cult Med Psychiatry 1995, among injection drug users. Curr HIV/AIDS Rep 2007,
19(1):39-72. 4(4): 158-164 .
10. Goffman E: Stigma: Notes on the Management of Spoiled Iden- 21. Kerr T, Small W, Moore D, Wood E: A micro-environ-
tity. New York: Simon and Schuster, Inc.; 1986 (1963). mental intervention to reduce harms associated with
11. Mora G: Vincenzo Chiarugi (1759-1820) and his drug-related overdose: evidence from the evaluation of
psychiatric reform in Florence in the late 18th century. Vancouver's safer injection facility. Int J Drug Policy 2007,
J Hist Med 1959, 14(October):424-433. 18:37-45.
8 Public Health
22. Kerr T, Stoltz J, Tyndall M Li K, Zhang R, Montaner J, 36. Wood E, Kerr T, Stoltz J, Qui Z, Zhang R, Montaner JSG,
Wood E: Impact of a medically supervised safer injec- Tyndall MW: Prevalence and correlates of hepatitis C among
tion facility on community drug use patterns: a before users of North America's first medically supervised safer
and after study. Br MedJ 2006, 332:220-222. injection facility. Public Health 2005, 119(12):1111-1115.
23. Kerr T, Tyndall MW, Lai C, MontanerJSG, Wood E: Drug- 37. Wood E, Kerr T, Tyndall MW, MontanerJSG: The Canadian
related overdoses within a medically supervised safer government's treatment of scientific process and evidence:
injection facility. Intl Drug Policy 2006, 17:436-441. inside the evaluation of North AmericA first supervised
24. Kerr T, Tyndall MW, Lai C, Montaner JSG, Wood E: injecting facility. Int J Drug Policy 2007, 19(3):220-225.
Circumstances of first injection among illicit drug users 38. Wood E, Montaner JS, Kerr T: Reflection and reaction: il-
accessing a medically supervised safer injecting facility. licit drug addiction, infection disease spread, and the need
Am J Public Health 2007, 97(7):1228. for an evidence-based response. Lancet 2008, 8:142-143.
25. Kerr T, Tyndall MW, Li K, Montaner JS, Wood E: Safer 39. Wood E, Tyndall MW, Lai C, Montaner JSG, Kerr
injecting facility use and syringe sharing among injec- T: Impact of a medically supervised safer injecting facil-
tion drug users. Lancet 2005, 366:316-318. ity on drug dealing and other drug-related crime. Subst
26. Kerr T, Wood E, Small D, Palepu A, Tyndall MW: Poten- Abuse Treat Prey Policy 2006, 1(13):1-4.
tial use of safer injecting facilities among injection drug 40. Wood E, Tyndall MW, Li K, Lloyd-Smith E, Small W,
users in Vancouver's Downtown Eastside. Can Med Assoc Montaner JSG, Kerr T Do supervised injecting facilities
J 2003, 169(8):1-5. attract higher-risk injection drug users? Am J Prey Med
27. McKnight I, Maas B, Wood E, Tyndall MW, Small W, Lai 2005, 29(2):126-130.
C, et al: Factors associated with public injecting among 41. Wood E, Tyndall MW, Montaner JS, Kerr T: Summary
users of Vancouver's supervised injection facility. Am J of findings from the evaluation of a pilot medically
Drug Alcohol Abuse 2007, 33(2):319-325. supervised injecting facility. Can Med Assoc J 2006,
28. Milloy MJ, Wood E, Small W, Tyndall M, Lai C, Mon- 175(11):1399-1404.
taner J, Kerr T: Incarceration experiences in a cohort 42. Wood E, Tyndall MW, Qui Z, Zhang R, Montaner JS,
of active injection drug users. Drug Alcohol Rev 2008, Kerr T: Service uptake and characteristics of injection
27(6):1-7. drug users utilizing North America's first medically
29. Petrar 5, Kerr T, Tyndall MW, Zhang R, McKnight B, supervised safer injection facility. Am J Public Health
Montaner JSG, Wood E: Injection drug users' percep- 2006, 96(5):770-773.
tions regarding use of a medically supervised safer in- 43. Wood E, Tyndall MW, Stoltz J, Small W, Lloyd-Smith E,
jetting facility. Addict Behav 2007, 32:1088-1093. Zhang R, Montaner JSG, Kerr T: Factors associated with
30. Stoltz JA, Wood E, Small W Li K, Tyndall M, Montaner syringe sharing among users of a medically supervised
J, Kerr T: Changes in injecting practices associated with injecting facility. Am J Infect Dis 2005, 1(1):50-54.
the use of a medically supervised injection facility. J Pub 44. Wood E, Tyndall MW, Stoltz J, Small W, Zhang R,
Health (Oxford) 2007, 29(1):35-39. O'Connell J, Montaner JSG, Kerr T: Safer injecting
31. Tyndall MW, Kerr T, Zhang R, King E, Montaner JG, education for HIV prevention within a medically super-
Wood E: Attendance, drug use patterns, and referrals vised safer injecting facility. Int J Drug Policy 2005,
made from North America's first supervised injection 16:281-284.
facility. Drug Alcohol Depend 2005, 83(3):193-198. 45. Wood E, Tyndall MW, Zhang R, Stoltz J, Lai C, Montaner
32. Tyndall MW, Wood E, Zhang R, Lai C, Montaner JS, Kerr JSG, Kerr T: Attendance at supervised injecting facilities
T: HIV sero-prevalence among participants at a super- and use of detoxification services. N Eng J Med 2006,
vised injection facility in Vancouver, Canada: implica- 354(23):512-514.
tions for prevention, care and treatment. Harm Reduction 46. Wood E, Tyndall MW, Zhang R, Montaner JS, Kerr T:
Journal 2006, 3(36):1-5. Rate of detoxification service use and its impact among
33. Wood E, Kerr T, Buchner C, Marsh D, Montaner JS, a cohort of supervised injection facility users. Addiction
Tyndall MW: Methodology for evaluating Insite: Can- 2007, 102:916-919.
ada's first medically supervised safer injection facility 47. Wood RA, Wood E, Lai C, Tyndall MW, Montaner JSG,
for injection drug users. Harm Reduction Journal 2004, Kerr T: Nurse-delivered safer injection education among
1(9):1-5. a cohort of injection drug users: evidence from the
34. Wood E, Kerr T, Montaner JS, Strathdee 5, Kerr T, evaluation of Vancouver's supervised injection facility.
Wodak A, Spittal P, Hankins C, Schechter MT, Tyndall Int J Drug Policy 2008, 19(3):183-188.
M: Rationale for evaluating North America's first medic- 48. Latour B: From the world of science to the world of re-
ally supervised injecting facility. Lancet Infect Dis 2004, search. Science 1998, 280(5361):208-209.
4(5):301-306. 49. Small D: Looking into the cultural mirror: addic-
35. Wood E, Kerr T, Small W, Li K, Marsh D, Montaner JS, tion, secret lives and lost personhood. Visions 2005,
Tyndall MW: Changes in public order after the opening 2(5):29-30.
of a medically supervised safer injection facility for injec- 50. Thoreau HD: Walden. Vermont: Orion Publishing
tion drug users. Can Med AssocJ 2004, 171(7):731-734. Group; 1995 (1910).
Christie/Groarke/Sweet Virtue Ethics as an Alternative to Deontological and Consequential Reasoning
Virtue Ethics as an Alternative to the three most influential models in contemporary eth-
ical thought and, second, it is more robust than simply
Deontological and Consequential referring to abstract principles and then using intuitions
Reasoning in the Harm Reduction to determine which principle is most important.
Debate
Harm Reduction
Timothy Christie, Louis Groarke, and William Sweet
To discuss harm reduction it is necessary to identify
(1) its fundamental assumptions and definition, (2)
Introduction specific kinds of harm reduction interventions, and (3)
the implications of the political advocacy of many harm
The purpose of this paper is to extract and analyze reduction proponents.
ethical theories that bear on two approaches to dealing The fundamental assumption of harm reduction,
with problematic substance use, i.e., abstinence-based which is the primary focus of this paper, is that it is
approaches and harm reduction approaches. The pri- important to try to reduce drug-related harm (Erick-
mary issue to be investigated is how harm reduction— son, Butters, & Walko, 2007). The harm reduction
because it does not require individuals to either reduce approach does not require individuals to reduce or ab-
their consumption of illicit substances or to abstain stain from drug use; rather, it attempts to mitigate the
from illicit substance use---can be ethically justified. negative consequences of drug use. Abstinence could
The harm reduction approach claims that many of the be an eventual outcome and is consistent with harm
negative consequences associated with problematic reduction; however, abstinence is not a condition of the
substance use are avoidable through specific interven- harm reduction approach.
tions, and these interventions can be effective regardless There are numerous kinds of harm reduction inter-
of whether substance use persists. The abstinence-based ventions, and they apply in different ways depending
approach, however, claims that it is important to get on the particular substances and activities involved: il-
individuals off drugs or at least to decrease consump- licit opiates, cocaine, ecstasy, alcohol, tobacco, etc. The
tion. This latter approach generally maintains that, specific interventions discussed in this paper are needle
since harm reduction interventions tolerate continued exchange programs and supervised injection sites. Not
substance use, they send the wrong message. all harm reduction interventions have the same ethical
We employ three different ethical models to analyze justification. For example, needle exchange programs
the values conflict between the harm reduction and the and supervised injection sites assist individuals who are
abstinence-based approaches. Our methodology in this already engaged in an activity. If an individual is going
paper is different than that usually employed in address- to inject drugs regardless of the harms involved, needle
ing ethical issues in the applied ethics field (Beauchamp exchange gives that person clean needles, and a super-
& Childress, 2001). Typically in applied ethics, the vised injection site provides a medical setting in which
principles of autonomy, beneficence, nonmaleficence, the person can inject the drugs, but in neither case
and justice are first applied to specific situations. Then does the intervention involve supplying the substance.
the analyst will use his or her moral intuitions to deter- The ethical justification for other harm reduction
mine which principle or principles are most important interventions would be different. For instance, heroin
given the particular facts of the situation. We approach prescription programs actually provide the substance
the debate in a different way because depending on to individuals, which is different than simply providing
moral intuitions is highly problematic. Specifically, we sterile equipment or supervising injections. Noting this
will focus on the philosophical theories from which the difference is not to make the judgment that there may be
above-mentioned principles have been abstracted. For something ethically problematic with heroin prescrip-
example, the principle of autonomy is abstracted from tion; it simply recognizes that the situation is different
Deontological ethics, beneficence and nonmaleficence and, therefore, the ethical justification may be different.
from Utilitarianism, and the principle of justice was For the sake of simplicity, we have chosen to focus on
originally articulated by Aristotle who is a key source the generic philosophical aspects of harm reduction,
of contemporary Virtue Ethics. The advantages of our with specific references to needle exchange programs
approach are, first, it directly addresses what are clearly and supervised injection sites, when necessary.
Source: From International Journal of Drug Policy 19, 1 (2008), 52-8. Reprinted with permission.
II JII III I.k.1
8 Public Health
Finally, as with any important development in essential to deter substance use (ONDP, 2007b) and (3)
public policy, there are elements of political advocacy it is imperative to disrupt illegal drug markets (ONDP,
within the harm reduction movement. Some advo- 2007c). The impetus for these objections is the belief
cates argue for complete legalization of illicit drugs that any policy that tolerates drug use sends a "mixed
and radical drug law reform (Hankins, 2000), while signal" and frustrates primary prevention efforts.
others argue for a medical model that is more cautious It seems, however, that these arguments are simply
(Anderson, 2000). We address the advocacy issue per- a restatement of the first objection, namely, that harm
ipherally in the sense that we think harm reduction reduction will encourage illegal drug use. For example,
makes ethical sense, is superior to alternative policy if the outcome of encouraging illegal drug use was
approaches to problematic substance use, and can sur- not relevant to this argument, it is difficult to under-
mount the standard ethical objections to it. stand what "sending a mixed signal" would mean or
why it would even be a relevant consideration. The
Criticisms of Harm Reduction significance of this objection is mitigated by the same
evidence that invalidates the first objection. A third
The most important ethical concern with harm reduc- criticism of harm reduction is found in the following
tion is related to the "value-judgment" that it is more quotation from the (current) Canadian Federal Minis-
important to reduce the harms associated with drug ter of Health, Tony Clement:
use than it is to reduce or prDhibit drug use. The con-
troversial character of this value judgment is amplified Do safe injection sites contribute to lowering drug
because the relevant drugs are illegal. Critics of harm use and fighting addiction? Right now the only
reduction have argued that (:.) it encourages drug use, thing the research to date has proven conclusively
(2) it sends a mixed message, and (3) it fails to get is [that] drug addicts need more help to get off
people off drugs. drugs. Given the need for more facts, I am unable
There is a significant amount of the literature, to approve the current request to extend the
however, supporting the claim that harm reduction Vancouver site for another three and a half years
interventions do not increase or encourage drug use. (Health Canada, 2006).
In fact, the evidence demonstrates the opposite. The
experience of the Canadian supervised injection site Scientific evaluation of the Canadian supervised
shows that some patients, who would not have other- injection site revealed that the facility attracted drug
wise sought treatment, evenually seek treatment, in- users who were formerly most likely to inject in public
cluding abstinence-based programs, as a result of using and be at highest risk of HIV infection and death due
the site (Tyndall et al., 2005). This is consistent with to overdose (Wood, Kerr, & Montaner, 2004a; Wood,
what has been observed in needle exchange programs Kerr, & Small, 2005a; Wood, Tyndall, & Li, 2005b).
as well; participants frequently seek referrals for treat- The opening of the supervised injection site was fol-
ment (Strathdee, Celentano, & Shah, 1999; Hagan, lowed by statistically significant reductions in public
McGough, & Thiede, 2000). Based on what is cur- drug use and publicly discarded syringes (Wood et al.,
rently known, the claim that harm reduction programs 2004b); was associated with reduced syringe sharing
encourage drug use is unsubstantiated. and other forms of HIV risk behaviour (Kerr, Tyndall,
The second criticism of harm reduction is that it Li, Montaner, &Wood, 2005); and was a key referral
"sends a mixed signal." This objection was made in a mechanism to addiction treatment and other com-
1996 letter to the Governor's Advisory Council on AIDS munity resources, including abstinence-based pro-
(Whitman, 1996). In this letter, Christine Whitman, grams (Tyndall et al., 2005). It was not associated with
the former Governor of New Jersey, acknowledges increased drug use or other drug-related problems
that the National Academy of Science and the Centers (Kerr et al., 2006).
for Disease Control and Prevention endorse needle The Minister's reasoning, however, suggests that it
exchange programs as effective interventions for inter- is more important to "contribute to lowering drug use
rupting the spread of HIV. She argues, however, that [,] fighting addiction," and getting people "off drugs"
needle exchange programs "send a mixed signal" and than it is to achieve these outcomes. This belief is so
that governments should not "be in the business of fa- strongly held that the Minister refused to renew the
cilitating illegal activity" These sentiments are echoed Section 56 exemption to the Controlled Drugs and
in the United States' National Drug Control Strategy, Substances Act given to the supervised injection site,
which states that (1) there should be no tolerance for which will make it illegal for the site to operate after
substance use whatsoever (ONDP, 2007a), (2) it is this exemption expires in June 2008.
Christie/Groarke/Sweet: Virtue Ethics as an Alternative to Deontological and Consequential Reasoning
This is a substantive objection to the theoretical pursue their own life goals as they see fit. The only
underpinnings of harm reduction and requires a care- justifiable limit to an individual's liberty occurs when
ful ethical analysis. If a program will get people off that person's actions will harm others. This is called
drugs, then the Minister would seem to approve of the "harm principle." (Mill did not consider merely
it; if it does not get people off of drugs, the Minister giving of fence or causing inconvenience as harm. He
is definitely against it, even if the program otherwise thought if harm were defined to include these ways,
avoids many of the negative consequences of illicit individuals could be prevented from criticizing social
drug use. Embedded in this objection is the values- conventions.) His contention was that by restraining
based preference for an abstinence-based approach the state's ability to interfere with the liberty rights of
to drug policy. The abstinence-based approach differs individuals, the long-term consequences for society
from harm reduction in a very important respect; it will be positive.
requires the individual either to abstain from drugs or Many liberal theorists root their theories in some
attempt to abstain from drugs (Christie & Anderson, conception of "natural" or "human rights." They main-
2003; Tammi & Hurme, 2007). The harm reduction tain that rights, such as the rights to life, liberty, and
approach, however, does not require the individual property, are natural, inalienable, and intrinsically valu-
to either abstain from or reduce consumption of illicit able. However, Mill thinks that individual liberty rights
drugs. It simply seeks to prevent the negative conse- are instrumental, i.e., good for the sake of achieving
quences of illicit substance use. something else. This idea is rooted in his commitment
to Utilitarianism. Utilitarianism is the theory which
Ethical Analysis contends that the foundation of morality is the greatest
happiness principle and holds that an action is right
In his recent article entitled "Ethics and Drug Policy," if it tends to promote the greatest happiness for the
Alex Wodak argues that the scientific debate about greatest number and wrong if it tends to produce the
harm reduction is over (Wodak, 2007a). Harm reduc- opposite of happiness. Happiness is defined as pleasure
tion works. The real conflict, which has largely been and the absence of pain, and unhappiness is defined as
overlooked, is about ethics. More specifically, it is a pain and the privation of pleasure. Mill contends that
revival of the debate between Utilitarian and Deonto- pleasure and freedom from pain are the only things
logical approaches to ethics. Wodak claims that, im- that are intrinsically valuable and that everything else
plicit in the reasoning of harm reduction advocates, is instrumental to them.
is a Utilitarian argument, which holds that the key When determining what will produce happiness,
guideline in ethics is that if negative consequences can Mill is not focusing just on the agent's own happiness
be avoided, they should be avoided. Abstinence advo- but on that of the entire community. In determining
cates, however, generally seem to employ a Deonto- this, each agent has one "vote." For the Utilitarian, the
logical ethic, which maintains that the moral worth of consequences for the community are of fundamental
one's actions has nothing to do with the consequences importance, and when calculating the consequences
of those actions but, rather, is determined by the in- we must consider all relevant variables, including the
tention of the actors. Our analysis will examine this difference between short-term and long-term conse-
debate in more depth. We will present and interpret quences. For example, an action might produce much
the ethical theories of the Consequentialist Utilitar- happiness in the short-term but, in the long-term,
ian, John Stuart Mill, of the founder of Deontological result in more harm than good; or the opposite might
ethics, Immanuel Kant, and Aristotle's Virtue Ethics, be true.
which many see as a remedy to the Consequentialist— With regard to harm, Mill's "liberal" view clearly
Deontological dilemma. makes allowances for individuals to harm themselves
(e.g., to consume alcohol to excess). But if drunken-
Utilitarianism (John Stuart Mill)
ness leads an individual to neglect his or her respon-
sibilities or to interfere with others, Mill would allow
John Stuart Mill was a 19th century British philoso- the state to intervene. The ethical issue is not the
pher who was seminal in the development of liberal- drinking or the harm to oneself but the individual ful-
ism (Arnold, 2006) and is still recognized as one of filling his or her responsibilities—although drinking is
the key proponents of Utilitarianism (Mill, 1863). the cause of the failure to carry out one's responsibil-
His theory of liberty holds that individuals in society ities. If a person could drink and not harm society, Mill
should have broad-ranging freedoms, such as liberty would certainly not allow the state to interfere with
of thought, discussion, association, and the right to that person's liberty.
AI II :Al 1111:1.1
8 Public Health
This bears on the cun•ent discussion about drug Deontological Ethics (Kant, 1785)
use. Consider, for example, the harms of injection-
related infections, overdoses, blood-borne disease, vio- Immanuel Kant was an 18th century German philoso-
lence, property crime, involvement in the sex trade, and pher who considered ethics an essential component
incarceration (Kerr & Palepu, 2001; Hunt, 2005) The of human life. He contrasted ethics with physics, ex-
cost for every untreated opiate user is estimated to be plaining that physics describes the laws of nature ac-
over $45,000 (Cdn) per year (Wall, Rehm, & Fischer, cording to which every material thing "acts" and ethics
2000), the lifetime cost of treating a person with HIV describes the laws of morality according to which every
exceeds $250,000 (Cdn) (Kuyper, Hogg, Montnaer, person "ought to act." He lamented the fact that fre-
Schecter, & Wood, 2004), a course of Hepatitis-C quently what "ought to happen" does not happen be-
treatment can range from $10,000 to $30,000 (Cdn) cause, unlike purely material objects, human agents are
per patient (Public Health Agency of Canada, 2003), free to choose how to act—i.e., whether to follow the
and Emergency Department utilization is significantly ethical law Since human beings are rational creatures,
higher among those with problematic substance use Kant thought the laws of ethics could be derived from
than those without (Palepu, Tyndall, & Leon, 2001; "reason" alone without having to appeal to specific cir-
McGeary & French, 2000). cumstances or experience. On this theory, whether an
Mills liberal view, then, would allow the state to action is right or wrong depends on the intrinsic nature
interfere with a person's libe7ty to use such substances of the action rather than on the situation or specific
when that use leads to harmful consequences for society. circumstances in which one finds him or herself. In
However, his liberalism does not answer the question of other words, for Kant, ethics is primarily concerned
what is the most appropriate state response: abstinence- with doing the right thing because it is the right thing
based policies or harm reduct on policies? The answer to to do, not because it is in the agent's self-interest, or
this question requires a Utilitarian analysis. Utilitarianism because it will produce good consequences, or for any
requires us to consider the current policy environment, other "instrumental" reason.
the scientific evidence for each approach, and whether it According to Kant, only actions motivated by
is effective or not. This will give us guidance as to which a sense of duty have moral worth, and duty consists
approach best promotes the general happiness. of acting out of respect for the moral law (or, as Kant
As discussed above, the policy environment—at calls it, the Categorical Imperative). The Categorical
least in the United States and Canada—is predomin- Imperative can be stated as follows: first, "act only on
ately abstinence-based; it is also influenced by the that maxim whereby thou canst at the same time will
criminal justice system in which prohibition is domin- that it should become a universal law," or, secondly, "so
ant. For example, in Canada, 73% of federal spending act as to treat humanity, whether in thine own person
in the National Drug Strategy is spent on enforcement or in that of any other, in every case as an end withal,
measures such as border control, police investigations, never as a means only"
and federal prosecution expenses, whereas less than For Kant, the Categorical Imperative is a test useful
3% is spent on harm reduction (deBeck, Wood, Mon- for determining duty. He formulated it in different ways
taner, & Kerr, 2006). This imbalance occurs despite but thought that each formulation stated essentially the
the fact that there is significant evidence indicating same insight and thus would always lead to the same
that abstinence-based policies have had many (albeit conclusion. The first test is whether the maxim—the
unintended) negative consequences (Hankins, 2000; principle of action—could be universalized without
Oscapella, 1996). Moreover, these negative outcomes contradiction. Kant gives the example of lying. If an in-
are largely preventable through harm reduction inter- dividual needs to borrow money to get out of a difficult
ventions, such as needle exchange programs and super- situation and is aware that he or she will not be able to
vised injection sites (Wodak, 2007b; Stoltz, Wood, & repay the money but is also aware that no one will lend
Small, 2007). him or her money unless a promise of repayment is
Since abstinence-based policies have many un- made, should that individual lie? Suppose that individ-
intended negative consequences, Mill would conclude ual were to say "yes." Kant would say that the universal-
that they would likely not promote the greatest happi- ized form of the principle of this action—i.e., "Making
ness for the greatest number. Moreover, he would note a false promise (i.e., lying) is a universal law"—would
that the harm reduction approach will avoid many make the concept of a promise incoherent; there would
negative consequences. Thus, a Utilitarian would likely be no sense in promising. Others could never know
arrive at the conclusion that ethics demands a policy of what a speaker was saying or committing him or herself
harm reduction and not simply abstinence. to when he or she uttered the words, "I promise . . ."
Christie/Groarke/Sweet: Virtue Ethics as an Alternative to Deontological and Consequential Reasoning
Likewise, on the second formulation, lying would be to have a good society populated with citizens of good
using another person as a means to achieve one's own moral character.
ends and thus not respecting that person as an end in In the Nicomachean Ethics, Aristotle famously offers
him or herself. the mean as a heuristic device to determine whether a
According to Kant's deontology consequences are specific character trait is a virtue (Aristotle, 350 Bc).
irrelevant when considering the moral quality of an Simply put, we can have too much or too little of a
action. But the entire debate between harm reduction character trait understood as a mid-point between two
and abstinence is one that focuses on consequences. opposites. Consider, for example, the virtue of courage
The major criticism of the abstinence-based approach understood as the correct balance of fear to fearlessness.
is that it has greater negative consequences than does People who suffer from cowardice have too much fear
the harm reduction approach. Similarly, the major and not enough fearlessness; those who suffer from rash-
justification for harm reduction is that it works (i.e., ness have too much fearlessness and not enough fear. Or
has fewer negative consequences). However, on a consider the virtue of honesty understood as the correct
Kantian model, an appeal to consequences is irrel- balance between hiding and revealing the truth. The liar,
evant when promoting or criticizing an approach; the person who is dishonest, hides the truth too much;
rather, one is restricted to focusing on the intention the blabbermouth, the person who lacks propriety, re-
of the agent. veals the truth too much. Moreover, Aristotle insists that
An abstinence-based approach would almost cer- morality depends upon a number of contextual factors.
tainly "pass" the Kantian test. An application of Kant's We need to do the right thing, to the right people, at the
Categorical Imperative would be: can we universalize right time, in the right way, for the right reasons. Thus,
the maxim that everyone everywhere should refrain a Virtue Ethics model takes account of context and con-
from engaging in illicit injection drug use? Clearly, we sequences without reducing ethics to simple matters of
could. A harm reduction approach, however, could promoting pleasure, avoiding pain, or doing one's duty.
not be justified on a Kantian model because its "raison In the context of the present discussion on harm
d'etre" is ultimately to avoid negative consequences. reduction and problematic substance use, we need to
For example, needle exchange programs and super- ask what virtue would require of a policy-maker (and
vised injection sites have value, not because illicit injec- by implication, of a society): implementing harm reduc-
tion drug use is good but rather to avoid more serious tion policies or implementing abstinence-based poli-
negative consequences. Kant would argue that this type cies. The positive character trait driving those involved
of instrumental reasoning does not meet the relevant in harm reduction programs might be called "compas-
ethical standard. For an action to pass the test of the sion." Compassion, defined in terms of the correct
Categorical Imperative it must be able to be rationally ratio between removing and not removing suffering or
willed by everyone and be able to be acted upon by pain or misery from others, is clearly a virtue. When,
everyone, that is, it would have to be seen to be good in however, it comes to helping those engaged in illegal
itself without appealing to consequences. The essence or unhealthy and destructive behaviour, the issue be-
of harm reduction is to appeal to good consequences as comes more complicated. What does it mean to be too
a justification for specific actions, e.g., needle exchange compassionate or not to be compassionate enough? In
and supervising injections. It is important to note, finding the mean, we can better gauge, from the per-
however, that this does not mean that Kant would find spective of Virtue Ethics, the morality of harm reduc-
harm reduction immoral but only that it is not moral. tion and abstinence-based programs.
On Kant's view, consequences are irrelevant to the mor- So how compassionate should a society be? Define
ality of actions and so reflect neither virtues nor vices. the virtue of compassion as having the right measure
of sensitivity or solicitude for suffering. Compassion is
Virtue Ethics then a mean between two extremes: being "too hard"
is not having enough sensitivity, and being "too soft" is
Virtue Ethics is most famously associated with Aris- having too much sensitivity. Thus abstinence advocates
totle. It has enjoyed resurgence in contemporary moral would argue that harm reduction is "too soft" and harm
philosophy as an alternative to the narrowness of the reduction advocates would accuse abstinence-based
Deontological—Consequentialist dichotomy. Virtue policies of being "too hard."
Ethics does not focus on isolated acts but on the Aristotle would caution that people with differ-
character of the agent—e.g., honesty, loyalty, courage, ent roles in society should aspire to different degrees
compassion, kindness, fairness, etc. Aristotle does not of compassion. A military drill sergeant who has
separate morality from politics. The point of politics is to toughen up his charges to face hardship in battle
II II I Nal
8 Public Health
should clearly be, so to speak, more hard than soft. The approaches, which are largely abstinence-based, are
same could be said of a coach preparing athletes for ineffective. This does not, however, settle the issue
grueling competition. One could even argue that par- of whether and how the harm reduction approach is
ents should practice a certain degree of "hardness," that ethically justifiable; this is the concern of the present
overly permissive parenting spoils the child. But these paper. Consequently, we conducted an ethical analysis
are cases where not relieving suffering (i.e., hardness) using three dominant ethical theories. While there are
has positive results. In the case of harm reduction and other important theoretical approaches, such as the
abstinence-based programs, the evidence suggests that "principle" approach of applied ethics, the feminist
hardness (the abstinence-based approach) does not approach, pragmatism, and ethical theories from other
seem to have any positive effe:ts and might have signifi- cultures, the three that we discuss are clearly the most
cant (unintended) negative consequences. Doing away influential today and are the source of many of the
with harm reduction programs, like supervised injec- principles referred to in contemporary applied ethics.
tion sites and needle exchange programs, means that In conclusion, the major ethical issue we have
more individuals die, become sick, contract HIV, are investigated is how the harm reduction tenet—that in-
involved in criminal activity, etc. As a result, one could dividuals need not be required to abstain or at least to
argue that the rationale for "hardness" is eliminated. attempt to abstain from substance use—might be eth-
There is, however, an equally important second ically justified. This value judgment is clearly justified
issue. It is virtuous to promote virtue; it is vicious to on Utilitarian grounds because, based on the evidence,
promote vice. By implementing harm reduction pro- harm reduction policies would produce the greatest
grams, are we helping people living with problematic good for the greatest number. However, Kant would
substance use become virtmus persons, or are we not think that the values guiding harm reduction meet
aiding and abetting vicious behaviour? From the needle the appropriate ethical standard, because the "raison
exchange literature and the literature on supervised in- d'etre" of harm reduction is exclusively to prevent nega-
jection sites, it is clear that harm reduction programs tive consequences. While he would likely embrace an
do not plausibly lead to an increase in drug use and, abstinence-based approach, consequences would be ir-
in at least some cases, lead to rehabilitation. Therefore, relevant to this judgment. Virtue Ethics takes account
one could argue that compassion is not aiding and of the social consequences but also the moral character
abetting and is the appropriate moral response. If harm of the agent. The virtue of compassion would provide a
reduction programs do not promote substance use but strong ethical foundation for at least some harm reduc-
merely tolerate it—there is a difference between these tion policies without requiring recourse to Utilitarianism.
features—and if, instead of making people ill, they
relieve the suffering of the ill, they are morally valuable.
Acknowledgments
We would like to thank Dr. Robert Larmer, Professor of

Conclusion
Philosophy at the University of New Brunswick, for his
This paper does not attempt to provide a critical review comments on earlier versions of this paper. We would
of the scientific evidence bearing on harm reduction also like to thank the peer-reviewers of this manuscript
and abstinence-based programs. Instead, we accept and the editors at the International Journal of Drug
the view that there is extensive evidence that the harm Policy. Their critical comments and careful review have
reduction approach is effective and that current policy improved this paper significantly.
References
Anderson, J. (2000). Interpreting the relation between in- Beauchamp, T. L., & Childress, J. F (2001). Principles of bio-
jection drug use and harm: A cautionary note. CMAJ, medical ethics (5th ed.). New York: Oxford University Press.
162(12), 1695-1696. Carpenter, C. (2007). Workplace drug testing and worker
Aristotle. (350 BC). Nicomachean ethics (selections). In S. M. drug use. Health Services Research, 42(2), 795-810.
Cahn, & P Markie (Eds.), Ethics: History, theory and con- Christie, T., & Anderson, J. (2003). Drug treatment courts are
temporary issues (3rd ed., pp. 124-178). Oxford: Oxford popular but do they work and are they ethical and appro-
University Press. priate for Canada? Health Law in Canada, 23(4), 70-78.
Arnold, M. (2006). Mill's On Liberty. Ezine articles. Retrieved deBeck, K., Wood, E., Montaner, J., & Kerr, T. (2006). Can-
4 September, 2007, from http://ezinearticles.com/?Mills- ada's 2003 renewed drug strategy—an evidence-based
On-Liberty&id= 181525. review. HIV AIDS Policy Law Review, 11(2/3), 1,5-12.
Christie/Groarke/Sweet: Virtue Ethics as an Alternative to Deontological and Consequential Reasoning 457
Erickson, P., Butters, J., Walko, K., Butterill, D., Reggie, C., strategy 2006. Retrieved 19 November 2007 from http://
Fischer, B., et al. (2007). CAMH and harm reduction: A www.whitehousedrugpoficy.goy
background paper on its meaning and application for sub- Oscapella, E. (1996). Illegal drugs policy, AIDS and hepa-
stance use issues. Centre for Addiction and Mental Health. titis: From prohibition to harm reduction. Med. Law, 15,
Retrieved 19 November 2007 from http://www.camh.net/ 391-398.
Public policy/Public policy papers/harmreductionback- Palepu, A., Tyndall, M. W. , Leon, H., Muller, J., O'Shaughnessy,
ground.html. M. V, Schecter, M. T., et al. (2001). Hospital utilization
Hagan, H., McGough, J. P, Thiede, H., et al. (2000). Reduced and costs in a cohort of injection drug users. CMAJ,
injection frequency and increased entry and retention in 164(4), 415-420.
drug treatment associated with needle-exchange partici- Public Health Agency of Canada, (2003). Hepatitis C preven-
pation in Seattle drug injectors. Journal of Substance Abuse tion, support and research program. Retrieved 19 November
Treatment, 19,247-252. 2007 from http://www. phac-aspc.gc.ca/hepc/pubs/
Hankins, C. (2000). Substance use: Time for drug law reform. psrpmideval-ppsrevalinter/i_problem_e. html.
CMAJ, 162(12), 1693-1694. Stoltz, J., Wood, E., Small, W, Li, K., Tyndall, M., Mon-
Health Canada, (2006). No new injection sites for addicts until taner, J., et al. (2007). Changes in injecting practices
questions answered says Minister Clement. News Release associated with the use of a medically supervised safer
2006-85 1 September 2006. Retrieved 19 November injection facility. Journal of Public Health (Oxford), 29(1),
2007 from http://www.hc-sc.gc.ca/ahcasc/media/nr- 35-39.
cp/2006/2006 85e.html. Strathdee, S. A., Celentano, D. D., Shah, N., Lyles, C.,
Hunt, N. (2005). A review of the evidence-base for harm re- Strambolis, V. A., Macalino, G., et al. (1999). Needle-
duction. Forward Thinking on Drugs, 55. exchange attendance and health care utilization pro-
Kant, I. (1785). Groundwork of the metaphysic of morals. mote entry into detoxification. Journal of Urban Health,
In S. M. Cahn & P Markie (Eds.), Ethics: History, theory 76(4), 448-460.
and contemporary issues (3rd ed., pp. 270-308). Oxford: Tammi, T., & Hurme, T. (2007). How the harm reduction
Oxford University Press. movement contrasts itself against punitive prohibition.
Kerr, T., & Palepu, A. (2001). Safe injection facilities in International Journal of Drug Policy, 18, 84-87.
Canada: Is it time? CMAJ, 165(4), 436-437. Tyndall, M. W , Kerr, T., Zhang, R., King, E., Montaner,
Kerr, T., Stoltz, J. A., Tyndall, M., Li, K., Zhang, R., Montaner, J. G., & Wood, E. (2005). Attendance, drug use pat-
J., et al. (2006). Impact of a medically supervised safer in- terns, and referrals made from North America's first
jection facility on community drug use patterns: A before supervised injection facility. Drug Alcohol Dependence,
and after study. 58 BMJ, 332(7535), 220-222. 6(238), 51-54.
Kerr, T., Tyndall, M., Li, K., Montaner, J., & Wood, E. (2005). Wall, R., Rehm, J., Fischer, B., Brands, B., Fliksman, L.,
Safer injection facility use and syringe sharing in injection Stewart, J., et al. (2000). Social costs of untreated opioid
drug users. Lancet, 366(9482), 316-318. dependence. Journal of Urban Health, 77(4), 688-722.
Kuyper, L. M., Hogg, R. S., Montnaer, J. S., Schecter, M. T., Whitman, C. (1996). Letter to Mr. David Troast, Chairperson of
& Wood, E (2004). The cost of inaction on HIv trans- Governor's Advisory Council on AIDS. Trenton, NJ.
mission among injection drug users and the potential Wodak, A. (2007a). Ethics and drug policy. Psychiatry, 6(2),
for effective interventions. Journal of Urban Health, 81(4), 59-62.
655-660. Wodak, A. (2007b). Health exchange and prevention of lily:
McGeary, K. A., & French, M. T. (2000). Illicit drug use and The evidence for effectiveness is beyond dispute. Addic-
emergency room utilization. Health Service Research, 35(1 tion, 102,161-163.
Pt 0,153-169. Wood, E., Kerr, T., Montaner, J. S., Strathdes, S. A., Wodak,
Mill, J. S. (1863). Utilitarianism. In S. M. Cahn & P. Markie A., Hankins, C. A., et al. (2004). Rationale for evaluating
(Eds.), Ethics: History, theory and contemporary issues North America's first medically supervised safer-injecting
(3rd ed., pp. 317-351). Oxford: Oxford University Press. facility. Lancet Infectious Disease, 4(5), 301-306.
ONDP (Office of National Drug Control Policy United States Wood, E., Kerr, T., Small, W, et al. (2005). Changes in public
Government). (2007a). Disrupting drug markets. National order after the opening of a medically supervised safer
drug control strategy 2006. Retrieved 19 November 2007 injecting facility for illicit injection drug users. CMAJ, 2,
from http://www.whitehousedrugpolicy.gov. 126-130.
ONDP (Office of National Drug Control Policy United Wood, E., Kerr, T., Small, W, Li, K., Marsh, D. C., Montaner,
States Government). (2007b). Healing America's drug J. S., et al. (2004). Changes in public order after the
users. National drug control strategy 2006. Retrieved opening of a medically supervised safer injecting facility
19 November 2007 from http://www.whitehousedrug for illicit injection drug users. CMAJ, 171(7), 731-734.
policy.gov. Wood, E., Tyndall, M. W, Li, K., Lloyd-Smith, E., Small, W,
ONDP (Office of National Drug Control Policy United States Montaner, J. S., et al. (2005). Do supervised injecting fa-
Government). (2007c). Stopping drug use before it starts: cilities attract higher risk injection drug users? American
Education and community action. National drug control Journal of Preventive Medicine, 2,126-130.
al I I dal MLLE
458 8 Public Health
8.5 Inequality in Health
When Are Health Inequalities about injustice. To know whether one inequality is a
cause or determinant of another and then to ascribe
Unjust?—The Social Determinants some judgment about inequity to it, we have to know
of Health something about the mechanisms at work. We need
Norman Daniels to know, for example, that although poor health or
disability can lead to lower income, which can be a
significant consequence in many settings, such "health
When is an inequality in health status between dif- selection," as it is called, explains only part of the cor-
ferent socioeconomic groups unjust? More generally, relation between low income and poor health in gen-
when is an inequality in health status between different eral,' and we have good reason to suppose the existence
demographic groups unjust? F. . .1 of mechanisms that work in the other direction, from
We have known for over 150 years that an individual's socioeconomic inequality to health differences. Being
chances of life and death are patterned by social class: The less well educated, for example, can lead to engaging
more affluent and better educated people are, the longer in some risky patterns of behavior, and these risks
and healthier their lives (Villerme 1840, cited in Lynch increase health inequalities.3 But even this mechan-
et al. 1998). When these inequalities are present in poor, ism fails to explain the general effect of socioeconomic
developing countries, it is tempting to think that poverty inequality on health (Marmot 2004), and researchers
and deprivation adequately explain them. We should are exploring more promising mechanisms. But even
avoid the temptation. Some rather poor countries pursue if more work is needed to clarify the exact mechan-
policies that produce excellent aggregate health outcomes isms, it is not unreasonable to talk here about the social
while moderating significantly the degree of health in- "determinants" of health (Marmot 1999). To the extent
equality. In addition, these effects of socioeconomic status that these social determinants are socially controllable,
(SES) are present as well in rich, developed countries, we clearly face questions of distributive justice.
where the sources of inequality are not deprivation and In some ways, gender and race or ethnic inequal-
deep poverty. More striking, these patterns persist when ities are easier to address than class inequalities. Since
there is universal access to health care—a fact quite sur- these kinds of inequalities are the result of social ex-
prising to those who think that financial access to medical clusion and other unjust practices aimed at vulnerable
services is the primary determinant of health status. groups, we are generally and justifiably inclined to
In addition to socioeconomic inequalities' that are view them as inequities. Identifying them, however,
correlated with health inequalities, we have extensive does not make them easy to remedy.
evidence about patterns of racial and ethnic disparities Socioeconomic inequalities pose harder ques-
in many countries. Many of these inequalities are due to tions, since few people are radical egalitarians opposing
the way in which social exclusion and patterns of dis- all forms of such inequality. Many who are not at all
crimination push ethnic mino::ities (or majorities, as in troubled by significant inequalities in income, wealth,
some former colonies, like South Africa) into the lowest or opportunities for a higher quality of life are particu-
socioeconomic groups. Still, there is often a significant larly troubled by health inequalities. They believe that
residual effect of race, as in the United States, even a socioeconomic inequality that otherwise seems just
when correlations are controlled for income, education, becomes unjust if it contributes to health inequalities.
and insurance levels. There are also striking patterns of Is every health inequality that results from unequally
gender inequality in health in some countries—leading distributed social goods unjust? If there is an irreducible
Sen (1990) to ask where the millions of "missing" health gradient across socioeconomic groups, does that
women are in some Asian countries, as compared to make the very existence of those inequalities unjust?
gender ratios in developed countries. F. . .1 Does a concern about health inequalities drive us
Of course, we cannot infer causation from correla- to even more radical forms of social and economic egali-
tion, and correlations alone will not support judgments tarianism than we had imagined we were committed to?
Source: From "Justice, Health, and Health Care," in Medicine and Social Justice: Essays on the Distribution of Health Care, ed R.
Rhodes, M. Battin, and A. Silvers (Oxford University Press, 2012), 10-14. Reprinted with permission.
Daniels: When Are Health Inequalities Unjust?—The Social Determinants of Health 459
Alternatively, are some health inequalities the when health inequalities are unjust. Specifically, each
result of acceptable trade-offs? Perhaps they are simply of his principles of justice as fairness governs a cluster
unfortunate by-products of inequalities that work in of key social determinants of health. General conform-
other ways to help worse-off groups. For example, it ance with these principles will consequently flatten
is often claimed that permitting inequality provides the socioeconomic gradient of health as much as we
incentives to work harder, thereby stimulating growth can reasonably demand (and, arguably, considerably
that will ultimately benefit the poorest groups. To more than we observe even in wealthy nations with
whom must these trade-offs be acceptable if we are to developed social welfare protections). In effect, social
consider them just? Are they acceptable only if they are justice in general is good for population health and its fair
part of a strategy aimed at making the situation more distribution. There will remain a problem of residual
just? In our judgments about justice, does it matter health inequalities produced by otherwise justifiable
exactly how social determinants produce inequalities inequalities, and we shall consider the extent to which
in health status? justice as fairness gives any guidance about them.
Unfortunately, these questions have been almost f. . .1 First, however, I shall describe some of the
totally ignored within the field of bioethics, as well findings from recent social epidemiological work and
as within ethics and political philosophy more gener- point to their serendipitous connection to Rawls's prin-
ally.4 With some significant exceptions, bioethics has ciples of justice.
not looked "upstream" from the point of delivery of
medical services to the role of the health-care system in
Social Determinants of Health:
improving population health. Bioethics has even more
Some Basic Findings
rarely looked further upstream to the distribution of
social goods that determine the health of societies (cf. Four central findings in the literature on the social
Marchand et al. 1998). This omission is quite striking, determinants of health have implications for an ac-
for bioethics here lags behind both social science and count of justice and health inequalities. First, if we
policy concerns. The social science literature on social look across nations, the national income/health gra-
determinants of health has grown impressively for a dients we observe are not the result of some fixed or
quarter of a century. More important, a concern about determinate laws of economic development but are
health equity and its social determinants has emerged influenced by social policy choices. The wealth of
as an important consideration in the policies of several nations matters up to a relatively modest point, but
European countries over the past two decades (Ben- policies always matter, and arguably they matter more
zeval et al. 1995; Ostlin and Diderichsen 2001). WHO in poor countries. Second, if we look within societies,
has devoted growing attention to inequalities in health the individual SES/health gradients we observe are not
status and the policies that cause or mitigate them. So just the result of the deprivation of the poorest groups.
have research initiatives, such as the Global Health Rather, a gradient in health operates across the whole
Equity Initiative, funded by the Swedish International socioeconomic spectrum within societies even where
Development Agency and the Rockefeller Foundation universal medical coverage and other welfare programs
(cf. Evans et al. 2001). In the past decade, there has are in place. Third, although there is broad agree-
emerged a growing social science and policy concern ment that the degree of social inequality contributes
about health disparities in the United States, although to the steepness of a society's health gradient, there
its focus has largely been on issues of race, not class, is conflicting evidence and a complex debate about
with a few important exceptions. whether the steepness of the gradient is affected by
To fill this bioethical gap, I shall argue that we the degree of income inequality in a society. Fourth,
must supplement a widely cited intuitive analysis of there are reasonable hypotheses identifying some social
what health inequalities are unjust—those that are and psychosocial pathways through which inequality
avoidable, unnecessary, and unfair (Dahlgren and affects health. These causal pathways can be molded
Whitehead 1991; Whitehead 1992)—with a more sys- or changed by specific policy choices that should be
tematic account. f. . .1 I turn f. . .1 to Rawls's theory guided by considerations of justice, and they point to
of justice as fairness [. . .1 to illustrate what an ac- institutions that are appropriately governed by Rawls's
count of the just distribution of the determinants of principle of justice as fairness.
health might look like. My contention is that—quite
unintentionally—Rawls's theory provides a defensible Cross-National Evidence on Health Inequalities
account of how to distribute the social determinants of The finding that prosperity is related to health, whether
health fairly and thus tells us something useful about measured at the level of nations or individuals, might
II all 1111.1.1
8 Public Health
lead one to the conclusion that these income/health aggregate wealth. Culture, social organization, and
gradients are inevitable. Bt .t evidence suggests other- government policies, among others, are important in
wise. Figure 1 shows the relationship between the determining population health and thus in explain-
wealth and health of nations as measured by per capita ing the differences in health outcomes among nations.
gross domestic product (GDPpc) and life expectancy. Variations in these socially controllable factors are thus
There is a clear association between GDPpc and life important from the perspective of social justice.
expectancy, but only up to a point. The relationship
levels off beyond about $8,000 GDPpc, with virtu- Individual SES and Health
ally no further gains in life expectancy. This leveling At the individual level, numerous studies have
effect is most apparent among the advanced industrial documented the "socioeconomic gradient." On this
economies that largely account for the upper tail of the gradient, each increment up the socioeconomic hier-
curve in Figure 1. archy is associated with improved health outcomes
Closer inspection of this figure points out some over the rung below (Black et al. 1988; Davey-Smith
startling discrepancies. Though Cuba and Iraq (before et al. 1990; Pappas et al. 1993; Adler et al. 1994).
the second Iraq war) are equally poor (GDPpcs about Note that this relationship is not simply a contrast
$3,100), life expectancy in Cuba exceeds that in Iraq between the health of the rich and the poor but is
by 17.2 years in 1995 (and no doubt by more now). observed across all levels of SES (remember that not
And while the difference between the GDPpc for all the components of SES—income, education, social
Costa Rica and the United States is enormous (about class—behave identically; see footnote 2). Though
$21,000), Costa Rica's life expectancy is nearly the the effects of income or wealth work across the SES
same as that of the United States. In fact, despite being spectrum, the gradient is steeper at lower income
the richest nation on earth, Lie United States performs levels, with considerable flattening out at the high-
rather poorly on health indicators (i.e., forty-fourth in est income levels. This "concavity" of the gradient
life expectancy at birth).5 means that transfers of resources from the best-off
Taken together, these observations support the to the worst-off SES groups would improve aggregate
notions that the relationship between economic de- health and would have little negative effect, if any, on
velopment and health is nor. fixed and that national the best-off groups. (I return to this point in my later
health achievement is mediated by factors other than discussion of Rawls's difference principle.)
90 —
Japan
80 — Cuba Costa Rica • Luxembourg
•
• Sc. o •-w-• d
Georgia , • • • ea • •
•
USA
• •
• 0, 340•ci• ° •
• •• Brunei
.t 70 —
• o. 0 .5 a • • . • •
co • Qatar
,eZertl. • •
- • -
• Libya
•
A
• 60 •
••40 Iraq
ai •
di • •
X
11.1 •el
.0.••
ti 50
•,.. • Botswana
40 —
• Sierra Leone
30
0 5,000 10,000 15,000 20,000 25,000 30,000 35,000 40,000
Per Capita Gross Domestic Product

(1995 USS purchasing power parities)
Figure 1 Relationship between Country Wealth and Life Expectancy

Source: United Nations Human Development Report Statistics 1998.
Daniels: When Are Health Inequalities Unjust?—The Social Determinants of Health
What is particularly notable about the SES gradi- Relative Income and Health
ent is that it does not appear to be explained by dif- A lively debate exists in the empirical literature about
ferences in access to health care. Steep gradients have whether income inequality, as opposed to absolute
been observed even among groups of individuals, such levels of income, helps to determine the steepness of
as British civil servants, with adequate access to health the gradient. Were this true, we might find that middle
care, housing, and transport (Davey-Smith et al. 1990; income groups in a country with high income inequal-
Marmot et al. 1998). The Whitehall studies of British ity have lower health status than comparable or even
civil servants provide strong support for such a gra- poorer groups in a country with low income inequality.
dient, not only for mortality rates and life expectancy There is some evidence for this pattern in the United
but also for morbidity rates across a range of diseases. States if variations among states in inequality are exam-
Strikingly some of these studies show that civil service ined (Kennedy et al. 1998; Lynch et al. 1998; Subra-
rank explains more of the risk of premature heart dis- manian and Kawachi 2006), but there are conflicting
ease among ranks than such standard risk factors as study results as well (Mellor and Milyo 2002; Deaton
smoking, serum cholesterol level, or blood pressure and Lubotsky 2003). This effect, if it can be established,
(Marmot et al. 1984). is apparent in Figure 2, where the prevalence of self-
Importantly, the steepness of the gradient varies reported fair/poor health is higher for almost every
substantially across societies. Some societies show income group (and the gradient is steeper) for those
a relatively shallow gradient in mortality across SES living in states with the highest income inequality (Ken-
groups. Others, with comparable or even higher nedy et al. 1998). This effect of shifting the curve, if it is
levels of economic development, show steeper gra- supported by the empirical findings, would suggest that
dients in mortality rates across the socioeconomic income inequality has an effect over and above what is
hierarchy. implied by the concavity of the SES gradient itself.
35 —
30 —
---- Low Inequality
—o— HighInequality
25 —
Percent reporting fair/poor health
20 —
15 —
10 — The income-poor health curve is shifted up in

the states with the highest income inequality
for almost the entire population
5—
0
0PC1 0PC1 0PC1 r,ce 0PCI
cl•
•scl' ;1.D'
_11 _,,,,,;r‘
_se sir
N.4r tir‘ t2V t'Sr
Household income
Figure 2 Self-Rated Health and Individual Household Income

II .111 111111
8 Public Health
Leaving aside the contentious results of these stud- and female adult literacy, which explains much of the
ies,6 it is worth noting why the income relativity thesis variation in health achievement among these countries
has attracted such interest. If it is true, it is not just after accounting for GDPpc. Similarly, in the United
the size of the economic pie but how the pie is shared States, differences between the states in women's
that matters for population health. It is not the absolute status—measured in terms of their economic auton-
deprivation associated with low economic develop- omy and political participation—are strongly correl-
ment (lack of the basic conditions necessary for health, ated with female mortality rates (Kawachi and Kennedy
such as clean water, adequate nutrition and housing, 1997; Kawachi et al. 1999). Even if we do not know
and general sanitary living conditions) that explains all the underlying mechanisms through which im-
health differences among developed nations but the proved education, including female literacy, improves
degree of relative deprivation within them. Relative dep- health outcomes, we have good evidence that invest-
rivation refers not to a lack of the goods that are basic ment in this socially controllable factor has measurable
to survival but rather to a lack of sources of self-respect health effects.
that are deemed essential for full participation in soci- The second empirical thesis I highlighted con-
ety. (These "recognitional" features of just arrangements cerned the strong effect of a socioeconomic gradient of
are emphasized in Rawls account, as we shall see.) If health. This gradient is demonstrated even where soci-
the income relativity thesis can be established, it would oeconomic inequality does not mean absolute depriva-
then require elaboration of the causal mechanisms that tion in material goods or access to health care—as in
underlie it.7 the Whitehall studies. Indeed, if we focus on the Brit-
The income relativity thesis is interesting and worth ish civil servants, one important variable in their job
further investigation, but it has only a modest bearing situations concerns their degree of control over work
on the claims I pursue [. . .1. As we shall see, there (Marmot 2004). Such control may affect the kinds of
is strong evidence that health inequalities are reduced stress people are exposed to, and this in turn may affect
if we concentrate first on making the worst-off groups basic immune mechanisms or may directly impact the
socially and economically as well off as possible, paying course of certain diseases. Uncovering mechanisms
special attention to improvements in human capital here is a complex research program that may require
and investments in public health. If true, the income work on both animals and humans, including some
relativity thesis would strengthen support for giving carefully selected natural experiments. How readily
priority to making the worst cff as well off as possible, we can convert this understanding of mechanisms into
for it would make an even stronger claim that it works modifications of social policy, including, for example,
to the advantage of all. In addition, it might point us to the organization of the workplace, remains to be seen.
other mechanisms that we may have to address. Our third thesis concerned the controversial
evidence about the degree of income inequality and
Pathways Linking Social Inequalities to its effects on health. It may be premature to look for
Health Inequalities mechanisms, since the basic association is controver-
To address key issues about sccial policy and distribu- sial. Still, we may better understand some of the mech-
tive justice, we must better understand the mechanisms anisms through which income inequality might work if
underlying the correlations between health outcomes we note some of the related studies.
and social inequalities in income, education, and other In the United States, the states with the most un-
factors. Unless we know something about these under- equal income distribution invest less in public educa-
lying causal mechanisms, we cannot focus properly on tion, have more uninsured persons, and spend less
the socially controllable factors that a just policy must on social safety nets (Kaplan et al. 1996; Kawachi and
regulate. Fortunately, there are plausible and research- Kennedy 1997). Differences in human capital invest-
able pathways through which social inequalities pro- ment are particularly striking. These are demonstrated
duce inequalities in health. for educational spending and, more importantly, for
The first empirical result I highlighted earlier was outcomes; even when controlling for median income,
the claim that social policy more than the wealth of income inequality explains about 40 percent of the
nations, explains variations in health outcomes across between-state variation in the percentage of fourth-
countries. Developing countries that invest heavily in grade children who are below the basic reading level.
human capital, for example in education, have better Similarly strong associations are seen for high school
health outcomes. Indeed, one of the strongest predict- dropout rates. These data clearly show that educa-
ors of life expectancy among developing countries is tional opportunities for children in states with high
adult literacy, particularly the disparity between male income inequality are quite different from those in
more egalitarian states. Furthermore, these early ef- and enforcement of traffic laws, but suppose there
fects not only have an immediate impact on health, remains an elevated mortality rate that seems unavoid-
increasing the likelihood of premature death during able, or perhaps necessary given the importance of
childhood and adolescence (as evidenced by the much trucking. Some may insist that the inequality is still
higher death rates for infants and children in the high- unfair, if unavoidable, even though they would con-
inequality states) but also have lasting effects showing clude that the inequality is not an inequity. If we can
up later in life as part of the SES gradient in health agree on what is avoidable, unnecessary, and unfair,
(Bartley et al. 1997; Davey-Smith et al. 1990). and if this analysis is correct, then we can agree on
These societal mechanisms are tightly linked to the which inequalities are inequitable.
political processes that influence government policy. The Whitehead-Dahlgren analysis is deliberately
For one thing, income inequality appears to affect broad. It not only includes socioeconomic differences
health by undermining civil society. Income inequal- but others as well. Age, gender, race, and ethnic dif-
ity erodes social cohesion, increasing social mistrust ferences in health status exist that are independent
and reducing participation in civic organizations—two of socioeconomic differences, and they raise distinct
features of civil society (Kawachi and Kennedy 1997; questions about equity or justice. For example, should
Kawachi et al. 1997). Lack of social cohesion, in turn, we view the lower life expectancy of men compared to
is reflected in significantly less political activity (e.g., women in developed countries—as much as six years'
voting, serving in local government, volunteering for difference in some observations—as an inequity?8 If it
political campaigns), thus undermining the responsive- is rooted in biological differences that we do not know
ness of government institutions in addressing the needs how to overcome, then, according to this analysis, it is
of the worst-off groups. This is demonstrated not only not avoidable and therefore not an inequity. Perhaps
by the human capital investment data presented earlier not all of it is unavoidable—higher smoking rates for
but also by the lack of investment in human security. men or higher workplace risks may explain some of the
States with the highest income inequality, and thus the difference. But how much? This is not an idle contro-
lowest levels of social capital and political participa- versy: Taking average rather than gender-differentiated
tion, provide far less generous social safety nets. life expectancy in developed countries as a benchmark
or goal will yield different estimates of the inequity
An Intuitive Analysis of Health Inequities
women face in some developing countries.9 In any
case, the analysis of inequity here is only as good as
In the public health literature (Braveman 1999), there our understanding of what is avoidable or unnecessary.
is an influential intuitive answer to [our starting ques- The same point applies to judgments about fair-
tion about justice,] namely, Whitehead's (1992) and ness. Is the poorer health status of a social class or
Dahlgren and Whitehead's (1991) claim that health ethnic group that engages in heavy drug and alcohol
inequalities count as inequities when they are avoid- use unfair? We may be inclined to say it is not unfair,
able, unnecessary, and unfair. How do these conditions provided that the use or avoidance of drugs and alco-
fit together? hol is truly voluntary. But if many people in an ethnic
Presumably, inequalities that are avoidable and group or class behave similarly there may also be fac-
unnecessary but not unfair do not count as inequit- tors at work that reduce the autonomous nature of their
able. For example, if lung cancer rates differed between behavior and how much responsibility they should
smokers and nonsmokers, and if smoking rates were bear (Wikler 1978; Roemer 1995). The analysis thus
not themselves strongly influenced by SES, ethnicity, or leaves us with judgments about responsibility that are
gender but seemed completely a matter of individual, complex and unresolved; as a result, we have continu-
voluntary, informed choice, then we might not think ing disagreements about fairness (or avoidability).
the cancer rate inequality unfair and thus not inequit- The poor in many countries have no clean water,
able. Similarly, an inequality may be unfair but also sanitation, adequate shelter, basic education, vaccina-
unavoidable or even necessary, given our inability to tions, and prenatal and maternal care. As a result of
change certain conditions. For example, suppose we some or all of these factors, there are infant mortality
learned that the mortality rate of certain workers, say differences between them and richer groups. Since
truck drivers, means that their life expectancy is lower social policies could supply the missing determinants
than that of other workers and many believe (intui- of infant health, the inequalities are avoidable.
tively) that this is unfair. We may be able to reduce Are these inequalities also unfair? Most of us would
that higher mortality rate through road and equipment immediately think that they are, perhaps because we
improvements and through better training of drivers believe that international (Pogge 2002) or domestic
II II 11.1,1
8 Public Health
policies that create and sustain poverty are unjust. We Rawls sought to leverage our relatively broad liberal
also believe that social policies that compound poverty agreement on principles guaranteeing certain equal
with lack of access to the determinants of health are basic liberties into an agreement on a principle limiting
doubly unfair. Of course, libertarians would disagree. socioeconomic inequalities, a matter on which liber-
They would insist that what is merely unfortunate is als considerably disagree (J. Cohen 1989). His strat-
not unfair; on. their view, we have no obligation of egy was to show that a social contract designed to be
justice, as opposed to charity, to provide the poor with fair to free and equal people ("justice as [procedural]
what they are missing. Many of us might reject the lib- fairness") would not only justify the choice of those
ertarian view as itself unjust because of this dramatic equal basic liberties but would also justify the choice of
conflict with our beliefs about poverty and our social principles guaranteeing equal opportunity and limiting
obligations to meet peopleb basic needs. inequalities to those that work to make the worst-off
The problem becomes more complicated, how- groups fare as well as possible.
ever, when we remember one of the basic findings from Rawls's account, though developed to answer this
the literature on social determinants: We cannot elim- general question about social justice, also provides
inate health inequalities sim?ly by eliminating poverty. principles for the just distribution of the social deter-
Health inequalities persist even in societies where poor minants of health, unexpectedly adding to its scope
people have access to all the determinants of health and power as a theory. The extra power of the theory
noted previously. Furthermore, these health inequi- is a surprise, since [. . .] Rawls deliberately simplified
ties persist as a gradient of health throughout all SES the construction of his theory by assuming that his
classes, not just between the very poorest groups and contractors are fully functional over a normal life span.
those immediately above them. Some of the mechan- Examining the social determinants of health inequal-
isms may be difficult to avoid and may depend on deep ities from the perspective of Rawls's theory is particu-
structural features of an economy. Thus, educational larly appealing because justice as fairness is egalitarian
and skill differences that steer people into hard manual in orientation (Daniels 2003) and yet justifies certain
labor rather than nonmanual labor significantly affect inequalities, such as those in income and wealth, that
the rate at which poor health is self-reported (Case and contribute to health inequalities. In addition, my ex-
Deaton 2003). Whether we consider these mechan- tension of Rawls links the protection of health to the
isms and their effects unfair may depend in part on our protection of equality of opportunity, again setting up
views about the fairness of overall social inequality. the potential for internal conflict. To see whether this
At this point, many of us are forced to re-examine combination of features leads to contradictions in the
what we believe about the justice of the remaining theory or to insight into the problem, we must examine
socioeconomic inequalities. Unless we believe that the issue in more detail.
all socioeconomic inequalities (or at least all inequal-
ities we did not choose) are unjust—and very few do Rawls on Allowable Inequalities
so—we must consider more carefully the problem How does Rawls justify socioeconomic inequalities?
created by the health gradient and the fact that it be- Why wouldn't free and equal contractors simply insist
comes steeper as inequality increases. Our judgments on strictly egalitarian distributions of all social goods,
about the intuitive notion of fairness in the Dahlgren- just as they insist on equal basic liberties and equal
Whitehead account provide less guidance in thinking opportunity?
about the broader issue of t he social determinants of Rawls's answer is that it is irrational for contractors
health inequalities even if we, rightly or wrongly felt to insist on equality if doing so would worsen their life-
confident in appealing to them when rejecting the time prospects. If incentives to develop skills and take
libertarian position. Indeed, we may even believe that risks increase social productivity, then the social pie
some socioeconomic inequality is unavoidable or even that gets divided may be increased so much that less
necessary and therefore is no: unjust. than equal shares for those who are worst off are still
absolutely larger than equal shares of a smaller social
A Rawlsian Analysis of Health Inequities pie. Specifically, Rawls argues that contractors would
choose the difference principle, which [. . .] permits in-
One reason we develop general ethical theories, includ- equalities provided that they make the worst-off groups
ing theories of justice, is to provide a framework for in society as well off as possible.rn The argument for
settling important disputes about conflicting moral the difference principle appears to suggest that relative
beliefs or intuitions of the sort facing the Whitehead- inequality is less important than absolute well-being,
Dahlgren analysis. For example, in A Theory of Justice, a suggestion that is in tension with other aspects of
Rawls's view. The tension is illustrated by Rawls's in- simpler case of income inequality. In doing so, he as-
sistence that inequalities allowed by the difference sumes that the level of income will correlate with the
principle should not undermine the value of political level of other social goods on the index.
liberty or the requirements of fair equality of oppor- This simplification can be misleading. In crucial
tunity. The priority given these other principles over cases—including those involving health—the correla-
the difference principle thus limits the inference that tion may not occur, and it will be essential to use the
Rawls has no concern about relative inequality. Specif- whole index, rather than just the income component,
ically, as we shall see, these principles work together to to measure how well off people are.0 For example,
constrain inequality and to preserve the social bases of let us suppose that having "democratic" control over
self-respect for all. one's workplace is crucial to self-realization and the
Two points will help avoid misunderstanding promotion of self-esteem (Cohen 2001). Marmot
of the difference principle and its justification. First (2004) documents how hierarchical workplaces may
[. . .] it is not a mere trickle-down principle but one create health inequalities through the effect of status
that requires maximal flow to help the worst-off on stress. Suppose further that hierarchical work-
groups. The worst off, and then the next worst off, places are more efficient than democratic ones, pro-
and so on (Rawls [1971] calls this "chain connect- viding higher incomes for the worst-off workers than
edness") must be made as well off as possible, not democratic workplaces. Then the difference principle
merely somewhat better off, as a trickle-down prin- does not clearly tell us whether the hierarchical work-
ciple implies. The difference principle is thus much place contains allowable inequalities since the worst
more demanding than a principle that would permit off are better off in some ways (economic) but worse
any degree of inequality provided that there was off in others (health and thus opportunity and self-
some trickle of benefits to the worst off." Indeed, respect). Without knowing the weighting of items
it is more egalitarian than alternative principles that in the index, we cannot use it to say clearly what
merely assure the worst off a "decent" or "adequate" inequalities are permitted. When we evaluate which
minimum. Part of the rationale for the more demand- income inequalities are allowable by asking which
ing principle is that it would produce less strain of ones work to make the worst-off groups as well off as
commitment, less sense of being unfairly left out, at possible, we must judge how well off groups are by
least for those who are worst off, than principles that reference to the whole index of primary goods and not
allow more inequality (Cohen 1989). Indeed, from simply the resulting income.
what we have learned about the social determinants This point is of particular importance in the cur-
of health, the more demanding difference principle rent discussion. My extension of Rawls [. . .] treats
would also produce less health inequality than any health status as a determinant of the opportunity
other principles that allow inequalities. By flattening range. Since opportunity is included in the index of
the health gradient, it also benefits middle-income primary social goods, the effects of health inequalities
groups, not simply the poorest. In this regard, its are thereby included as well. I also noted earlier that
benefits are important beyond the level where we the social bases of self-respect are among the primary
have helped the worst off to achieve "sufficiency." social goods on the index. From Marmot's (2004)
This point provides a reply to those who suggest work, we may infer that hierarchical workplaces not
that the difference principle has no appeal once the only undermine health, and thus opportunity, but
worst off are sufficiently provided for (Gutmann and do so because there is inadequate protection of self-
Thompson 1996). respect. Clearly, we must use the whole index, not the
Second, when Rawlsian contractors evaluate how simplifying assumption that income correlates with all
well off the principles they choose will make them, they other items on it.
are to judge their well-being by an index of "primary Unfortunately, Rawls says very little about how
social goods" [. . .]. The primary social goods, which items in the index are to be weighted. This is one of
Rawls thinks of as the "needs of citizens," include lib- the crucial points on which the theory says less than
erty, powers, opportunities, income and wealth, and we might have wished. Therefore, we have little guid-
the social bases of self-respect. (These objective meas- ance about how these primary goods are to be traded
ures of well-being should be contrasted with measures off against each other in constructing the index. This
of happiness or desire satisfaction that are familiar from silence pertains not only to the use of the index in the
utilitarian and welfare economic perspectives.) In his contract situation but also to its use by a legislature
exposition of the difference principle, Rawls illustrates trying to apply the principles of justice in a context
how it will work by asking us to consider only the where many features of a society are known.
Ii JI 111111
8 Public Health
Flattening the SES Gradient: of Health including access to graduate and professional educa-
Now that we have seen why Rawls believes that in- tion and continuing adult education in a world where
equalities are allowable under some conditions, we work skills must be continuously upgraded.
can consider how those general inequalities affect the The equal opportunity principle also requires
distribution of health and contribute to health inequal- extensive public health, medical, and social support
ities. If allowable general inequalities magnify health services aimed at promoting normal functioning for all
inequalities, we might conclude that justice, as Rawls 1. . .1. It even provides a rationale for the social costs of
conceives it, is not good for our health. Instead, we reasonable accommodation of persons with incurable
shall see why social justice, as described by Rawls's disabilities, 1. . .1. Because the principle aims at promo-
principles, is good for our health and promotes its fair ting normal functioning for all as a way of protecting
distribution.° 1. . .1 opportunity for all, it aims at both improving popula-
To understand this claim, let us start with the ideal tion health and reducing health inequalities. Obviously,
case, a society governed by Rawls's principles of jus- this requires universal comprehensive health care, in-
tice as fairness that seeks to achieve what Rawls calls cluding public health, primary health care, and med-
"democratic equality" (Rawls 1971; Daniels 2003). ical and social support services.
Consider what the society requires in distributing the To act justly in health policy, we must understand
social determinants of health. In such a society all are how socioeconomic (and other) inequalities work to
guaranteed equal basic liberties (Rawls 1982a), includ- produce differential health outcomes. Suppose that, as
ing effective exercise of (or worth of) political participa- Marmot (2004) argues, structural and organizational
tion (Daniels 1975; cf. Rawls 1995). In addition, there features of the workplace that induce stress and loss
are institutional safeguards aimed at assuring all, richer of control tend to promote health inequalities. If this is
and poorer alike, the value of effectively exercisable true, then those features should be modified to reduce
political participation rights. Without such protection, their negative effects on health as a public health re-
basic capabilities of citizens cannot develop. Recogniz- quirement of the equal opportunity approach; this is
ing that all citizens have these capabilities protected is on a par with the requirement to reduce exposure to
critical to preserving self-esteem, on Rawls's view. In toxins in the workplace [. . .1. Moreover, modifying
requiring institutional support for political participa- such features would be favored by Rawls's emphasis on
tion rights, Rawls (1995) rejects the claim that freedom securing the social bases of self-respect.
of speech of the rich is unfairly restricted by limiting Finally in the ideal Rawlsian society, the difference
their personal expenditures on their own political cam- principle significantly reduces allowable inequalities in
paigns 1. . .1. After all, the limitation does not unduly income and wealth and, more generally, in the index
burden the rich compared to others. There is evidence of primary social goods.14 The inequalities allowed by
that political participation is itself a social determinant this principle (together with the principles assuring
of health. Therefore, the Rawlsian ideal assures insti- equal opportunity and the value of political participa-
tutional protections that counter the usual effects of tion) are probably more constrained than those found
socioeconomic inequalities on participation and thus in even the most industrialized societies. If so, then the
on health. inequalities that conform to the difference principle
The Rawlsian ideal of democratic equality also would produce a flatter gradient of health inequality
involves conformity with a principle guaranteeing fair than we currently observe in even the more extensive
equality of opportunity 1. . .1. Not only is discrimina- welfare systems of Northern Europe.
tion prohibited by the principle, but it requires strong Earlier, I noted the lively ongoing debate about the
measures to mitigate the effects of socioeconomic effects of income inequality on health inequalities. I be-
inequalities and other social contingencies on oppor- lieve my claim that the difference principle would flat-
tunity. In addition to equitable public education, such ten gradients more than principles calling for weaker
measures would include developmentally appropriate constraints on inequality is true even if the income
day care and early childhood interventions to promote relativity thesis turns out not to be supported by the
the development of capabilities independently of the evidence. The difference principle is about how well off
advantages of family background. Such measures the worst off are in absolute terms, and though applying
match or go beyond the best models of such interven- it may reduce relative inequality it would affect mainly
tions we see in European efforts to provide day care health inequalities even if the income relativity thesis
and early childhood education. We also note that the does not hold. Those who oppose the income relativ-
strategic importance of education for protecting equal ity thesis in any event endorse the basic facts about a
opportunity has implications or all levels of education, socioeconomic gradient of health. Pulling the bottom
up in absolute terms—coupled with other social poli- a benefit workers gain only because their opportunities
cies of the sort assured by the other principles of justice are unfairly restricted, and when is it an appropriate
as fairness—would therefore flatten the gradients. If exercise of their autonomy E. . .1? Many such trades are
income relativity does have a further effect on health ones we think it unjustifiably paternalistic to restrict;
inequalities, then my claim is strengthened further. others we see as unfair.
In short, Rawls's principles of justice as fairness Rawlsian contractors, however, cannot base such
regulate the distribution of the key social determinants trades on specific knowledge of their own values. They
of health, including the social bases of self-respect. cannot decide that their enjoyment of skiing makes it
Nothing about the theory, or my extension of it E. . .1, worth the risks to their knees or necks. To make the
should make us focus narrowly on medical services. contract fair to all participants, and to achieve im-
Properly understood, justice as fairness tells us what partiality, Rawls imposes a thick "veil of ignorance"
justice requires in distributing all social determinants of that blinds them to all knowledge about themselves,
health. Nor, I should add, does the fact that we should including their views of the good life. Instead, they
focus on the distribution of goods in addition to health must judge their well-being by reference to the index
care mean that health care is no longer of special moral of primary social goods (noted earlier) that includes
importance. Not only does health care make its own a weighted measure of rights, opportunities, powers,
significant contribution to population health, but even income and wealth, and the social bases of self-respect.
if all the other socially controllable factors were prop- But our theoretical question about residual health
erly distributed, some pathology would remain and we inequalities reminds us that the theory says too little
would have to respond to its impact on opportunity. about the construction of the index to provide us with
Health care may not be the only good of special moral a clear answer.
importance, given the moral importance of health, but One of Rawls's (1971) central arguments for sin-
it remains a good of special moral importance. gling out a principle protecting equal basic liberties
and giving it (lexical) priority over his other principles
Residual Inequalities: Unjust or Not? of justice is his claim that once people achieve some
We still face an interesting theoretical issue, as well threshold level of material well-being, they will not
as many practical issues in our nonideal or partially trade away the fundamental importance of liberty for
just world. First, the theoretical issue. Even if the other goods. Making such a trade might deny them the
Rawlsian distribution of the determinants of health liberty to pursue their most cherished ideals, includ-
flattens health gradients further than those observed ing their religious beliefs, whatever they turn out to be.
in the most egalitarian developed countries, we must Can we make the same argument about trading health
still expect some health inequalities. In part, this may for other goods?
happen because we may not understand all the relevant There is some plausibility to the claim that rational
causes or the interventions for modifying them. The people should not trade health for other goods. Loss
theoretical issue is whether the theory requires us to of health may prevent us from pursuing what we most
reduce further those otherwise justifiable inequalities value in life. We do, after all, see people willing to trade
because of the inequalities in health status they create. almost anything to regain health once they lose it.
We should not further reduce those inequalities if If we take this argument seriously we might con-
doing so reduces productivity so much that we can no clude that Rawls should give opportunity including
longer support the institutional measures we already the effects of health status, a heavier weighting in the
employ to promote health and reduce health inequality. construction of the index than income alone." Such
Our commitment to reducing health inequality should a weighting would mean that absolute increases in
not require steps that threaten to make health worse income for the worst off that might otherwise have
off for those with less-than-equal health status. So, the justified increasing relative income inequality, accord-
theoretical issue reduces to this: Would it ever be rea- ing to the difference principle, now fail to justify those
sonable and rational for contractors to accept a trade- inequalities because of the negative effects on oppor-
off in which some health inequality is allowed in order tunity. Although the income of the worst off would
to produce some nonhealth benefits for those with increase, they are not better off according to the whole
the worst health prospects? We know that in real life (weighted) index of primary social goods, and so the
people routinely trade health risks for other benefits. greater inequality is not permitted. Rawls's simplifying
They do so when they commute longer distances for a assumption that income correlates with other goods
better job or take a ski vacation. Some such trades raise fails in this case (as it did in the hypothetical example
questions of fairness. For example, when is hazard pay of workplace democracy cited earlier).
II 411 IIILIJ
8 Public Health
Nevertheless, there is also strong reason to think emerge with conformance to the principles are not a
that the priority given to health, and thus opportunity, compromise with what justice ideally requires; they are
is not as clear-cut as the previous argument implies, acceptable as just.
especially where the trade is between a risk to health So far, we have been considering whether the deci-
and other goods that peop: e highly value. Refusing sion about such a trade-off can be resolved from the
to allow any (ex ante) trades of health risks for other perspective of individual contractors. Instead, suppose
goods, even when the background conditions are that the decision is to be made by the legislature in a so-
otherwise fair, may seem unjustifiably paternalistic, ciety that conforms to Rawls's principles. Because those
perhaps in a way that refusal to allow trades of basic principles require effective political participation by all
liberties is not. socioeconomic groups, we can suppose that the groups
I propose a pragmatic route around this problem, most directly affected by any trade-off decision have a
one that has a precedent elsewhere in Rawls. Fair equal- voice in the decision. Since there is a residual health
ity of opportunity, Rawls admits, is only approximated gradient, groups affected by the trade-off include not
even in an ideally just system, because we can only only the worst off but those in the middle as well. De-
mitigate, not eliminate, the effects of family and other veloping a democratic process that involved delibera-
social contingencies (Fishkin 1983). For example, only tion about the trade-off and its effects might be the best
if we were willing to violate widely respected parental we could do to answer the theoretical question E. . .1.
liberties could we intrude in family life and "rescue" In contrast, where the fair value of political partici-
children from parental values that arguably interfere pation is not adequately assured—and we doubt that it
with equal opportunity. Similarly, though we give is so assured in even the most democratic societies—we
general priority to equal opportunity over the differ- have much less confidence in the fairness of a demo-
ence principle, we cannot achieve complete equality in cratic decision about how to trade health against other
health any more than we can achieve completely equal goods. It is much more likely that those who benefit
opportunity. Even ideal theory does not produce per- most from the inequalities, that is, those who are better
fect justice. Justice is always rough around the edges. off, also wield disproportionate political power and
Specifically, if we had good reason to think that "demo- will influence decisions about trade-offs to serve their
cratic equality" had flattened inequalities in accord interests. It may be that the use of a democratic process
with the principles of justice, then we might think we in nonideal conditions is the fairest resolution we can
had done as much as was reasonable to make health practically achieve, but it still falls well short of what
inequalities fair to all. The residual inequalities that an ideally just democratic process would involve. E. . .1
Notes
1. Deaton (personal communicE.tion, July 13, 2006) cautions 5. CIA Factbook, https://www.cia.gov/cia/publications/
against thinking that the various components of SES factbookkankorder/2102rank.html, accessed July
(income, education, social class) all work the same way 18, 2006.
and that the convenience of combining them may be 6. Sparked by Wilkinson's (1992) paper showing a
outweighed by the inaccuracy that can result. See also correlation between national measures of income
Deaton (2002a,b). inequality and the gradient of health inequality among
2. Deaton (personal communication, July 13, 2006) and a select group of Organisation of Economic Cooperation
Case and Deaton (2003) poir t to the impact of ill health and Development (OECD) countries, researchers in the
on early retirement and thus on income reduction; Case United States and elsewhere found conflicting results
and Deaton argue that the significantly worse self-reported during the subsequent decade. The initial positive results
health of those in the lowest income quartile compared to from the OECD countries disappear when a full rather than
those in the highest is primarily driven by the ill health of a selected set of them is studied. Negative results are also
those out of work, presumably because they are ill. Health found primarily in countries more egalitarian than the
selection may have a greater effect on income than class. United States (Sweden, Japan, Canada, Denmark,
Cf. Marmot (1994). New Zealand), as noted by Subramanian et al. (2003) and
3. Behavior risk factors nevertheless contribute significantly Subramanian and Kawachi (2004). Some negative results
to inequality: Jha et al. (2006) show that smoking are also found in the United States, especially if units
contributes nearly half of the male mortality in the lowest of aggregation are smaller, such as municipalities rather
social class in four countries. than states. A significant number of US studies support
4. Sen (2004) is an important exception. the claim, especially at the state level. Positive results
I.. .1 are also found in a recent study in Chile, a country with
Daniels: When Are Health Inequalities Unjust?-The Social Determinants of Health
more income inequality than exists in the United States 10. A careful discussion of Rawls's argument for the
(Subranamian et al. 2003). difference principle and the extensive critical literature
This pattern of seemingly conflicting results might be it has generated is beyond the scope of this chapter. It
explained if income inequality affects health inequalities is important, however, to distinguish Rawls's own social
only above some threshold level of inequality, a threshold contract argument from the many informal and intuitive
crossed in the United States and more unequal countries reformulations of it. See Barry (1989) and G. A. Cohen
but not in more egalitarian countries (Subramanian (1992; 1995) Some controversy is fed by Rawls's (1971:
and Kawachi 2004). Alternatively, the effect of income Sect. 13) different formulations of the difference principle;
inequality might be countered by more egalitarian some comes from the complexity of other assumptions
social policies, such as the more redistributive health involved in his argument, such as "chain connection" and
and welfare policies in other OECD countries. This latter "close-knittedness"; some follows from disagreement with
alternative highlights an important point about the claim giving such strong priority to the worst off, regardless of
generally made by supporters of the income relativity the benefits lost to others and regardless of how well off the
thesis: Income inequality is a causal factor that works worst off are. For a useful discussion, see Williams (1995).
through various mechanisms, political, social, and 11. What was said previously applies to the ideal case, where
possibly individual (e.g., stress), and is not claimed to be a society is in general conformance with the principles of
"by itself" or "per se" a cause of health inequalities. justice. In a nonideal setting, where we may be making
7. In the United States, in those studies that support the thesis, a system more just, Rawls is willing to say that an
the strength of this relationship between social inequality improvement is not unjust as long as it moves the worse-
and health inequality is not trivial. One study of US off groups closer-even if not maximally closer-to being
metropolitan areas, rather than states, found that areas with as well off as possible.
high income inequality had an excess of death compared to 12. In thinking about the impact of income inequality on
areas of low inequality that was equivalent in magnitude to health, Deaton (personal communication, July 2006) makes
all deaths due to heart disease (Lynch et al. 1998). a related point when he cautions that we should think more
8. Sen (2004) points to the gender inequality and denies that broadly about social inequality and not income alone.
it is an inequity, not because we could not do something 13. Sen (2004) too makes the point that social justice broadly
about it but because what we could do, e.g., give men understood is crucial to identifying and addressing
more health care than women, would be discriminatory, health inequity.
violating "process" constraints on fair treatment. 14. G A Cohen (1992) has argued that a strict interpretation
9. Murray's reduction of the male-female gender gap to 2.5 of the difference principle would allow few incentive-based
years from an observed 6 years, based on extrapolation inequalities; for a more permissive view, see Daniels (2003).
from some cases and assumptions about greater exposure 15. Rawls (1971: 93) does suggest that since fair equality of
to risk, was criticized as possible gender bias in the opportunity is given priority over the difference principle,
construction of DALYs. See Anand and Hanson (1998). within the index we can assume that opportunity has a
1. . .1 heavier weighting.
References
Adler, N. E., Thomas, B., Chesney, M. A., Cohen, S., Folkman, Braveman, P 1999. Monitoring Equity in Health: A Policy-
S., Kahn, R. L., and Syme, S. L. 1994. Socioeconomic Oriented Approach in Low- and Middle-Income Countries.
status and health: The challenge of the gradient. American Geneva: World Health Organization.
Psychologist 49: 15-24. Case, A., and Deaton, A. 2003. Broken Down by Work and Sex:
Anand, S., and Hanson, K. 1998. DALYs: Efficiency versus How Our Health Declines. Cambridge, MA NBER Working
equity. World Development 26: 307-10. Paper No. 9821.
Barry, B. 1989. Theories of Justice. London: Harvester Wheat- Cohen, G. A. 1989. On the currency of egalitarian justice.
sheaf, pp. 213-34. Ethics 99: 906-44.
Bartley, M., Blane, D., and Montgomery, S. 1997. Socio- . 1992. Incentives, inequality, and community, in G.
economic determinants of health: Health and the life Petersen (ed.), The Tanner Lectures on Human Values, vol.
course: Why safety nets matter. British Medical Journal 13. Salt Lake City: University of Utah Press, pp. 262-329.
314: 1194. . 1995. The Pareto argument for inequality. Social Phil-
Benzeval, M., Judge, K., and Whitehead, M. (eds.). 1995. osophy and Policy 12 (Winter): 160-85.
Tackling Inequalities in Health: An Agenda for Action. Cohen, J. 2001. Taking people as they are? Philosophy and
London: Kings Fund. Public Affairs 30(4): 363-86.
Black, D., Morris, J. N., Smith, C., Townsend, P. and White- Dahlgren, G., and Whitehead, M. 1991. Policies and Strategies
head, M. 1988. Inequalities in Health: The Black Report; The to Promote Social Equity in Health. Stockholm: Institute of
Health Divide. London: Penguin Group. Future Studies.
II au !ULU
8 Public Health
Daniels, N. 1975. Equal liberty and unequal worth of lib- and mortality in metropolitan areas of the United States.
erty. In N. Daniels (ed.), Recding Rawls. New York Basic American Journal of Public Health 88: 1074-80.
Books, pp. 253-81. Marchand, S., Wikler, D., and Landesman, B. 1998. Class,
. 2003. Democratic equality: Rawls's complex egalitar- health, and justice. Milbank Quarterly 76: 449-68.
ianism. In S. Freeman (ed.), Companion to Rawls. Oxford: Marmot, M. 1994. Social differentials in health within and
Blackwell, pp. 241-76. between populations. Daedalus 123: 197-216.
Daniels, N., Kennedy, B., and Kawachi, L 1999. Why justice . 1999. Social causes of social inequalities in health.
is good for our health: The social determinants of health Cambridge, MA: Harvard Center for Population and
inequalities. Daedalus 128 (4): 215-51. Development Studies, Working Paper Series 99.01.
2000. Is Inequality Bad for Our Health? Boston: . 2004. The Status Syndrome: How Social Standing
Beacon Press. Affects Our Health and Longevity. New York Henry Holt,
Daniels, N., Light, D., and Caplan, R. 1996. Benchmarks Times Books.
of Fairness for Health Care Reform. New York: Oxford Marmot, M. G., Fuhrer, R., Ettner, S. L., Marks, N. F., Bumpass,
University Press. L. L., and Ryff, C. D. 1998. Contribution of psychosocial
Davey-Smith, G., Shipley, M. J., and Rose, G. 1990. Magni- factors to socioeconomic differences in health. Milbank
tude and causes of socioecor.omic differentials in mortal- Quarterly 76: 403-48.
ity. Further evidence from the Whitehall Study. Journal of Marmot, M. G., Hemingway, B. H., Brunner, E., and Stan-
Epidemiology and Community Health 44: 265-70. field, S. 1997. Contribution of job control and other
Deaton, A. 2002a. Policy implications of the gradient of risk factors to social variations in coronary heart disease
health and wealth. Health Affairs 21 (2): 13-29. incidence. The Lancet 350: 235-9.
. 2002b. The convoluted story of international Marmot, M., Shipley, M. J., and Rose, G. 1984. Inequalities
studies of inequality and health. International Journal of in death-specific explanations of a general pattern. The
Epidemiology 31 (3): 546-9. Lancet 5: 1003-6.
Deaton, A., and Lubotsky, D. 2003. Mortality, inequality Mellor, J., and Milyo, J. 2002. Is exposure to income inequal-
and race in American cities and states. Social Science and ity a public health concern? Lagged effects of income
Medicine 56: 1139-53. inequality in individual and population health. Health
Fishkin, J. 1983. Justice, Equal Opportunity, and the Family. Services Research 38: 1.
New Haven, CT: Yale University Press. Ostlin, P., and Diderichsen, E 2001. Equity Oriented National
Gutmann, A., and Thompson, D. 1996. Democracy and Dis- Health Strategy for Public Health in Sweden. Policy Learning
agreement. Cambridge, MA: Harvard University Press. Curve Series No. 1. Brussels: European Center for Health
Jha, P., Peto, R., Zatonski, W , Boreham, J., Jarvis, M. J., and Policy.
Lopez, A. D. 2006. Social inequalities in male mortality, Pappas, G., Queen, S., Hadden, W , and Fisher, G. 1993. The
and in male mortality frorr smoking: Indirect estima- increasing disparity in mortality between socioeconomic
tion from national death rates in England and Wales, groups in the United States, 1960 and 1986. New England
Poland, and North America. Lancetonline D01:10.1016/ Journal of Medicine 329(2): 103-9.
So140-6736 (06)68975-7. Pogge, T. W. 2002. World Poverty and Human Rights: Cosmo-
Kaplan, G. A., Pamuk, E. R., Lyr.ch, J. W, Cohen, R. D., and politan Responsibilities and Reforms. Cambridge: Blackwell.
Balfour, L. 1996. Inequality in income and mortality in Potts, M., and Walsh, J. 2003. Tackling Indian HIV epidemic:
the United States: Analysis of mortality and potential Lessons from Africa. British Medical Journal 326: 1389-92.
pathways. British Medical Journal 3(12): 999-1003. Rawls, J. 1971. A Theory of Justice. Cambridge, MA: Harvard
Kawachi, I., Daniels, N., and Robinson, D. 2005. Health dis- University Press.
parities by race and class: Why both matter. Health Affairs 1982a. The basic liberties and their priorities, in The
24(2): 343-4. Tanner Lectures on Human Values, vol. III. Salt Lake City:
Kawachi, I., and Kennedy, B. P. 1997. Health and social co- University of Utah Press.
hesion: Why care about income inequality? British Medical . 1993. Political Liberalism. New York: Columbia Uni-
Journal 314: 1037-40. versity Press.
Kawachi, I., Kennedy, B. P., Lochner, K., and Prothrow-Stith, . 1995. Political Liberalism (paperback ed.). New York:
D. 1997. Social capital, income inequality and mortality. Columbia University Press.
American Journal of Public Health 87: 1491-8. Roemer, J. 1995. Equality of opportunity, Boston Review,
Kawachi, I., Kennedy, B. P., Prod- row-Stith, D., and Gupta, V April-May. Available at http://www.bostonreview.net/
1999. Women's status and the health of women: A view BR20.2/Roemer.html.
from the states. Social Science and Medicine 48: 21-32. Sen, A. K. 1990a. Justice: Means versus freedoms. Philosophy
Kennedy, B. P., Kawachi, I., Glass, R., and Prothrow-Stith, & Public Affairs 19(2): 111-21.
D. 1998. Income distribution, socioeconomic status, . 1990b. More than 100 million women are missing.
and self-rated health: A U.S. multi-level analysis. British New York Review of Books 37: 30
Medical Journal 317: 917-21. . 2004. Why health equity? In S. Anand, E Peter, and
Lynch, W , Kaplan, G. A., Pamuk, E. R., Cohen, R. D., Heck., A. Sen (eds.) Public Health, Ethics, and Equity. New York:
K. E., Balfour, J. L., and Yen, I. H. 1998. Income inequality Oxford University Press, pp. 21-34.
Daniels: When Are Health Inequalities Unjust?—The Social Determinants of Health 471
Subramanian, S. V, Delgado, I., Jadue, L., Vega, J., and Kawa- E Northridgei, C. Phelan, and M. L Ganz, Social epi-
chi, I. 2003. Income inequality and health: Multilevel demiology and the fundamental cause concept: On the
analysis of Chilean communities. Journal of Epidemiology structuring of effective cancer screens by socioeconomic
and Community Health 57: 844-8. status. Milbank Quarterly 76 (1998): 375-402.
Subramanian, S. V., and Kawachi, I. 2004. Income inequality Whitehead, M. 1992. The concepts and principles of equity
and health: What have we learned so far? Epidemiologic and health. International Journal of Health Services 22:
Reviews 26: 8-91. 429-45.
. 2006. Whose health is affected by income inequal- Wikler, D. 1978. Persuasion and coercion for health: Issues in
ity? A multilevel interaction analysis of contemporaneous government efforts to change lifestyle. Milbank Memorial
and lagged effects of state income inequality on individual Fund Quarterly: Health and Society 56(3): 303-38.
self-rated health in the United States. Health and Place Wilkinson, R. G. 1992. Income distribution and life expect-
12(2): 141-56. ancy. British Medical Journal 304: 165-8.
Villerme, L. 1840. Tableau d'Etat Physique et Moral des Williams, A. (Andrew). 1995. The revisionist difference prin-
Ouvriers, vol. 2. Paris: Renouard. Cited in B. G. Link, M. ciple. Canadian Journal of Philosophy 25: 257-82.
8.6 Cases
Case 1 shot or wear a protective mask. VCH explained that it

Health Care Workers and Flu Shots had tried to persuade and encourage health care work-
ers to be vaccinated or wear a mask, but this effort had
Vancouver Coastal Health (VCH) is the local health au- been a failure.2•3.4 Before the rule became mandatory,
thority for Vancouver, BC. It employs thousands of health many health care workers were already getting vaccin-
care workers at many levels in many kinds of health care ated, but many others at every level still weren't. The
facilities.' They include doctors and nurses in hospitals only alternative, VCH says, was to make flu vaccination
and occupational therapists and assistants in long-term or the wearing of an appropriate mask mandatory.
residential care facilities and everything in between. VCH Not everyone is happy about this policy. Alan
employees see hundreds of thousands of patients every Cassells, a respected journalist who writes about
year with many different kinds of health conditions. Since health matters, has said the scientific research does
2012, VCH has required all of these health care workers to not prove that having all or a very high percentage of
get flu vaccine shots or wear a protective mask at all times health care workers vaccinated against the flu actually
when in contact with patients and clients during winter results in fewer infections in patients.5 The BC Nurses
flu seasons—at least if they want to keep their jobs.2 Union, however, has said scientific evidence shows that
Why did VCH make this policy mandatory? VCH nurses and other health care workers should choose
says that health care workers have a moral responsibil- to be vaccinated because this stops infections and so
ity not to harm their patients and clients.2•3 This is a prevents harms to patients.° But the BC Nurses Union
special moral duty health care workers have that is also believes that health care workers should not be
greater than the one ordinary people have not to harm required as a condition of continued employment to
others. If a health care worker at any level had the flu, have the flu vaccine or wear a mask. Such a require-
they could easily infect a patient. Doing this would ment wrongly violates an individual's moral right to
harm the client, and in some cases it could even lead decide for themselves whether the scientific evidence
to his or her death. VCH says the best way not to harm shows that they should either be vaccinated or wear a
patients in this way is to get a preventative flu vaccine protective mask.3.6
Notes
1. Vancouver Coastal Health, 'Quick Facts," http://www.vch.ca/ 3. E Galankis et al., "Ethics of Mandatory Vaccination
about-us/quick-facts. for Health Care Workers," Eum Surveillance 18 (2013),
2. Vancouver Coastal Health, "Health Care Worker article 20627.
Influenza Control Program—Questions and answers," 4. Kumanan Wilson, "Case Discussion in Response to
https://www.vch.ca/media/Staff-Flu-Shot-QA.pdf. Mandatory Immunization of Local Health Authorities,"
8 Public Health
in Population and Public Heali h Ethics (Toronto: University 6. BC Nurses Union, "Position Statement Influenza
of Toronto Joint Centre for Bioethics, 2012), 118-21. Control Policy," https://www.bcnu.org/AboutBcnu/
5. Alan Cassells, "Will a Flu Shot Keep You Healthy?' FOCUSonline Documents/position-statement-influenza-
October 2012, http://wwwfocusonline.caficrnode/447. control.pdf.
Case 2 And according to Cynthia Forlini, a research fellow

Students and Study-(Enhancing Drugs at the University of Queensland in Australia, "There
seems to kind of be a ceiling of enhancement." Stu-
dents who are already doing well benefit less. There
Because of intense competition for awards and admis- are also bad side-effects.'" Most users reported an un-
sion to graduate programs, students at Canadian col- pleasant "come-down," and some said their sleep was
leges and universities are using cognition-enhancing disrupted. Experts say other effects include loss of ap-
drugs like Adderall, Ritalin, Concerta, and Modafinil to petite and dehydration. Adderall can cause a change in
increase their concentration during study sessions and the rhythm of the heart and even a heart attack.'"
help them complete assignments on time.' •2 An article In addition, some wonder if using these drugs
in the Canadian Medical Association Journal in 2011 esti- is cheating."9 Students who use these drugs have a
mated that 8 to 35 per cent of Canadian university stu- competitive advantage only when most other students
dents are taking these kinds of drugs for non-medicinal aren't using them. If most or all students used them,
purposes.3 Their use may even be more common then no one would have gained an edge, and every-
among pre-med students.4 The drugs are prescribed one would be worse off because of the bad health side
to treat things like attention-deficit hyperactivity or effects.m In 2011, Duke University in North Carolina
the sleeping disorder narcolepsy. Possessing and using made "the unauthorized use of prescription medication
them without a prescription is illegal in Canada, but to enhance academic performance" an offence under its
this hasn't stopped students not diagnosed with these rules on academic dishonesty. n It's a form of cheating. '2
issues from getting and using the drugs. According to a This was a policy supported by the student newspaper.
CTV News report from February 2013, these drugs can In September 2016, its editors wrote that "Prescription
easily be purchased on websites like Craigslist and from drug abuse on campus should be treated as a public
sellers who visit campus libraries or from students who health . . . issue." 13 Britain's Academy of Medical
have a prescription.5 (Their prices go up at exam time!) Sciences compared the use of these drugs to steroid use
The popularity of these drugs has raised a variety in athletes.° It has called on universities there to con-
of questions. Do they really work in ways users hope? sider a policy like Duke's. Many students disagree. It's
Modafinil has been demonstrated to improve memory not cheating "because it's still your work, . . . you're just
by 10 per cent.° A third-year arts and economics stu- doing it with an increased mental capacity," as a second-
dent at McGill University who used Ritalin said, "You year University of British Columbia student put it.2 Ac-
sit down and concentrate for eight hours or so without cording to UBC neuroscientist Peter Reiner, "There is a
feeling the need for breaks or wanting to stop." But widely held perception that taking the drugs is [ethic-
an assignment won't get done if, rather than helping ally] bad whereas coffee is OK. In principle, there's no
you to concentrate on the assignment, using Aderall real difference between the two, except all the evidence
means you become focused on Facebook for hours. is that caffeine is a little more effective than Adderall."2
Notes
1. Rhys Edwards, "This Is Your Brain o n Adderall," The Ubyssey 4. Robyn M. Emanuel et al., "Cognitive Enhancement Drug
7 Apr 2013; Vassilena Sharlandjieva and Samantha Use among Future Physicians," The Journal of General
McCabe, "The High in Higher Education," The Ubyssey 24 Internal Medicine 28 (2012), 1028-34.
February 2016. 5. ovNews.ca Staff, "Canadian Students Abusing Adderall to
2. James Bradshaw, "Students Reaching for ADHD Drugs to Deal Get Edge in Studying," 4 February 2013.
with Academic Stress," The Globe and Mail 18 October 2013. 6. R.M. Battleday and A.-K. Brem, "Modafinil for
3. Daniel Rosenfield et al., "Time to Address Stimulant Abuse Cognitive Neuroenhancement in Healthy Non-Sleep-
on Our Campuses," Canadian Medical Association Journal Deprived Subjects: A Systematic Review, European
163 (6 September 2011), 1345. Neuropsychopharmacology 25 (November 2015), 1865-81.
Cases 1 473
7. Cynthia Forlini, Serge Gauthier, and Eric Racine, "Should 10. Veljko Dubljevid, "Cognitive Enhancement,
Physicians Prescribe Cognitive Enhancers to Healthy Rational Choice and Justification," Neuroethics
Individuals? Canadian Medical Association Journal 185 6 (April 2013), 179-87.
(3 September 2013), 1047-50. 11. Jenna Johnson, "College Administrators Worry
8. Man Lamkin, "A Ban on Brain-Boosting Drugs Is That Use of Prescription Stimulants Is Increasing,"
Not the Answer," The Chronicle of Higher Education Washington Post 27 November 2011.
27 February 2011. 12. Duke University, "Academic Dishonesty," haps://
9. Barbara Sahakian and Jamie Nicole LaBuzetta, studentaffairs.duke.edu/conduct/z-policies/
"Professor's Little Helper—The Ethics of Enhanced academic-dishonesty.
Cognition," in Bad Moves (Oxford University Press, 13. Editorial Board, "Duke and Dope," Duke Chronicle
2013), ch. 5. 22 September 2016.
Case 3 of breast cancer or not.3.4.5 But a Canadian randomized

Breast Cancer Screening control trial reported in 2014 demonstrated that there
is no difference between those who had mammograms
and those who didn't in the breast cancer death rate in
If you're a woman over 50 in Ontario, you will start to women aged 40-49 without a family history of breast
receive letters from Cancer Care Ontario (a provincial cancer.3,8 A summary of all the evidence in 2012 said
agency responsible for improving cancer services) en- that if 2000 women over 50 with no family history of
couraging you to have a breast cancer screening mam- breast cancer had breast cancer screening for 10 years
mogram.' The letters will tell you that doing so is part of or more, the life of one of these women would be saved
taking care of your health. They will continue to arrive by the screening who would otherwise have died.'
until the agency learns that you have had a mammogram. The summary also pointed out that there were
Breast cancer is the most common cancer among some downsides or harms to the screening.' Of those
Canadian women.2 About one in nine Canadian women 2000 women who were screened, about 100 would
will end up with it during their lifetime, and on aver- have a lump or mass show up in the mammogram
age 14 Canadian women will die each day from breast x-ray that turned out on a biopsy examination not to
cancer.2.3 A mammogram screening test is an x-ray that be a cancer. These are false positives. Ten women in
allows a doctor to inspect breast tissue for suspicious the 2000 would have a cancer tumour that was treated
areas. It is not a test that determines the presence of with surgery and perhaps chemotherapy or radiation
cancer. Additional testing is needed to tell this. These tests but unnecessarily. It would be a tumor that is grow-
will include a biopsy examination, meaning that some of ing so slowly that it would not have been a problem
the lump has to be removed for laboratory testing.3 The during their lifetime. These are overdiagnosis and
idea behind this screening test is to catch a cancer early in overtreatment cases. But as science technology and so-
its history and especially before it has spread to the rest of ciety professor Alan Richardson points out, these facts
the body. Once cancer has spread, it is often fatal. and numbers by themselves won't decide that a breast
Does the screening test save lives? Starting in cancer screening program is being oversold.8 Accord-
the 1970s and until a few years ago, it was believed ing to Richardson, determining whether a screening
obvious that mammogram screening saved lives for program like the one in Ontario is worthwhile requires
women of any age whether their family had a history considering social and moral values.
Notes
1. Cathy Risdon, Breast Cancer Screening: Making Sense of the 4. Siddhartha Mukherjee, The Emperor of All Maladies:
Evidence, McMaster University Health Forum Presentation, A Biography of Cancer (New York: Scribner, 2010), p. 395.
17 April 2015. 5. Canadian Task Force on Preventive Health Care,
2. Canadian Cancer Society, Breast Cancer in Canada 2016, "Recommendations on Screening for Breast Cancer in
http://www.cbcf.org/ontario/AboutBreastCancerMain/ Average-Risk Women Aged 40-74 Years," Canadian Medical
FactsStats/Pages/Breast-Cancer-Canada.aspx. Association Journal 183 (November 2011), 1991-2001.
3. Jonathan Sussman, Breast Cancer Screening: Making 6. Anthony B. Miller et al., "Twenty Five Year Follow-up for
Sense of the Evidence, McMaster University Health Breast Cancer Incidence and Mortality of the Canadian
Forum Presentation, 17 April 2015. National Breast Screening Study: Randomised Screening
Trial," British Medical Journal 348 (February 2014), 348-58.
uw I II All 111111
8 Public Health
7. P.C. Gotzsche and K.J. Jorgenson, "Screening for Breast 8. Alan Richardson, "The Breast Cancer Screening's Debate
Cancer," Cochrane Database of Systematic Reviews 2013, Is about Values and Resources," The Globe and Mail
issue 6, article no. CD001877. 29 November 2011.
Case 4 According to a 2016 investigation by The Globe and

Water in First Nations Communities Mail newspaper, one third of Indigenous people living
on reserves—just over 150,000 people—used drink-
ing water from treatment systems that threatened their
Most Canadians expect that they will be able to get health.3 An earlier investigation by CBC News said
water from their kitchen and bathroom taps and that that two-thirds of First Nations communities had at
it will be safe to drink, wash food with, and bathe and least one drinking water advisory from Health Canada
shower in. For the most part this happens.' But in between 2004 and 2014.2
some parts of Canada—in particular, on First Nations For non-Indigenous communities, centralized
reserves—water isn't so easily available or, if it is avail- water systems are governed by local town or city or
able, it isn't safe to drink or bathe in.'•2'3.4 While some municipal authorities.3 These authorities can find the
other Canadians sometimes have the same kinds of financial resources needed to construct and adequately
problems with the availability or safety of their water, maintain and staff water systems through local taxes or
the rate at which these issues happen in First Nations utility fees.4 First Nations water systems, however, are
communities is disproportionately greater.25 Accord- mostly the responsibility of the Indigenous and North-
ing to environmental justice lawyer Kaitlyn Mitchell, ern Affairs Department of the federal government.3.4
this difference "contributes to the perpetuation of According to a recent report by the David Suzuki Foun-
social, political, and legal disadvantage" for people dation, while federal money is often there for the initial
who live in these communities.5 It is, she says, "an on- construction of water treatment facilities, the problem
going human rights violation that serves to compound is that funding resources are not enough for the con-
current and historic injustices First Nations face." tinued proper maintenance of these facilities and the
Some Indigenous communities, for example, are training and salaries of the staff that run them.4 One of
without any running drinking water. They rely on the department's current goals is to improve the avail-
delivery of water by truck, sometimes in big heavy ability and safety of fresh water in these communities.
jugs. Some rely on untreated water from contamin- Recent federal budgets have promised to devote more
ated wells or lakes. But even when they receive water money to the issue.4•' But it has also been decided that
treated in water plants, it can still be unsafe. The water the best way to go about achieving the goal is to allocate
can cause gastrointestinal illnesses and increase the whatever financial resources there are—whether they
risk of cancer if drunk or used to wash food. If people are increased or not—to improving the worst central-
bathe or shower in it, skin infections like eczema and ized water systems in Indigenous communities.3,6 For
psoriasis can develop or worsen.'•5 By August 2016, example, reserves that have had boil-water health noti-
for example, there were 15.3 drinking water safety fications for a long time will have greater priority over
warnings in 114 First Nations reserves across Canada.4 those where the water issues are less significant.
Notes
1. CBC News, "Clean Running Water Still a Luxury o n Many 6. Indigenous and Northern Affairs Canada,
Native Reserves," 12 April 20:L4. "Department's Response to the National Assessment
2. CBC News, "Bad Water: 'Third World' Conditions on First of First Nations Water and Wastewater Systems,"
Nations in Canada," 15 Octoter 2015. https://www.aadnc-aandc.gc.ca/eng/1314383237592/
3. Matthew McCleam, "Water Systems at Risk," The Globe and 1314383364994.
Mail 26 August 2016. 7. Water Canada, "Federal Budget 2016 Promises
4. David Suzuki Foundation and the Council of Canadians, Dedicated Funds for Water Infrastructure, Stewardship,
Glass Half Empty? David Suzuki Foundation, February 2017; and Science," http://watercanada.net/2016/canadas-budget-
David Suzuki, "Government Must Do More to Address First 2016-promises-dedicated-funds-for-water-infrastructure-
Nations' Water Woes," The Georgia Straight 14 February 2017. stewardship-and-science.
5. Kaitlyn Mitchell, "Drinking Water Crisis in First
Nations Communities Violates Human Rights" Ecojustice
23 February 2017.
Study Questions
8.7 Study Questions
1. What is the difference between public health and clinical health topics? Consider some
examples like abortion services. Do these belong to public health or clinical health?
2. What are collective action problems in health care? What is the difference between
someone being a free rider and someone being an overconsumer? Is it always morally
wrong to be a free rider or an overconsumer?
3. One writer in an article not included here has suggested that public health inter-
ventions are just those that deal with collective action problems in health (Johnny
Anomaly, "Public health ethics and public goods," Public Health Ethics vol 4 (2011),
251-259). This is certainly a different definition than the one provided in the intro-
duction to this chapter . Is it a better definition?
4. What is the difference between a screening test and a diagnostic test for an illness?
Could one test be both a screening and a diagnostic test?
5. What is the difference between a false positive result from a screening test and an
overdiagnosis result? Will overdiagnosis always lead to overtreatment?
6. When several different screening tests are available for an illness, they usually differ in
their false positive and their false negative rates. Sometimes both are worse on one test,
but often one is worse and the other better. What's worse in a population screening
test: a higher false positive rate or a higher false negative rate?
7. Are there moral problems with harm reduction programs? What are they? Are they
ones that mean that these kinds of programs should not exist?
8. Are all average differences in health between population groups unjust or unfair? What
makes them unjust or unfair? What about differences in health between individuals?
Are these always unjust or unfair? Are the differences between population groups
unjust or unfair only when those differences between the individuals in one group and
the other are unjust or unfair?
Armstrong, Pat, and Ann Pederson, eds. 2015. Women's Health, 2nd edn. Toronto: Women's Press.
Bernheim, Ruth, James Childress, Richard Bonnie, and Alan Melnick. 2015. Essentials of
Public Health Ethics. Burlington, MA: Jones & Bartlett Learning.
Bognar, Greg, and Iwao Hirose. 2014. The Ethics of Health Care Rationing. New York:
Routledge.
Dawson, Angus, ed. 2011. Public Health Ethics: Key Concepts and Issues in Policy and Practice.
Cambridge: Cambridge University Press.
and Marcel Verweij, eds. 2009. Ethics, Prevention, and Public Health. Oxford: Oxford
University Press.
Greenwood, Margo, Saraj de Leeuw, Nicole Lindsay, Charlotte Reading, eds. 2015.
Determinants of Indigenous Peoples' Health in Canada. Toronto: Canadian Scholars' Press.
Holland, Stephen. 2014. Public Health Ethics, 2nd edn. Cambridge: Polity Press.
Holmes, Morgan, Linda Mooney, David Knox, and Caroline Schact. 2013. Understanding
Social Problems, 5th Canadian edn. Toronto: Nelson Education.
Mankiw, Gregory. 2009. "Public Goods and Common Resources," ch. 11 of Microeconomics,
5th edn. Mason, OH: South-Western Cengage Learning.
Mikkonen, Juha, and Dennis Raphael. 2010. Social Determinants of Health: The Canadian
Facts. Toronto: York University School of Health Policy and Management.
Peckham, Stephen, and Alison Hann, eds. 2009. Public Health Ethics and Practice. Bristol:
Policy Press.
Rossiter, Kate, and Kate Reeve. 2004. The Last Straw! A Board Game on the Social Determinants
of Health. Toronto: Brunswick Books.
l'iessavch vitih Humans
9.1 Introduction
Health research is essential to advancing clinical patient care and human well-being gener-
ally. It is also needed to test "common-sense" and accepted medical practice, which can be
in error, including in cases of invasive or risky treatments. For example, landmark studies
from the 1980s found—contrary to widely held medical belief—that mastectomy is not
supericr to lumpectomy in treating some breast cancers.' Similarly, more recent health
research has been essential in establishing that arthroscopic knee surgery in osteoarthritis
patients is not preferable to a combination of non-invasive physical and medical therapy2
But while scientific evaluation of new and current treatments is vital to advancing human
health and well-being, it poses distinct, sometimes more complex, ethical challenges from
those encountered in day-to-day patient care.
Because of the distinctive nature of health research, we begin with an account of the
differences between health research and health care to make sure those terms are under-
stood at the outset. We then review the history of research scandals involving human re-
search participants (sometimes referred to as "the dark history" of research) and describe
regulatory responses in Canada and elsewhere. These preliminaries sketch out a range of
ethical issues and problems raised by health research. Discussion of the main readings ad-
dressing general and specific ethical issues in health research is set against this background.
The Challenges of Conducting Ethical Health Research
Therapeutic Misconception
Understanding the distinct ethical challenges posed by health research begins with an ap-
preciation of the ways in which health research differs from health care and how they are
sometimes confused. Health research, like clinical care, aims to improve human health. It is
understandable then that the testing of new therapies and collection of observational health
data may often be thought by research participants as intended to make a direct difference
therapeutic misconception to their care. This is often referred to as the therapeutic misconception. This error has
the failure to understand conceptual and empirical aspects and leads to underestimating risks and overestimating
that health research is aimed
the prospects of direct personal benefits of research participation.
primarily at producing
knowledge and is not
The conceptual therapeutic misconception consists of not recognizing that the specific
conducted specifically for the purpose of health research is to advance knowledge. Thus, unlike every treatment in a clin-
sake of research participants' ical setting, research testing of new drugs, medical devices, new ways of managing illness,
health. the collection of health care data, and so on is not done for the sake of the individuals being
studied but rather to advance knowledge of human illness and injury and therapeutic
and other means for restoring and fostering human health. This knowledge is intended
foremost to help future patients. Often, there may be no, or no reasonable, expectation that
participants in a research study will ever benefit directly.
The empirical therapeutic misconception consists in failing to recognize the often
remote chance of actually receiving any direct therapeutic benefit from an experimental
therapy or collection of personal health information for research purposes. Recent re-
views of research involving investigational (that is, new) drugs report that only about
10 per cent are ever licensed as therapies.3 A large majority of research participants in
clinical studies of new drugs is therefore exposed to research-related risks and never
benefit directly from those experimental therapies. Even the 10 per cent figure is likely
to mislead. Only a much smaller overall percentage of research participants is likely to
receive any direct benefit from experimental drugs that are later licensed. Some phases
of experimental research do not look principally, or at all, for therapeutic benefits; some-
times research is about a range of dosages for drugs, not all of which are likely to be
both safe and effective. Further, in most experimental research there is normally up to a
50 per cent prospect of being randomized into a study arm where research participants study arm participants in
will receive no new treatment at all—for example, they receive a placebo or a treatment clinical trials are randomly
assigned to different groups
available outside the study against which the experimental treatment is compared (re-
(arms) receiving different
ferred to as "the control arm"). interventions to compare the
Research participants are not the only individuals who can fall prey to therapeutic mis- effectiveness and safety of
conception. Researchers can, too. This is partly because many of them are also health care those interventions.
providers and may see research as an extension of clinical care. We shall also see below that
the therapeutic misconception may also be subtly present in how researchers conceptualize
the design of studies as meeting duties of care to research participants.
Informed Consent
We have seen that it is not the primary aim of health research to provide therapeutic
benefits to research participants and that this will not be statistically likely in most cases.
Failure to recognize this undermines a research participant's understanding of a study and
their role in it and so is inconsistent with informed consent to participate in research. An
obvious solution is to recognize that these misconceptions are based on informational
deficiencies that should be addressed as part of the process of obtaining informed consent
from potential research participants. But addressing therapeutic misconceptions is by no
means the only special challenge of obtaining informed consent in health research. Indeed,
health care research poses other burdens on informed consent that may be even more
difficult to address.
To make informed choices, research participants have to understand clearly the dif-
ferences between research and clinical care. They also have to be informed about their
clinical options, how the research differs from those options, and what the related potential
risks and benefits of each are. Since less is typically known about the risks and benefits of
experimental interventions, potential participants often have to make decisions under a
greater burden of uncertainty about potential outcomes than in clinical care. They will also
need to understand the research methodology in order to know what will happen to them
during the research. This is often no small matter. Trying to understand what is involved
in a randomized, double-blind, placebo-controlled trial and how a participant will fit randomized, double-blind,
into this methodology can be difficult for many. Potential research participants also need placebo-controlled trial
to be informed about research-related tests, which can be extensive, what their risks are, a study in which research
and how they differ from tests that are part of standard care. They should also be informed participants are randomly
assigned to either an
of special issues around confidentiality. In clinical care, access to patient information is
experimental treatment or a
limited to a patient's physician and some members of the patient's circle of care, whereas placebo and neither they nor
research data may be accessed by sponsors, government regulatory bodies, research ethics the researchers know what
boards, and sometimes other researchers. Sometimes research participants' data, biological they are receiving.
II ill II IGLI
9 Research with Humans
samples, and personal information must be stored or analyzed in other countries where
there are different standards for protecting confidentiality.
The requirements of informed consent in health care research are often more ex-
tensive and demanding of research participants than those for patients receiving clin-
ical care. There is also the further problem that informed consent cannot be obtained
for some types of research. For example, research in emergencies must often be done
without consent because participants may lack capacity and there may be no time to
obtain consent from substitute decision-makers. In other cases, the consent process may
undermine research for other reasons. For example, disclosing to nurses that they are
being observed to determine how well they implement a new or experimental protocol
for, say, triaging patients in emergency rooms may bias study results. If the nurses knew
they were being observed, they could alter their behaviour to better comply with a new
protocol, and so an accurate picture of its workability and usefulness would not emerge.
Hawthorne Effect the (This is sometimes referred to as the Hawthorne Effect.) Sometimes obtaining consent
alteration of research is impractical for other reasons. For example, obtaining consent for collecting clinical
participants' behaviour as
chart data accumulated over many years may be impractical because of difficulties in
a result of knowing that
they are being observed
contacting patients or because consent may introduce bias into the research because only
by researchers, typically by healthy or living patients will respond.
improving their behaviour.
Conflicts of Interest
Potential participants should also be aware that the aims of research and health care can
conflicts of interest conflict and that the research enterprise itself contains a variety of conflicts of interest.
situations in which a person They need to be informed about those conflicts and how they will be managed. Suppose
(for example, a researcher) has
researchers are trying to answer a scientific question about the safety and effectiveness of
an obligation to a person or
an institution that is in tension
a therapeutic intervention. If a participant is not doing well in a study, he or she may wish
with another interest or to withdraw. But that will sometimes limit information about side effects, a clearer picture
obligation and could corrupt of which may emerge if the participant remains in the study. There is thus a potential
the decision-making of that conflict between the scientific needs of a study and the needs of a participant. The conflict
person (that is, the researcher).
can become more complicated if the researcher is also the participant's caregiver (e.g.,
personal physician or nurse), since the researcher will have conflicting roles and duties.
Sometimes researchers also have financial or career interests in the outcomes of studies.
Special processes are needed, then, to manage conflicts of interest. One way to address
them, at least partially, is to disclose them to potential participants and emphasize the
voluntariness of research participation, including that participants can withdraw at any
time without having to give reasons. As well, researchers should often play a limited role
in recruiting participants and obtaining informed consent, particularly when researchers
are also caregivers. Researchers with financial interests in the interventions being studied
should play a limited, if any, part in the collection and analysis of the data.
observational research
attempts to advance
knowledge by collecting data Designing Ethical Studies
from research participants
Recognizing and responding to therapeutic misconception, the special challenges of seeking
without making any change to
their care, treatment, or lives
informed consent, and conflicts of interest do not nearly exhaust the ethical problems in-
in relation to what is being volved in. research. Arguably, the overarching ethical issue in clinical research is how to design
studied. experimental and observational studies to advance knowledge of human health and at the
same time respect the rights and well-being of research participants. This is the main conflict
of values in health research. How it is resolved defines the ethical character of the research.
experimental (or A main challenge for designing ethical research is for researchers to create scientific
interventional) research studies that appropriately minimize risks to participants—that is, valid research that ob-
attempts to advance
knowledge by studying the
tains reliable, useful results and at the same time observes principles of non-maleficence
effects of research interventions and harm prevention. For observational research, the main ethical requirement is to pro-
on research participants. tect the confidentiality of the research data that is collected. Experimental research must
protect confidentiality but faces a further range of ethical problems related to minimizing
risks of research interventions.
An intuitively attractive solution for experimental studies is to accept research design
as ethical where there is clinical equipoise between the study arms. Clinical equipoise clinical equipoise the
means that an experiment on humans is only justified if there is genuine uncertainty among state of genuine uncertainty
within the expert community
the relevant expert community as to which arm in a study is better.4 Clinical equipoise has
about which study arm is
seemed to many to answer important questions about the ethical design of experimental therapeutically superior
health research. In particular, it seems to meet a duty of care researchers have to par- in a multi-arm study of an
ticipants. If experts cannot say which arm of a study is superior, then administering the experimental or existing
treatments does not involve researchers knowingly exposing anyone to unnecessary harm. therapy.
It also implies that there can be an acknowledged prospect of a direct benefit in the experi-
mental arm. This appears to provide an important rule, then, for when a multi-arm study
design can be ethically justified. It also appears to explain when a study must stop, since
if evidence emerges during a study that shows one arm to be clearly inferior to another,
clinical equipoise has been disturbed, and the study must stop in order not to expose some
participants to a known inferior intervention.
The problem with clinical equipoise, however, is that it is extremely difficult to know
when it exists, particularly when untested treatments are compared against standard treat-
ments with known effectiveness. Moreover, there are many accepted examples of experi-
mental research that simply do not seem to meet standards of clinical equipoise (discussed
below). And a general, often unacknowledged issue is how to determine when there is
genuine uncertainty within the relevant expert community as to which arm in a proposed
study is better. It will not be when all researchers are indifferent about competing arms,
since researchers often strongly believe one arm to be superior to others. Must a majority
of relevant experts be indifferent, then? Or is the idea of indifference irrelevant and to
be replaced by the idea of respectable disagreement among relevant experts? If so, what
constitutes respectable disagreement, and when is it overturned?
One of the ethical principles that underlies clinical equipoise, that of not exposing
research participants to known unnecessary risks, has further important implications for
the design of ethically sound clinical research. It can apparently provide ethical guidance
about when use of placebos and standard treatments as controls is justified. It implies that
if there is a standard treatment, it would be ethically wrong to use a placebo control, since
that would knowingly expose some research participants to a known inferior therapy, thus
violating a duty of care. This principle also explains why health care research must satisfy
high standards for scientific integrity, particularly for research that exposes participants
to more than minimal (or everyday) risk. If research lacks scientific value, its results will
not help to answer questions related to health care, and the participants will have been
exposed to risk without any prospect that the research will contribute to knowledge about
human health. Bad science is always unethical in health research. It wastes scarce research
resources, wastes the time and abuses the trust of participants (who mistakenly think they
are contributing to science), and exposes participants to risk without any prospect that the
research will contribute anything of scientific or social value.
The Dark History of Research
An International Problem
The need to take account o f the special ethical requirements of conducting health research
was brought to a head after the discovery of Nazi medical research atrocities during the
Second World War. The subsequent trial and conviction of many Nazi doctors for war
crimes led to the creation of the first internationally recognized code of research ethics,
the Nuremberg Code (later replaced by the World Medical Association's Declaration
of Helsinki).' The Nazis were not alone in participating in ethical violations in health
iI JII 1111
research. The Japanese conducted similar experiments in Manchuria during the Second
World War.° Developed modern democracies also engaged in unethical medical experi-
ments before, during, and after the Second World War, including after the Nuremberg
Code and Declaration of Helsinki were adopted. The most famous of these research
scandals is probably the Tuskegee Syphilis Study, which observed the natural progress of
syphilis in a population of poor, mostly illiterate, black agricultural workers in Alabama
between 1932 and 1972.7 In order to conduct the study, participants were not told they
had syphilis and were denied access to penicillin once it became a recognized effective
treatment for the disease. They were enticed to participate with promises of free medical
care, meals, and burial insurance.
The Tuskegee study was not particularly exceptional. In 1966, American researcher Henry
K. Beecher reported 22 examples of research that failed to meet adequate ethical standards,
often involving vulnerable individuals, including children and the mentally ill.8 Neither the
Tuskegee study nor the studies Beecher described were hidden from public view. They were
known to medical researchers, and their results were often published in leading medical jour-
nals. Further evidence of US research scandals emerged through the 1960s. In 1972, public
outrage as a result of news reports finally stopped the Tuskegee study. Government hearings
that were held into the scandal led to The Belmont Report,' the first systematic attempt to sum-
marize ethical principles and provide guidance for human research. The Belmont Report led to
adoption in 1981 of a set of US federal regulations now in place for conducting clinical trials
known as "the Common Rule."10 The Common Rule was then, and arguably remains today,
the leading and most comprehensive set of research ethics regulations.
Canada, too, has its own dark history of medical experimentation, part of which is
reviewed in this chapter. Colin A. Ross's article "Dr Ewen Cameron" (see page 485) describes
how a prominent Canadian psychiatrist engaged in "unethical, unscientific, and inhumane"
brainwashing and later mind-control research in Canadian mental health institutions from
the 1930s to the 1960s, including at McGill University. Dr Cameron was not alone among
Canadian doctors in conducting this research, much of which was also published in medical
journals. Some of the research was funded by the US military and Central Intelligence Agency.
More recently, historians have begun to report controlled nutritional and other experiments
undertaken without consent in First Nations communities and residential schools."
The Canadian Response

Research scandals have not been as prominently discussed in Canada as in the United
States, and they may not have been as extensive. That may be due more to lack of
funding and opportunity than to any greater ethical sensitivity on the part of Canadian
physicians and researchers. However, it is very unlikely that these types of research
could be conducted today in Canada. Although there is no specific Canadian law gov-
erning health research in Canada, there is, in effect, a set of national standards estab-
lished in The Di-Council Policy Statement: Ethical Research Involving Humans (second
edition, usually referred to as "TcPS2").'2 TCPS2 establishes ethical rules and guidelines
for regulating research along with processes for ethical review of research by independ-
ent "research ethics boards." The Tri-Council is a federally supported group of research
funding councils that includes the Canadian Institutes of Health Research, the Natural
Sciences and Engineering Research Council, and the Social Sciences and Humanities
Research Council. TCPS2 operates much as a de facto Common Rule in Canada in that
to be eligible for Tri-Council funding, Canadian universities (including hospitals with
university affiliations) and colleges are required to adhere to TCPS2 in all their human re-
search, whether or not it is Tri-Council-funded. This encompasses most health research
that is conducted in Canada. (Students who are interested in receiving a certificate of ac-
creditation for completing a course on research ethics can take the TCPS2 online course.
Links to TCPS2 and the tutorial are in Suggested Further Reading at the end of this
The Dark History of Research 481
chapter.) TCPS2 is notably unique among research ethics guidelines and codes in that
it combines statements of standards with extensive discussion of their justification and
meaning as a way of assisting interpretation and application of the standards to specific
research contexts.
Often, research ethics regulations from other countries have to be taken into account
as well. Research originating in other countries must also often satisfy those countries'
ethical requirements. For example, US research that is conducted abroad, including
in Canada, must nevertheless satisfy the standards of the Common Rule. Physician re-
searchers are also expected to observe the standards in the World Medical Association's
Declaration of Helsinki in virtue of Canada's membership in the WMA. Often, these differ-
ent codes and regulations do not fully agree, posing further challenges for ethical review
and compliance.
Problems Still Not Over

Despite the more careful regulation now of the ethical conduct of research studies, ethical
scandals still arise from time to time. In "Biomedical Conflicts of Interest: A Defence of the
Sequestration Thesis" (see page 489), Canadian bioethicist Arthur Schafer highlights how
commercial pharmaceutical funding of research can present financial conflicts of interest
that can interfere with the ethical treatment of human research participants. The leading
recent example of this is the case of Dr Nancy Olivieri, who was testing an experimental
drug developed by the pharmaceutical company Apotex at the Toronto Hospital for Sick
Kids. Olivieri found evidence in 1996 that the treatment could cause harm and regarded
it as her duty to inform participants of the potential risks. Her contract with Apotex,
however, stated that she could only release information about her findings with Apotex's
consent, which was refused. Dr Olivieri disclosed the information anyway. Instead of being
supported by the hospital, she was fired from her position as head of the Hemoglobin
Research Program and was the general subject of a smear campaign. Olivieri argued that
she had obligations to disclose this information to her participants, the hospital research
ethics board, and other researchers. For a long time, she received no effective support from
the University of Toronto where she held a related appointment. Commentators generally
view this case as a reflection of the power large pharmaceutical companies have to pres-
sure hospitals and universities to tolerate unethical conduct through their prerogative to
withdraw research funding.
Olivieri was later vindicated in a variety of reviews of the case. The obligation to
disclose to research participants evidence of harm without the consent of drug sponsors
is now accepted. Nevertheless, many commentators are still doubtful that issues regarding
industry-sponsored research have been fully addressed."
Designing Ethical Research Studies
In "An Ethical Framework for Biomedical Research" (see page 495), Ezekiel J. Emanuel,
David Wendler, and Christine Grady identify eight ethical principles of health care research
that they claim "delineate a systematic and comprehensive way of assessing the ethics of
particular clinical research'" These principles are an attempt to provide a general ethical
framework to address the challenges just reviewed and to make sure that the dark history
of research is not repeated.
The eight principles are: (1) The research must be a collaborative partnership be-
tween researchers and the communities in which the research is done. (2) The research
must have social value. (3) The research must have scientific validity without denying
generally available health services or requiring access to health services that are not prac-
tically available. (4) The selection of research participants must be fair. (5) There must be
a favourable risk-benefit ratio. (6) The research must receive independent ethical review.
(7) Informed consent must be appropriately obtained. (8) Research participants must be
treated with respect throughout their participation in research and after they have finished
their participation.
Emanuel et al.'s discussion is an excellent starting place for thinking about the ethical
requirements of health research. But while their eight principles recognize and address
many of the ethical challenges facing research mentioned earlier, putting them into effect
is frequently not straightforward. For example, it can be difficult to determine whether
research will contribute to improvements in health care, whether the science is valid,
whether important populations of interest will be excluded, what counts as an appropriate
collaborative partnership between researchers and the communities they are investigating,
what determines an appropriate balance of risks and benefits, whether ethical review has
been independent and sufficient, whether the elements of informed consent have been
properly fulfilled or waived on appropriate grounds, and what measures should be taken
to respect participants after their involvement in research has finished.
Some of Emanuel et al.'s discussed requirements are controversial. It might be asked
whether requiring fulfillment of all these standards in the detail provided can be too
demanding and unduly discourage or prevent needed health research. Also, Emanuel
et al. do not directly address important controversies about designing ethical research.
For example, there is no mention of recognizing the role of clinical equipoise in design-
ing ethical research. In this respect, they may have been influenced by Franklin G. Miller
and Howard Brody who have provided an acute challenge to the role of clinical equi-
poise in health research in "A Critique of Clinical Equipoise: Therapeutic Misconception
in the Ethics of Clinical Trials" (see page 509) and critiqued an earlier defense of clinical
equipoise by Emanuel et al. Miller and Brody point out that the recent history of clinical
research has been mistakenly dominated by what they call "the similarity position"—
namely, that the ethics of clinical equipoise requires upholding duties of care similar to
those of clinical care, or generally putting care of research participants ahead of research
considerations where they conflict. Miller and Brody argue in response that the ethics
of research is fundamentally different from clinical care and that the notion of clinical
equipoise should be abandoned. They say their position is reflected by research ethics
boards' and regulators' recognition of the methodological importance of sometimes
having placebo-controlled studies even when standard therapies are available for use as
controls. They also point to the extra-therapeutic risks involved in research-related tests
and other research procedures as being inconsistent with the similarity position, includ-
ing invasive tests like lumbar punctures and "wash out periods" (when active treatments
for the condition under study are discontinued to establish a non-treatment baseline).
Miller and Brody argue that failure to recognize the fundamental differences between
research and clinical care is in effect an extension of the therapeutic misconception men-
tioned earlier. In this context, however, it is the researchers who accept the idea of equipoise
who have a misconception that research requires no known risks over and above those of
standard clinical care. Miller and Brody's alternative is that scientific research should be
driven by considerations of what methodologies will best contribute to scientific know-
ledge, unless they pose unacceptable levels of risk to participants One important practical
shortcoming of their discussion is that they do not say what unacceptable levels of risk are.
If researchers do not owe research participants a duty of care that is comparable to clinical
duties of care, what can participants reasonably be asked to consent to? It is important to
ask, however, whether questions about acceptable levels of risk raise objections to their
view or whether they highlight another difficult area of research ethics deliberation, sup-
porting, in effect, robust ethical review of research via independent research ethics boards.
If Miller and Brody are right that the notion of clinical equipoise is inappropriate and often
rightly ignored, a reconsideration of the nature and extent of a duty of care to research
Designing Ethical Research Studies 483
participants will be necessary to clarify the boundaries of ethically acceptable research. For
many ethicists, Miller and Brody's article poses a strong challenge to conventional wisdom
regarding the ethics of clinical research. It is fair to say, however, that their arguments
are controversial and have not yet significantly influenced official attitudes toward clinical
studies, which retain today a strong similarity position in research ethics codes and regula-
tions. The use of clinical equipoise as a principle for guiding clinical studies is retained, for
example, in Canadian research ethics guidelines in TCPS2.
Obviously, scientifically and ethically well-designed studies as envisaged by Emanuel
et al. are meaningless if research is not reported with integrity. In "Repairing Research
Integrity" (see page 523), Sandra L. Titus, James A. Wells, and Lawrence J. Rhoades discuss
the prevalence of researcher misconduct involving misrepresentation of research results
and what measures can be taken to combat it. See also Case 1 in this chapter for a famous
Canadian example of researcher misconduct (see page 547).
Research on Special Populations
One important issue addressed b y Emanuel et al. that is now beginning to find its way
into research ethics guidelines is to what extent interests of communities should be taken
into account in clinical research. This has been a pressing issue in Canada because of
the practical realities and legal status of First Nations peoples. They have interests in
knowing that research will serve community interests, that they will not be exploited by
researchers as they have been in the past, and that dissemination of research results will
not promote discrimination against their members. Researchers, by contrast, are often
interested in studying particular communities because they afford useful opportunities
to understand the etiology of disease, especially genetic determinants, by looking at
groups where prevalence of certain diseases is known to be high. The article by Charles
Weijer and Ezekiel J. Emanuel, "Protecting Communities in Biomedical Research"
(see page 518), is an attempt to develop a consultative framework for resolving these
often competing interests. Weijer and Emanuel's ideas are broadly reflected in Chapter 9 of
TCPS2, which is addressed primarily to research involving First Nations, Inuit, and Metis
communities. Weijer and Emanuel are clear, however, that respect for communities should
be extended further. This reflects the position taken by Emanuel et al., discussed earlier. In
many respects, these authors' proposals are now frequently reflected in work researchers
do with certain communities that share illness or disease, for example, HIv patients or
drug addicts. However, consultations with such groups pose different sorts of challenges
because of difficulties defining what constitutes a community and the frequent absence of
geographically defined communities with identifiable leadership or political institutions.
The chapter ends with two articles that discuss the challenges of doing research
on vulnerable persons and populations. There is general consensus that the vulnerable
have claims to special protection in clinical research, though there has been no agree-
ment in research ethics guidelines as to what this means and, therefore, who it covers.
The article by Samia A. Hurst, "Vulnerability in Research and Health Care; Describing
the Elephant in the Room?" (see page 526), identifies the limitations of efforts to clarify
what makes a person or population vulnerable. Hurst argues that vulnerability consists
of "an increased likelihood of incurring additional or greater wrong"15 (that is, moral
mistreatment) and that this definition should guide the application of research ethics
guidelines. The definition can also inform the conduct of clinical care involving vulner-
able persons or populations and so is relevant for other chapters in this textbook. Hurst's
article can be usefully compared with the discussions of vulnerability in TCPS2.16 In
"Best Practices for Health Research Involving Children and Adolescents" (see page 537),
D. Avard et al. discuss ethical treatment of young persons in clinical research. These best
II Wu IIILL.I
practices were intended to assist with the application of TCPS2 to pediatric research and
assent a right that are particularly notable for their presentation of the notion of assent. Research ethics
incompetent persons can guidelines now universally recognize that incompetent patients, including minors, have
have to accept or refuse certain rights to refuse to participate in research even if substitute decision-makers—for
treatment or participation in
research. It is to be contrasted
example parents—are prepared to consent on their behalf. A right to assent/dissent
with a competent person's is thr.s to be contrasted with rights to consent. But as Avard et al. acknowledge, the
right to consent. extent of a child's right to overrule his or her parent is controversial, and there is no
accepted international standard in this respect. TCPS2 takes an unusually strong pos-
ition, however, which is also reflected in Avard et al.'s discussion: namely, if the child
lacks capacity for consent but has some ability to understand the research and dissents
to participation, that decision must be respected despite a substitute decision-maker's
willingness to consent on the child's behalf.
Notes
1. See Case 1 in this chapter.
2. Carly Weeks, 2008, "Athroscopic Knee Surgery Ineffective: Study," The Globe and Mail 11 September,
ht:piJwwwtheglobeandmail.com/life/arthroscopic-knee-surgery-ineffective-study/article 18 137799.
3. Biotechnology Innovation Organization, "Clinical Development Success Rates 2006-2015" (2016),
hups://www.bio.org/sites/default/files/Clinical%20Development%20Success%20Rates%202006-
2015%20-%20BIO,%20Biomedtracker,%20Amplion%202016.pdf.
4. Benjamin Freedman, "Equipoise and the Ethics of Clinical Research," New England Journal of
Me dicine 317 (1987), 141-5.
5. The Nuremberg Military Tribunal, The Nuremberg Code, from US v. Karl Brandt et al. (1947), https://
history.nih.gov/research/downloads/nuremberg.pdf; World Medical Association, 2013, Declaration of
Helsinki (2013), http://www.wma.net/en/30publications/lOpolicies/b3.
6. Sheldon H. Harris, Factories of Death: Japanese Biological Warfare 1932-45 and the American Cover Up
(New York: Routledge, 1994).
7. Gregory E. Pence, "The Tuskegee Study," in Classic Cases in Medical Ethics, 2nd edn (New York:
McGraw Hill, 1995).
8. Henry K. Beecher, "Ethics and Clinical Research," New England Journal of Medicine 274 (1966),
1354-60.
9. Naiional Commission for the Protection of Human Subjects of Biomedical and Behavioral Research,
Department of Health, Education and Welfare, The Belmont Report: Ethical Principles and Guidelines
for 'he Protection of Human Subjects of Research (Washington, DC: United States Government Printing
Office, (1978).
10. US Department of Health and Human Services, National Institutes of Health, and Office of Human
Research Protections (revised), The Common Rule, Title 45 (Public Welfare), Code of Federal Regulations,
Pan: 46 (Protection of Human Subjects), (2009).
11. Ian Mosbey, "Administering Colonial Science: Nutrition Research and Human Biomedical
Experimentation in Aboriginal Communities and Residential Schools, 1942-1952." Social Histoiy 46
(2013) no. 91,145-72.
12. Canadian Institutes of Health Research, Natural Sciences and Engineering Research Council of
Canada, and Social Sciences and Humanities Research Council, Tri-Council Policy Statement: Ethical
Conduct for Research Involving Humans (2014), referred to as `TCPS2".
13. Jocelyn Downie, "The Power of Money: Commercialisation of Research Conducted in Public
Institutions," University of Otago Law Review 11 (2006) no. 2,305-25.
14. Ezekiel J. Emanuel, David Wendler, and Christine Grady, "An Ethical Framework
for Biomedical Research," in The Oxford Textbook of Clinical Research Ethics, edited by Ezekiel, J.
Emanuel, Christine C Grady, Robert A. Crouch, Reider K. Lie, Franklin G. Miller, and David D.
Wendler (New York: Oxford University Press, 2011) p.131.
15. Sam: a A. Hurst, "Vulnerability in Research and Health Care', Describing the Elephant in the Room?"
Bioethics, (2008) 22(4): p. 195.
16. TCPS2, page 8 and 54.
Ross: Dr Ewen Cameron
9.2 The Dark History of Human Research in Canada
Dr Ewen Cameron was not a relevant variable in the academic equation.

What counted for the psychiatrists, I think, was money,
Colin A. Ross power, perks, academic advancement, and the thrill of
being a spy doctor.
Born in Bridge of Allan, Scotland on 24 December Despite the code of silence, and despite later
1901, [Ewen] Cameron immigrated to Canada in claims by the Canadian Psychiatric Association that
1929 to take a job as a psychiatrist at Brandon Mental Dr Cameron was unaware he was working for the CIA,
Hospital in Brandon, Manitoba. He was recruited unwitting investigator status for Cameron is implaus-
by Dr Thomas Pincock; one of the buildings in the ible for several reasons. He was far too politically con-
Department of Psychiatry at the University of Manitoba nected to be unwitting. At various times, Dr Cameron
in Winnipeg in the 1980s was the Pincock Building. was president of the Quebec, Canadian, American
Dr George Sisler and Dr John Matas, both of whom re- and World Psychiatric Associations, the Society of
ferred patients to Dr Cameron when he was at the Allan Biological Psychiatry, and the American Geriatrics
Memorial Institute in Montreal, taught at the University Society. Dr Cameron was one of four co-founders
of Manitoba into the 1980s, as did Dr Gordon Lambert, of the World Psychiatric Association; another was
who treated one of Dr Cameron's mind control victims Dr William Sargant,1 the foremost British authority on
on her return to Winnipeg. brainwashing. Many Board Members and Presidents
Despite these historical connections, I heard no of the Society of Biological Psychiatry were LSD re-
conversation about Dr Ewen Cameron or CIA mind searchers, funded by the military or otherwise in the
control while a resident and then a staff psychiatrist mind control network.
in the Department of Psychiatry in Winnipeg from A letter from the CIA to Senator Pete Wilson
1981 to 1991, despite the fact that plaintiffs, in- dated 11 December 1985 states that the CIA contacted
cluding Val Orlikow from Winnipeg, settled a suit Dr Cameron directly. On page 4, the correspondent says:
with the CIA in 1988. Mrs Orlikow's husband, David
Orlikow, had been a prominent Member of Parlia- First, the CIA did not instigate this research,
ment from Winnipeg for many years. There was si- create the protocol, or supervise the work. Rather
lence in psychiatry about CIA mind control, but no CIA contacted a prominent and highly respected
conspiracy of silence. No one was told to be quiet. Canadian psychiatrist, Dr Ewen Cameron,
From the perspective of academic psychiatry, mind who was conducting research into treatment of
control experimentation didn't exist, so there was no mental illness with drugs such as LSD, and the
need to cover it up. CIA provided minimal and partial funding for
Throughout the twentieth century, academic a short time period. In return, the CIA received
psychiatry provided no public commentary, ethical periodic reports on his research into behavioural
guidance, peer review, or moral oversight of any kind modification through a process which he termed
concerning mind control experimentation, despite the "psychic driving."
fact that the leading psychiatrists and medical schools
were well funded by the CIA and military for mind Dr Cameron was eulogized in obituaries in the
control research. Mental patients, cancer patients, Canadian Psychiatric Association Journal,2 the Canadian
prisoners, and unwitting citizens were experimented Medical Association Journal,3 the American Journal of
on by mind control doctors at Yale, Harvard, McGill, Psychiatry,4 and Recent Advances in Biological Psych-
Stanford, UCLA, and the other major universities. iatry,5 the latter written by Hudson Hoaglund, PHD,
These human guinea pigs were never told that they who was personally referred to J. Edgar Hoover by G.H.
were subjects in military and CIA mind control experi- Estabrooks. Dr Cameron received many awards includ-
ments, and they never gave informed consent. They re- ing the Adolph Meyer Award, the Samuel Rubin Award,
ceived no systematic follow-up to document the harm and the Montreal Mental Hygiene Institute Award,
done to them. The welfare of the "human subjects" given to "a scientist who has made an outstanding
Source From The CIA Doctors: Human Rights Violations by American Psychiatrists (Richardson, TX: Manitou Communications, 2006).
Reprinted by permission of the author.
II ail 111111
contribution to the mental health of the Canadian describes his brainwashing techniques and says, "An-
people." alogous to this is the breakdown of the individual
In a n article entitled "McGill University Department under continuous interrogation."
of Psychiatry 50th Anniversary" Pinard and Young6 Psychic driving was a procedure carried out in
echo the sentiments of the eulogist: two stages; in the first stage, patients were depatterned,
which meant they were reduced to a vegetable state
Since the department's inception in 1943, research through a combination of massive amounts of elec-
has been a preponderant part of its mission; this troconvulsive shock, drug-induced sleep, and sensory
was stated in the very first reports to the university isolation and deprivation. When fully depatterned,
by the department's founder, Ewen Cameron. . . . patients were incontinent of urine and feces, unable to
The departmentb record has not been one of feed themselves, and unable to state their name, age,
unblemished success. Cameron's drive led to the location, or the current date.
foundation and growth of the department, but also In the second stage, psychic driving was introduced.
led him to perform much publicized experiments of This consisted of hundreds of hours of tape loops being
doubtful ethical or scientific value in which patients played to the patient through earphones, special hel-
received multiple courses of ECT or doses of LSD. mets or speakers in the sensory isolation room. The
tape loops repeated statements of supposed psycho-
In his obituary in the Canadian Psychiatric Association logical significance. If such procedures were carried out
Journal2 Cameron is eulogized as follows: under third world dictators, they would be denounced
as human rights violations by American and Canadian
As a diligent seeker after new knowledge, a gifted psychiatry, and would be called brainwashing.
author, a renowned administrator, and inspiring There is a further reason to conclude that Ewen
teacher he brought, not only to his professional Cameron had a security clearance and was witting of
colleagues but also to the community at large, a CIA funding of his research; Dr Cameron definitely had
wider and deeper understanding of the import- a security clearance with the US government. In 1945
ance and significance of the emotional life of man. he was part of an American team that did psychiatric as-
sessments of German War criminals including Rudolph
Dr Cameron began conducting unethical, unscien- Hess, who was examined at the request of the Military
tific, and inhumane brainwashing experiments at Tribunal in Nuremberg. Dr Cameron must have heard
Brandon Mental Hospital in :he 1930s. He continued about the mescaline research done in the death camps
this work into the 1960s. In one paper' Dr Cam- by Nazi psychiatrists. He himself instituted similar
eron describes treating schizophrenics with red light work at McGill when he began experimenting with LSD.
produced by filtering light from 15 200-watt lamps Rather than being the object of suspicion and
through an inch of running water and a layer of sodium investigation in the 1950s, Dr Cameron was well re-
salt of ditolyl disazo-bis-napthylamine and sulphuric garded in the Canadian media. Favourable articles
acid impregnated into cellophane. about him were entitled "Canadian Psychiatrists De-
The colour red was chosen because it is the colour velop Beneficial Brainwashing,"10 "New 'Personalities'
of blood. In these experiments, schizophrenic patients Made to Order,"u and "Two-Month Sleep, Shock New
were forced to lie naked in red light for eight hours a Schizophrenic Cure."12 Similarly as recently as 6 June
day for periods as long as eight months. Another experi- 1987, the official position of the Canadian Psychiatric
ment involved overheating patients in an electric cage Association on Dr Cameron's brainwashing experi-
until their body temperatures reached 102 degrees F. ments was far from negative:
After leaving Brandon Mental Hospital in 1936,
Cameron took a job at Worcester State Hospital in . . . the fact that Dr Cameron's research would not
Massachusetts. The Worcester Foundation for Experi- be accepted by today's standards of ethical and sci-
mental Biology received CIA money through MKUL- entific inquiry cannot be used as a retrospective cri-
TRA Subproject 8, and was tie professional home of tique of his work. What has to be recognized clearly
Dr Cameron's eulogist, Hudson Hoaglund.8 At Worces- is that in the intervening 20 to 30 years there has
ter State Hospital, Dr Cameron massively over-utilized been a continuing progression of scientific and eth-
insulin coma therapy by putting patients in coma for ical research standards that included much more so-
two to five hours per day for up to 50 days in a row. phisticated peer review and ethical approval review
In a paper published in the American Journal of now in place as part of standard practice. This rep-
Psychiatry entitled "Psychic Driving," Dr Cameron9 resents the evolution of concern and control for all
medical research using human subjects deriving in radiation experiments and the Tuskegee Syphilis Study,
part out of concerns experienced in several fields of and the Canadian Government has established a fund
medicine. Such experiments would not be permit- that compensates victims of unethical experiments by
ted in today's research climate. Dr Cameron at the Allan Memorial Institute; compen-
sation of $100,000.00 can be activated by documented
The position on Dr Cameron taken by the Canadian victims by calling a toll-free number provided by the
Psychiatric Association is mistaken for several reasons. Canadian Government. Given the positions taken by
Dr Cameron received a grant from Canada's Depart- two federal governments on such medical experiments,
ment of Health and Welfare for $57,750.00 for the years the position of the Canadian Psychiatric Association on
1961 to 1964 for "A Study of Factors Which Promote Dr Ewen Cameron requires revision.
or Retard Personality Change in Individuals Exposed to Dr Cameron was not the only researcher at McGill
Prolonged Repetition of Verbal Signals." The Helsinki funded by the CIA and the military. Another psychiatrist
Declaration governing ethical rules for medical research at McGill, Dr Raymond Prince,13 was funded through
was adopted in 1964; Dr Cameron's brainwashing ex- MKULTRA Subproject 121. Dr Prince was an unwitting
periments clearly violated the principles of informed investigator and is the only psychiatrist to have written
consent and protection of the patient from undue harm about DA mind control in the peer-reviewed medical
contained in the Helsinki Declaration. literature. He is the only MKULTRA contractor to have
Dr Cameron's experiments alsoviolated the informed publicly identified himself to date. No other MKULTRA
consent provisions of the Nuremberg Code, which arose contractor has engaged in any public discussion of
out of the war crime trials of the Nazi doctors, in which psychiatric participation in CIA and military mind control.
Dr Cameron participated as a member of the American Dr Hassan Azima was a young McGill psychiatrist
psychiatric team. He thus had direct knowledge of the who was being groomed as a military mind control
medical atrocities the Nuremberg code was designed to contractor prior to his death from cancer in his early
prevent. The Canadian Psychiatric Association's position forties. A colleague, Dr Sarwer-Foner,'4 gave the Hassan
that Dr Cameron's research would "not be permitted in Azima Memorial Lecture at a meeting of the Society of
today's research climate" is correct, but ignores the fact Biological Psychiatry; Dr Cameron was a Past President
that the rules of ethical conduct in medical research have of the Society.
not changed since Nuremberg. Dr Azima15,16 worked at the Allan Memorial Insti-
The fact that medical schools were routinely lax in tute, where he gave psilocybin to patients; psilocybin is
ensuring that prevailing ethical codes were adhered to in the active ingredient of "magic mushrooms." He also at-
the 1950s and 1960s is a condemnation of the medical tended LSD symposia and performed sensory isolation
schools, not a vindication of Dr Cameron. I consider the experiments17 that caused damage to patients. Two pa-
Canadian Psychiatric Association's official position on tients with "obsessional neuroses manifested acute psy-
the mind control experiments conducted by Dr Ewen chotic episodes. They were treated with electric shock,
Cameron to be a violation of the Hippocratic Oath. Lies which resulted in improvement in both paranoid and
and silence concerning psychiatric mind control experi- obsessional features." Azima and Cramer18 write:
mentation are a betrayal of the physician's ethical duty
The fact that Dr Cameron's unethical, inhumane, Contrary to the above case, a hebephrenic-
and grossly damaging experiments were published catatonic girl who remained in isolation for six
in the psychiatric literature is a condemnation of the days showed no perceptual alteration. Behaviour-
editorial standards of the journals, not a vindica- ally, she manifested overt hostility, became quite
tion of Dr Cameron. The only argument protective talkative and self-assertive. Her F.D. [figure draw-
of the psychiatric journals is the fact that Dr Cam- ings] revealed gradual, but definite emergence of
eron whitewashed the experiments for publication. aggressive tendencies. She also experienced several
Dr Cameron's brainwashing experiments stopped in spontaneous orgasms, and verbalized memories of
1964, whereas the Tuskegee Syphilis Study continued her "sexual adventures."
until 1972. The continuation of the Tuskegee Syph-
ilis Study under the auspices of the Center for Disease Another patient in the series is described as follows:
Control until 1972 does not provide vindication for
Dr Cameron, rather it provides further grounds for Another case of obsession neurosis, suffering severe
criticism of organized medicine. motor compulsions, who had not responded to any
The US Government has officially apologized form of treatment, was put in isolation with the ex-
to and financially compensated the victims of the plicit aim of provoking a psychotic disorganization.
Al II AEI IIILLI
He remained five days in isolation, began to Another McGill psychiatrist, Dr James Tyhurst,
manifest signs of depersonalization on the second worked at the Allan Memorial Institute and received
day, and showed several acute psychotic episodes, funding from Canada's Defence Research Board for
lasting about three hours on the fourth and fifth studies of individual reactions to community disasters.2°
days. The disorganization manifested itself, in part, Disaster studies were also the subject of investigation
as a marked disinhibition. He experienced many in MKULTRA Subproject 126, which was approved by
spontaneous orgasms, and manifested overt erotic the CIA in 1960. Dr Tyhurst attended a meeting with
behaviour toward the nurses. His eating habits de- CIA personnel in 1951 in Montreal devoted to over-
teriorated, and his behaviour was like that of a very sight of BLUEBIRD and ARTICHOKE.21 He also worked at
hungry child during the feeding periods. In the post- Hollywood Hospital in Vancouver, where hundreds of
isolation period he showed some reorganization and patients were treated with LSD.22
lost some of his motor compulsions. But because of Dr Donald Hebb, Head of the Department of
the appearance of some paranoid tendencies, he was Psychology at McGill during the 1950s, received fund-
put on electric shock therapy, which resulted in con- ing from Canada's Defence Research Board for experi-
siderable improvement and subsequent discharge. ments on sensory isolation.23 The network of doctors
with CIA and military funding at McGill included Dr
Dr Azima19 published a paper with Dr Eric Cameron, Dr Hebb, Dr Tyhurst, Dr Wittkower, and
Wittkower, who worked at the Transcultural Psychiatry Dr Prince, and in addition Dr Azima was firmly estab-
Institute at McGill, where Dr Prince was employed. lished in the mind control network and using many
Dr Wittkower founded and. edited The Transcultural of the same experimental procedures. LSD research was
Psychiatric Research Review. The Review was funded by also done at McGill and Montreal General Hospital by
CIA cutout the Society for the Investigation of Human Dr J.H. Quastel.24 Any claim that Dr Cameron's CIA
Ecology, which lists a payment to Dr Wittkower of funding was an anomaly or isolated incident is there-
$7,500.00 in its 1961 Annual Report. The Board of fore incorrect.
Advisors for the Review inc:_uded Dr Ewen Cameron Medical experimentation by the Department of
and Margaret Mead, who received CIA money for her Psychiatry at McGill resulted in death, psychosis, vege-
anthropology research, and who was married to Greg- table states, organic brain damage, and permanent loss
ory Bateson. Bateson took LSD supplied to him by a of memory among other damages. It resulted in the cre-
psychiatrist, and both Bateson and Mead were mem- ation of amnesia, identity disturbance, and depersonal-
bers of the Cybernetics Group, which was funded by ization among other dissociative symptoms. Dr Ewen
CIA cutout The Josiah Macy, J r. Foundation. Cameron was the main figure in these activities.
Notes
1. Sargant, W. 1957. Battle For the Mind. Doubleday: Garden 11. Cahill, B. 1956. "New 'Personalities' Made to Order," in
City, NY. Montreal Gazette (18 June).
2. Anonymous. 1967. "In Memoriam. Donald Ewen Cameron: 12. Cahill, B. 1957. "Two Month Sleep, Shock New
1901-1967," in Canadian Psychiatric Journal 12: 475. Schizophrenic Cure," in Montreal Gazette (2 Sept.).
3. Cleghorn, R.A. 1967. "D. Ewen Cameron, MD FRCP [C]," in 13. Prince, R. 1995. "The Central Intelligence Agency and the
Canadian Medical Association journal 97: 985. Origins of Transcultural Psychiatry at McGill University,"
4. B., E J. 1967. "D. Ewen Cameron: 1901-1967," in in Annals of the Royal College of Physicians and Surgeons of
American Journal of Psychiatry 124: 168-9. Canada 28 (7): 407-13.
5. Hoagland, H. 1967. "Donald Ewen Cameron: 1901-1967," 14. Sarwer-Foner, G. J. 1963. "On the Mechanisms of
in Recent Advances in Biological Psychiatry 10: 321-2. Action of Neuroleptic Drugs: A Theoretical
6. Pinard, G., and S.N. Young. 1993. "McGill University, Psychodynamic Explanation. The Hassan Azima
Department of Psychiatry 50th Anniversary," in Journal of Memorial Lecture," in Recent Advances in Biological
Psychiatric Neuroscience 4: 141-2. Psychiatry 6: 244-57.
7. Cameron, D.E. 1936. "Red Light Therapy in 15. Azima, H. 1958. "Sleep Treatment in Mental Disorders," in
Schizophrenia," in British Journal of Physical Medicine 10: 11. Diseases of the Nervous System 19: 523-30.
8. Hoagland, H., and Donald Ewen Cameron: 1901-1967. 16. Azima, H. 1962. "Psilocybin Disorganization," in Recent
1967. Recent Advances in Biological Psychiatry 10: 321-2. Advances in Biological Psychiatry 5: 184-98.
9. Cameron, D.E. 1956. "Psychic Driving," in American 17. Azima, H., and E J. Cramer. 1956. "Effects of Partial
Journal of Psychiatry 112: 502-9. Perceptual Isolation in Mentally Disturbed Individuals," in
10. Moore, J. 1955. "Canadian Psychiatrists Develop Beneficial Diseases of the Nervous System 17: 117-22.
Brain Washing," in Weekend Magazine 5: 40. 18. Ibid.
19. Azima, H., E.D. Wittkower, and J. LaTendresse. 1958. 22. MacLean, J.R., MacDonald, D.C., Ogden, F, and E. Wilby.
"Object Relations Therapy in Schizophrenic States," in "LSD 25 and Mescaline as Therapeutic Adjuvants," in
American Journal of Psychiatry 115: 60-2. The Use of LSD in Psychotherapy and Alcoholism, ed. H.
20. Tyhurst, J.S. 1951. "Individual Reactions to Community Abramson. Bobbs-Merrill Company: New York, 407-29.
Disaster," in American Journal of Psychiatry 107: 764-9. 23. See note 21.
21. Gillmor, D. 1987. I Swear By Apollo. Dr Ewen Cameron and 24. Author. 1959. "Clues to Biochemistry o f Schizophrenia: May
the CIA Brainwashing Experiments.: Eden Press: Montreal. Lead to Rational Therapy of Disease," in Factor 8-9 (Dec.).
Biomedical Conflicts of Interest: • Each of them appealed for assistance, unavailingly,

to the senior administrators of the University of To-
A Defence of the Sequestration ronto and its Faculty of Medicine. Although there
Thesis—Learning from the Cases of had been a changeover of university presidents
Nancy Olivieri and David Healy and medical faculty deans in the interval between
these two scandals, personnel changes made very
Arthur Schafer little difference to the university's official response.
• In both scandals, university and hospital officials
The leading individual roles in this diptych are taken by failed to recognize that there had been a funda-
two internationally eminent medical researchers, hema- mental violation of the principle of academic free-
tologist Nancy Olivieri and psychiatrist David Healy. dom at the affiliated hospitals.
The institutional players include one research-intensive • In both cases, the whistle-blowing physicians
university (the University of Toronto) and two affiliated found themselves removed from their positions:
research-intensive teaching hospitals (the Hospital for Olivieri was fired from her position as director of
Sick Children, referred to as "Sick Kids" or "Sick Kids' the Hemoglobinopathy Research Program at Sick
Hospital" and the Centre for Addiction and Mental Kids' Hospital; Healy's employment contract with
Health, referred to as "CAMH"). The cast of supporting both CAMH and the University of Toronto's Depart-
characters is large. On one side are senior hospital ad- ment of Psychiatry was terminated.
ministrators and medical faculty deans, together with • Both hospitals and the university denied strenu-
hospital and university presidents and boards of direc- ously that these "firings" were in any way related
tors. On the other side is to be found a small group of to the whistle-blowing
medical scientists, supported primarily by the Canadian • Damaging rumours were circulated among Olivi-
Association of University Teachers (CAUT). eri's colleagues, including allegations that she
Not coincidentally, the Olivieri and Healy scandals was scientifically incompetent, guilty of stealing
share in common a number of key elements money from her research grants, unethical in her
patient care, and sleeping with some of the scien-
• Wealthy and powerful drug companies hover in tists who looked favourably on her research find-
the background of both, and sometimes occupy a ings;' damaging rumours were circulated about
good deal of the foreground, as well: Apotex in the Healy that he was a bad clinician, and both a racist
case of Olivieri, Eli Lilly in the case of Healy. and a member of a cult known as Scientology. A
• These drug companies not only fund university journalist who telephoned me for an interview at
and hospital researchers, they are also major the height of the Healy controversy asked whether
donors to the institutions within which researchers I knew that Healy was a prominent Scientologist.
carry out their clinical studies. Her previous interviewee had been a hospital
• Neither Apotex nor Eli Lilly was happy to have spokesperson who was circulating that piece of
adverse information about their drugs publicized. disinformation among the media, presumably in
• Both Olivieri and Healy personally experienced an effort to discredit Dr Healy.
serious negative consequences from their willing- • The perpetrators of these false but damaging ac-
ness to speak publicly about potential dangers cusations against Olivieri and Healy mostly pre-
to patients. ferred to remain anonymous. . . .
Source: From Journal of Medical Ethics 30 (2004) 1, 8-24. Copyright © BMJ Publishing Group Ltd with permission from BMJ Pub-
lishing Group Ltd. Reprinted with permission.
II aIi 111111
Nancy Olivieri and the Hospital for possible risks of harm. Respect for the value of patient
Sick Children autonomy clearly requires that those patients who are
also research subjects be given all materially relevant
. . . Once Dr Olivieri came to believe, based on scientific- information in order to enable them to decide whether
ally credible preliminary evidence, that the experimental they wish to continue participating in a clinical trial.
treatment she was administering might cause unantici- It should go without saying that the information to
pated harm to some of her patients/research subjects, which patients are morally and legally entitled includes
she was duty-bound to disclose those risks. (The risks information about risks of harm which comes to light
of harm were discovered by serial liver biopsy but actual during the course of a clinical trial.
harm, were it to occur, would be expected to occur very Olivieri also had ancillary obligations to report any
gradually over a period of many years.) Olivieri's uni- newly discovered risks to the research ethics board of
versity and her hospital had a corollary duty to support her hospital and to share her findings with other re-
her request for assistance in this exercise of academic searchers, both at scientific meetings and in peer re-
freedom and in the performance of her obligations as a viewed journals. Only in this way could her colleagues,
physician and a researcher. Their failure to provide this worldwide, test and assess her conclusions and properly
support in an effective manner raises important ques- inform their own thalassemia patients of newly discov-
tions about the way in which society funds biomedical ered potential risks. In every case, Dr Olivieri behaved
research institutions and biomedical research. in the manner required by her professional obligations,
Every version of the Hippocratic Oath, from an- though she, and the core group of colleagues who sup-
cient times down to the present day, has had, as its ported her, paid a heavy career and personal price for
leading principle, some version of the maxim that "the doing her/their duty. Dr Brenda Gallie, Dr Helen Chan,
life and health of my patient will be my first considera- Dr Peter Dune, and Dr John Dick were all colleagues of
tion." Thus, whether or not Dr Olivieri is ultimately Olivieri at the Hospital for Sick Children. All supported
proven to have been correct in her negative interpreta- Olivieri in her struggles with Apotex, Sick Kids' Hospi-
tion of the preliminary scier tific data, once her data tal, and the University of Toronto, despite serious risks
indicated the possibility of unanticipated harm, she of harm to their own careers.3 It is difficult not to em-
was morally obliged to inform her patients of this pathize with Olivieri when she laments: "It should not
risk. Writing in the New England Journal of Medicine, be so hard to protect children at Sick Kids' Hospital."
two blood science researchers, David G. Nathan, of This sentence was quoted by a colleague and supporter
Harvard's Dana-Farber Cancer Institute and David J. of Olivieri, Dr Paul Ranalli, in a letter published by the
Weatherall, o f Oxford University, comment that: "as Globe and Mail, headed "Courage under Fire."4
of this writing, the safety and efficacy of deferiprone It is true, of course, that Apotex had a legal con-
have not been established." They suggest that it takes tract with Dr Olivieri, which was signed in 1995. That
years of careful monitoring before the effectiveness contract contained a confidentiality provision—one
of any iron chelator is clinically established; but they that prohibited her from disclosure "to any third party"
then go on to remark that: "Suffice it to say, when the of data from her Apotex-sponsored clinical trial of the
dispute began, Olivieri had good reason to believe that drug deferiprone, without the express permission of the
deferiprone was neither safe nor effective."2 (Since company, for a period of three years after the termina-
Apotex discontinued the clinical trial prematurely, no tion of the trial. The non-disclosure clause of the LA—Ol
conclusive scientific evidence exists, and the scientific contract between Olivieri and Apotex reads as follows:
aspect of the controversy remains unresolved.)
One important qualification should be appended All information whether or not obtained or gener-
to this claim. Given that the stakes were high, both ated by the investigators during the term of this
for the patients/research subjects and for the drug agreement and for a period of one year thereafter,
company, Olivieri had an obligation to exercise due shall be and remain secret and confidential and
diligence by consulting qualified colleagues about her shall not be disclosed in any manner to any third
interpretation. This she did, and they supported her party, except to an appropriate regulatory agency
concerns. (Apotex, however, most definitely did not for the purposes of obtaining regulatory approval
agree with Olivieri's interpretation of her data, and the to manufacture, use or sell Ll unless the informa-
company was supported in its favourable interpretation has been previously disclosed to the public
tion by a number of scientists receiving financial sup- with the consent of Apotex. The investigator shall
port from them.) It would then be the responsibility not submit any information for publication with-
of patients to weigh the hoped-for benefits against the out the prior written approval of Apotex.
Schafer: Biomedical Conflicts of Interest
This clause, it should be noted, does not specif- them, and warned her not to disclose information "in
ically list "patients," but they would clearly appear to any manner to any third party except with the prior
be covered under the phrase "any third party." Olivieri written consent of Apotex," and warned further that it
claims, supported by tapes of telephone conversations would "vigorously pursue all legal remedies in the event
with Apotex, that Apotex threatened repeatedly to sue that there is any breach of these obligations."5 Just prior
her if she breached the confidentiality clause of the to their suing her, she sued them for defamation.
contract, and that they warned her not to disclose her According to one standard account of heroism, the
concerns to patients and others. hero is a person who acts far beyond the call of duty. By
At the time Olivieri signed the contract such non- this test, Olivieri's actions would not count as heroic.
disclosure provisions were common. Olivieri readily She only did that which it was her duty to do. But there
admits that she failed to appreciate the potential sig- is another account of heroism according to which the
nificance of that contract, and concedes that she should hero is a person who does her duty, at great risk to her
never have signed it. The University of Toronto admits own self-interest, when most others would resist from
that it was guilty of an institutional oversight by per- fear. Olivieri relates a story about her personal fears.
mitting its researchers to agree to such terms, and the Sitting in a restaurant with her scientific collaborator,
university subsequently took steps to preclude repeti- Dr Garry Brittenham, she raised the concern with him
tion by any of its faculty. that if they were to break their contract with Apotex by
Because Apotex refused, repeatedly, to give per- disclosing to patients the risks they had newly discov-
mission for disclosure, Olivieri might have been found ered, Apotex might act on its threats to sue, and the enor-
legally liable for significant damages arising out of her mous costs of fighting such a lawsuit could mean that
disclosure of risks to her patients and colleagues. they would each lose their homes. Brittenham replied:
Since the protection of human life is, other things "Red wine or white?" At that moment, Olivieri reports,
being equal, a higher value than respect for the sanctity it became obvious to her that there was no decision to
of contracts, it is possible that the legal system would make concerning whether or not to disclose. They had
have "thrown out of court" any lawsuit for breach of to disclose the risks. Better, therefore, to concentrate
contract brought by Apotex against Olivieri, as being on those matters that still required a decision, such as
against public policy and, hence, unenforceable. the colour of the wine they were to consume with their
For our purposes, it matters little whether the non- dinner (The Current, CBC radio interview, 2 Mar 02).
disclosure provision of the contract Olivieri signed Apotex is currently suing Olivieri for damages,
with Apotex would have been found by the courts to be claiming that she defamed both the company and their
nugatory. Nor, for our purposes, does it matter much drug (deferiprone). Olivieri is suing Apotex for defama-
whether the information Olivieri disclosed to patients tion. For the benefit of those who have had the good
and colleagues was information actually covered by the fortune never to be involved in a legal action of this sort,
terms of the confidentiality agreement she signed—an it is perhaps worth noting that the costs of defending
issue also in dispute. Even if Nancy Olivieri were legally such an action (at least in North America and England)
bound to keep confidential all information about the tend to be ruinously expensive; hence, utterly beyond
risks of deferiprone, she was morally obliged to disclose the means of any except the wealthiest individuals.
that information to her patients and to her colleagues When Olivieri turned to her hospital and university
worldwide. It could be argued, of course, that one has for financial and other help in the face of intimidat-
a moral obligation to keep the contracts one signs. This ing threats of legal action against her, they provided
moral obligation is prima facie, however, rather than little effective assistance.° Instead, both the University
absolute, and should surely be overridden where the of Toronto and the Hospital for Sick Children "took ac-
lives and health of patients are at stake. tions that were harmful to Dr Olivieri's interests and
Apotex did not agree with Olivieri's interpretation professional reputation, and disrupted her work."7 In
of her data and they refused her request to disclose their public pronouncements about the case, none of
these risks to her patients. They also threatened to take the senior administrators of the university, the medical
legal action against her if she were to violate the non- faculty, or the hospital gave any sign that they recog-
disclosure clause of the contract. Olivieri proceeded nized that the case was one involving a serious issue of
anyway, in the face of these threats, to disclose her find- academic freedom. They justified their official "tread
ings, and some time after these events, the company lightly" policy in part by characterizing the conflict as
did take legal action against her. On 24 May 1996, for a "scientific dispute," to be resolved primarily between
example, Apotex wrote to Olivieri that it was terminat- the parties themselves. Some University of Toronto of-
ing both of the clinical trials she was conducting for ficials did make efforts, behind the scenes, to promote
MI III II AI
II j11 111111
a settlement between Olivieri and Apotex but, as the are typically viewed, at least in part, as an investment.
Thompson report found, the support which they of- This, in turn, raises the questions (to which an answer
fered was "not effective."8 is supplied later): What exactly is being bought by such
It was discovered during this period of conflict investments? What exactly is being sold?
and controversy that the University of Toronto was
negotiating for a twenty million dollar donation from David Healy and the Centre for
Apotex (with additional millions promised for its af- Addiction and Mental Health
filiated hospitals). Some were led to speculate that the
university's failure to recognize and support Olivieri's In December of 2000, while the Olivieri affair was
academic freedom might not have been unconnected still capturing attention, both within and without the
to its eagerness to secure financial support from Apotex University of Toronto, a second major scandal, also
for the university's proposed molecular medicine build- raising basic issues of academic freedom and patient
ing project. Indeed, it was subsequently revealed that safety, was brewing at the same university.
the university's then-president had gone so far as to Some months previously, the Centre for Addic-
lobby the Government of Canada on behalf of Apotex. tion and Mental Health had hired Dr David Healy to
In a private letter to the Prime Minister of Canada, become the new director of its Mood and Anxiety Dis-
President Robert Prichard stated that the government's orders Clinic. After accepting their offer and the offer
proposed changes to drug patent regulations would of a joint appointment in the university's Department of
adversely affect Apotex revenues and could thereby Psychiatry, Healy notified his employer in Wales of his
jeopardize the building of the university's new medical intention to resign, and prepared to move his family to
research centre. President Prichard was unsuccessful in Canada to take up this new appointment. Then, Healy's
persuading the federal government to change its drug career plans came dramatically unstuck.
patent laws, but his action demonstrated the lengths On 30 November 2002, some months before his
to which the university was prepared to go in appeas- new appointment was officially scheduled to begin,
ing the company or promoting its interests. When Healy made a conference presentation at CAMH. (The
Prichard's conduct became public knowledge, he symposium, called "Looking Back, Looking Ahead,"
apologized to the executive committee of the university was held to mark the 75th anniversary of the university
for acting inappropriately in this matter.9 Department of Psychiatry, as well as the 150th anni-
This embarrassing episode illustrates the dangers versary of the Queen Street Mental Health Service.) In
that can ensue from university reliance upon industry this lecture, which he subsequently delivered at Cor-
"philanthropy." When career success for university/ nell, and in Paris, Minneapolis, and Cambridge, Healy
hospital presidents and deans is measured in significant raised the question of whether the drug Prozac, manu-
part by their ability to raise vast sums of money from factured by Eli Lilly, might be responsible for increasing
corporate donors, such fundraising can easily become the risk of suicide among certain kinds of patients. This
a dominating priority. In North America, top univer- issue was by no means the principal theme of Healy's
sity and hospital officials are now required to ride two talk, but the potential link of Prozac to patient sui-
horses their fundraiser's horse and, simultaneously, cides, and the call for further research on this matter,
their academic horse (as guardians of core university was almost certainly regarded by CAMH officials as the
values). Unfortunately, those who attempt to ride two most controversial part of Healy's presentation. Healy
horses can come to grief when, as sometimes happens, reports that his talk was well received in all the places
the horses pull in opposite directions. Perhaps it is time where he presented it, and it is noteworthy that the
for a radical rethinking of the competing role respon- audience at the CAMH conference honoured his lecture
sibilities of top university and hospital officials. with the highest rating for content.1° Despite this fact,
The word "philanthropy" is placed above in warn- senior administrators of CAMH were not well pleased.
ing quotes, not to suggest that big pharma never behaves Within 24 hours of the talk they were trying to contact
in a genuinely philanthropic manner but, rather, to flag him. Within a week he received an email unilaterally
the point that when corporate donors make substantial rescinding their offer of employment.
donations they often expect to gain substantial influ- Why was David Healy's employment terminated
ence. Indeed, it is the legally mandated duty of corpor- so precipitately by both the Centre for Addiction and
ate executives and board members to act in the "best Mental Health and the university Department of Psych-
interests" of the corporation, which is commonly inter- iatry? No one disputes that Healy is an internationally
preted to mean that they have a legal duty to maximize distinguished psychiatrist and researcher. The uni-
overall profitability. Corporate donations to universities versity and CAMH recruited him with enthusiasm and
persistence. Since he was unhired almost immediately University of Toronto officials later denied that
after he gave his conference lecture at CAMH, the infer- Healy was unhired because of fears on their part that
ence is inescapable that his contract for employment if Dr Healy were allowed to take up his position drug
was cancelled because of the contents of his lecture that companies might be reluctant in future to donate money
day. In this lecture, Healy expressed the view, referred to or fund research at the centre. Notwithstanding their
to above, that the antidepressant drug Prozac might strenuous denials, however, many people understood
cause some patients to commit suicide. Although Healy the above quoted words to mean". . . the university was
did not condemn Prozac outright, he did advocate cau- worried about the risk to the financial inflows to the
tion on the part of doctors who prescribe this drug, department from pharmaceutical company sources."13
and he called for further research into possible adverse In September 2001, an international group of
side effects. He was also critical of the practice whereby physicians published an open letter to the president
drug companies are engaged in ghostwriting some of of the University of Toronto, in which they protested
the therapeutic literature. against what they termed the "maltreatment" of Dr
Some time prior to Healy's conference presen- Healy. In their open letter they concluded: "To have
tation, Eli Lilly had donated 1.5 million dollars to sullied Dr Healy's reputation by withdrawing the job
CAMH, and a new wing of the hospital, built with their offer is an affront to the standards of free speech and
financial assistance, was scheduled to have its official academic freedom." The signatories, who included two
opening soon after. There is no evidence that Eli Lilly Nobel Prize winners, chose not to focus on the possible
attempted to have Dr Healy fired from his new appoint- involvement of a drug company in university affairs,
ment at CAMH. The incident raises legitimate questions, but they nevertheless insisted that the central issue in
however, about whether those involved with rescinding the case was the failure of the University of Toronto and
his contract offer were affected, consciously or uncon- CAMH to uphold "the standards of open discussion and
sciously by the relationship between CAMH and Eli Lilly. frank exchange in university life."14 That is, the issue
In this connection, it is worth noting that six was essentially one of academic freedom.
months before Healy delivered his fateful presentation at The university's official response to the concerns
CAMH, he had published an article on Prozac in the bio- expressed by this international group of scholars was
medical ethics journal, the Hastings Center Report.11 In dismissive: they (the protesting scholars) were ill-
this article, Healy developed several of the themes which informed outsiders, unaware of all the pertinent infor-
later became controversial at the University of Toronto, mation. University of Toronto spokespeople went even
namely suicide and Prozac, and ghostwriting of scientific further in their defence of the unhiring of Dr Healy by
articles by drug companies. Eli Lilly, which had hitherto suggesting that his publicly expressed concerns were
been the largest annual private donor to the Hastings dangerously irresponsible. On the University of To-
Center, publisher of the Hastings Center Report, subse- ronto's website, Healy's warnings about the potential
quently withdrew its financial support for the centre. hazards of Prozac were compared to the "fool" who
The administrations of both the University of Toronto cries "fire" in a crowded theatre.'5 To this accusation,
and CAMH claim that the unhiring of Dr Healy had noth- Healy responds: "But what if there is a fire in the
ing to do with academic freedom. Instead, they contend, theatre?"16 It is worth bearing in mind that Prozac or
his lecture gave rise to "clinical concerns" and revealed other drugs of its class, known as SSR1s, are often pre-
that he would be a "bad fit" with his new colleagues. It scribed to healthy patients with problems in living. If,
may be worth quoting a key paragraph from the email as Healy believes the evidence indicates, some of these
which the University of Toronto sent to Healy by way of healthy patients become suicidal because of their in-
explaining their decision to rescind his contract: gestion of SSRI-type drugs, then a failure to warn them
and their physicians of this potential side effect would
Essentially, we believe that it is not a good fit be- be grossly irresponsible.
tween you and the role as leader of an academic The argument underlying such an analogy is, pre-
program in mood and anxiety disorders at the sumably that Healy's warnings (of possible adverse side
centre. While you are held in high regard as a effects from taking Prozac) might deter some depressive
scholar of the history of modern psychiatry, we patients from using Prozac or other SSRI drugs and this,
do not feel your approach is compatible with the in turn, might result in their committing suicide. In
goals for development of the academic and clinical other words, the university's position seems to be that
resource that we have. This view was solidified by when the values of clinical care clash with the values
your recent appearance at the centre in the context of science, the former should trump the latter. The
of an academic lecture.0 problem with this argument, however, is that if valid it
41
II del IIILL1
proves too much. It proves that researchers ought never A short time after the university's dismissive rejec-
to warn patients of potentially harmful side effects lest tion of the open letter, described above, Dr Healy initi-
some patients thereupon forgo an effective medication. ated what might have been the first legal action in the
This manifestation of an approach often labelled English-speaking world based, in part, on the alleged
"physician paternalism" would be morally objection- tort of violating academic freedom. ("Alleged" because
able because it would usurp the patient's right to give until there is a legal precedent in which the courts find
informed consent to treatment. How can patients weigh that such a tort exists, one cannot be sure of its validity)
and balance the benefits and harms of treatment options A settlement was subsequently negotiated, which in-
(including the option of not taking any antidepres- cluded the appointment of Healy as visiting professor
sant medication) if evidence about potential harms is in the Department of the History of Medicine (with
deliberately withheld from them? There is by now a unrestricted academic freedom to speak out publicly
vast literature, both legal and ethical, in which the near about any of the issues). The appointment as visiting
universal consensus of philosophers and jurists is that professor of the history of medicine was for one week
competent adult patients have a fundamental right to a year during each of the following three years. He was
give informed consent to treatment. In practice, this not permitted, however, to assume the position for
means that research scientists must make the results of which he had originally been hired, as director of the
their research public, so that physicians can adequately Mood Disorders and Anxiety Clinic of CAMH.
inform their patients about potential risks. The duty to David Healy, himself, feels little doubt about the
warn would seem, then, to be a fundamental obliga- most important lesson to be learned from his experience
tion of every research scientist. Both Healy and Olivieri at CAMH, and he insists that it is the same lesson that
were alerting patients and the scientific community to should be learned from the experience of Nancy Olivieri
the need for further researc a into potentially serious at Sick Kids' Hospital: "What is involved is a contrast be-
adverse consequences of the drugs they were investi- tween the values of science and the values of business.""
gating. For either to have remained silent about their Although the Thompson report dealt only with the Olivi-
preliminary adverse data would surely have been a vio- eri case, the Thompson authors, like Healy, conclude
lation of their legal, as well as their moral, duty. that the problem is system-wide: "The safety of research
Interestingly, on 10 June 2003, the Medicines and subjects in clinical trials and the integrity of the research
Healthcare Products Regulatory Agency [MHRA] of the project are more important than corporate interests.'
UK issued a caution to physicians that Seroxat (Paxil) Nathan and Weatherall, in their NEJM commentary on
was "contra-indicated" in children under 18 for the the Olivieri case, reach a similar conclusion,19 as does
treatment of major depressive disorder. Potential side ef- Somerville writing in Nature,2° and as do the authors of
fects include dramatically increased risk of "potentially the CMAJ article "Dancing with porcupines."21 Together,
suicidal behaviour" (G. Duff, personal communication, the Healy and Olivieri cases have forced both the uni-
2003). Thus, it seems that recent evidence further con- versity community and the wider public to confront
firms the wisdom of Healy's warnings about drugs of the ways in which university-industry partnerships can
the SSRI category. Sadly, his scrupulous caution appears imperil the fundamental values of academic freedom,
to have cost him his job at the University of Toronto. research integrity, and patient safety . . .
Notes
1. O'Hara, J. 1998. "Whistleblowing," in Maclean's Magazine: 66. 6. Ibid., 29.

2. Nathan, D.G., and Weatherall, D.J. 2002. "Academic 7. Ibid., 32.
Freedom in Clinical Research," in New England Journal of 8. Ibid.
Medicine 347: 1368-70. 9. Ibid., 13.
3. Naimark, A., Knoppers, B.M., and Lowry, EH. 1998. 10. Healy, D. 2000. "Good Science or Good Business?,"
Clinical Trials of Li (Deferiprone) at The Hospital for Sick in Hastings Center Report 30.
Children: A Review of the Facts and Circumstances. Hospital 11. Ibid., 19-22.
for Sick Children: Toronto. 12. Ibid., 6.
4. Ranalli, P 1998. "Courage Ur. der Fire" [letter], in Globe 13. Ibid., 6.
and Mail. 14. Ibid.
5. Thompson, J., Baircl, P., and Downie, J. 2001. Report of the 15. Clarke, C. 1998. 'Top Scientists Allege U of T Academic Chill,"
Committee of Inquiry on the Case Involving Dr Nancy Olivieri, the in The Globe and Mail.
Hospital for Sick Children, the Un iversity of Toronto, and Apotex 16. Healy, "Good Science or Good Business?," 11.
Inc. Canadian Association of Uriversity Teachers: Toronto, 143. 17. Ibid., 11.
18. Thompson, Baird, and Downie, Report of the Committee of 21. Lewis, S., Baird, P, and Evans, R.G., et al. 2001. "Dancing
Inquiry, 17. with the Porcupine: Rules for Governing the University/
19. Nathan and Weatherall, "Academic Freedom in Clinical Industry Relationship," in CMAJ 165: 783-5.
Research."
20. Somerville, M.A. 2002. "Post-Modern Tale: The Ethics of
Research Relationships," in Nature 1: 316-20.
9.3 Ethical Issues in the Design and Conduct of Clinical Research
An Ethical Framework for 22 principles subsequently expanded to 32, contain

no elaboration." Such sparse, oracular statements lack
Biomedical Research an overarching framework to ensure that all relevant
Ezekiel J. Emanuel David Wendler, and Christine Grady ethical issues are addressed. They also lack justifica-
tions for their claims, implying that the ethical guidance
is either self-evident or beyond debate. Consequently,
Over the last 60 years or so, there has been myriad guid- when controversies arise about whether the principle
ance on the ethical conduct of research with humans1-13 itself is valid or how a principle should be applied to
Despite the profusion, the extant guidance seems flawed a case, there is nothing to appeal to other than the au-
in several respects. First, most guidance was "born in thority of these documents. Agreement can frequently
scandal."14 That is, the guidelines or reports were a re- be secured on the broad principles, but this often hides
sponse to a specific controversy, and therefore tend to deep disagreements about how they should be inter-
focus on what was perceived to be the transgression of preted and applied to specific situations.1°
that scandal. The Nuremberg Code directly addressed Finally, and maybe most important, the existing
the atrocities of the Nazi physicians; the Belmont guidance seems mistaken on some important issues.
Report was a response to the Tuskegee Syphilis Study For instance, the Nuremberg Code's strong statement
and other scandals;4 and the Advisory Committee on that "the voluntary consent of the human subject is
Human Radiation Experiments responded to covert absolutely essential" seems to prohibit all pediatric
radiation experiments during the Cold War and there- research.2 Yet this seems wrong. Similarly, the 1993
fore emphasized deception.15 Second, regulatory guid- Council for International Organizations of Medical
ance tends not to examine the overall ethics of research Sciences (CIOMS) guidelines recommended that Phase
but to have a specific practical purpose. For instance, I or II studies of drugs and vaccines should be con-
the International Conference on Harmonisation has the ducted first in sponsoring countries before being done
purpose of creating common rules across developed in developing countries.17 Because of strong objections,
countries for the "registration of pharmaceuticals for especially by developing countries, a decade later this
human use."8 The aim is more to enhance the efficiency was deleted from the revision.° The most recent ver-
of drug approval than to protect research participants, sion of the Declaration of Helsinki addresses conflicts
for which it defers to the Declaration of Helsinki.3 In of interest through disclosure, requiring that potential
general, these regulatory guidelines emphasize the pro- research participants be adequately informed about
cedural safeguards of informed consent and independ- "any possible conflict of interest" and that these "should
ent review by an institutional review board or research be declared in the publication."3 The value and import-
ethics committee because these leave "paper trails" that ance of disclosing conflicts of interest to research par-
can subsequently be audited. ticipants is controversia1.18 More important, exclusive
Both of these deficiencies contribute to a third: ex- reliance on disclosure in the absence of prohibitions on
isting guidance is neither comprehensive nor systematic. certain conflicts of interest seems inadequate.19.20
The guidelines tend to be lists of claims or principles. Because of the deficiencies of existing research
For instance, the Nuremberg Code with its 10 state- ethics guidance, there is a need for a broader, system-
ments and the Declaration of Helsinki, originally with atic, and comprehensive framework that includes an
Source: From The Oxford Textbook of Clinical Research Ethics (2011), Chapter 11:123-135. Reprinted with permission.
II .111 1111.11
496 9 Research with Humans
ethical justification and specification for how each not necessarily include all of them. In addition, they
principle is to be fulfilled in practice.21'22 Among other are presented sequentially, going from the develop-
goals, this framework should incorporate those con- ment of research proposals to the conduct of research
cerns that overlap in the existing guidance and organ- to monitoring during research.
ize them into a coherent whole. Each principle is specified by benchmarks that offer
a specific elaboration and understanding of each prin-
Fundamental Ethical Purpose ciple.22 The benchmarks are practical interpretations of
what is required to fulfill each principle.22,28.29 In this
Informing this overarching framework is the under- sense, the benchmarks should clarify and focus the
standing that the fundamental ethical challenge of all kinds of values and considerations at stake in fulfilling
research with humans is to avoid exploitation.21.22 Re- each principle. No matter how specific and detailed,
search aims at obtaining generalizable knowledge that the benchmarks cannot eliminate all controversy over
can be used to improve health and health care. Partici- the principles.16.22 However, by specifying and clarify-
pants in research are a necessary means to obtaining ing the eight principles, these benchmarks should help
this knowledge. Consequently, participants are used in to narrow any disagreement related to specific cases,
the research process for the benefit of others and are making it easier to focus on the substance of the dis-
at risk of being exploited. The fundamental purpose agreement, assess the importance of the problems and
of research guidelines is to minimize the possibility of concerns, and even identify potential solutions.22
exploitation in clinical research.
There are two distinct conceptions of exploitation. Collaborative Partnership
Both are important in protecing research participants. Clinical research is meant to serve a social good, to en-
One is the traditional, Kantian notion of exploitation as hance the health and health care of people. It is part of
using an individual merely as a means and not simultan- the way people collectively improve their well-being.
eously as an end in itself.23,24 This Kantian conception Clinical research is not meant to be done to people but
of exploitation is grounded in the use of individuals for done with people." The principle of collaborative part-
an end they do not agree with or to which they have nership recognizes that the community in which re-
not consented. Using individuals without their consent search is conducted should collaborate in the research
violates their autonomy.25 The remedy for the Kantian endeavor.22,27 Seeking the community's agreement
type of exploitation is obtaining informed consent and and input helps ensure that the particular commun-
sometimes ensuring collaborative partnership with a ity will not be exploited.27 In addition, collaboration
larger community that agrees to the research. helps ensure—although it does not guarantee—that
A second conception of exploitation elaborated by the community will receive fair benefits from the
Alan Wertheimer rests on the unfair distribution of the conduct of the research.27." Collaborative partnership
benefits and burdens of an interaction.26,27 This is dis- helps ensure that the community determines for itself
tinct from the Kantian conception because it concerns whether the research is acceptable and responsive to
the distribution of benefits—who benefits and how its health problems. Finally, collaborative partnership
much they benefit—rather than autonomy. Import- is practically important. Without the engagement of
antly, this type of exploitation can occur even when the researchers and community members, research is un-
interacting parties provide valid consent.26 Minimizing likely to have any lasting impact. Without the invest-
this type of exploitation is moire complex, requiring the ment of health policy makers, the research results are
fulfillment of multiple principles.27 unlikely to influence policy making and the allocation
of scarce health-care resources.22
Principles and Benchmarks of Ethical Collaborative partnership can be fulfilled through
Clinical Research myriad formal and informal mechanisms. For in-
stance, establishment of community advisory boards,
The following eight ethical principles provide a com- consultations with advocacy groups, public meetings
prehensive and systematic framework to guide the with community members, and advocacy for fund-
ethical conduct of clinical research and thereby mini- ing of research are approaches to developing collab-
mize the possibility of exploitation21.22 (see Table 1). orative partnerships.3o.32 Which method is preferred
These principles are general and identify considera- depends upon the nature of the particular research
tions necessary to justify research as ethical. They study. Because many of these mechanisms exist in the
are conceptually included in most of the previously background without the need to launch explicit initia-
mentioned guidance, although existing guidelines do tives or are just part of "doing business," collaborative
Emanuel/Wendler/Grady: An Ethical Framework for Biomedical Research
partnership has infrequently been included as an ex- Social Value

plicit ethical requirement of clinical research?' One Clinical research is not an end in itself. It has instru-
example of research that fails on collaborative partner- mental value because it generates knowledge that leads
ship grounds includes "helicopter research" in which to improvement in health or health care.33,34 It is such
researchers arrive in a community, take samples, and improvements in health that ultimately constitute the
leave, never to return. social value of research. Unfortunately, the emphasis
Several benchmarks are essential to fulfilling the on protection of research participants has displaced
principle of collaborative partnership.22 First, collab- the importance of assessing research's social value.
orative partnership obviously requires partners. This Without social value, research exposes participants to
means identifying representatives of the target com- risks for no good reason and wastes resources.21•22•33-35
munity to be involved in the research. Second, it re- However, the process of translating research results
quires collaboration. This entails sharing responsibility into health improvements is complex, incremental,
for assessing the importance of the health problem and and haphazard.36 Typically, early studies are valuable
the value of the research to the community, as well as because the data they generate informs additional re-
for planning and conducting the study, disseminating search that ultimately could improve health. Priorities
the results, and ensuring that the results are used for may change while a study is being conducted, and the
health improvements. cooperation of diverse groups is often needed to make
Third, a collaborative partnership requires changes based on research results. This makes the
mutual respect. This entails recognition of and respect process of going from research to health improvement
for a community's distinctive values, circumstances, uncertain and arduous. Assessment of the value of
culture, and social practices.3° Importantly, respect research is made prospectively before any data are col-
does not mean uncritical acceptance of practices lected. Consequently, determinations of social value are
that might be oppressive or coercive. Indeed, some uncertain and probabilistic, entailing judgments about
of these practices may be challenged in research. A the usefulness of a sequence of research and chances
true collaborative partnership based on respect also of implementing the results.35,36 Even in wealthy coun-
aspires toward equality between the partners. In this tries with well-established research studies and health
sense, collaborative partnership aspires to minimize system infrastructures, research results are imperfectly
the deprived circumstances of the involved commun- incorporated into clinical practice.
ity. Research aims to ameliorate deprivations usually Certain kinds of research clearly lack social value:
of disease and sometimes of social circumstances. for example, research that is nongeneralizable, that ad-
This could occur through a number of interventions dresses a problem of little relevance to anyone, that will
directly related to the goals of the research project not enroll sufficient numbers of patients, that assesses
or ancillary mechanisms such as developing the gen- proven or empirically well-established results, and re-
eral infrastructure necessary to actually conducting search that could never be practically implemented to
ethical research. improve health or health care even if effective in the
Fourth, the community in which the research is research setting.37.38
being conducted should receive fair benefits from the Consideration of four benchmarks helps to ensure
conduct and/or results of the research.27,3' What level of fulfillment of the principle of social value.2',22 First, to
benefits is fair depends upon the burdens the commun- whom will the research be valuable? It is important
ity bears for the conduct of the research.26 Such benefits to delineate both the short-term and long-term pro-
might include direct benefits to the research partici- spective beneficiaries of the research study, specifying
pants as well as more indirect benefits such as employ- whether they include a specific group, similarly situ-
ment and training for community members to augment ated groups, a larger community from which research
health care services for the entire community.27•3' participants will be recruited, the country hosting the
Finally, collaborative partnership requires a fair research, or people outside the host country.22
distribution of the tangible and intangible rewards of Second, what is the potential value of the research
research among the partners. Very little can generate for each of the prospective beneficiaries? Potential
more resentment, mistrust, and sense of exploitation beneficiaries may rank the health problem's importance
than an unfair distribution of the benefits of collab- differently and may receive different benefits from the
oration. This may require agreements regarding sharing research results. Factors to be considered might include
intellectual property rights, royalties, and other sources how widespread the disease or condition is, the impact
of financial profit as well as appropriate authorship and of the disease on individuals and communities, and
other credit for contributions to the research.27,3' the extent to which the research is likely to offer an
Table 1 Principles and Benchmarks for Ethical Clinical Research

Principles Benchmarks
Collaborative partnership • Which community representatives will be partners, involved in helping to plan and conduct the
research, disseminate the results and use the results to improve health?
• How will responsibility be shared with these partnersfor planning and conducting the research,
disseminating the results and using the results to improve health?
• How will respect for the community's values, circumstances, culture, social practices, and so forth,
be demonstrated?
• How will fair benefits for the community from the conduct and results of the research be assured?
• How will the tangible benefits of the research, such as authorship credit and intellectual property
rights, be distributed to ensure fairness?
Social value • Who will benefit from the conduct and results of research?
• What is the potential value of the research for each of the prospective beneficiaries?
• How will the social value of the research be enhanced?
• How can adverse impacts, if any, of conducting the research be minimized?
Scientific validity • Do the scientific and statistical design and methods satisfy generally accepted standards and
achieve the objectives of the study? If not, is there clear justification for the deviations?
• Will the research results be interpretable and useful in the context of the health problem?
• Does the study design ensure participants health-care services they are entitled to? If not, are
there methodologically compelling reasons and are participants protected from serious harm?
• Is the research design practically feasible given the social, political, economic, and cultural
environment?
Fair participant selection • Is the research population selected to ensure that the research complies with scientific norms and
will generate valid and reliable data?
• Is the research population selected to minimize risks to the participants?
• Are the individual research participants selected to maximize social value and enhance the
possibility of benefits to the participants?
• Are the participants vulnerable based on age, clinical status, social marginalization, economic
deprivation, and so forth? If so, what safeguards are included to protect the participants?
Favorable risk-benefit ratio • Are the potential physical, psychological, social, and economic risks of the research for the
individual participants delineated and their probability and magnitude quantified to the extent
possible given the available data?
• Are the potential physical, psychological, social, and economic benefits of the research for the
individual participants delineated and their probability and magnitude quantified to the extent
possible given the available data?
• When compared, do the potential benefits to the individual participants outweigh the risks?
If not, does the knowledge gained from the study for society justify the net risks to the
individual participants?
Independent review • Are the procedures for independent review established by law and regulation being
properly followed?
• Is the review body both independent and competent?
• Is the review process transparent, and are reasons given for the review committee's decisions?
• Are multiple reviews minimized and reconciled if they conflict?
Informed consent • Are recruitment procedures and incentives consistent with cultural, political and social practices of
the potential participants and their community?
• Are disclosure forms and verbal disclosure procedures sensitive to participants' culture, language,
and context?
• Is the information presented to participants complete, accurate, and not overwhelming?
• Are there appropriate plans in place for obtaining permission from legally authorized
representatives for individuals unable to consent for themselves?
• Are supplementary consents or permissions, for example, from spouses or community leaders,
obtained? If so, are there ways to ensure that the individual participant can still decide whether to
participate independent of the spouse or community leader?
• Are the mechanisms to symbolize consent consistent with participants' culture and context?
• How will individual participants be made aware of their right to refuse to participate and are they
actually be free to refuse?
Respect for participants • How will the health and well-being of participants be monitored to minimize harms? Are the criteria
for changing doses or procedures for stopping the study for the health of participants adequate?
• How will the confidentiality procedures actually be implemented?
• How will it be ensured that participants who want to withdraw can withdraw without penalty?
• How will results of the research be disseminated?
• What are the plans for care of the participants after the research is completed?
intervention or information useful to the beneficiaries. statistical design and methods of the research must
For example, because malaria is a substantially greater plausibly realize the objectives of the research and must
health problem for certain developing countries than for also satisfy the generally accepted norms of research.
developed countries, research on cerebral malaria may Research must have clear, justifiable objectives, an ad-
be of substantial value to people in developing coun- equate sample size, and unbiased and reliable outcome
tries. Conversely, research on prophylactic medications measures and statistical analyses. Deviations from such
for malaria is likely to be more valuable for tourists, standards, such as innovative designs, must be plaus-
whereas research on a malaria vaccine may be perceived ibly justifiable to the research community.
as valuable to everyone, but to a different degree. Simi- Second, a research study must be designed to
larly, research on new HIV/AIDS medications in a devel- generate results that will be interpretable and useful
oping country, although needed in that country, could in the context of the health problem.15 Interventions
benefit those outside the host country more than the should be selected to ensure that the design is useful
community in which the research is being conducted if in identifying ineffective or appropriate interventions;
the ultimate cost of the medication is high. implementing socially, culturally, and economically
Third, it is important to develop mechanisms to appropriate changes in the health-care system; or pro-
enhance the social value of research. Through collab- viding a reliable foundation for conducting subsequent
orative partnerships, strategies should be devised to research. Interventions should be selected to ensure
disseminate results in appropriate ways to key stake- that the design will realize social value and that the data
holders including people with the disease, practicing are generalizable.21,2241
clinicians, advocacy groups, health policy makers, and Third, the study design must realize the research
sometimes international health-care organizations.22,3° objectives while neither denying health-care services
In addition to presentations at scientific conferences that participants are otherwise entitled to nor requir-
and journal publications, this may require novel forms ing services that are not feasible to deliver in the con-
of dissemination such as letters to patients, articles text.37,38,42 However, studies can be ethically designed
in advocacy publications, presentations at commun- yet not provide a service or intervention individuals are
ity gatherings, public service announcements in the entitled to under certain, restrictive conditions.41,43- 45
media, or letters to clinicians. Social value can also be Specifically, it is ethical to use placebo or less than the
enhanced when research is integrated into a long-term diagnostic tests or treatments to which individuals are
collaborative strategy, so that one research project forms entitled when two conditions are fulfilled: (1) there is a
part of a more comprehensive research and health de- methodologically compelling reason to do so, and (2)
livery strategy to address significant health problems." there is only minimal chance of serious harm—such as
Finally, consideration should be given to the suffering irreversible morbidity or disability, or revers-
impact of the research on the existing health-care ible but serious injury.41,43-45
infrastructure. The conduct of the research should not Determining entitlement to medical services in
undermine a community's existing health-care services studies is challenging because entitlements differ among
or social structures and leave it worse off at the end of countries, and may differ among groups within a coun-
the research. Supplementing the existing system and try.46,47 Even in wealthy countries, participants are not
contributing to sustainable improvements in health entitled to every available or effective medical service,
through the provision of additional resources, equip- because justice necessitates establishing priorities for
ment, medications, or training appropriate to the re- the distribution of scarce resources.460 For instance,
search can enhance value. some developed countries may not guarantee expen-
sive drugs when inexpensive but more inconvenient
Scientific Validity yet effective drugs are available. Similarly, it is widely
Contrary to many claims, in research, science and accepted that cardiac research conducted in developing
ethics do not conflict.21,22,34 Valid science is a funda- countries need not be designed to require a coronary
mental ethical requirement.21.22,35 Unless research gen- care unit because participants would not necessarily
erates reliable and valid data that can be interpreted be entitled to this service under a just distribution of
and used by the specified beneficiaries of the research, scarce resources in those countries.37,38,42,46,49 Con-
it will have no social value and participants may be versely, in a study evaluating interventions to reduce
exposed to risks for no benefits.39,4° Research must be mortality from cerebral malaria conducted in rural set-
designed in a way that provides valid and reliable data. tings in which travel to hospitals is impracticable, pro-
Four benchmarks are important in fulfilling the vision of bed nets may be part of a valid design even if
principle of scientific validity. First, the scientific and participants may not otherwise have them.5° However,
II 111111
even if the study's objective is deemed socially valuable, infections are important considerations to minimize
especially to the enrolled participants' community, it is risk Similarly, individuals with high creatinine clear-
not ethically necessary to provide more comprehensive ance may be appropriately excluded from a trial of a
interventions beyond those to which participants are potentially renal toxic drug in order to reduce risk.
entitled, especially interventions that may not be feas- Third, individuals should be selected in order
ible and sustainable. Doing so may even be unethical if to enhance both the social value of the research and
it undermines the scientific objectives or makes the re- the possibility of benefits to participants.22'55-57 For
sults irrelevant to the enrolled participants' community. example, assuring an adequate number of women in
Finally, the study must be designed in a way that a study of a disease largely affecting women enhances
is practically feasible given the social, political, and benefits to women. Selecting individuals who are able
cultural environment in which it is being conducted.' to comply with the study requirements will enhance
Ensuring feasibility might require extensive com- the chances that they will benefit from the intervention
munity education and outreach as well as sustainable and that the study will yield valid data. Communities
improvements to the health-care infrastructure, such as should be selected in which a collaborative partnership
training of personnel, construction of additional facili- can be developed and in which social value can be real-
ties, or provision of an affordable drug. Feasibility also ized. Consequently, it is preferable to select commun-
requires that it be possible tc achieve the sample size in ities that have, or can establish, a system for identifying
a reasonable time frame. legitimate representatives and that will share respon-
sibility for planning and conducting the study and
Fair Participant Selection ensuring that results are implemented through health
Historically, populations that were poor, uneducated, system improvements or additional research.
or powerless to defend their own interests were targeted Finally, factors such as cognitive ability, age,
for high-risk research, whereas promising research was clinical status, familial relationships, social marginal-
offered to more privileged individuals.15,34,52 Fair selec- ization, political powerlessness, and economic dep-
tion of participants requires that the research objectives rivation should be considered in order to determine the
be the primary basis for determining eligibility.4,15,21,22,34 vulnerability of individuals or groups.58 For instance, if
Once a target group is identified based on scientific ob- health policy makers suggest a particular group for re-
jectives, considerations of minimizing risk, enhancing search participation, the researchers should determine
benefits, minimizing vulnerability, feasibility, as well whether the group has been selected for good reasons,
as facilitating collaborative partnership, become deter- such as a high incidence of disease, or because of social
minative.22 Factors extraneous to the objectives, risks, subjugation. If scientifically appropriate individuals or
benefits, and feasibility of conducting the research groups are identified as vulnerable, specific safeguards
should not be the basis for selecting target commun- to protect the population should be implemented,
ities or excluding individuals or communities.4,15,22,34 such as consent monitoring or independent capacity
Four benchmarks are necessary to fulfill the assessment, independent clinical monitoring, ensuring
principle of fair participant selection. First, the study confidentiality, and ensuring that potential research
population should be selected to ensure valid sci- participants are free to decline joining the study.
ence .21,22,34,53 Scientific reasons for choosing a particu-
lar group of individuals or a community might be high Favorable Risk-Benefit Ratio
prevalence or incidence of a disease, the magnitude of Like life itself, all research entails some risks. How-
harms caused by the disease, high transmission rates of ever, clinical research typically should offer individual
an infection, special drug resistance patterns, deprived participants a favorable net risk-benefit ratio 21,22,34 In
social circumstances that increase susceptibility to a cases in which potential risks outweigh benefits to in-
disease, or particular combinations of diseases. Social dividual participants, the social value of the study must
status that is irrelevant to the research objectives should be sufficient to justify these net risks.4'59 Because clin-
not influence selection. Scientific considerations alone, ical research involves drugs, devices, and procedures
however, will usually underdetermine which commun- about which there is limited knowledge, uncertainty
ity or individuals are selected. about the degree of risks and benefits is inherent. And
Second, selecting participants in a way that mini- the uncertainty is greater in early phase research.
mizes risk is essentia1.54 For instance, in selecting a target The principle of a favorable net risk-benefit ratio
population for an Hiv vaccine study, the extent to which requires fulfilling three benchmarks. First, the risks
a community protects Hiv-infected persons against dis- of the research should be delineated and minimized.
crimination and provides treatment for opportunistic Researchers should identify the type, probability, and
Emanuel/Wendler/Grady: An Ethical Framework for Biomedical Research 501
magnitude of the risks of the research. The risks are not into account the context in which the participants live
limited to physical risks, but should also encompass po- and the risks they actually face. The underlying risks
tential psychological, social, and economic risks. To the of a particular disease can vary because of differences
extent possible, the assessment of risks should be based in incidence, drug resistance, genetic susceptibility,
on available empirical data, not intuition or specula- or social or environmental factors. When participants
tion. Within the context of good clinical practice, these confront a higher risk of disease, riskier research may
risks should be minimized "by using procedures which be justifiable.61 Similarly, the net risk-benefit ratio for
are consistent with sound research design and which a particular study may be favorable in communities in
do not unnecessarily expose subjects to risk, and when- which the social value of the research is high, yet may
ever appropriate, by using procedures already being be unfavorable in communities in which the potential
performed on the subjects for diagnostic or treatment value is lower.
purposes."5 In addition, research procedures should be When potential benefits to participants from the
performed by trained and competent individuals who research are proportional to the risks they face, then
adhere to the standards of clinical practice.3 the additional social value of the research, assured by
Second, the type, probability, and magnitude of the fulfillment of the value and validity requirements,
the benefits of the research should be identified. The implies that the cumulative benefits of the research
benefits to individual participants, such as health outweigh its net risks.
improvements, are relevant. The specification of poten- The notions of "proportionality" and potential
tial benefits to individual participants should consider benefits "outweighing" risks are metaphorical!' Yet
only health-related potential benefits derived from the the absence of a mathematical formula to determine
research intervention itself.21,22,34 The benefits to society when the balance of risks and potential benefits is
through the generation of knowledge are assumed if proportionate does not connote that such judgments
the research is deemed to be of social value and sci- are inherently haphazard or subjective. Instead, as-
entifically valid. Secondary benefits, such as payment, sessments of risks and potential benefits to the same
or adjunct medical services, such as the possibility of individuals can appeal to explicit standards, informed
receiving a hepatitis vaccine not related to the research, by existing data on the potential types of harms and
should not be considered in the risk-benefit evaluation; benefits, their likelihood of occurring, and their long-
otherwise simply increasing payment or adding more term consequences.4 Evaluations of the quality of
unrelated services could allow the benefits to justify books are not quantifiable either, but neither are they
even the riskiest research.22,6° Furthermore, although merely matters of subjective taste; comparing the qual-
participants in clinical research often receive some ity of Shakespeare or Dostoevsky with Danielle Steel
health services and benefits, the purpose of clinical entails judgments based on shared standards that can
research is not the provision of health services. Servi- be justified to others. Similarly people routinely make
ces directly related to clinical research are necessary to discursively justifiable intrapersonal comparisons of
ensure scientific validity and to protect the well-being risks and benefits for themselves, and even for others,
of the individual participants. such as children, friends, and employees without the
As a matter of general beneficence, consideration aid of mathematical formulae.62
should be given to enhancing benefits to participants Finally a more complex evaluation is necessary
and their community, especially when such benefits when clinical research presents no or few potential
can be provided easily and will not compromise the benefits to individual participants, such as in Phase I
scientific validity of the study. However, such enhance- safety and pharmacokinetic studies, and even in some
ments of benefits are not to be considered in the as- epidemiology research, or when the risks outweigh the
sessment of the risk-benefit ratio—or even of the social potential benefits to individual participants. In this
value—of the research study itself. case, a more complex evaluation, what Charles Weijer
Third, the risks and potential benefits of the clin- calls a "'risk-knowledge' calculus," is necessary.'
ical research interventions to individual participants This calculus assesses whether the societal benefits in
should be compared. In general, the more likely and/ terms of knowledge gained justify the "excess" risks
or more severe the potential risks, the greater in likeli- to individual participants.63 Determining when po-
hood and/or magnitude the prospective benefits must tential social benefits outweigh net risks to individual
be; conversely, research entailing potential risks that are participants requires interpersonal comparisons that
less likely and/or of lower severity can have more uncer- are conceptually and practically more difficult than
tain and/or circumscribed potential benefits. Import- intrapersonal comparisons.62 However, policy makers
antly, this comparison of risks and benefits should take are often required to make these kinds of comparisons,
II :ill 111111
for example, when considering whether pollution members of society can have confidence that they will
and its attendant harms to some people are worth the not benefit from the exploitation of other humans.
potential benefits of higher employment and tax rev- Four benchmarks help in fulfilling this principle.
enues to others. There is no settled framework for how First, procedures established by law and regulation
potential social benefits shculd be "balanced" against should be followed. Research has not revealed the
individual risks. Indeed, the appeal to a utilitarian ap- best mechanism to conduct independent review.72
proach of maximization, as in cost-benefit analysis, is Consequently, the actual review mechanisms are usu-
quite controversial both morally and because many ally determined by laws and regulations that vary both
risks and benefits of research are not readily quantifi- internationally and locally. For instance, some countries
able on commensurable scalzs.64-66 Nevertheless, these and institutions separate scientific and ethical review,
comparisons are made,67 and regulations mandate that whereas others integrate scientific and ethical assess-
investigators and research review committees make ments into a single review. Similarly, some countries
them with respect to clinical research.4'5 When research have ethics review committees composed only of lay-
risks exceed the combination of potential medical persons, whereas others have committees dominated
benefits to individuals and the benefit of useful know- by medical scientists and physicians. Nevertheless,
ledge to society, clinical research is not justifiable. prevailing laws and regulations establish the standards
that should be followed for independent review They
Independent Review should be amended as better processes are identified.
Second, whatever the process, the review must
Independent ethical review of all clinical research be independent and competent. Members of the
protocols is necessary for two reasons: (1) to minimize review committees must be free of any conflicts with
concerns regarding researchers' conflicts of interest and the researchers or the research study. The reviewers
(2) to ensure public accountability.21'22 Investigators in- should not be collaborators on the research or with
herently have multiple, legitimate interests—interests the researchers, and should not have any financial in-
to conduct high quality research, to complete the re- terests in the outcomes of the study. Similarly, review-
search expeditiously, to protect research participants, ers should be excluded from the review if they have
to obtain funding and advance their careers, and so other conflicting interests, such as responsibility for
forth.'" Even for well-intentioned investigators, these the financial interests of the institution in which the
diverse interests can generate conflicts that may unwit- research is conducted, that might preclude them from
tingly distort or undermine their judgments regarding evaluating the protocols according to ethical principles
the design, conduct, and analysis of research, as well and without bias. Similarly, the reviewers should have
as adherence to ethical requirements.190-7° Wanting sufficient expertise—or be able to access advice—in
to complete a study quickly may lead to the use of the scientific, clinical, and statistical areas necessary to
questionable scientific methods or to the use of read- assess the research protocol. Training in research ethics
ily available participants rather than fairer participant for the reviewers may be necessary.
selection criteria; enthusiasm for and commitment to Third, the review should be transparent. This is
the research project may lead to overemphasis of po- especially important in multinational research in which
tential benefits and under emphasis of potential harms differences in culture, practices, and understandings
to participants. Independent review by individuals may yield different judgments. One fundamental aspect
unaffiliated with the clinical research study helps to of transparency is that the reasons for decisions of the in-
minimize the potential impact of such conflicts of independent review committee are explained. This allows
terest.21,22,34,71 In this way, in dependent reviewers can observers to assess whether the reasons are appropriate
assure potential research participants that the study and relevant considerations have been addressed.
they are considering is ethical—that is, it will generate Finally given the increasing complexity of research,
socially valuable information and the risk-benefit ratio multiple independent reviews frequently occur 73-75
is favorable. Multiple independent reviews may seem to be required
Independent review of clinical research is also by law or regulation for multisite studies or studies
important for a second, less emphasized, reason: social conducted by investigators from multiple institutions.
accountability.21 Clinical research imposes risks on Importantly, however, the ethical principle of independ-
participants for the benefit of society. An independent ent review does not require multiple reviews.76 The
review of a study's compliance with ethical requirements only requirement is that the reviewers competently and
assures members of society that people who enroll in independently assess relevant scientific and ethical con-
trials will be treated ethically. Based on this review, siderations. Indeed, multiple reviews may have no added
EmanueVWendler/Grady: An Ethical Framework for Biomedical Research
value or may even be counterproductive, by taking time level of compensation depends upon the local economic
and requiring adjudication without added protections.72 and social context.°7 Although concerns about undue
Such situations are unethical—resources are expended inducement are frequently raised,45,84 high potential
that produce no value or even waste value.40 social value and a favorable risk-benefit ratio—implying
If there is disagreement among such reviews, it is minimal net risks to the participants—dispel these
important to clarify its nature. Disagreement may re- concems.88-91 Indeed, worry about undue inducement
flect different ways of balancing various principles and could reduce compensation and some other benefits for
benchmarks, or the appropriateness of different ways participants and host communities. Paradoxically, bal-
of fulfilling them. That is, disagreement might reflect ancing fair compensation and undue inducement may
how the ethical principles are met, rather than whether result in less compensation for members of impoverished
they are met .77 Conflicts may also arise because of dif- communities and raise the specter of exploitation.26.88
ferent guidelines or regulatory requirements, which Second, both written and verbal disclosure of infor-
themselves may not have good ethical justification mation should be sensitive to participants' culture and
or may be insensitive to particular cultural or social context. Disclosures should use the language, culturally
circumstances.78 Only rarely are there fundamental appropriate idioms, and analogies of the prospective
disagreements about whether ethical principles and participants at a level they can understand. This entails
benchmarks are fulfilled. Unfortunately, there is no a need for collaborative partnership. After disclosure,
widely accepted procedure for adjudicating such con- investigators should feel confident that participants
flicts. In practice, the requirements specified by the understand the information and are consenting with-
sponsor's review board are often determinative. This out any pressure or major misconceptions. In some
contravenes the principle of collaborative partnership cases, a formal assessment of understanding, monitor-
and the notion that the community that assumes the ing of the consent process, or independent assessment
risks of the research should make the assessment about of participants' capacity to consent may be warranted.92
the research protocol." Third, the disclosure of information relevant to the
research study must be complete and accurate, but not
Informed Consent overwhelming. Providing less than complete and accur-
No requirement has received as much explica- ate information raises concerns about potential decep-
tion as informed consent. The purpose of informed tion of participants. However, complete information
consent is to show respect for the autonomy of does not imply lengthy or exhaustive disclosure forms
individuals.4,6,15,25,34,80-85 To enroll individuals in clin- detailing every aspect of the research study, which may be
ical research without their authorization is to treat them overwhelming to the participants. Indeed, shorter, more
merely as a means to purposes and ends they may not focused forms, without repetition and boilerplate disclo-
endorse or even know about, denying them the op- sures, may be more effective.93 Disclosure forms must
portunity to choose what projects they will pursue and balance completeness with not being overwhelming.
subjecting them to Kantian-type exploitation.23-25 By Fourth, some research entails enrollment of
allowing individuals to decide if—and how—they con- individuals unable to consent because of their age,
tribute to research, informed consent respects persons permanent mental incapacity, an acute loss of mental
and their autonomy.4,25 functions, or other reasons. In these cases, researchers
Valid informed consent requires that the consenting must have a strategy for obtaining permission from
person has the capacity to understand and make deci- legally authorized representatives of the potential
sions, receives relevant information about the research participants.15,83,84,94-99
study, understands that information, and consents In some cases, "spheres of consent" ranging from
voluntarily and without coercion.4,15,25,34,80-84 Each of spouses to heads of households to school principals to vil-
these elements is necessary to ensure that individuals lage elders or community leaders may be required before
make rational and free determinations of whether the researchers can invite individual participation.30.100,'0i
research trial is consonant with their interests.86 With a few exceptions, such as emergency research, it
Seven benchmarks are necessary to fulfill the prin- is unacceptable to supplant individual consent of com-
ciple of informed consent. First, recruitment procedures petent adults by family or community consent.1°2 The
and incentives for participants should be consistent with family or community gives permission only to approach
cultural, political and social practices of the potential partici- individuals. When family or community permission to
pants. In some communities, compensation for partici- approach individuals is reasonable, special care should
pation in research may be expected, whereas in others, it be given to assure that the individual can still refuse
may be considered offensive. The appropriate form and participation—that is, that there is no coercion.
II JII 11111.1
Sixth, researchers should utilize consent pro- Third, respect includes permitting participants to
cedures that are acceptable within the local context, change their minds, to decide that the research does
while ensuring that an independent observer could not comport with their interests or preferences, and to
verify voluntary participation by the individuals. For withdraw without penalty. Fourth, as new information
instance, US regulations require a written signature.5 about the impact of the intervention or about the par-
In many cases, this is an acceptable and efficient way ticipant's clinical condition is gained during the course
to document consent authorization. However, in some of the research, respect requires providing this new in-
cases, because of limited literacy or cultural differ- formation to the participants. Researchers should also
ences, such requirements may be inappropriate and develop explicit strategies to inform participants and
unethical." Alternative methods to express consent, host communities of the results of the research. Having
such as handshakes, embracing, or sharing a meal, participated in research and assumed risks, the partici-
are known." Appropriate alternative procedures for pants and host communities have a right to know what
documenting informed consent might include tape re- was found and its implications for public health and
cordings or witnessed written documentation of these health-care policies.
methods of consent. Finally, plans should be made regarding the care of
Finally, special attention must be given to ensure participants when the trial is over. In some cases, this
that individuals are aware of their right to, and are may simply involve referral to a primary care provider.
actually free to, refuse to participate or to withdraw In other cases, this may require researchers to find
from research. A key element of informed consent is creative strategies for providing access to treatments
the ability to refuse or withdraw participation without benefiting the participants, even when these interven-
penalty.m3 Prorating offered compensation and other tions are unlicensed.
research-related benefits may help to obviate possible
familial or community coercion or retribution. Characteristics of the Principles
Respect for Participant; The eight general principles and the benchmarks de-
lineate a systematic and comprehensive way of assess-
The ethical conduct of clinical research does not end ing the ethics of particular clinical research.21,22 They
when informed consent is obtained.21,22,104 Research- provide a coherent and organized way for researchers,
ers have ongoing obligations to treat individuals with ethics reviewers, participants, and others to evaluate a
respect from the time they are approached—even if research protocol and to determine whether it fulfills
they refuse enrollment—throughout their partici- ethical standards. They should not be seen as adding
pation and even after their participation ends. Re- ethical requirements, but rather distilling and coher-
specting potential and enrolled participants entails ently articulating the ethical norms underlying much of
multiple activities. First, and arguably most import- the prevailing guidance. These principles and bench-
ant, this principle requires monitoring the health and marks offer a more organized and systematic delinea-
well-being of participants, and intervening to prevent tion of what many researchers, ethics reviewers, and
or treat harms that might result from the adverse others already do.
reactions, untoward events, or changes in clinical Importantly, these principles are not independent
status associated with the research.10" In some cases, of all other ethical principles. They operate within and
research studies need to include procedures to adjust presume compliance with more general moral norms,
drug doses and even withdraw study participants such as honesty and promise keeping.22 Similarly, these
because of adverse events. Furthermore, specific principles focus on what is required to evaluate re-
stopping rules may be necessary if excessive adverse search studies, not on the enforcement or proper con-
events or benefits are identi:ied. duct of the research itself. Having ethical researchers
Second, pledges of confidentiality should be hon- is important for implementation of the framework but
ored and procedures to protect confidentiality imple- not a requirement for evaluating the research protocol.
mented. Such procedures ir.clude securing databases, Determining what is ethical and what needs to be en-
locking file cabinets containing data, coding specimens forced must be done prior to and should not be con-
and data forms, as well as interviewing participants fused with how to implement an ethical protocol or to
in private spaces where they cannot be overheard. In enforce the requirements.21,22
addition, it is important to alert participants that de- These eight principles are necessary. The presump-
spite researchers' best efforts, absolute confidentiality tion is that they must all be fulfilled for a research proto-
cannot be guaranteed. col to be ethical. There is no picking and choosing.
EmanueVWendler/Grady: An Ethical Framework for Biomedical Research 505
However, in specific cases, such as emergency research, are independent of reliable facts and plausible
informed consent may be legitimately waived. These assumptions about particular societies. The argu-
principles are justified by ethical values that are widely ments begin from where we are, and appeal to
recognized and accepted, that reasonable people would those with whom we now live. This is why moral
want to be treated in accordance with—avoidance relativism is seldom as important an issue in prac-
of exploitation, the just distribution of benefits and tical as it is in theoretical ethics.107
burdens, beneficence, respect for persons, and so
forth.105,106 These requirements are precisely the types Importantly, that there are eight principles sug-
of considerations that would be invoked to justify gests that the ethics of research is complex. Adher-
clinical research if it were challenged. The benchmarks ence to a single ethical principle rarely provides a
provide more practical considerations for discerning complete solution; most situations implicate multiple
satisfaction of the general principles. principles.48,62,64.105,107-110 Consequently, the various
The principles are sufficient. Fulfilling these eight principles and benchmarks will sometimes conflict.
principles means the research is ethical. Failing on any What is fair participant selection could at times increase
one principle—except for waiving informed consent risk; what is required for informed consent may some-
in specific cases, in which waiving consent must be times compromise scientific validity. Unfortunately,
justified—makes the research unethical. The proposed there is no simple algorithm for determining how to
benchmarks, however, may not be sufficient, and may balance or weigh these principles when they conflict.
need revision with experience and time. They certainly Different researchers and communities will balance
provide a useful first estimation of the kind of specific the principles in different ways, some emphasizing in-
elements that need to be fulfilled. formed consent, others the importance of minimizing
These eight principles are universal; they apply in risks or enhancing social value. Ignoring or rejecting
all countries and contexts, regardless of sponsorship. basic principles in designing or conducting a research
The principles are general statements of value; they study could render it unethical. Conversely, accepting
must be elaborated by traditions of interpretation and the principles and benchmarks, yet disagreeing about
require practical interpretation and specification. The how to balance them in a particular case, highlights
benchmarks offer a first level of specification, indicat- the intricacies of ethical judgments entailing multiple
ing how to fulfill these principles. However, the details considerations. Disagreement on the balancing of the
of this specification will inherently be context and cul- various benchmarks does not necessarily make one
ture dependent. This does not make them relativistic or assessment ethical and the other unethical. Rather, it
less universal. It simply recognizes that applying ethical reflects different, but perhaps legitimate, ways of re-
principles in the world requires taking facts into ac- solving competing ethical claims.m In fact, this frame-
count, and these facts depend upon the context. work can help narrow disagreements and elucidate the
different underlying views. When conflicts between
Moral arguments take place in context, and they principles and benchmarks occur, or when different
therefore depend at least implicitly on matters groups weigh the principles differently the important
of fact, estimates of risk, suppositions about point is to be clear about the reasons for the evaluation
feasibility, and beliefs about human nature and and the differences. Ultimately, a thoughtful process of
social processes. . . . Even those who rely on balancing ethical considerations can be as important
what they regard as universal moral principles as any particular judgment in the effort to ensure that
do not presume that their practical conclusions research is conducted ethically.
References 3. World Medical Association. Declaration of Helsinki: Eth-

ical Principles for Medical Research Involving Human Sub-
1. S. International Guidelines on Bioethics. Geneva,
Fluss, jects. Tokyo, Japan: WMA; October 2004. [Online] 2004.
Switzerland: European Forum on Good Clinical Prac- Available: http://wwwwma.net/e/policy/b3.htm.
tim/cioms, 1998. 4. The National Commission for the Protection of
2. The Nuremberg Code. In Trials of War Criminals Before Human Subjects of Biomedical and Behavioral Re-
the Nuremberg Military Tribunals Under Control Council search. The Belmont Report: Ethical Principles and
Law No. 10. Volume 2. Washington, DC: US Government Guidelines for the Protection of Human Subjects of Re-
Printing Office; 1949:181-2. [Online] Available: httpl/ search. Washington, DC: Department of Health,
ohsr. od.nih.gov/guidelines/nuremberg.html. Education and Welfare; DHEW Publication OS
11 11 111111
78-0012 1978. [Online: April 18, 1979. Available: 17. Council for International Organizations of Medical Sci-
http i/www. hhs gov/oht p/hum ansu bj e c ts/g uid a nce/ ences, in collaboration with the World Health Organ-
belmont.htm. ization. International Ethical Guidelines for Biomedical
5. Department of Health and Human Services, National Research Involving Human Subjects. Geneva, Switzerland:
Institutes of Health, and Office for Human Research CIOMS and WHO, 1993.
Protections. The Common Rule, Title 45 (Public Welfare), 18. Forster, H.P., Emanuel, E., Grady, C. The 2000 revision
Code of Federal Regulations, Part 46 (Protection of Human of the Declaration of Helsinki: A step forward or more
Subjects). [Online) June 23, 2005. Available: http://www confusion. Lancet 2001, 358: 1449-53.
.hhs.gov/ohrp/humansubjects/guidance/45cfr46.htm. 19. Thompson D.F. Understanding financial conflicts of in-
6. Council for International Organizations of Medical Sci- terest. New England Journal of Medicine 1993, 329: 573-6.
ences, in collaboration with the World Health Organiza- 20. American Society of Clinical Oncology. Revised Conflict
tion. International Ethical Guidelines for Biomedical Research of Interest Policy (adopted on November 7, 2002, by
Involving Human Subjects. Geneva, Switzerland: CIOMS the American Society of Clinical Oncology). Journal of
and WHO; 2002. [Online) November 2002. Available: Clinical Oncology 2003, 21: 2394-6.
http://wwwcioms.ch/framc_guidelines_nov_2002.htm. 21. Emanuel EJ., Wendler D., Grady C. What makes clin-
7. Council of Europe, Directorate of Legal Affairs Convention for ical research ethical? JAMA 2000, 283: 2701-11.
the Protection of Human Rights and Dignity of the Human Being 22. Emanuel E.J., Wendler D., Killen J., Grady C. What
With Regard to the Applicatior of Biology and Medicine: Conven- makes clinical research in developing countries ethical?
tion on Human Rights and Biomedicine. Strasbourg, France: The benchmarks of ethical research. Journal of Infectious
Council of Europe; 1997. [Online) April 4, 1997. Available: Diseases 2004, 189: 930-7.
httpl/conventions.coe.int/tnaty/en/treaties/html/164.htm. 23. Wood A.W. Exploitation. Social Philosophy and Policy
8. InternationalConf erence on Harmonisation of Technical Re- 1995, 12: 135-58.
quirements for Registration of Pharmaceuticals for Human 24. Buchanan A. Ethics, Efficiency and the Market. Lanham,
Use. The ICH Harmonised Tripartite Guideline-Guideline for Md.: Rowman & Littlefield, 1985.
Good Clinical Practice. Geneva: ICH; 1996. [Online) Avail- 25. Beauchamp T.L., Childress J. The Principles of Biomedical
able: http//wwwichorg/LOB/media/MEDIA482.pdf. Ethics. 5th edn. New York: Oxford University Press, 1996.
9. UK Medical Research Council. Guidelinesfor Good Clinical 26. Wertheimer A. Exploitation. Princeton, NJ: Princeton
Practice in Clinical Trials. London, England: MRC; 1998. University Press, 1996.
[Online) Available: http://www.mrc.ac.uk/pdf-ctg.pdf. 27. Participants in the 2001 Conference on Ethical Aspects
10. Uganda National Council of Science and Technology (uNCST). of Research in Developing Countries. Moral standards for
Guidelines for The Conduct of .qealth Research Involving Human research in developing countries: From "reasonable avail-
Subjects in Uganda. Kampala, Uganda UNCST, 1998. ability" to "fair benefits." Hastings Center Report 2004,
11. Canadian Institutes of Health Research, the Nat- 34(3): 17-27.
ural Sciences and Engineering Research Council, 28. Daniels, N., Bryant, J., Castano, R.A., et al. Benchmarks
and the Social Sciences and Humanities Research of fairness for health care reform a policy tool for devel-
Council. Tri-Council Policy Statement: Ethical Con- oping countries. Bulletin of the World Health Organization
duct for Research Involving Humans. [Online) 1998 2000, 78: 40-50.
(with 2000, 2002 and 2005 amendments). Available: 29. Daniels, N., Light D.W., Caplan R.L. Benchmarks of Fair-
http://www.pre.ethics.gc.ca/english/policystatement/ ness for Health Care Reform. New York: Oxford Univer-
policystatement.cfm. sity Press, 1996.
12. National Health and Medical Research Council, Com- 30. Weijer, C., Emanuel, E.J. Protecting communities in bio-
monwealth of Australia. National Statement on Ethical medical research. Science 2000, 289: 1142-4.
Conduct in Research Involving Humans. Canberra: NHMRC, 31. Participants in the 2001 Conference on Ethical Aspects
1999. [Online) Available: http://www.nhmrc.gov.au/ of Research in Developing Countries Fair benefits for re-
publications/ files/e35.pdf search in developing countries. Science 2002, 298: 2133-4.
13. Department of Health and Human Services, Office for 32. Diallo, D.A., Doumbo, O.K., Plowe, C.V., et al. Com-
Human Research Protecticns. International Compilation munity permission for research in developing countries.
of Human Subject Research Protections. 2nd edn [Online) Clinical Infectious Diseases 2005, 41: 255-9.
October 1, 2005. Available: http://www.hhs.gov/ohrp/ 33. Vanderpool, H.Y., ed. The Ethics of Research Involving
international/HSPCompilation.pdf. Human Subjects: Facing the 21st Century. Frederick, Md.:
14. Levine, C. Has AIDS changel the ethics of human subjects University Publishing Group, 1996.
research? Law Medicine and Health Care 1988, 16:167-73. 34. Levine, R.J. Ethics and Regulation of Clinical Research, 2nd
15. Advisory Committee on Human Radiation Experiments edn. New Haven, CT: Yale University Press, 1988.
Final Report of the Advisory Committee on Human Radiation 35. Freedman, B. Scientific value and validity as ethical re-
Experiments. New York, N.Y.: Oxford University Press 1996. quirements for research. IRB: A Review of Human Subjects
16. Macklin, R. After Helsinki: Unresolved issues in inter- Research 1987, 9(6): 7-10.
national research. Kennedy Institute of Ethics Journal 36. Black, N. Evidence based policy: Proceed with care. Brit-
2001, 11:17-36. ish Medical Journal 2001, 323: 275-8.
37. Grady, C Science in the service of healing. Hastings 57. Weijer, C. Evolving issues in the selection of subjects
Center Report 1998, 28(6): 34-8. for clinical research. Cambridge Quarterly of Healthcare
38. Crouch, R.A., Arras, J.D. AZT trials and tribulations. Ethics 1996, 5. 334-5.
Hastings Center Report 1998, 28(6): 26-34. 58. Kipnis, K. Vulnerability in research subjects: A bioethical
39. President's Commission f or the Study of Ethical Problems taxonomy. In National Bioethics Advisory Commission.
in Medicine and Biomedical and Behavioral Research. Ethical and Policy Issues in Research Involving Human Partici-
Summing Up. Washington, DC: US Government Printing pants, Vol. II. Rockville, Md.:NBAC, 2001. [Online] Available:
Office, 1983. [Online] Available: httpi/www.bioethics http://www.georgetown.eduhrsearch/nrcbl/nbac/human/
.gov/reports/past_commissions/summing_up.pdf. overv012.pdf.
40. Rutstein, D.D. The ethical design of human experiments. 59. Weijer, C. The ethical analysis of risk. Journal of Law,
In Freund, PA., ed. Experimentation with Human Subjects. Medicine and Ethics 2000, 28: 344-61.
New York: George Braziller, 1970: 383- 401. 60. Freedman, B., Fuks, A., Weijer, C. Demarcating research and
41. Emanuel, E.J., Miller, F.G. The ethics of placebo- treatment: A systematic approach fbr the analysis of the ethics
controlled trials-A middle ground. New England Jour- of clinical research. Clinical Research 1992, 40: 653-60.
nal of Medicine 2001, 345: 915-9. 61. Weijer, C. The future of research into rotavirus vaccine.
42. Freedman, B. Placebo-controlled trials and the logic of British MedicalJournal 2000, 321: 525-6.
clinical purpose. IRB: A Review of Human Subjects Research 62. Anderson, E Value in Ethics and Economics. Cambridge,
1990, 12(6): 1-6. MA. Harvard University Press, 1993: chapter 9.
43. Temple, R., Ellenberg, S.S. Placebo controlled trials and 63. Weijer, C. Thinking clearly about research risks: Im-
active control trials in the evaluation of new treatments. plications of the work of Benjamin Freedman. IRB: A
I. Ethical and scientific issues. Annals of Internal Medicine Review of Human Subjects Research 1999, 21(6): 1-5.
2000, 133: 455-63. 64. Sen, A., Williams, B. Introduction. In Sen, A., Williams,
44. Wendler, D., Emanuel, E.J., Lie, R.K. The standard of B., eds., Utilitarianism and Beyond. New York: Cambridge
care debate: Can research in developing countries be University Press, 1982: 1-22.
both ethical and responsive to those countries' health 65. Leonard, H., Zeckhauser, R. Cost-benefit analysis and
needs? American Journal of Public Health 2004, 94: 923-8. the management of risk: Philosophy and legitimacy. In
45. Nuffield Council on Bioethics. The Ethics of Research Related MacLean, D., ed. Values at Risk. Totowa, NJ: Rowman &
to Healthcare in Developing Countries. London, England: Littlefield Publishers, 1986: 31- 48.
Nuffield Council on Bioethics, 2002. [Online] Available: 66. Gold, M.R., Siegel, J.E., Russell, L.B., Weinstein, M.C.,
httpi/nuffieldbioethics.org/fileLibrary/pdf/errhdc_f ull eds. Cost- Effectiveness in Health and Medicine. New York:
report001.pdf. Oxford University Press, 1996.
46. Emanuel, E.J. The Ends of Human Life. Cambridge, MA: 67. Sen, A. Interpersonal comparisons of welfare. In Sen,
Harvard University Press, 1991: chapters 4 and 5. A., Choice, Welfare, and Measurement. Cambridge, MA:
47. Daniels, N., Sabin, J.E. Setting Limits Fairly. New York Harvard University Press, 1982: 264-84.
Oxford University Press; 2002: chapter 2. 68. Relman, A.S. Economic incentives in clinical investiga-
48. Walzer, M. Spheres of Justice. New York: Basic Books, 1983. tions. New EnglandJournal of Medicine 1989, 320: 933- 4.
49. Faden, R.R., Kass, N. HIV research, ethics and the developing 69. Porter, R.J., Malone, T.E. Biomedical Research: Collabora-
world. American Journal of Public Health 1998, 8a 548-50. tion and Conflicts of Interest. Baltimore, MD: Johns Hopkins
50. Kidane, G., Morrow, R.H. Teaching mothers to provide University Press, 1992.
home treatment of malaria in Tigray, Ethiopia: A ran- 70. Spece, R.G., Shimm, D.S., Buchanan, A.E., eds. Conflicts
domized trial Lancet 2000, 356: 550-5. of Interest in Clinical Practice and Research. New York:
51. Bloom, B.R. The highest attainable standard: Ethical Oxford University Press, 1996.
issues in AIDS vaccines. Science 1998, 279: 186-8. 71. National Commission for the Protection of Human
52. Lederer, S.E. Subject to Science. Baltimore, Md.: Johns Subjects of Bio- medical and Behavioral Research. In-
Hopkins University Press, 1995. stitutional Review Boards: Report and Recommendations.
53. National Institutes of Health. NIH policy and guidelines on Washington, DC: US Government Printing Office, 1978.
the inclusion of children as participants in research involving 72. Emanuel, E.J., Wood, A., Fleischman, A., et al. Oversight
human subjects. [Online] March 6, 1998. Available: http:// of human participants research: Identifying problems to
grants2.nihgov/grants/guide/notim-filestriot98-024.htmL evaluate reform proposals. Annals of Internal Medicine
54. Weijer, C., Fuks, A. The duty to exclude: Excluding 2004,141: 282-91.
people at undue risk from research. Clinical and Investi- 73. McWilliams, R., Hoover-Fong, J., Hamosh, A., et al. Prob-
gative Medicine 1994, 17: 115-22. lematic variation in local institutional review of a multicen-
55. DeBruin, D. Justice and the inclusion of women in clinical ter genetic epidemiology study. JAMA 2003, 290: 360-6.
studies. Kennedy Institute of Ethics Journal 1994, 4: 117-46. 74. Roberts, L.M., Bowyer, L, Homer, C.S., Brown, M.A. Multi-
56. Mastroianni, A.C., Faden, R.R., Federman, D.D. Women centre research Negotiating the ethics approval obstacle
and Health Research: Ethical and Legal Issues of Including course [letter]. MedicalJournal of Australia 2004, 180: 139.
Women in Clinical Studies. Washington, DC: National 75. Stair, T.O., Reed, C.R., Radeos, M.S., et al., for the MARC
Academies Press 1994. Investigators. Variation in Institutional Review Board
al II ;iii IIIIJJ
responses to a standard protocol for a multicenter clin- 93. Flory, J.H., Emanuel, EJ. Interventions to improve
ical trial. Academic Emergency Medicine 2001, 8: 636- 41. research participants' understanding in informed
76. Wood, A., Grady, C., Emanuel, E.J. Regional ethics consent for research: A systematic review. JAMA
organizations for protection of human research partici- 2004, 292: 1593-1601.
pants. Nature Medicine 2004, 10: 1283-8. 94. National Bioethics Advisory Commission. Research Involv-
77. Mulholland, K., Smith, P.G., Broome, C.V., et al. A ran- ing Persons with Mental Disorders That May Affect Decision-
domized trial of a Haemophilus influenzae type b conju- making Capacity. Rockville, MD: NBAC, 1998. [Online]
gate vaccine in a developing country for the prevention Available: http://www.georgetown.edu/research/nrcb1/
of pneumonia-Ethical considerations. International nbac/capacityfrOC.htm.
Journal of Tuberculosis and Lung Disease 1999, 3: 749-55. 95. Buchanan, A.E., Brock, D.W. Deciding for Others:
78. Wendler, D., Rackoff, J.E. :nformed consent and respect- The Ethics of Surrogate Decision Making. New York:
ing individual autonomy: What's a signature got to do Cambridge University Press, 1990: chapter 2.
with it? IRB: Ethics and Human Research 2001, 23(3): 1- 4. 96. Dresser, R. Mentally disabled research subjects:
79. White, M.T. Guidelines for IRB review of international The enduring policy issues. JAMA 1996, 276: 67-72.
collaborative medical research: A proposal. Journal of 97. Michels, R. Are research ethics bad for our mental health?
Law, Medicine and Ethics 1999, 27: 87-94. New EnglandJournal of Medicine 1999, 340: 959-61.
80. Freedman, B. A moral theory of informed consent. Hastings 98. Capron, A.M. Ethical and human-rights issues in
Center Report 1975, 5(4): 32-9. research on mental disorders that may affect decision-
81. President's Commission for the Study of Ethical Prob- making capacity. New England Journal of Medicine
lems in Medicine and Biomedical Research. Making 1999, 340: 1430- 4.
Health Care Decisions: Ethical and Legal Implications of 99. Wendler, D., Prasad, K. Core safeguards for clinical
Informed Consent in the Physician-Practitioner Relationship. research with adults who are unable to consent. Annals
Washington, DC: US Government Printing Office, of Internal Medicine 2001, 135: 514-23.
1982. [Online] Available http: //www.bioethics.gov/ 100. Weijer, C., Goldsand, G., Emanuel, E.J. Protecting
reports/past_commissions/making_health_care_ communities in research: Current guidelines and limits
decisions.pdf. of extrapolation. Nature Genetics 1999, 23: 275-80.
82. Donagan, A. Informed cons= in therapy and experimen- 101. Macaulay, A.C., et al. Participatory research with native
tation. Journal of Medicine ar. d Philosophy 1977, 2: 318-29. community of Kahnawake creates innovative code of
83. Faden, R.R., Beauchamp, TL., with King, N.M.P. A History research ethics. Canadian Journal of Public Health 1998,
and Theory of Informed Consent. New York: Oxford Uni- 89: 105-8.
versity Press, 1986: chapters 5-9. 102. IJsselmuiden, C.B., Faden, R.R. Research and informed
84. Berg, J.W., Applebaum, P.S., Lidz, C.W., Parker, L.S. In- consent in Africa: Another look. New England Journal of
formed Consent: Legal Them), and Clinical Practice. 2nd Medicine 1992, 326: 830- 4.
edn. New York: Oxford University Press, 2001: chapters 103. Karim, Q.A., Karim, S.S.A., Coovadia, H.M., Susser, M.
2, 3, 11, and 12. Informed consent for HIV testing in a South African
85. Dworkin, G. The Them), and Practice of Autonomy. New York. hospital: Is it truly informed and truly voluntary?
Cambridge University Press; 1988: chapters 1, 6, and 7. American Journal of Public Health 1998, 88: 637- 40.
86. Sreenivasan, G. Does informed consent to research 104. Weijer, C., Shapiro, S., Fuks, A., Glass, K.C., Skrut-
require comprehension? Lancet 2003, 362:,2016-8. kowska, M Monitoring clinical research: An obligation
87. Dickert, N., Grady, C What's the price of a research unfulfilled. Canadian Medical Association Journal 1995,
subject? Approaches to payment for research partici- 152: 1973-80.
pation. New England Journal of Medicine 1999, 341: 105. Scanlon, T.M. What We Owe to Each Other. Cambridge,
198-203. MA: Harvard University Press; 1999: chapters 1 and 8.
88. Emanuel, E.J., Currie, X.E., Herman, A„ on behalf of 106. Kymlicka, W. Liberalism, Community and Culture.
Project Phidisa. Undue inducement in clinical research New York: Oxford University Press, 1989.
in developing countries: Is it a worry? Lancet 2005, 366: 107. Gutmann, A., Thompson, D. Democracy and Disagree-
336- 40. ment. Cambridge, MA: Harvard University Press, 1996:
89. Emanuel, E.J. Ending concerns about undue induce- chapter 1.
ment. Journal of Law, Medicine and Ethics 2004, 32: 100-5. 108. Nagel, T. The fragmentation of value. In Nagel, T.,
90. Harris, J. Wonderwoman and Superman: The Ethics of Mortal Questions. New York: Cambridge University
Human Biotechnology. New York: Oxford University Press, 1979: 128- 41.
Press, 1992: chapter 6. 109. Temkin, L. Inequality. New York: Oxford University
91. Wilkinson, M., Moore, A. Inducement in research. Press, 1993: chapter 2.
Bioethics 1997, 11: 373-89. 110. Richardson, H.S. Specifying norms as a way to resolve
92. Grisso, T., Applebaum, P.S. Assessing Competence to Consent concrete ethical problems. Philosophy and Public Affairs
to Treatment. New York: Oxford University Press, 1998. 1990, 19: 279-310.
Miller/Brody: A Critique of Clinical Equipoise
A Critique of Clinical Equipoise: could two such radically divergent viewpoints exist,
without research ethics being widely regarded as in a
Therapeutic Misconception in the state of crisis?
Ethics of Clinical Trials
Franklin G. Miller and Howard Brody Therapeutic Misconceptions
The prevailing ethical perspective on clinical trials
The Hypericum Depression Trial Study Group pub- holds that physician-investigators can discharge their
lished in 2002 the results of a randomized trial com- "therapeutic obligation" to patients in the context of
paring hypericum (St. John's Wort), sertraline (Zoloft), randomized clinical trials (RCTs) as long as treatments
and placebo in the treatment of major depression.' In being tested scientifically satisfy clinical equipoise. We
the study, funded by the National Institutes of Health, contend that this ethical perspective is fundamentally
340 subjects from twelve participating centers were flawed. An ethical framework that provides normative
randomized to three trial arms for an eight-week guidance about a practice should accurately charac-
period, with careful monitoring to assure that patients terize the practice. The prevailing ethical perspective
who worsened significantly or who became suicidal fails this test: All sound ethical thinking about clinical
were removed from the study and received adequate research, and the regulatory framework for review of
treatment. Neither hypericum nor sertraline was found protocols for clinical investigation, depends on a basic
to be superior to placebo on the primary outcome distinction between research and therapy. But the
measures. The authors noted, "From a methodological claims in the prevailing ethical perspective on clinical
point of view, this study can be considered an example trials conflate research and therapy. These claims are
of the importance of including inactive and active com- that the ethics of the physician—patient relationship
parators in trials testing the possible antidepressant must govern RCTs, that physicians who conduct these
effects of medications. In fact, without a placebo, hy- trials have a "therapeutic obligation" to patients en-
pericum could easily have been considered as effective rolled in them, and that RCTs must be compatible with
as sertraline."2 some form of equipoise.
What can we conclude about the ethics of this trial? Certainly, investigators and ethicists recognize that
One dominant viewpoint in research ethics would have clinical trials are scientific experiments, which differ
prohibited the study. On this viewpoint, a random- from standard medical care. They also recognize that
ized trial is ethical only in circumstances of "clinical they are subject to regulatory requirements which
equipoise"—a genuine uncertainty within the medical do not apply to routine medical practice. However,
community as to whether (in this case) any of the three the prevailing ethical framework views clinical trials
treatment arms are superior to the other two. No such through a therapeutic lens. The mainstream ethical ap-
uncertainty exists. Approximately twenty-five clinically proach to clinical trials attempts to have it both ways: to
available antidepressants, including sertraline, have view the clinical trial as a scientific experiment, aimed
been shown to be superior to placebo.3 Moreover, the at producing knowledge that can help improve the
majority opinion within psychiatry probably holds care of future patients, and as treatment conducted by
that sertraline is definitely superior to hypericum for physicians who retain fidelity to the principles of thera-
major depression, even if hypericum has potential for peutic beneficence and therapeutic non-maleficence
the treatment of mild to moderate depression. But an- that govern the ethics of clinical medicine. The doc-
other widespread viewpoint would hold that the trial trine of clinical equipoise has emerged as the bridge
was ethically sound. Depressed individuals widely between medical care and scientific experimentation,
use hypericum, a "natural" agent, despite the lack of allegedly making it possible to conduct RCTs without
proven efficacy. Accordingly, a rigorous evaluation of- sacrificing the therapeutic obligation of physicians to
fered scientific, clinical, and social value. According to provide treatment according to a scientifically valid-
the report of trial results, the study was approved by ated standard of care. This constitutes a "therapeutic
institutional review boards (IRBs) at twelve sites, and misconception" concerning the ethics of clinical trials,
subjects provided written informed consent. analogous to the tendency of patient volunteers to con-
But if clinical equipoise is a basic requirement for fuse treatment in the context of RCTs with routine med-
ethical research, how could all these review boards be ical care.4 As Paul Appelbaum has recently observed,
blind to the unethical nature of this trial? And how "In fact, this confusion between the ethics of research
Source: From The Hastings Center Report 33 (2003) no. 3, 19-28. Reprinted with permission of the Hastings Center.
II 311 111111
and of ordinary clinical care appears rampant in the abandoning the distinction between therapeutic and
world of clinical trials. "5 nontherapeutic research.
The therapeutic misconception in the ethics of Clinical research shares with medical care the fact
clinical trials is reflected in the language commonly that both are performed by physicians in clinical set-
used within the clinical research enterprise. Clinical tings and both often use similar diagnostic and treat-
trials are often described as "therapeutic research," ment interventions. When the commission began its
and investigators are regarC ed as having a "therapeutic work, physicians commonly regarded clinical research
intent." Research participants who are being studied and medical therapy as inextricably connected. One
because they have a medical condition under investiga- authority quoted by Levine claimed that "Every time
tion are referred to as "patients" and investigators as a physician administers a drug to a patient, he is in a
"physicians" or "doctors," without qualification. sense performing an experiment." But the commission
To demonstrate our contention about the main- recognized the importance of determining the bound-
stream approach to the ethics of clinical trials, we will aries between routine medical practice and research.
offer an intellectual reconstruction of some of the his- For Levine, the commission's conceptual breakthrough
tory of research ethics since the 1970s. This history is came with the realization that the physicians of the day
characterized by incoherence resulting from commit- were thinking about clinical research in the wrong way
ment to two incompatible positions, each approaching and that the boundary between research and therapy
research ethics in a fundamentally different way. The was clear rather than fuzzy. The commission came to
therapeutic misconception about the ethics of clinical hold that clinical research is fundamentally different
trials has emerged from the "similarity position," which from medical practice.°
argues that ultimately, the ethics of clinical trials rest Clinical medicine aims at providing optimal medical
on the same moral consithrations that underlie the care for individual patients. Ethically, it is governed by the
ethics of therapeutic medicine. The "difference pos- principles of therapeutic beneficence and therapeutic non-
ition" argues that the ethics of clinical trials must start maleficence. Therapeutic beneficence directs physicians
with the realization that medical research and medical to practice medicine with primary fidelity to promoting
treatment are two distinct forms of activity, governed the health of particular patients. According to therapeutic
by different ethical principles. nonmaleficence, the risks of medical care to which a
The reigning ethical paradigm for clinical trials has patient is exposed are to be justified by the prospect
coexisted with clinical trials practice that departs from of compensating medical benefits for that patient. The
its guidance. Clinical equipcise, the cornerstone of the physician uses scientific knowledge to care for the patient
similarity position, rules out placebo-controlled trials and engages in therapeutic experimentation with the aim
whenever there is a proven effective treatment for the only of finding optimal treatment. It is not part of the
disorder under investigation.° However, TRBs have rou- role of the physician in providing medical care to develop
tinely approved such placebo-controlled trials. These scientific knowledge that can help future patients.
two anomalies—unappreciated theoretical incoherence Clinical research, in contrast, is not a therapeutic
and conflict between the theoretical paradigm and the activity devoted to the personal care of patients. It is
practice of ethical review of clinical trials—call for designed for answering a scientific question, with the
critical examination of the similarity position and the aim of producing "generalizable knowledge." The in-
doctrine of clinical equipoise. vestigator seeks to learn about disease and its treatment
in groups of patients, with the ultimate aim of improv-
The Distinction between Research ing medical care. Scientific interest in any particular
and Therapy patient concerns what can be learned that is applicable
to other patients. In view of the nature and purpose
In 1979, Robert Levine summarized "the most im- of clinical research, the principles of beneficence and
portant achievements of the National Commission" nonmaleficence applicable to clinical research lack
for the Protection of Human Subjects of Biomedical the therapeutic meaning that guides their application
and Behavioral Research in "correcting the conceptual to medical care. Clinical research is dedicated primar-
and semantic errors that had undermined virtually all ily to promoting the medical good of future patients
previous attempts to develop rational public policy by means of scientific knowledge derived from ex-
on research involving human subjects."' Two portions perimentation with current research participants—a
of Levine's summary capture the essential ingredients frankly utilitarian purpose.
of the difference position: recognizing the distinc- A major reason for distinguishing research from
tion between research and therapy and, accordingly, therapy is to underscore that clinical research has an
inherent potential for exploiting research participants.9 which test the safety or efficacy of investigational and
Exploitation also may occur in clinical medicine—venal standard treatments. Since patients may derive medical
physicians sometimes perform medically unnecessary benefit from trial participation, especially in phase III
procedures for the sake of profit, for example. Yet when RCTs (the final stage of testing, which many investiga-
physicians of integrity practice medicine, physicians' tional drugs never even reach), clinical trials are often
and patients' interests converge. The patient desires characterized as "therapeutic research."
to regain or maintain health or to relieve suffering the Nonetheless, the process of treatment in RCTs dif-
physician is dedicated to providing the medical help fers radically from routine clinical practice.14 Consider
that the patient needs. the contrast between the hypericum—sertraline trial
In clinical research, by contrast, the interests of in- and routine medical care for depression. If a phys-
vestigators and patient volunteers are likely to diverge, ician treated 340 patients for major depression, she
even when the investigator acts with complete integrity. would not decide which drug to administer by flipping
Patient volunteers, especially in clinical trials, typically a coin. If the physician elected to use sertraline, she
seek therapeutic benefit, though they also may be mo- would judge each case individually to determine dose,
tivated by altruism.1° Investigators are interested pri- when to change the dose, and whether to prescribe a
marily in developing scientific knowledge about groups second antidepressant or recommend other treatment.
of patients. Regardless of investigators' motivations, We would expect to find considerable variation in the
patient volunteers are at risk of having their well-being treatment administered to those 340 patients after eight
compromised in the course of scientific investigation. weeks or so. From the vantage point of therapy, this is
Clinical research involves an inherent tension between what it means to provide care to patients.
pursuing rigorous science and protecting research par- From the vantage point of research, such variation
ticipants from harm.il would wreak havoc on experimental design and the
Historically, the ethical distinction between research validity and generalizability of findings. So when pa-
and therapy emerged out of concern about exploitive tients are randomized to one or another experimental
abuses of patients in clinical research. Reflection on drug and are treated according to relatively inflexible
this dark history gave rise to a major development in protocols, the activity is very different from therapeutic
the ethics of clinical research: the requirement for in- medicine.
dependent, prospective review and approval of research In many other ways, too, routine aspects of re-
protocols.0 Prior independent review was considered search deviate from what would be required by the
necessary for clinical research because of the divergence duties of therapeutic beneficence and nonmaleficence.
between the interests of the investigator and the research Volunteer patients and physician investigators are often
participant. Self-regulation by physician-investigators ignorant of assignment to the experimental or control
could not be trusted in the research context to the same treatment, which may be a placebo. Trials often include
extent that self-regulation by physicians was appropri- interventions such as blood draws, lumbar punctures,
ate in the therapeutic context. The basic rationale for radiation imaging, or biopsies that measure trial out-
prospective, independent research review depends on comes but in no way benefit participants. RCTs often
the distinction between research and therapy. contain a drug "washout" phase before randomization
The point of distinguishing research and therapy to avoid confounding the evaluation of the investiga-
is not to make an invidious comparison, implying that tional treatment with the effects of medication that
clinical trials are more risky or ethically problematic than patients were receiving prior to the trial. These various
routine clinical practice. Indeed, there is some evidence features of research design promote scientific validity,
that patients receive more favorable medical outcomes in they carry risks to participants without the prospect of
many clinical trials,13 and clinical medicine is certainly compensating therapeutic benefit.
rife with ethical problems. Further, since research is For these reasons, Levine argued that the second
more carefully regulated than medical practice, it is quite major contribution of the commission was to abandon
likely that fewer ethical violations occur in research. To the "illogical" distinction between therapeutic and
say that two activities are ethically different is not to say nontherapeutic research, which previous policymakers
that either is inherently better than the other. thought was essential to the proper regulation of re-
search and the protection of human subjects.15 Because
Abandoning the Distinction research and therapy are distinct activities, and the
ethics of therapeutic medicine therefore cannot be
The distinction between research and therapy is most automatically extended to guide research, it is mistaken
likely to be obfuscated in the context of clinical trials, to label research as "therapeutic" or "nontherapeutic,"
Al
as if that made any funda::nental ethical difference. individual patient against a model in which "medicine
Many research trials consist of a complex mix of thera- is to be viewed as caring for populations."18What made
peutic and nontherapeutic elements—the placebo- the RCT ethically suspect was that it seemed to him a
controlled trial being only one obvious example—such prime example of population-focused—rather than
that labeling the trial as a whole as "therapeutic" individualized—and utilitarian medicine.
or "nontherapeutic" is misleading. In addition, the Fried devoted most of his book to defending pa-
therapeutic—nontherapeutic distinction diverts atten- tients—rights in personal care."19 Returning to medical
tion from key ethical issues. Consider a nontherapeutic research, he took issue with trials in which patients
trial in which one interviews subjects and takes saliva were randomized to receive either the experimental
samples and a therapeutic trial in which one is testing intervention or standard care. Fried coined the term
a new cancer drug that has some promise for creating "equipoise" to describe the ethically necessary condition
remission but also has potentially life-threatening tox- for conducting an RCT: physician-investigators must be
icity. Is the latter trial less in need of stringent regula- indifferent to the therapeutic value of the experimental
tory oversight because it is "therapeutic"? Or does the and control treatments evaluated in the trial. The basic
therapeutic—nontherapeutic distinction distract the idea of equipoise had previously been articulated by
observer from those aspects of the trials that assume far Bradford Hill, a pioneer in the development of RCTs.2°
greater moral weight, such as the level of risks and the But what Fried objected to primarily in RCTs was not
potential vulnerability of subjects? randomization per se but the fact that no informed
Once one understands the distinction between consent had been obtained. Fried saw the threat of
research and therapy, one realizes that "therapeutic" "care for groups" (instead of "care for individuals") as
research is still research and that the ethical rules ap- residing primarily in the idea that it was legitimate to
propriate to it are those appropriate for clinical research enroll subjects in an RCT without explicit, informed
generally. Even though the patient may derive benefit consent because the results of the trial would provide
from treatment being evaluated, the basic goal of the new medical knowledge that would improve the lot of
activity is not personal therapy, but rather the acquisi- future patients.21 Because Fried was concerned chiefly
tion of generally applicable scientific knowledge. The about informed consent, an essential ingredient of both
basic goal and nature of the activity determines the medical research and therapeutic medicine, he saw no
ethical standards that ought to apply. problem in applying the ethics of medical therapy to
Writing in 1993, Jay Katz affirmed the vital import- medical research.
ance of the distinction between research and therapy In the 1970s, the "respect for patient autonomy"
and deplored its blurring in practice: "The astronomical movement was gaining steam as a replacement for the
increase in clinical research has, in practice, not led to old Hippocratic ethic of paternalistic beneficence. Since
a clear demarcation between therapy and research, bio- both Fried and the National Commission seemed on
ethical theories notwithstanding. This vital distinction the surface to be championing patient autonomy, it was
remains blurred when physician-investigators view easy to miss the point that they were proposing two
subjects as patients, and then believe that patients' in- fundamentally different strategies for approaching the
terests and not science's are being served by participa- ethics of clinical trials. Put another way, so long as the
tion in randomized clinical trials that are so commonly bioethics debate of the moment has to do with whether
conducted in today's world."15 One of the reasons in- research ethics requires all competent subjects to give
vestigators (and bioethicists) 1-tave failed to appreciate fully informed consent, any fundamental divergence
the distinction between research and therapy is that between the similarity and the difference positions is
the similarity position has conceived the ethics of clin- likely to be obscured.
ical trials within the context of the physician—patient
relationship. The Emergence of Clinical Equipoise
Charles Fried and the Similarity Position During the 1980s, philosophers interested in research
ethics recognized a tension between the obligation of
In 1974, Fried published Medical Experimentation: physicians to offer optimal care to their patients ("the
Personal Integrity and Social Po icy, which launched the therapeutic obligation") and the provision of medical
similarity position within bioethics." Fried assumed treatment in the context of clinical trials. Don Marquis
that answers to ethical dilemmas in research would addressed this problem in a 1983 essay, "Leaving Ther-
have to be found within the ethics of therapeutic apy to Chance."22 The title is significant, suggesting that
medicine. He defended fidelity to the interests of the the RCT is a form of therapy rather than an ethically
distinct activity. Marquis began his essay "Consider ethically entitled to expect treatment from his or her
this dilemma: according to an argument that is hard physician—an entitlement that cannot be sacrificed to
to refute, the procedure for conducting randomized scientific curiosity."27
clinical trials of anticancer drugs is incompatible with The bioethics community perceived Freedman's
the ethics of the physician—patient relationship. If this concept of clinical equipoise as both a theoretical and
problem is to be resolved, then either a key proced- a practical advance. Theoretically, it appeared to offer
ure for achieving scientific knowledge in medicine a more intellectually compelling argument than Fried's
must be given up or unethical behavior by physicians initial formulation. Practically, it would permit useful
must be tolerated."23 In framing this "RCT dilemma," RCTs that would otherwise be ethically proscribed to
Marquis assumed that the appropriate ethic for clinical go forward. Since it appeared to solve the RCT dilemma
trials was that of the (therapeutic) physician—patient by accommodating the conduct of clinical trials with
relationship. the therapeutic obligation of physicians to offer optimal
Fred Gifford, following the lead of Marquis, exam- medical care, clinical equipoise gained wide currency as
ined the RCT dilemma in greater depth "The central a fundamental concept of the ethics of clinical trials.28
dilemma concerning randomized clinical trials (RcTs) The persuasive way in which Freedman fortified the
arises out of some simple facts about causal methodol- similarity position diverted attention from the fact that
ogy (Ras are the best way to generate the reliable causal clinical equipoise collapsed the distinction between
knowledge necessary for optimally-informed action) and research and therapy.
a prima facie plausible principle concerning how phys- The similarity position and clinical equipoise have
icians should treat their patients (always do what it is been popular not only among bioethicists but also
most reasonable to believe will be best for the patient)."24 among investigators. We speculate that this ethical
Neither Marquis nor Gifford found what they regarded perspective helps to address investigators' psycho-
as a satisfactory solution, and neither considered the logical needs. Physician-investigators, after all, went
possibility that the difference position could dismiss the to medical school, not investigator school. To think of
"RCT dilemma" as misguided to begin with. research with patients outside the ethical framework
In a landmark 1987 article, Benjamin Freedman of the physician—patient relationship, as the difference
offered a solution to the RCT dilemma that gained position requires, may be difficult and threatening to
widespread acceptance within bioethics. He argued them. Clinical equipoise offers a formula that seems
that the tension between ethically legitimate scien- to allow them to mix both physician and investigator
tific experimentation and the therapeutic obligation roles—even if the psychological comfort is purchased
of physicians could be overcome by the principle of at the price of ethical obfuscation.
"clinical equipoise."25 Freedman agreed with Fried and The anomaly therefore exists that much of today's
Marquis that ethical clinical trials had to be compatible bioethical thinking accepts clinical equipoise as an
with therapeutic beneficence and nonmaleficence. But outgrowth of the similarity position, while the Federal
he argued that Fried's formulation of equipoise was too regulations grew out of the work of the National Com-
constraining. Freedman called Fried's original concept mission, which largely endorsed the difference position.
"theoretical equipoise" (sometimes called "individual One would imagine that sooner or later proponents of
equipoise") and contrasted it with his favored concept clinical equipoise would realize the need to defend this
of "clinical equipoise" (sometimes called "collective doctrine from the charge that it conflates the ethics of
equipoise"). In the latter sense of equipoise, any in- clinical trials with the ethics of medical care. But this is
dividual investigator or physician might have reasons precisely what has not yet happened.
to believe that one arm of the RCT offers a therapeutic
benefit over the other arm, but the medical profession The Case of Placebo-Controlled Trials
as a whole remains divided. According to Freedman,
an RCT is ethical so long as the professional community Although the similarity position, bolstered by clin-
has not yet reached a consensus, which recognizes that ical equipoise, became the reigning paradigm in the
"medicine is social rather than individual in nature."26 ethics of clinical trials, its dominion over practice was
When, and only when, clinical equipoise is satisfied limited. This divorce between theory and practice has
will patients enrolled in a clinical trial be assured that been particularly pronounced in the case of placebo-
they will not be randomized to treatment known to controlled trials. Freedman and his colleagues argued
be inferior. Freedman thus asserted in a later article that the use of placebo controls is unethical whenever
that clinical equipoise is "grounded in the normative proven effective treatment exists for the medical con-
nature of clinical practice, the view that a patient is dition under investigation in a clinical trial because
■ II I II 411 IIILLI
those randomized to placebo would receive treatment they contravene clinical equipoise.34 Other commen-
known to be inferior.29 tators, writing in leading medical journals, defended
Despite the clear implications of clinical equipoise more or less extensive use of placebo-controlled trials
for the ethics of placebo-controlled trials, numerous on methodological and ethical grounds.35 Without
trials, such as the hypericum—sertraline trial, continued directly challenging the doctrine of clinical equipoise,
to use placebo controls despite proven effective treat- they implied that clinical equipoise provides erroneous
ment. Placebo controls have typically been used in ethical guidance for placebo-controlled trials. Accord-
trials of new treatments for a wide range of chronic ingly, the debate over placebo-controlled trials jeop-
conditions—including mood and anxiety disorders, ardizes the reigning ethical paradigm of the similarity
asthma, stable angina, hypertension, and migraine position and clinical equipoise.
headaches—all of which can be treated with medica-
tion of proven efficacy. Critique of the Similarity Position and
There are two explanations for this incoherence Clinical Equipoise
between theory and practice. First, the FDA has encour-
aged the use of placebo controls in trials concerning Our reconstruction of the recent history of the ethics
these and other chronic conditions.3° Active-controlled of clinical trials has traced the emergence and dom-
trials designed to test the equivalence of the experi- inance of the similarity position. This history also
mental treatment with a standard treatment suffer reveals cracks in the foundation of this ethical para-
from serious methodological limitations. Whenever digm. Simultaneous endorsement of the difference
active-controlled trials show no statistically significant position, reflected in the federal regulatory system
difference between the invest gational treatment and an and the Belmont Report, and the similarity position,
active comparator, two conclusions are possible. Either which invokes the doctrine of clinical equipoise, has
both were effective in the trial sample of patients, or left the ethics of clinical trials in a state of incoherence.
neither was effective. Withou: the use of a placebo con- Although this incoherence has not received critical
trol, such trials lack internal validity. Accordingly, the attention, it becomes apparent once the assumptions
FDA has insisted that pharmaceutical companies use underlying the similarity position and clinical equi-
placebo controls in trials of new treatments for condi- poise are challenged. In addition, the divorce between
tions characterized by fluctuating symptoms and high research ethics theory and clinical trials practice in
rates of placebo response.31 Second, the U.S. federal the case of placebo-controlled trials suggests that a
regulations governing human subjects research do not critique of the similarity position and clinical equi-
provide any explicit guidance on the use of placebo poise is overdue.
controls.32 utBs have been free to approve such placebo- We contend that clinical equipoise is fundamen-
controlled trials, provided that they meet regulatory tally mistaken because "the RCT dilemma," for which
requirements for a favorable risk-benefit ratio, includ- it was proposed as a solution, is false. Clinical equi-
ing the potential value of knowledge to be gained and poise and all other forms of equipoise make sense as
informed consent. a normative requirement for clinical trials only on the
For the most part, this lack of fit between theory assumption that investigators have a therapeutic obli-
and practice received little critical attention until the gation to the research participants. The "therapeutic
publication in 1994 of an article in the New England obligation" of investigators, forming one horn of the
Journal of Medicine entitled "The Continuing Unethical RCT dilemma, constitutes a therapeutic misconception
Use of Placebo Controls."33 Kenneth Rothman and Karin about the ethics of clinical trials. The presumption
Michels castigated the practice of placebo-controlled that RCTs must be compatible with the ethics of the
trials in the face of proven ef 'ective treatment and the physician—patient relationship assumes erroneously
role of the FDA in encouraging these trials. They cited that the RCT is a form of therapy, thus inappropriately
the Declaration of Helsinki, which relies heavily on the applying the principles of therapeutic beneficence and
similarity position, as prohibiting this widespread "un- nonmaleficence that govern clinical medicine to the
ethical" practice. fundamentally different practice of clinical research. It
Their article stimulated a lively debate over the is impossible to maintain fidelity to doing what is best
ethics of placebo-controlled trials. Freedman and medically for patients in the context of RCTs because
his colleagues attacked "the placebo orthodoxy" in a these are not designed for, and may conflict with, per-
two-part article that challenged the scientific value of sonalized care. Although ethically appealing, the pro-
placebo-controlled trials and reiterated that they are un- ject of bridging the gap between therapy and research
ethical when proven effective treatments exist because via the doctrine of clinical equipoise is doomed to fail.
The insight that the RCT contravenes the ethics It is worth pondering, however, the practical con-
of the physician—patient relationship led Samuel sequences that might ensue if physicians, investigators,
Hellman and Debra Hellman to argue that the RCT is patients, and ethicists understood clinical trials with-
unethical and that other methods of evaluating treat- out distortion by therapeutic misconceptions. Would
ments should be employed.36 This stance, however, recruitment of participants for valuable clinical trials
would deprive patients and society of the benefits become substantially more difficult, slowing progress
that flow from rigorous scientific evaluation of experi- in medical care? The fact that clinical trials are no
mental and standard treatments. The more reasonable longer seen as a mode of therapy leaves unchanged the
conclusion is that Rcrs should be governed by eth- real prospect of therapeutic benefits offered to patients
ical norms appropriate to clinical research, which are from trial participation, including the opportunity
distinct from therapeutic beneficence and therapeutic to receive promising investigational agents, ancillary
nonmaleficence. medical care, expert diagnostic evaluations, and educa-
Clinical equipoise is neither necessary nor suffi- tion about their disorder. Nonetheless, some patients
cient for ethically justifiable Rcrs. The use of placebo might be less inclined to participate in clinical trials
controls when proven effective treatment exists violates when they appreciate the differences between these
clinical equipoise; however, when methodologically scientific experiments and medical care.
indicated, their use is no different in principle from To attract enough subjects, researchers might have
any research intervention that poses risks to subjects to pay people for their participation, as researchers in
without the prospect of benefiting them.37 In many industry-sponsored clinical trials already do with in-
cases, the risks of withholding effective treatment are creasing frequency Payments would add to the cost of
excessive, and the use of placebo controls would thus conducting clinical trials, but it might help prevent the
be unethicaL Nevertheless, it is the unacceptable level therapeutic misconception among trial participants.38
of risk, not the violation of investigators' alleged "thera- To be paid signifies that the trial participant is not
peutic obligation," that makes these trials unethical. In merely a patient seeking therapy. If additional expendi-
other cases, including the hypericum—sertraline trial, ture is necessary to motivate clinical trial participation,
use of placebo controls when proven effective treat- then this is a price worth paying for enhanced profes-
ment exists is ethically justifiable. sional integrity and informed consent.
By conflating the ethics of clinical trials with the
ethics of therapeutic medicine, proponents of the An Alternative Ethical Framework
similarity position may also contribute to the lack of
adequate informed consent. If investigators view the In view of the theoretical and practical problems as-
ethics of clinical trials through a therapeutic lens, sociated with the similarity position and its logical
they may explicitly or implicitly foster the therapeutic offspring, clinical equipoise, an alternative framework
misconception among research participants—that is, for the ethics of clinical trials is needed. The most
the tendency of participants in trials to confuse clin- promising recent treatment of research ethics has
ical trials with medical care. Research participants been developed by Ezekiel Emanuel, David Wendler,
need to know that the overall activity is aimed not and Christine Grady.39 They proposes seven ethical
at their own ultimate benefit but at discovering new requirements for all clinical research: (1) scientific or
knowledge to help future patients. If they think that social value; (2) scientific validity; (3) fair subject se-
clinical trial participation is a form of therapy, then lection; (4) favorable risk-benefit ratio; (5) independ-
they cannot give informed consent. Moreover, unlike ent review; (6) informed consent; and (7) respect
the therapeutic context, the patient-subject cannot for enrolled research participants. This framework is
delegate the decision to the physician-researcher. built on the difference between research and therapy
In the therapeutic setting, a patient can decide to and on the core value of protecting research participants
trust the physician to choose the best treatment from exploitation.
because the physician has the patient's best interests Yet even this formulation of an ethical framework
at heart. The investigator has the interests of future appropriate to clinical research testifies to the hold of
patients at heart and so cannot decide for the subject the similarity position. The authors endorse clinical
whether or not to participate in the research. To be equipoise, claiming it is implied by the requirements
trustworthy, investigators must themselves under- of value, validity, and risk-benefit ratio. We contend,
stand clearly the ways in which clinical research by contrast, that the endorsement o f clinical equipoise
differs from clinical practice and convey this forth- renders incoherent any account that arises from the
rightly to potential research subjects. difference position. The most important next step for
II 3111 1111:61
research ethics is to develop this "non-exploitation" not significantly different for those receiving sertraline
framework systematically in a way that avoids any con- would not validly support the inference that hyperi-
flation of clinical research wit:1 medical care. cum was effective.'" It would remain possible that nei-
Those who agree that physician-investigators who ther treatment was effective in the study sample—as
conduct clinical trials are not governed by therapeutic was in fact shown. The study, therefore, was properly
beneficence still might argue that clinical equipoise pro- designed as a three-arm placebo-controlled trial.
vides important methodological guidance for justifying
clinical trials. Freedman and his colleagues have argued Fair Subject Selection
that clinical equipoise is both an ethical and a scientific There is no evidence to suggest that particularly vul-
principle: "That principle can be put into normative or nerable patients were recruited inappropriately for
scientific language. As a normative matter, it defines this study, which included a sample representative of
ethical trial design as prohibiting any compromise of depressed patients.
a patient's right to medical treatment by enrolling in a
study. The same concern is often stated scientifically Favorable Risk-Benefit Ratio
when we assert that a study must start with an honest Risk-benefit assessment of research protocols ul-
null hypothesis, genuine mechcal uncertainty concern- timately comes down to a matter of judgment. With
ing the relative merits of the various treatment arms respect to the use of the placebo control—the aspect
included in the trial's design."i° Nevertheless, whatever of the trial that violated clinical equipoise—the risks
is valid methodologically in clinical equipoise—the to participants from an eight-week trial, with careful
honest null hypothesis—can be stated more clearly exclusionary criteria and monitoring, were not exces-
and without confusion with the therapeutic obligation, sive and were justifiable by the anticipated value of the
by appeal to the requirement of scientific value: no re- knowledge to be gained from the research. Hence, the
search participants should be exposed to the risks of placebo component of the study had a favorable risk-
valueless research. Clinical trials must be designed to benefit ratio. Eliminating the placebo would have made
answer valuable scientific questions. If the answer is the risk-benefit ratio unfavorable by virtue of under-
already known or the question is trivial, then there is mining the scientific validity of the research.
no honest null hypothesis, arid a clinical trial should
not be conducted. But this is logically independent of Independent Review, Informed Consent, and
whether all the patients enrolled in the trial would re- Respect for Enrolled Research Participants
ceive medical treatment that is believed by the expert The report of the study asserted that IRB approval was
medical community to be at least as good as the stan- obtained at all sites and that all subjects gave informed
dard of care. consent. In addition, the described procedures for
This alternative framework provides accurate eth- monitoring subjects for possible risk of harm indicated
ical guidance concerning clinical research without pre- an acceptable level of respect.
suming that the ethics of therapeutic medicine should In sum, this study was ethically justifiable despite
govern clinical trials. We illustrate this by applying violating clinical equipoise; moreover, had it been de-
the seven ethical requirements to the example of the signed in accordance with clinical equipoise, it would
hypericum—sertraline trial. have been methodologically deficient and therefore
ethically questionable.
Scientific or Social Value and Scientific Validity Charles Weijer, a leading advocate of clinical
The study has social value owing to the widespread use equipoise and the similarity position, has recently
of herbal remedies. Since the efficacy of hypericum in claimed that "Placebo-controlled trials in the context
treating depression (especially major depression) was of serious illnesses such as depression or schizophrenia
uncertain, there was an honest null hypothesis that are ethically egregious precisely because no competent
hypericum would be no better than placebo. It would physician would fail to offer therapy to a patient with
have been unreasonable to design the trial as an active- the condition."'" Although we agree that depression is
controlled superiority trial, since it is highly unlikely a serious illness, the hypericum—sertraline trial demon-
that hypericum could be shown to be more effective strates that there is nothing "ethically egregious" about
than sertraline. An active-con:rolled equivalence trial the use of placebo controls in trials of treatment for
would lack "assay sensitivity" because the finding that depression, as long as the ethical requirements for clin-
the reduction in symptoms of depression experienced ical research are satisfied. Whether or not one agrees
by those trial participants receiving hypericum was that, all things considered, the placebo control was
ethical in this trial, the ethical justification of placebo adequate guidance for clinical trials without appeal
controls has nothing to do with the therapeutic practice to the incoherent doctrine of clinical equipoise and
of competent physicians. In any case, the alternative without conflating the ethics of research with the ethics
ethical framework with its seven requirements provides of therapy.
References
1. Hypericum Depression Trial Study Group, Effect of 19. Ibid., 94.

Hypericum Perforatum (St John's Wort) in Major Depres- 20. A.B. Hill, "Medical Ethics and Controlled Trials," British
sive Disorder: A Randomized Controlled Trial," JAMA Medical Journal 1 (1963):1043-1049.
287 (2002):1807-1814. 21. C. Fried, Medical Experimentation: Personal Integrity and
2. Ibid., 1813. Social Policy (New York: American Elsevier, 1974):8.
3. S.M. Stahl, Essential Psychopharmacology of Depression 22. D. Marquis, "Leaving Therapy to Chance," Hastings
and Bipolar Disorder (New York: Cambridge University Center Report 13, no. 4 (1983):40-47.
Press, 2000). 23. Ibid., 40.
4. P.S. Appelbaum, L.H. Roth, C.W. Lidz, P. Benson, and 24. F. Gifford, "The Conflict between Randomized Clinical
W. Winslade, "False Hopes and Best Data: Consent to Trials and the Therapeutic Obligation," Journal of Medi-
Research and the Therapeutic Misconception," Hastings cine and Philosophy (1986):347-366.
Center Report 17, no. 2 (1987):20-24. 25. B. Freedman, "Equipoise and the Ethics of Clinical Re-
5. P.S. Appelbaum, "Clarifying the Ethics of Clinical Re- search," NEJM 317 (1987):141-145.
search: A Path toward Avoiding the Therapeutic Miscon- 26. Ibid., 144
ception," American Journal of Bioethics 2, no. 2 (2002):22. 27. B. Freedman, "Placebo-Controlled Trials and the Logic
6. B. Freedman, "Placebo-Controlled Trials and the Logic of Scientific Purpose," IRB 12, no. 6 (1990):5.
of Clinical Purpose," IRB 12, no. 6 (1990):1-6. 28. T.L. Beauchamp and J.E. Childress, Principles of Biomedical
7. R.J. Levine, "Clarifying the Concepts of Research Ethics," Ethics, 5th ed. (New York: Oxford University Press,
Hastings Center Report 9, no. 3 (1979):21-26. 2001):323-327.
8. National Commission for the Protection of Human 29. B. Freedman, K.C. Glass, and C. Weijer, "Placebo
Subjects of Biomedical and Behavioral Research, The Orthodoxy in Clinical Research. II: Ethical, Legal and
Belmont Report (Washington, D.C.: U.S. Government Regulatory Myths," Journal of Law, Medicine & Ethics 24
Printing Office, 1979) p. 3. (1996):252-259.
9. E.J. Emanuel, D. Wendler, and C. Grady, "What Makes 30. R. Temple and S.E. Ellenberg, "Placebo-Controlled Trials
Clinical Research Ethical?" JAMA 283 (2000):2701-2711. and Active-Control Trials in the Evaluation of New
10. J. Sugarman, N.E. Kass, S.N. Goodman, P. Perentesis, P. Treatments: Part 1: Ethical and Scientific Issues," Annals
Fernandes, and R.R. Faden, "What Patients Say about of Internal Medicine 133 (2000):455-463.
Medical Research," IRB 20, no. 4 (1998):1-7. 31. T.P. Laughren, "The Scientific and Ethical Basis for Placebo-
11. E.G. Miller, D.L. Rosenstein, and E.G. DeRenzo, "Pro- Controlled Trials in Depression and Schizophrenia: An
fessional Integrity in Clinical Research," JAMA 280 FDA Perspective," European Psychiatry 16 (2001):418-423.
(1998):1449-1454. 32. Department of Health and Human Services. "Protec-
12. R.R. Faden and T.L. Beauchamp, A History and Theory tion of Human Subjects. Code of Federal Regulations."
of Informed Consent (New York: Oxford University Press, 45 CFR 46,1991.
1986):200-232. 33. K. J. Rothman and K.B. Michels, "The Continuing Unethical
13. D.A. Braunholtz, S.J.L. Edwards, and RJ. Lilford, "Are Use of Placebo Controls, NEJM 331 (1994):394-398.
Randomized Clinical Trials Good for Us (in the Short 34. See B. Freedman, K.C. Glass, and C. Weijer, "Placebo
Term)? Evidence for a 'Trial Effect,"' Journal of Clinical Orthodoxy in Clinical Research. I: Empirical and Meth-
Epidemiology 54 (2001):217-224. odological Myths," Journal of Law, Medicine & Ethics 24
14. J.W. Berg, P.S. Appelbaum, C.W. Lidz, and L.S. Parker, In- (1996):243-251; and B. Freedman, K.C. Glass, and C.
formed Consent: Legal Theory and Clinical Practice, 2nd ed. Weijer, "Placebo Orthodoxy in Clinical Research. II:
(New York: Oxford University Press, 2001):280-283. Ethical, Legal and Regulatory Myths," Journal of Law,
15. R.J. Levine, Ethics and Regulation of Clinical Research, 2nd Medicine & Ethics 24 (1996):252-259.
ed. (New Haven: Yale University Press,1986):8-10. 35. R. Temple and S.E. Ellenberg, "Placebo-Controlled
16. J. Katz, "'Ethics and Clinical Research' Revisited: A Tribute Trials and Active-Control Trials in the Evaluation of
to Henry K. Beecher," Hastings Center Report 23, no. 5 New Treatments: Part 1: Ethical and Scientific Issues,"
(1993):36. Annals of Internal Medicine 133 (2000):455-463; E.J.
17. C. Fried, Medical Experimentation: Personal Integrity and Emanuel and E.G. Miller, "The Ethics of Placebo-
Social Policy (New York: American Elsevier, 1974). Controlled Trials-A Middle Ground," NEJM 345
18. Ibid., 5 (2001):915-919.
II J11 1111.1j
36. S. Hellman and D.S. Hellman, "Of Mice but Not Men: 40. B. Freedman, K.C. Glass, and C. Weijer, "Placebo
Problems of the Randomized Controlled Trial," NE JM Orthodoxy in Clinical Research. II: Ethical, Legal and
324 (1991):1585-1589. Regulatory Myths," Journal of Law, Medicine & Ethics
37. E.G. Miller and H. Brody, "What Makes Placebo- 24 (1996):253.
Controlled Trials Unethical?" American Journal of Bio- 41. R. Temple and S.E. Ellenberg, "Placebo-Controlled
ethics 2, no. 2 (2002):3-9. Trials and Active-Control Trials in the Evaluation of
38. N. Dickert and C Grady, "What's the Price of a Research New Treatments: Part 1: Ethical and Scientific
Subject? Approaches to Payment for Research Participa- Issues," Annals of Internal Medicine 133 (2000):
tion," New EnglandJournal of Medicine 341(1999):198-203. 455-463.
39. See E.J. Emanuel, D. Wendler, and C. Grady, 'What Makes 42. C. Weijer, "When Argument Fails," American Journal of
Clinical Research Ethical?" JAMA 283 (2000):2701-2711. Bioethics 2, no. 2 (2002):10.
Protecting Communiities in new Canadian research guidelines (the Tri-Council

policy statement) applied guidelines for the protec-
Biomedical Research tion of aboriginal communities in biomedical research
Charles Weijer and Ezekiel J. Emanuel to a wide variety of other communities, including
Ashkenazi Jews and families." This effort was cur-
tailed as it soon became apparent that many of the
Geneticists and other scientists have increasingly specific protections could not be applied to these
targeted communities for biomedical research into other communities.
the etiology, especially the genetic determinants, Rather than simply applying existing guidelines,
of common diseases and have met with some well- a rational strategy for the development of protections
known successes. For example, particular mutations for communities in biomedical research must entail a
predisposing to breast, ovarian, and colon cancer have series of steps. In discussing this strategy, we will de-
been identified through studies of Ashkenazi Jews.'"Z scribe characteristics and types of communities; these
Although these discoveries will undoubtedly have im- are not meant to be exhaustive lists but to be the ones
portant implications for cancer prevention and treat- relevant to medical research.
ment, the community has expressed concern that they
may become the target of discrimination, and there Community Characteristics
is growing public concern that added protections for
communities in biomedical research are required.3 The term community delineates a wide variety of
Protections for communities in biomedical research human associations, from tribes to municipalities to
have been developed in limited circumstances. The U.S. religious adherents. A single set of regulations to fit
Food and Drug Administration has issued regulations all types of communities is doomed to failure. What
allowing for a waiver of informed consent in certain is needed are morally relevant criteria that distinguish
emergency room research, provided there is public communities. Characteristics of particular importance
disclosure of the research plans and consultation with or relevance to communities in biomedical research can
community representatives.4 Guidelines on consulting be identified and used to delineate seven types of com-
communities involved in research on HIV/AIDS have munities [Table 1; for definitions see Web table 119
been proposed and implemented.5 The most developed Communities may be arrayed along a spectrum
protections for communities in biomedical research are of cohesiveness, from those that have all the charac-
found in guidelines for research involving aboriginal teristics to those that have only a few. At one end of
communities, exemplified by those of the Australia the spectrum, a cohesive aboriginal community often
National Health and Medical Research Council.° (or nearly always) has all of the characteristics listed.
Attempts to generalize these approaches by ex- Conversely, a less-cohesive occupational community
tending them from one community to another have embodies only two of the characteristics: common cul-
been problematic.' For example, a 1996 draft of the ture and a communications network.
Source: From Science 18 (2000) 289,1142-4. Reprinted with permission.

Weijer/Emanuel: Protecting Communities in Biomedical Research
Table 1 Characteristics of Types of Communities in Biomedical Research. Examples are

aboriginal, Kahnawake; geographic/political, Jackson, MI, and Iceland; religious, Amish; disease,
HIV; ethnic/racial, Ashkenazim; occupational, nurses; and virtual, e-mail discussion group
Type of community
Geographic/ Ethnic/
Community characteristic Aboriginal Political Religious Disease Racial Occupational Virtual
Aboriginal Geographic/Political Religious ++ ++ +/- ++
Disease Ethnic/Racial Occupational Virtual
Comprehensiveness of culture ++ +/- ++ +/-
Health-related common culture ++ ++ ++ +/-
Legitimate political ++ ++ +/- +/-
Authority
Representative group/individuals ++ ++ ++ ++ +/- +/-
Mechanism for priority setting in health care +/- +/- +/-
Geographic localization ++ +/- +/- +/-
Common economy/ shared resources ++ ++ +/- +/- +/-
Communication network ++ + +/- +/- ++
Self-identification as community ++ ++ ++ +/- +/-
++The community nearly always or always possesses the characteristic. +The community often possesses the characteristic. +/-The community
occasionally or rarely possesses the characteristic. -The community very rarely or never possesses the characteristic.
Potential Protections for comment, acknowledge community contributions,

seek consent to identify the community, provide a final
Potential community protections extend from the genesis report to the community, and obtain community con-
of the research project to the publication of the results.' sent for media interviews. The researcher must report
Consultation in protocol development. The researcher compliance with guidelines to the Institutional Review
must show respect for the community culture, seek Board (IRB) and the publication.
community input on protocol development, ensure
research is useful to the community, and respect the Connecting Guideline Requirements to
community's knowledge and experience. Community Characteristics
Information disclosure and informed consent. Disclo-
sure to the community should be nontechnical, and It is possible to identify particular characteristics of
appropriate face-to-face meetings are encouraged. The communities that are necessary for the implementation
community ought to have adequate time for review, of specific protections. In this way, each of the identi-
the researcher must obtain the consent of the com- fied community characteristics is linked to one or more
munity, and community consent is required for proto- of the protections (Table 2).
col changes. For example, if the community is to have input on
Involvement in research conduct. The researcher the protocol, the community must have representatives
should ensure that skills and research expertise are who can provide this input on behalf of the community.
transferred to members of the community, employment Similarly, if community consent is to be sought before
is offered to members of the community the commun- individuals are approached for study participation, the
ity is reimbursed for research costs, and the community community must have a legitimate political authority
is informed about research progress. that is empowered to speak authoritatively for, and
Access to data and samples. The researcher must make binding decisions on behalf of, the community,
seek community consent for further use of samples, more than mere representation is required. In addi-
and storage of data should be negotiated. tion, if the community is to be reimbursed for research
Dissemination and publication of results. The re- costs, the community must have a common economy
searcher should involve the community in manuscript or shared resources. Furthermore, community consent
preparation, transmit a draft report to the community for further use of samples requires not only that the
II an II Lid
community have a legitimate political authority but individuals within it may have strong feelings. Finally,
also that they have a health-related common culture. if a draft report for comment is to be provided mean-
Unless there are shared ideas about health, the dispos- ingfully to the community, a communication network
ition of samples is likely to be of little or no relevance must be in place so that the report can be distributed to
to the community as a whole, although particular community members.
Table 2 Appropriate Protections for Communities Depend on Their Characteristics. Italics

indicate community protections that require consent; nonitalics, protections that require only
community consultation
COMMUNITY CHARACTERISTICS REQUIRED FOR PARTICULAR PROTECTIONS
Community characteristics
Proposed protections H LPA Rep. PS GL CE/SR CN SI
Consultation in protocol development
Respect for culture
Input on protocol
Research useful
Respect for knowledge and experience
Process of providing information and obtaining informed consent
Nontechnical and appropriate disclosure
Face-to-face meetings V
Adequate time for review
Consent
Consent required for protocol change.;
May withdraw consent
Involvement in research conduct
Transfer of skills and expertise
Employment
Reimbursement for research costs
Informed about research progress V
Access to data and samples
Consent for further use of samples V
Storage of data negotiated
Dissemination and publication
Involvement in manuscript preparation
Draft report for comment
Acknowledgment V
Consent to identify
Final report V
Consent for researcher media interview
Synthesizing Appropriate Protections consultation, (ii) community consultation alone, and

(iii) no added protections. [See Web table 2.9
Three general regimes of protection can be delineated, Community consent is only possible if the com-
based on the specific protections appropriate to the dis- munity has a legitimate political authority, which could
tinct types of communities: (i) community consent and be a legislative assembly, mayor, or tribal council that
Weijer/Emanuel: Protecting Communities in Biomedical Research
has the authority to make binding decisions on behalf develops, necessitating reconsideration of the level of
of its members. For instance, the Ashkenazim have no protections. For example, it is not meaningful to speak
legitimate political authority, and hence, suggesting of many disease groups, such as asthmatics, as com-
that community consent be sought from them is nei- munities. However, at some time in the future they may
ther morally nor pragmatically justifiable. However, come together for support and to advocate for more
this does not undermine the importance of respect for research funding and a voice in setting research agenda.
communities or the possibility that community con- In this evolution, such a group accrues these same mor-
sent is appropriate for some. ally relevant characteristics that confer the ability and
Not surprisingly, communities that have legitimate obligation to enact protections when the community
political authorities are among the most cohesive com- is targeted for research. Similarly a relatively cohesive
munities and have all or most of the characteristics rel- community may experience disagreements, rupture,
evant to the implementation of guidelines requirements and disintegration over time, losing characteristics and
(Table 1). Thus, protections for aboriginal and geo- thereby losing attendant protections.
graphic or political communities in research indude the
full list of guideline requirements; in other words, com- Possible Questions
munity consent and consultation are required (Table 2).
Even though community consent is not possible How do community protections relate to individual informed
for the Ashkenazi Jewish community, other protections consent? Ultimately, no person can be enrolled in research
that may be characterized as "community consultation" without his or her individual consent. Properly under-
are appropriate. Community consultation encompasses stood, community consent is an additional protection; a
the involvement of community representatives to a lim- study may not proceed without informed consent from
ited degree in study planning, informing the commun- both the community and the individual research subject.
ity as a whole of the study at its start and as progress However, protections for communities are asymmetric
is made, consulting with community representatives If the community consents to research participation, in-
regarding the disposition of data, and providing them dividuals may still refuse to participate; if the commun-
with a draft report on which to comment. Communities ity does not consent, then individuals who are identified
that share a religion, a disease, ethnicity, or race may be because they are members of the community should not
relatively cohesive and share many although not all, be approached for study enrollment. Conflicts may arise
of the characteristics required for the implementation when individual and community interests conflict. For
of guideline requirements. Reflecting the intermediate example, what if the community withdraws consent
degree of cohesiveness, the list of potential protections for study participation when individual research sub-
listed is roughly half that for aboriginal and geograph- jects seem to be deriving medical benefit and wish to
ical or political communities (Table 2). continue participating? Inevitably, the answer to such
Occupational and virtual communities are the least conflicts will depend on the circumstances.
cohesive of the communities in the typology (Table 1). Is it more appropriate to conceive of a community as
Generally, they have few of the morally relevant com- a vulnerable group protected by current regulations? Re-
munity characteristics, and accordingly no added pro- search ethics and protections have largely been shaped
tections are required for research involving them. in reaction to instances of unethical and exploitative
However, there may be exceptions to these guide- research on prisoners, children, the elderly, the poor,
lines. Actual communities are diverse and can deviate and racial or ethnic minorities." Vulnerable groups are
from the ideal types. For instance, for historical and socially, economically, and otherwise disadvantaged and,
social reasons, farm workers or coal miners with strong therefore, are more susceptible to exploitation or harm.
union representation, geographic localization, union- Regulations protecting such groups include added con-
based health insurance, and other social security pro- sent requirements and limits on the nontherapeutic re-
grams may be more cohesive than typical occupational search risk to which they may be exposed.0 Conversely,
communities. Their cohesiveness may be so extensive the driving issue for protections for communities is not
that they have all or most of the characteristics legitim- vulnerability but rather that communities have interests
izing the additional protections of community consent that are entitled to respect and protection.° Respecting
and consultation. and protecting the interests of a community call for a
Furthermore, it is important to recognize that partnership between community and researcher. This is
human associations are not static but dynamic; bonds a fundamentally different relation than with vulnerable
within a group may strengthen over time, and novel groups, for which there is more of a protective guardian-
social structures may emerge as a new community ship.'4 Consequently, the protections typically afforded
IIA
vulnerable groups, such as limits upon risk, would be interaction regarding reimbursement for use of resour-
inappropriate for communities in research. ces, whereas religious leaders have a stake in cases in-
Might a community use added protections for re- volving the disposition of tissue samples.
search to legitimize the oppression of groups within What if the community wants to suppress adverse or
the community? For example, a community with a undesirable research findings? This problem is not re-
male-dominated leadership may silence the voices stricted to research with communities but also exists
of women within the community. Such a community in relation to research funded by pharmaceutical
may be reluctant to permit research into the preva- companies, managed care and other health care insti-
lence and causes of spousal abuse, because it may tutions, or other research organizations.18 Experience
reflect poorly on the community. Careful reflection is in research with the aboriginal community provides a
needed to ensure that the desire to protect the com- useful guide to the negotiation of disparate interests.
munity in research does not perpetuate oppression. Researchers and the Kahnawake community have ne-
All the relevant values, including respect for persons, gotiated a mechanism in which consensus between the
beneficence, justice, and respect for communities, researcher and the community on data interpretation is
must be used in the assessment of potential oppres- sought.19 If consensus cannot be attained within a rea-
sion. Viewed in isolation, any of the ethical principles sonable amount of time, the competing interpretations
articulated in the "Belmont report"15 may be used to of the study will both be published. Further experience
justify unethical ends. One formulation of justice may generate other examples of creative and equitable
requires that the burdens and benefits of research solutions to these problems that are less threatening
participation be distributed equitably; another aspect to academic freedom than many existing agreements
of justice calls for the elimination of domination.16,17 between researchers and for-profit companies.
Thus, a community that seeks to perpetuate oppres-
sion might be legitimately criticized on grounds of Putting Principle into Practice
justice; community safeguards used to perpetuate
such oppression have no moral force. Undoubtedly, difficult questions do remain. By pro-
Who counts as the community leader? In some com- viding precision in distinguishing different types of
munities, a multiplicity of legitimate leaders may make communities in research, their characteristics, and pro-
it difficult to discern with whom researchers ought to tections appropriate for each, this analysis should make
interact; an aboriginal community may have both a tribal discussion of community consent and consultation
council and an elected mayor. The decision will depend more focused. By distinguishing between community
on the values and traditions of particular communities consent and community consultation, skepticism as to
and whose authority encompasses the questions raised. the feasibility and appropriateness of additional protec-
For instance, political leaders may be appropriate for tions for communities in research ought to be allayed.
References and Notes humans" (Public Works and Government Services Canada,
Ottawa, 1998).
1. J. P Streuwing e t al., Nature Genet. 11, 198 (1995). 10. Supplemental material is available to Science Online
2. S. J. Laken et al., Nature Genet. 17, 79 (1997). subscribers at www.sciencemag.org/feature/data/
3. S. Lehrman, Nature 389, 322 (1997). 1050363.shl.
4. 61 Code of Federal Regulations 51497. 11. R. J. Levine, Ethics and Regulation of Clinical Research
5. C. Levine, N. N. Dubler, R. J. Levine, IRB: Rev. Human Subj. (2nd ed.) (Yale Univ. Press, New Haven, CT, 1988).
Res. 13, 1 (January-April 1991). 12. C Weijer, Accountability Res. 7, 21 (1999).
6. Australia National Health and Medical Research Council, 13. C. Weijer, Cambridge Q. Healthcare Ethics 8, 501 (1999).
"Guidelines on ethical matters in aboriginal and Torres 14. B. Freedman, A. Fuks, C. Weijer, Hastings Center Rep. 23,
Strait islander research" (NHMRC, 1991). 13 (March-April 1993).
7. C. Weijer, G. Goldsand, E. J. Emanuel, Nature Genet. 23, 15. National Commission for the Protection of Human Subjects
275 (1999). of Biomedical and Behavioral Research, "Belmont report"
8. Canada Tri-Council Working Group on Ethics, "Code of (Department of Health, Education, and Welfare, 1979).
conduct for research involving humans (draft)" (Minister 16. J. Rawls, A Theory of Justice (Harvard Univ. Press,
of Supply and Services, Ottawa, 1996). Cambridge, MA, 1971).
9. Medical Research Council of Canada, Natural Sciences and 17. M. Walzer, Spheres of fustice (Basic Books, New York, 1983).
Engineering Research Council of Canada, Social Sciences 18. D. G. Nathan and D. J. Weatherall, Lancet 353, 771
and Humanities Research Council of Canada, "Tri-council (1999).
policy statement: Ethical conduct for research involving 19. A. C. Macaulay et al., Can. J. Public Health 89, 105 (1998).
Titus/Wells/Rhoades: Repairing Research Integrity
Repairing Research Integrity have addressed research misconduct incidence with

limited results because of methodological problems,
Sandra L. Titus, James A. Wells, such as applying inconsistent definitions of mis-
and Lawrence J. Rhoades conduct or not accounting for duplicate reports of
the same incident.3.4,5 So, we used the US federal
Misconduct jeopardizes the good name of any institu- definition of research misconduct6—fabrication,
tion. Inevitably, the way in which research misconduct falsification, or plagiarism in proposing, performing
is policed and corrected reflects the integrity of the or reviewing research, or in reporting research
whole enterprise of science. The US National Academy results—and verified whether reports accurately
of Sciences has asserted that scientists share an "obliga- fitted that definition. The possibility of duplicate
tion to act" when suspected research misconduct is ob- reports was virtually eliminated by selecting only
served.' But it has been unclear how well scientists are one National Institutes of Health (NIH)—funded
meeting that obligation. In the United States, the Office researcher in a given department to respond. We
of Research Integrity (ORI) evaluates all the investiga- asked about events only from the past three aca-
tion records submitted by institutions and plays an demic years to avoid inclusion of distant events
oversight role in determining whether there has been and to have a consistent time parameter. We used
misconduct at institutions that receive support from frequent and varied reminders to secure a high
the Department of Health and Human Services (DDHS). response rate to the survey. Previous research has
The reported number of investigations submitted to treated survey reports of misconduct as if the ob-
ORI has remained low: on average 24 institutional in- server could make the determination that they had
vestigation reports per year.2 observed misconduct. Instead, we consider the ob-
ORI focuses resources, not only on evaluating servations to be "possible research misconduct" and
institutional reports of research misconduct but also not all such observations will result in a finding of
on preventing misconduct and promoting research misconduct. In all we asked 4,298 scientists holding
integrity through deterrence and education. To evalu- NIH extramural research funds at 605 institutions to
ate these initiatives, we investigated whether the low respond to the survey so that our findings would be
number of misconduct cases reported to ORI is an representative of a broad spectrum of research fields
accurate reflection of misconduct incidence, or the as well as varied sizes of institutions.
tip of a much larger iceberg. The latter seems to be
the case. What Scientists Saw
The 2,212 researchers we surveyed observed 201
instances of likely misconduct over a three-year period. In 2006, we asked participants to indicate the number
That's 3 incidents per 100 researchers per year. A con- of times they had observed suspected research mis-
servative extrapolation from our findings to all DHHS- conduct in their own department in the past three
funded researchers predicts that more than 2,300 academic years (2002-05). A total of 2,212 scientists
observations of potential misconduct are made every provided complete responses to questions concerning
year. Not all are being reported to universities and few research misconduct (51 percent response rate). Of
of these are being reported to the ORI. these, 192 scientists (8.7 percent) indicated that they
No regulatory office can hope to catch all research had observed or had direct evidence of researchers in
misconduct and we think that the primary deterrent their own department committing one or more inci-
must be at the institutional level. Institutions must dents of suspected research misconduct over the past
establish the culture that promotes safeguards for three academic years. The 192 scientists described a
whistleblowers and establishes zero tolerance both for total of 265 incidents.
those who commit misconduct and for those who turn Scientists were asked to indicate how they became
a blind eye to it. aware of the possible misconduct and were told to
report observations and not hearsay. Suspected mis-
conduct was observed at all scientific ranks including
Defining Misconduct
postdocs, students, and tenured faculty members. The
A first step in developing that culture is taking stock following are examples of how scientists described
of misconduct's frequency. Several investigators such incidents. We used these descriptions to validate
Source: From Nature 453 (19 June 2008), 980-2, doi:10.1038/453980a.

II JE
whether the observation met the federal definition of to 155,000 researchers suggests that there could be,
research misconduct. minimally, 2,325 possible research misconduct ob-
servations in a year. If 58 percent of these cases were
"A post doc changed the numbers in assays in reported to institutional officials as in our survey, ap-
order to 'improve' the data." proximately 1,350 would have been reported whereas
almost 1,000 could be assumed to go unreported to
"A colleague duplicated results between three dif- any official.
ferent papers but differently labelled data in each These numbers indicate a sizeable disconnect
paper." between what universities are seeing and the 24 in-
vestigations evaluated by the ORI annually. Could all
"A co-investigator on a large, interdisciplinary the predicted cases be found to lack evidence? Could
grant application reported that a postdoctoral all the cases be concluded at the inquiry stage? Could
fellow in his laboratory falsified data submitted as the cases be primarily occurring in research that is not
preliminary data in the grant. As principal investi- funded by the Public Health Service and hence not
gator of the grant, I submitted supplementary data reportable to the ORI? Can duplicate observations of
to correct the application " misconduct account for this disparity?
We doubt that affirmative answers to these ques-
"A colleague used Photoshop to eliminate back- tions could sufficiently explain the discrepancy. We
ground bands on a western blot to make the data recognize that this estimate is not perfect. First we are
look more specific than they were." applying our findings from a defined context to a much
larger context and one that also includes the staff of
Two people independently coded and evaluated the investigator. Another weakness of the prediction is
the 265 descriptions to determine whether each met the fact that scientists in our study would have been
the federal definition of research misconduct. In all, narrowly reporting observations restricted to their own
64 reports (24 percent of the total) did not meet the experience. A single observer in a department cannot
threshold of the federal definition—which left 201 ob- be expected to have been exposed to all instances of
servations of potential misconduct made by 164 scien- misconduct. Thus, our estimate may be off by an order
tists (7.4 percent). These 201 misconduct observations of magnitude in either direction.
included fabrication or falsification (60 percent) and On an individual level, many reasons for under-
plagiarism only (36 percent). reporting are easy to understand because they in-
According to our respondents, 58 percent of the volve motivations we might all have experienced.
observed incidents had been reported to officials at For example, one does not want to accuse falsely.
their institutions. In 24 perce:at of incidents it was the One may also fear that reporting would take time
survey respondent who reported it and in 33 percent of away from research, or have concerns and fears
the incidents it was someone other than the respondent. about possible retaliation. One may assume some-
Responses indicated that 37 percent of incidents were one else will or should report it. Or one may have
not reported by anyone and for 5 percent of the cases sympathy toward a researcher, and might think "it's
respondents did not know not too bad," it can be sorted out without a career-
damaging investigation. Reporting also necessitates
Study Limitations
confidence that the issue will be examined carefully
and thoroughly.
Extrapolating the survey results—even conservatively—
projects an alarming picture of under-reporting. NIH Keeping It Quiet
extramural research grants in 2007 supported an esti-
mated 155,000 people, which includes principal inves- The leaders of institutions may also have concerns
tigators and other research personnel.' In our survey, about handling research misconduct. Because public
201 cases were observed over three years by 2,212 re- image is important to institutions, some may try
spondents, essentially 3 cases per 100 people per year. to minimize reporting and keep unfavourable in-
Most conservatively, we assumed that non-responders formation from reaching the ORI and the press. An
(roughly half of our sample) did not witness any mis- institution may choose to ignore conducting an in-
conduct. Thus, applying 1.5 cases in 100 scientists vestigation and instead they may simply dismiss an
Titus/Wells/Rhoades: Repairing Research Integrity
accused person or even a whistleblower in the hope Train the Mentors

that the problem will go away without needing further If we want to build a stronger culture of integrity,
examination. Additionally, institutional leaders may then the current generation of researchers has to be
wish to ignore or minimize allegations of possible educated to pay more attention to how they work
research misconduct to protect the revenue that the with their junior team members. Social science has
researcher generates; some may avoid investigations a long history of describing how group standards
because they are costly in terms of time and money. affect individual behaviour. Mentors specific-
Administrators may not recognize the significance of ally need to become more aware of their roles in
evaluating research misconduct and of course they establishing and maintaining research rules and
may be poorly equipped to conduct an investigation minimizing opportunities to commit research mis-
in an appropriate manner. conduct.10 Only 34 percent of scientists in a study
Fundamentally all explanations seem to share a with 2,206 laboratory directors strongly agreed that
common denominator—the failure to foster a culture their mentor had prepared them to be a good mentor
of integrity. An analysis commissioned by the ORI found to others.11 An institutional investment in building
in 2000 that only 29 percent of institutional miscon- better mentors is an important vehicle to promoting
duct polices explicitly obligate members to report sci- research integrity.
entific misconduct.8 Individuals and institutions, not
the federal government, are the guardians of research Use Alternative Mechanisms
integrity. Therefore, we urge action and recommend six Institutions must start to use other means to protect
strategies to champion integrity. the integrity of their studies. The Institute of Medicine
recommends that "Universities should not rely upon
Adopt Zero Tolerance formal complaints of scientific misconduct as the sole
To create a zero-tolerance culture, we think that it is source of monitoring the integrity and quality of the
essential that an institution specifies and implements research conducted under their auspices. They need
the requirements that all suspected misconduct continuing mechanisms to review and evaluate the re-
must be reported, and all allegations must be thor- search and training environment of their institution."12
oughly and fairly investigated. Social responsibility Auditing research records would be one such means.
to the academic community and to the public who Mechanisms of review are needed to reduce deficient
fund the research will be strengthened when it is record keeping, improper protection of human or
apparent that an institution has a real commitment animal subjects or the utilization of questionable re-
to integrity. search behaviour.13
Protect Whistleblowers Model Ethical Behaviour

Careful attention must be paid to the creation and People imitate the behaviour of powerful role
dissemination of measures to protect whistleblowers. models. Institutions successfully stop cheating, for
Responders to our survey said that reporting would be example, when they have leaders who communi-
most likely to improve if institutions and the federal cate what is acceptable behaviour, encourage faculty
government increased the whistleblower protection. members and staff to follow the policies, develop fair
Indeed, more than two-thirds of whistleblowers, in a and appropriate procedures for handling misconduct
Research Triangle Institute study, experienced at least cases, focus on ways to develop and promote eth-
one negative outcome as a direct result of their actions.9 ical behaviour, and provide clear deterrents that are
Plus, 43 percent reported that institutions encouraged communicated.14
them to drop the allegation. Nearly one generation after the effort to reduce
misconduct in science began, the responses by NIH
Clarify How to Report scientists suggests that falsified and fabricated re-
Researchers in our study also emphasized what would search records, publications, dissertations, and grant
promote reporting: establishing a reporting system that applications are much more prevalent than has been
clearly identifies the individuals to whom allegations suspected to date. Our study calls into question the
should be brought, and establishing clear policies, effectiveness of self-regulation. We hope it will lead
procedures and guidelines related to misconduct and individuals and institutions to evaluate their commit-
responsible conduct. ment to research integrity.
II VIII IIILLI
References
1. http://www.nap.edu/html/obas/. 8. http://ori.dhhs.gov/documents/institutional_policies.
2. httpi/ori.dhhs.gov/research/intra/documents/Investig pdf.
ations1994-2003-2.pdf. 9. http://ori.hhs.gov/documents/consequences.pdf.
3. Swazey, J., Anderson, M. & Lewis, K. Am. Sci. 81, 542-553 10. Adams, D. & Pimple, K D. Account. Res. 12, 225-240 (2005).
(1993). 11. http://ori.dhhs.gov/documents/research/intergity_meas-
4. St James-Roberts, I. New Sci. 72, 466-469 (1976). ures_final_report_11_07_03.pdf.
5. Rankin, M. & Esteves, M D. Nurs. Res. 46, 270-276 12. Institute of Medicine, The Responsible Conduct of Research
(1997). in Health Sciences (National Academies Press, Washing-
6. Department of Health and Human Services Public ton DC, 1989).
Health Service Policies on Research Misconduct 42 CFR 13. Martinson, B., Anderson, M. & DeVries, R. Nature 435,
93 (2005). 737-738 (2005).
7. Lederhendler, I. National Institutes of Health, personal 14. McCabe, D. L., Trevino, L. K. & Butterfield, K. D. Ethics
communication. Behay. 11, 219-232 (2001).
9.4 Use of Vulnerable Populations in Research
Vulnerability in Research and requires us to respect "the right balance between this
logic of the struggle for immortality and the finitude
Health Care; Describing the of the earthly presence of human suffering.'3 This and
Elephant in the Room? other such broad definitions4 encompass humanity
in its entirety. At the restrictive end of the spectrum,
Sarnia A.Hurst
"vulnerability" in research on human subjects is often
applied to individuals who are unable to give informed
Introduction
consent or who are more likely to be exploited.5 These
Broadly, we agree that the vulnerable should be af- restrictive definitions have been critiqued, even in this
forded some kind of special attention, or protection. context, as both "too broad and too narrow" by authors
Defining vulnerable persons or populations, however, who proposed to replace special scrutiny for vulnerable
has proved more difficult than we would like. This is populations with ongoing protection of individuals ac-
both a theoretical and a practical problem. On a theor- cording to existing regulations, and focused attention
etical level, uncertainty as to what we mean by vulner- on "characteristics of the research protocol and en-
ability is unsatisfactory because although we agree that vironment that present ethical challenges." 6
this notion has a strong pull., we cannot account for The definition of vulnerability for the purposes
this pull, justify it, or define its limits. On a practical of health care and research with human participants
level, we cannot know who should be afforded the is thus an unanswered question. That such an im-
protection due to vulnerable persons, or what form portant question is open should be regretted. Giving
this protection should take. Contradictory definitions up the concept could represent abandonment of a
can lead to confusion for those who are supposed to needed moral safeguard.' A definition that includes
protect the vulnerable, and wrong definitions may be humanity itself does not provide such an account,
acted upon. as it cannot provide reason for special protection. A
Attempts to define vulnerability have differed in restrictive definition will thus be required. It cannot,
their scope.' At the broad end, we find a European however, be restrictive to the point of forcing us
"principle of vulnerability," which should be considered either to exclude persons that should validly be con-
as a universal expression of the human condition2 and sidered vulnerable, or to pretend they fulfill some
Source From Bioethics 22, 4 (2008), 191-202. Reprinted with permission

Hurst: Vulnerability in Research and Health Care
other criteria for vulnerability when in fact they do anchoring vulnerability to consent is tempting. We let
not. For example, if we consider that vulnerability people make their own choices and count on this to
in research ethics is centred on the inability to give protect them. So we should only afford better protec-
voluntary informed consent, we may have a convin- tion to those less able to protect themselves in this
cing case for excluding terminally ill patients from way. Human subjects research, however, present us
vulnerable groups. If we remained convinced that with examples of human activities where consent is
they are nevertheless vulnerable, our argument could a necessary but insufficient condition to ethical prac-
only be that they are indeed incapable of giving valid tice.12 The same is true to a degree in health care as
informed consent. As it has been pointed out, this is well. As a patient, I will often lack crucial information
unsatisfactory.8 and will thus need health care providers to have my
Fortunately existing definitions seem mistaken interest at heart more than in other types of choices.
only in part and may be mutually complementary. De- In some clinical situations, decisions have the struc-
scriptions sometimes resemble those of the proverbial ture of a "prisoners' dilemma": what we consider to
elephant described by people with only partial views. be our best option can only be chosen if we are suf-
In this paper, I will show how existing definitions of ficiently confident that others will choose the same
vulnerability in medical research and clinical care are option. If everyone chooses generic substitution for
insufficient, and attempt a definition of vulnerability a benign ailment, the money available will benefit
as a claim to special protection that is both compre- everyone including those in situations of greater need.
hensive and usable. I will also address some possible If, however, I am the only one to sacrifice the small
concerns with this proposal. degree of comfort involved by, for example, taking a
pill less comfortable to swallow, no significant benefit
will accrue and I will lose out.13 In both of these ac-
What Is Vulnerability in Research
and Health Care?
tivities, we do not, in fact, expect consent to do all the
protective work. Whether this is because we believe
Restrictive definitions of vulnerability in research and that consent is often flawed to the degree of requir-
health care can be roughly described as consent-based, ing other safeguards,14 or because we believe it to be
harm-based, or comprehensive. intrinsically insufficient, does not change the impact
Consent-based definitions include that pro- this has on vulnerability. If additional safeguards
posed by the ICH tripartite guidelines,9 or by CIOMS, other than consent are required for everyone, and if
which defines the vulnerable as "Those who are vulnerability can exist in relation to these other safe-
relatively (or absolutely) incapable of protecting guards, then consent-based accounts of vulnerability
their own interests." 1° The Belmont report defines are insufficient.
the vulnerable explicitly on grounds of "their de- A variant of this view expands consent-based
pendent status and their frequently compromised vulnerability to include limits on the ability to avoid
capacity for free consent." 11 Perhaps in an attempt exploitation.15 But this sort of definition turns out
to complement their definitions and give a more to be either too broad or too narrow, depending on
comprehensive picture, guidelines for ethical con- what we mean by exploitation. If we include harm,
duct of human subjects research have also provided disrespect, and injustice° within the definition of
lists of vulnerable groups (Table 1). Although they exploitation, we are left with a poor understanding
do provide useful examples, these lists can be long, of the sort of wrong exploitation is. A stricter defin-
and lack an organizing principle. It is not clear that ition such as an "unfair distribution of burdens and
they are based on the definitions offered by the same benefits from an interaction," 17 or treating someone
guidelines, or even on a solid family resemblance "in a way to which he could not possibly consent." 18
between the listed groups. is more useful but means that the consent-based
More basically, however, it is not clear that the concept of vulnerability is not expanded very much
sort of thing we mean by vulnerability should refer by the inclusion of an added risk of exploitation.
strictly to being at risk of giving faulty consent. Moreover, if exploitation is a consent-based kind of
This understanding of vulnerability is appealing in wrong, then the problems outlined above will also
situations where we do, indeed, count on people's apply here. For example, if there is indeed a right to
choices to protect them. If we believe that informed access to health care, then being denied such access
consent is the principal protection of human subjects would constitute a wrong. It would, however, cer-
of research, and crucial in clinical care as well, then tainly not amount to exploitation. In research, not
II ,MI nu.1.1
Table 1 Examples of Vulnerability in International Guidelines for Research Ethics

Source Cited examples of vulnerability in human subjects research
Belmont report • Racial minorities
• The economically disadvantaged
• The very sick
• The institutionalized
45 CFR 46 • Children
• Prisoners
• Pregnant women and fetuses
Declaration of Helsinki • Incompetent persons

• Persons susceptible to coercion
• Persons who will not derive direct benefits from participation
• Persons for whom research is mixed with clinical care
CIOMS • Those with limited capacity or freedom to consent or to decline to consent ... [including] children,
and persons who because of mental or behavioural disorders are incapable of giving informed
consent
• Junior or subordinate members of a hierarchical group ... [such as] medical and nursing students,
subordinate hospital and laboratory personnel, employees of pharmaceutical companies, and
members of the armed forces or police
• Elderly persons
• Residents of nursing homes
• People receiving welfare benefits or social assistance and other poor people
• The unemployed
• Patients in emergency rooms
• Some ethnic and racial minority groups
• Homeless persons
• Nomads
• Refugees or displaced persons
• Prisoners
• Patients with incurable disease
• Individuals who are politically powerless
• Members of communities unfamiliar with modern medical concepts
ICH tripartite guidelines • Members of a group with a hierarchical structure such as medical, pharmacy, dental, and nursing
students, subordinate hospital and laboratory personnel, employees in the pharmaceutical
industry, members of the armed forces, and persons kept in detention
• Patients with incurable diseases
• Persons in nursing homes
• Unemployed or impoverished persons
• Patients in emergency situations
• Ethnic minority groups
• Homeless persons
• Nomads
• Refugees
• Minors
• Those incapable of giving consent
giving research subjects the result of the project they to compound additional harms.2° This susceptibility is
participated in can signify a lack of respect but it present when we become biologically weak or diseased,
would be a stretch to call it exploitation. Whether we and this is what justifies additional protection. This,
count on Institutional Review Boards ORB), research- however, only recognizes added likelihood of additional
ers,19 or clinicians to provide additional safeguards harm. To give a counter-example, women and children
for non consent-based wror gs, we do recognize that may frequently be at greater risk of incurring harms
such wrongs exist. but—a ccording to this view—this may not count
Alternatives to consent-based views of vulner- unless they have already incurred a first harm. Further-
ability include harm-based definitions. One such view more, it is not clear that people who are at higher risk of
accepts a broad definition of vulnerability as universal being disrespected or of giving invalid consent would
human fragility, and goes on to define the sort of con- be included under this definition. What these persons
cept useful to health care and research as susceptibility risk is being wronged, rather than being harmed. Thus,
Hurst: Vulnerability in Research and Health Care 529
though it captures important aspects of the intuitions could not ground any claim that ins give greater than
that drive our notion of who is vulnerable and re- usual protection in this case to the specific interest of
sembles safeguards sometimes intended specifically to living a longer life. Moreover, in some instances our
exclude persons at increased risk of harm in research,2' vulnerability may lie in a greater likelihood of requiring
this definition is too narrow. In a way it is the mirror protection in the first place. This will happen whenever
image of consent-based views of vulnerability and our interests are discounted, however easy they may be
shares the same fault: insufficient comprehensiveness. to protect if the concern is present. Garcia proposes to
Faced with the risk that a definition of vulnerabil- base the requirement for special scrutiny on the con-
ity may be too narrow, some have proposed combined cept of equal protection, which requires justifications
or otherwise more comprehensive definitions.22 Vul- for unequal treatment.26 Affording equal ethical protec-
nerability may, for example, be considered to include tion, however, is not limited to avoiding unequal treat-
"groups of people (i) whose capacity to safeguard their ment per se and can include using different treatments
own interests as research participants, through the pro- to compensate higher risks.
cess of informed consent or refusal, is compromised; Mixed definitions have been offered for vulner-
or (ii) who are more likely to take on the burdens of ability in clinical care also, and tend to focus on the
participation in research, in virtue of some feature they risk of neglect and lack of access to care.27 The Agency
share, and this is not compensated for by other suit- for Healthcare Research and Quality defines vulnerable
ably related benefits (not money), or, (iii) who are less populations as those less able to safeguard their own
likely to gain the benefits of participation in research, needs and interests adequately, a view based on con-
in virtue of some feature they share, and this is not sent or self-determination, but also as populations who
compensated for by other benefits to them (or to others may incur different health outcomes traceable to un-
similarly situated)."23 This bases vulnerability on both warranted disparities in their care, or stemming from
consent and fairness in subject selection. This view is special needs for care or barriers to care (Table 2).28
appealing because it covers both equity and freedom. The prominence of barriers to care, or the risk of
Inasmuch as there are wrongs other than transgressing neglect in this wording, reflects the visibility of lack of
the requirement for consent or fairness, it will however access as a potential wrong incurred in clinical care.
be insufficient as well. If we are more likely to have our Lack of access to health care, or even to health, is how-
confidential information disclosed, for example, this ever not the only source of vulnerability in clinical care.
definition will not consider us to be vulnerable. Individuals can be wronged without incurring loss of
A promising broad definition is offered by Agrawal, their opportunity to achieve maximum possible health,
who views vulnerability as "increased potential that if they are deprived of the means for self-determination.
one's interests cannot be protected."24 As he correctly People at greater risk of having confidential information
notes, labelling the inclusion of vulnerable subjects as disclosed are also left out of these definitions.
automatically unethical is incorrect. It can put these The difficulties involved in navigating between in-
people at a disadvantage, for example if their systematic sufficient comprehensiveness and excessive broadness, if
exclusion from research leads to missed opportunities all are to be considered vulnerable,29 have fuelled a cri-
to gain knowledge useful to them.25 This definition tique of using the concept of vulnerability in research at
recognizes that we have all sorts of interests that may al1.3° The authors of this critique argue that the concept
require protection, and that we may be at risk as re- of vulnerability has lost force through the inclusion of
gards each of them. Ethical research with vulnerable too many groups identified as vulnerable. This may be
populations requires more of investigators and IRBs. In true, but only insofar as individuals who belong to such
some cases, this greater effort will be successful, and groups are systematically and sometimes inaccurately
the interests of vulnerable persons will be protected in labelled as vulnerable. This is an avoidable practice.3' In
the end. Sometimes, however, this will not be feasible addition, classifying groups as vulnerable can be stereo-
without, for example, excluding them from a proto- typing: for example classifying the poor or pregnant
col. In some instances, however, the reason why our women as vulnerable is insulting if we mean that they are
interests cannot be protected is that some of them are not capable of decision-making. If we recognize that their
limited by human finitude. In such cases, our interests vulnerability is due to greater likelihood of being offered
will be truly impossible to protect in a way that places a bad risk-benefit ratio or of being disrespected, then this
them clearly outside the responsibility of clinicians and classification constitutes a justified attack on the perpe-
researchers. Someone who is at a high risk of dying trators rather than a slur on the victims. The authors also
of a terminal disease is certainly at "increased poten- note that vulnerability is often understood in relation
tial" that her "interests cannot be protected." But this to enrolment in research, whereas certain individuals
uu I II .111 II I N.L1
Table 2 Examples of Vulnerability in Clinical Care

Source Cited examples of vulnerability in health care
Agency for Healthcare • Populations less able than others to safeguard their own needs and interests adequately
Research and Quality • Populations who may incur different health outcomes traceable to unwarranted disparities in their
care or stemming from special needs for care of barriers to care
Aday • Social status
• Age: infants, children, adolescents, elderly
• Gender: females
• Race and ethnicity: African Americans, Hispanics, native Americans, Asian Americans
• Social capital
• Family structure: living alone, female head
• Marital status: single, separated, divorced, widowed
• Voluntary organizations: non-member
• Social networks: weak
Human capital
• School: less than high school
• Jobs: unemployed, blue collar
• Income: poor, low income
• Housing: substandard
Danis and Patrick Those at risk at any particular point in time for unequal opportunity to achieve maximum possible
health and quality of rife because of differences in intrinsic and extrinsic resources that are associated
with good health
• Financial circumstances
• Place of residence
• Cultural background and ethnicity
• Age
• Health conditions (such as terminal illness or mental illness, impairments, including psychological
and cognitive ones, and functional status or disability, such as inability to communicate effectively)
require ongoing protection. Although it does not tell us wrong means that we expect self-determination to do
what vulnerability is, this is an important point. more work in their case. If a claim exists that we should
afford the same protection to all regarding a claim we
consider valid, then additional requirements other than
Vulnerability as a Claim to Special Protection
complementing faulty consent will be required.
I propose that vulnerability as a claim to special protec- This definition is restricted to wrongs, including
tion should be understood as an identifiably increased wrongful harms and the wrongs that we incur when
likelihood of incurring additional or greater wrong. Vul- something to which we have a valid claim is denied us.
nerability in this sense is not restricted to the likeli- It cannot extend to any additional harm, or any inter-
hood of faulty consent or even to the limited capacity est more likely to be difficult to protect, because it is
to defend one's own interests. If we understood the not the case that we have a duty to protect all interests
worst, or perhaps the only, wrong to be lack of respect from all harms. I could, for example, decide to enrol
for self-determination, and that no harm is done to in research as a sales representative to be introduced
the willing, then this definition would be identical to to potential future customers. If this did not function
previous ones. The examples of research and clinical as expected, one of my interests would certainly have
care, however, show us situations where we do not been harmed, but we would hardly expect an IRB to
expect even ordinary patients or subjects to protect all protect me from this sort of frustration.
their interests themselves. When even ordinary, non- This definition requires vulnerability as a claim to
vulnerable people are not expected to protect their own special protection, to be defined starting from the sorts
interest, then additional elements become visible. Some of wrongs likely to occur. We thus agree with Levine
interests are more likely than others to be placed at and colleagues,32 who propose that attention must be
risk. They may be harder to achieve and thus to defend. focused on the characteristics of the research protocol
This affects some individuals even when they would and environment, rather than restricted to characteris-
be completely capable of defending their own interests tics of potential subjects, and that it should regard the
in other settings. Even for those able to consent, an ongoing conduct of research as well as enrolment. In the
increased likelihood of incun-ing additional or greater definition proposed here, there is no single specific
transgression linked with vulnerability. The substantial Applying this definition of vulnerability could take
contents both of vulnerability itself and of the transgres- the form of a four step approach:
sions associated with it will change with the nature of
the wrongs involved. In one sense or another, many 1. Is there an identifiable potential wrong?
individuals will be vulnerable, but this will not mean 2. If yes, are some people identifiably more likely
that they are vulnerable in identical ways. This could than others to incur this wrong, or likely to incur
partly explain the difficulties in defining vulnerability it to a greater degree?
for the purposes of research with human subjects and 3. Who shares in the duty to minimize, or avoid, this
clinical care. We seem to be caught between only two wrong, and does it include us in any way?
alternatives: defining everyone as vulnerable or sticking 4. What should we do to minimize this increased
to notions based on a very limited number of wrongs. likelihood or degree, or to compensate for it in
These in turn either fail to recognize some valid claims ethically justifiable ways?
to special protection, or require conceptual contortions
such as that which consists in describing someone as The mere definition of vulnerability does not
vulnerable on a consent-based view when their capacity identify those with a share in the responsibility for
to consent should really not be questioned. This prob- protecting the vulnerable. Applying this definition thus
lem is solved if we define vulnerability as proposed here. requires the addition of step 3 (Figure 1). Depending
This definition is limited in the sense that it does on the sort of wrong identified, those responsible for
not identify all forms of vulnerability, such as those as- preventing it will vary.
sociated in general with being human, fallible, mortal,
and capable of suffering. It is, however, relevant in an
Application to Research Ethics
important way, as it circumscribes the forms of vulner-
ability requiring additional attention as compared with If we accept the proposed definition, it is not surprising
the care we usually take to avoid perpetrating wrongs. It that vulnerability proves hard to define: it is as multiple
is also limited as a definition in that it does not provide as potential wrongs and as sources of greater likelihood of
a clear cut-off line between the vulnerable and the non- suffering them. It also involves some judgment, because
vulnerable. Inasmuch as some individuals and groups it is a matter of degree. This, however, does not void
will indeed be identifiably more likely to suffer wrongs, the concept of its usefulness. We may disagree in a grey
however, it provides a framework both for recognizing zone; but differences in the likelihood of suffering wrong
these groups systematically and for designing ways to will often be sufficiently marked to be uncontroversial.
address their specific kinds of vulnerability. Using the concept of vulnerability in research
If an identifiably increased likelihood of incur- ethics has proved difficult for the same sort of reason.
ring additional or greater wrong exists, including any General principles of research ethics exist to protect
wrongful harm, then there is an increased risk of moral subjects from wrongs, including wrongful harms.
transgression. Vulnerability in this sense is thus a two- Thus, the special scrutiny required by IRBs to deal
way street and affects those who practise health care with vulnerability is not a difference in kind but a
and research as well. This will be the case any time that difference in degree of care for ethical criteria that
an identifiable agent acts in way that predictably affects are, indeed, the same. Again, the concept of vulner-
vulnerable persons. ability is not rendered useless by this, as it serves to
identify groups of individuals that do, in fact, need
.........................
..•••
.•".
• •
Identifiably Application of .4.„ A share
greater vulnerability in the duty
likelihood or as a to avoid
likely degree of claim to identifiable
wrongs pecial protection/ wrongs
...............................
.
Figure 1 Two Ingredients to Apply Vulnerability as a Claim to Special Protection
1 ,111 IIIISI
and merit this special care in the application of cri- As outlined above, the relevant question for an IRB at
teria for ethical research. the third step will be "does this include us"? rather than
One application to research ethics, then, is to aid "are we solely responsible for this?." This is an important
IRBs in applying special scrutiny to protect vulnerable point, as understanding protection of the vulnerable in
subjects but in a targeted way. The first of our four research in this way would expand the responsibilities
questions would thus be what IRBs examine as part of of IRBs. Rather than checking a list of predefined vul-
their regular work: nerable groups, they would have to identify who was
vulnerable based on the wrongs likely to occur in the
1. Are any potential research subjects at risk of being case of each protocol they reviewed. This is consistent
wronged in any way by this research project? with their role. Before being regulatory institutions,
For each potential wrong, the other three ques- IRBs primarily have a moral function to protect human
tions would then become: subjects of research. Moreover, this expansion of their
2. Are some potential subjects identifiably more role would remain limited to wrongs actually linked to
likely than other persons to incur this wrong, or research. They would not become responsible for pro-
likely to incur it to a greater degree? tecting the vulnerable from any kind of wrong whatso-
3. Is our IRB among those who share i n the duty to ever. Rather than going through a list of pre-identified
minimize, or avoid, this wrong? vulnerable populations, then, IRBs would go through a
4. If yes, what should we do to avoid this wrong, or list of potential research-related wrongs (Table 3).
minimize this increased likelihood or degree, or Another application to research ethics could be in
ensure it is compensated in ethically justifiable ways? thinking through double standards. Critiques have put
forward that lower ethical standards are used in develop-
Based on recognized requirements for ethical re- ing countries.36 While this is often true, this accusation
search with human subjects,I5 examples of vulnerabil- is sometimes leveled at protocols that actually use all the
ity are shown in Table 3. Actions required for special same standards used in rich countries. The Tenofovir trial
protection would need to be tailored to the sort of was one such example.37 A possible interpretation using
wrong to be avoided, and to the source of the specific the definition of vulnerability proposed here could run
vulnerability (Figure 2). like this. If subjects in developing countries are, on the
Table 3 Using This Definition in Research with Human Subjects

Requirements Examples of vulnerability
Social or scientific value • Lack of access to either benefit or knowledge derived from research
Scientific validity • Rare disease, leading to difficulties in reaching statistical power to demonstrate therapeutic
effectiveness
Fair subjects selection • All persons likely to be victims of discrimination
Favorable risk-benefit ratio • Potentially higher risks: unstable patients, emergency research, fetuses, pregnant women
• Potentially lower benefits: subjects in phase I studies, terminally ill patients
• Subjects whose risk-benefit ratio might sometimes be the object of lesser concern to those
responsible for protection:terminally ill patients, disenfranchised persons, poor subjects in
developing countries, subjects without access to health care outside of research
Independent review All persons likely to be victims of discrimination, if those responsible for review share
discriminatory views
Informed consent Difficulties in receiving or understanding the relevant information: not knowing the language
used, or how to read
Lack of decision-making capacity: some children, some patients with mental disorders,
comatose patients
Lack of freedom to make a voluntary choice
• Through limited freedom: prisoners
• Through social weakness: minorities, refugees, sometimes women
• Through hierarchical weakness lab employees, students
Respect for potential and enrolled Health care providers, researchers and students close to the study team who are at increased
subjects risk of faulty confidentiality
Groups and communities at risk of stigmatisation in the interpretation of study results
• Health care providers are at greater risk.

Are potential research
Example 1: • IRBs share in the duty of protection.
subjects at risk of
breach of • Minimization: could require specific
being wronged by this
confidentiality anonymisation of data to limit colleagues'
research project?
access to their personal information.
Are some potential • If they stand to benefit less, terminally ill

subject identifiably patients may be at greater risk.
Example 2:
more likely than other • IRBs share in the duty of protection.
unfavourable
persons to incur this • Their risk/benefit ratio should be specifically
risk/benefit
wrong, or likely to examined by researchers and IRBs rather than
ratio
incur it to a greater assumed to be the same as for other potential
degree? subjects.
• Subjects of emergency research lack time to

think through the options.
• IRBs share in the duty of protection.
Is our IRB among • This can be minimized if consent is asked at that
Example 3:
those who share in the time onlyfor those parts of the protocol that are
duty to minimize, or being enrolled
truly urgent.
avoid, this wrong? without valid
consent • The remaining problems with consent at that
time can be compensated by including a
requirement that an independent clinician
confirm that enrolment is not contrary to the
potential subject's interest.
• Patients in developing countries who lack access

to care are excluded from an important part of
the social benefits of research.
If yes, what should we
• Although IRBs are not alone in bearing some
do to avoid this wrong, Example 4:
responsibility for this, it is among the points they
or minimize this being denied
should examine in general, and thus also for the
increased likelihood or the benefit of
purposes of protecting the vulnerable.
degree, or ensure it is research
• Minimization: reasonable availability33 aims to
compensated in
minimize this problem.
ethically justifiable
ways? • Compensation:fair benefits34 aim to compensate
it.
Figure 2 Examples of Application of This Approach by IRBs
whole, less likely than subjects in Western countries to Application to the Ethics of Clinical Care
see their claims met, additional efforts may be required
for them. Not just equivalent efforts, but actually greater Basic requirements for the ethical conduct of research
ones. This also helps to distinguish problematic double with human subjects are relatively uncontroversial. The
standards based on research-related wrongs from those nature of valid claims in clinical care is debated to a
that may be problematic based on other sorts of wrongs. greater degree. Application of the concept of vulnerability
Not providing additional protection against the risk proposed here to this field can be expected to reflect this.
that dangers to the poor could be discounted is clearly Importantly, however, this does not invalidate the use of
problematic and within the remit of IRBs. Not provid- this notion. We should expect greater controversy as to
ing adequate roads may very well be problematic, but what constitutes a valid claim; but once we admit a claim
lack of roads is clearly not a research-related risk and as valid, we become able to identify vulnerable popula-
thus outside the remit of IRBs. Not providing access to tions as those more likely to be denied fulfillment of this
treatment will be harder to attribute precisely because it claim, and to identify measures likely to prevent this.
is not quite clear to what extent it can be considered a Examples of requirements proposed for ethical
research-related wrong. Once we know this, however, ap- clinical care are shown in Table 4. They include having
plying the definition of vulnerability presented here does access to health care, adequate financial coverage, not
distinguish double standards that IRBs should concern being harmed, self-determination, confidentiality, get-
themselves with from those outside their scope. ting fair consideration in resource allocation, and having
I I ill 111111
Table 4 Using This Definition in Clinical Care

Examples of requirements Examples of vulnerability
Having access to health care
1. Availability • Patients with rare diseases, need for interventions requiring expensive technology
2. Accessibility • Distance from health service, responsibility for dependent relatives
3. Accommodation • Long or inflexible working hours
4. Affordability • Poverty, uninsurance, underinsurance, distance (high transportation costs)
5. Acceptability • Populations with reason to distrust the health care system
Adequate financial coverage • Greater likelihood of being denied insurance, such as pre-existing conditions, or risk factors
Not being harmed • Patients more likely to be treated in unusual ways, such as health care providers and their
families, or patients from whom litigation is feared
Self-determination/Autonomy • Difficulties in receiving or understanding the relevant information: not knowing the
language used, or how to read
• Difficulties in requesting a role in decision-making
• Lack of decision-making capacity: some children, some patients with mental disorders,
comatose patients
• Lack of freedom to make a voluntary choice
• through limited freedom: prisoners
• through social weakness: minorities, refugees, sometimes women
• through hierarchical weakness: hospital employees, students
Confidentiality • Public figures, health care providers, families of health care providers
Getting fair consideration in • Patients at risk of seeing their interest discounted, such as:
resource allocation. • terminally ill patients
• elderly patients
• cognitively impaired or handicapped persons
• disenfranchised or socially marginalized persons
Having a voice as a stakeholder in • Disenfranchised or socially marginalized persons
health care.
Avoiding illness? • Persons with less access to health literacy
• Persons at the lower end of the socio-economic spectrum
a voice as a stakeholder in heiIth care. Access can be fur- for people with short life-expectancies were likely to be
ther divided into the elements proposed by Penchansky based on an evaluation that their needs should count
and Thomas: availability, or the degree to which the pro- for less, then this would constitute vulnerability. If the
vider has the necessary resources to meet the patient's same decisions were systematically made with the same
needs, accessibility, or the ease with which the patient sort of considerations afforded anyone, such as likeli-
can physically reach the location of health services, ac- hood of benefit and quality of life, then they might still
commodation, or the degree to which the health service be allocated intensive care less frequently, but their
is organized in ways that mee: the constraints and pref- claim would not have been discounted.
erences of patients, affordability, or how the provider's Clearly some populations will only be considered
charges fit with the patients' ability and willingness to vulnerable in the sense proposed here if they do, indeed,
pay, and acceptability, or the extent to which patients are have a valid claim to whatever it is they are more likely
comfortable with the characteristics of the health servi- to be denied. On some counts, this may be more con-
ces and vice versa.38 Examples of vulnerability linked to troversial than in others. For example, some popula-
these requirements are shown in Table 4. tions are more at risk of becoming ill in the first place.'
Whenever resources are allocated, vulnerability If we have a right limited to access to health care, then
also applies to those more likely to see their claims this would constitute a part of normal human fragility,
transgressed or discounted in allocation processes. not vulnerability as a claim to special protection. If,
Importantly this may not mean that anyone less likely however, the claim to provide health care is based on a
to have resources allocated :o them is vulnerable in requirement to equalize health itself as a precondition of
this way: it would depend on their risk of having their fair equality of opportunity,4° then a greater likelihood
claims discounted, rather than on the final result. For of becoming ill would constitute vulnerability, a claim
example, if decisions about admission to intensive care to special protection as outlined here.
Some Possible Concerns then added difficulty would not affect it directly. How
much added effort is required when more is needed is
Importantly, this definition accepts that the claim to pro- indeed an open question, but this is due to the need
tection is based not on vulnerability itself, but on some to balance claims against those of others,44 rather than
other valid source. Attempts to ground obligations dir- because the claim itself is diminished.
ectly in vulnerability have been made both in continental This definition of vulnerability is silent as to whose
philosophyà and in bioethics.42 Concern could exist that duty it is to fulfill existing claims. The simple answer
if we do not have a requirement to respect a principle of might be that whoever had this duty in the first place
vulnerability, then the vulnerable could lack protection. still has it in the case of vulnerable persons. This, how-
This concern, however, presupposes an existing claim to ever, should mean anyone who shares in the duty to
protect the vulnerable. Clearly, then, such a claim cannot avoid the identifiable wrong (Figure 1). Asking who is
originate in the principle itself, as it seems to ground the the agent responsible, as if there had to be a single one,
very need for it. The present proposal accepts that the is simplistic, as we consider that different people may
vulnerable have a daim to protection; but this daim is have different sorts of duties to fulfill the same claim. A
grounded in other claims that we recognize anyway. If a child's parents may have a duty to make sure she does
daim for anything exists, then the higher likelihood that not fall into a pond but this does not relieve me of a
this claim will be transgressed generates a requirement for duty to rescue her if I happen to be there when she does.
greater attention that this claim be fulfilled whatever this, The problem of requirements for ancillary care, which
pre-existing, daim may be. The obligation to avoid wrong- has proved particularly thorny in research ethics, is an
ing is not derived directly from a principle of vulnerability example. If governments have a duty to provide health
but from another source; specifically, from a valid claim care to their citizens but fail to do so, how much of this
that some wrong should be avoided, including the wrong duty falls to researchers? Although we will not attempt to
we incur when a good to which we have a valid claim answer this question here, it is noteworthy that there are
is denied us. If vulnerability increases the likelihood of four counts on which researchers may have such a duty
being wronged, it also increases the attention required to to some degree. Entrustment of their health by research
avoid any wrong that we should avoid for other reasons. subjects was proposed as grounds for this." That they
This both clarifies and strengthens the claim for protec- are on the spot and able to help, as in the rule of rescue,
tion: we do not need to recognize a specific requirement could constitute another. If we should do our share of
based on vulnerability but only a situation where fulfilling a collective duty to fulfil such claims," then research-
existing requirements requires additional care. ers and sponsoring institutions could do their share by
Some may say that defining vulnerability in this providing ancillary care. Finally, the advantages that
way makes the concept superfluous, since it means researchers and sponsoring institutions sometimes reap,
nothing further than a claim to fulfill duties that we from the very fragilities that make their subjects vulner-
have anyway. This does not, however, void the notion able, could also ground such a duty. Increasing recogni-
of its use. An increased risk is morally relevant. It can tion that some claim to ancillary care exists47 could be
change the actions required to provide protection in understood as a growing realization of just this problem.
degree (for example providing more security for con- Finally, the definition of vulnerability proposed
fidentiality) or in nature (as when potential research here does not address differences in the way that people
subjects should be excluded from a protocol). Vulner- may have become vulnerable. As has been proposed,
ability thus truly means something different from the differences between vulnerabilities that originate in in-
mere existence of a pre-existing claim. justice, or misfortune, or that are the fault of the vulner-
We may also wonder whether the strength of the able person herself, could well be relevant.48 According
pre-existing claim might not be affected by the added to the view proposed here, however, this would affect
difficulty. There are, after all, instances where duties the legitimacy of the claim being considered, rather
exist in part because the burden to the agent is not too than the definition of vulnerability itself.
great, as in the rule of rescue.43 As this suggests, how-
ever, this would depend on the sort of claim considered. Acknowledgments
In research with human subjects, for example, a claim The author warmly thanks Bernard Baertschi PhD,
that was too difficult to fulfill might have to lead to the Marion Danis MD, and Alex Mauron PhD, for very in-
exclusion of potential subjects from the protocol. This sightful and constructive comments, as well as her stu-
is because we do not actually have a claim to partici- dents. This work was funded by the Bioethics Institute
pate in research per se. In health care, this would play at the University of Geneva, and by the Swiss National
out very differently. If we have any claim to health care, Science Foundation.
II ,111 1111j41
Notes
1. M.C. Ruof. Vulnerability, vulnerable populations, and 20. Kottow 2004 op. cit. note 4, M.H. Kottow. The vulnerable
policy. Kennedy Inst Ethics J 2004, 14: 411-25. and the susceptible. Bioethics 2003, 17: 460-71.
2. E. Levinas. 1961. Totalite et infini. Den Haag: 21. D. Wendler. When should 'riskier' subjects be excluded
Phenomenologica. from research participation? Kennedy Inst Ethics J 1998, 8:
3. J.D. Rendtorff. Basic ethical principles in European 307-27.
bioethics and biolaw: autonomy, dignity, integrity and 22. P.J. Nickel. Vulnerable populations in research: the case
vulnerability-towards a foundation of bioethics and of the seriously ill. Theor Med Bioeth 2006, 27: 245-64. R.
biolaw. Med Health Care Philos 2002, 5: 235-44. Macklin. Bioethics, vulnerability, and protection. Bioethics
4. D. Callahan. 2000. The Vulnerability of the Human 2003, 17: 472-86.
Condition. In Bioethics and Biolaw, Volume II: Four Ethical 23. Nickel op. cit. note 22.
Principles. P Kemp, et al., ec.s. Copenhagen: Rhodos 24. Agrawal op. cit. note 8.
International Science and AA Publishers; and Centre for 25. V. Merton. The exclusion of pregnant, pregnable, and
Ethics and Law in Nature and Society. A. Maclntyre. 1999. once-pregnable people (a.k.a. women) from biomedical
Dependent Rational Animals. Chicago and La Salle, Illinois: research. AmJ Law Med 1993, 19: 369-451.
Open Court. M.H. Kottow. Vulnerability what kind of 26. S.A. Garcia Equal protection clause enforcement as a
principle is it? Med Health Care Philos 2004, 7: 281-7. model for protecting vulnerable human research subjects.
5. J.P. Lott. Module three: vulnerable/special participant AmJ Bioeth 2004, 4: 81-2; discussion W32.
populations. Developing World Bioeth 2005, 5: 30-54. 27. Agency for Healthcare Research and Quality. 1999.
6. C. Levine, et al. The limitatims of 'vulnerability' as a Request for applications on measures of quality of care
protection for human researzh participants. AmJ Bioeth for vulnerable populations. LA. Aday: 2001. At Risk in
2004, 4: 44-9. Amedica; The Health and health Care Needs of Vulnerable
7. J.P. DeMarco. Vulnerability a needed moral safeguard. Am Populations in the United States. San Fransisco, CA: Josey-
J Bioeth 2004, 4: 82-4; discussion W32. Bass, M. Danis, & D.L. Patrick. 2002. Health Policy,
8. M. Agrawal. Voluntariness in clinical research at the end of vulnerability, and vulnerable populations. In Ethical
life. J Pain Symptom Manage 2003, 25: S25-S32. Dimensions of Health Policy. M. Danis, et al., eds. Oxford,
9. ICH Steering Committee. 1956. ICH Harmonized Tripartite New York Oxford University Press.
Guideline. In Guideline for Good Clinical Practice E6. 28. Agency for Healthcare Research and Quality op. cit. note 27.
10. CIOMS. 2002. International Ethical Guidelines for 29. F.J. Leavitt. Is any medical research population not
Biomedical Research Involving Human Subjects. Geneva, vulnerable? Cambridge Quarterly of Healthcare Ethics 2006,
Switzerland: CIOMS. 15: 81-8.
11. The National Commission for the Protection of Human 30. Levine op. cit. note 6 p. 4.
Subjects of Biomedical and Behavioral Research. 1979. 31. DeMarco op. cit. note 7.
The Belmont Report: Ethical Principles and Guidelines for 32. Levine op. cit. note 6.
the Protection of Human Subjects of Research. Washington 33. World Medical Association. Declaration of Helsinki:
DC: Department of Health, Education, and Welfare. Ethical Principles for Medical Research Involving Human
12. E.J. Emanuel, et al. What makes clinical research ethical? Subjects. Edinburgh; 2000. http://www.wma.net/e/policy/
Jama 2000, 283: 2701-11. b3.htm (accessed July 17th 2007).
13. S.A. Hurst, et al. Conserving scarce resources: willingness 34. Participants in the 2001 Conference on Ethical Aspects
of health insurance enrollees to choose cheaper options. J of Research in Developing Countries. Ethics. Fair benefits
Law Med Ethics 2004, 32: 496-9. for research in developing countries. Science 2002,
14. S.J. Edwards. Restricted treatments, inducements, and 298(5601): 2133-4.
research participation. Bioethics 2006,20: 77-91. 35. Emanuel op. cit. note 12.
15. K. Kipnis. 2001. Vulnerability in Research Subjects: A 36. P Farmer & N.G. Campos. New malaise: bioethics and
Bioethical Taxonomy. In Ethical and policy issues in research human rights in the global era J Law Med Ethics 2004, 32:
involving human research participants. National Bioethics 243-51,190-241.
Advisory Commission, ed. Bethesda, MD. 37. K. Page-Shafer, et al. HIV prevention research in a
16. D.B. Resnik. Exploitation in biomedical research. Theor resource-limited setting: the experience of planning a trial
Med Bioeth 2003, 24: 233-59. in Cambodia. Lancet 2005; 366: 1499-503.
17. E.J. Emanuel, et al. Undue inducement in clinical research 38. R. Penchansky & J.W. Thomas. The concept of access:
in developing countries: is it a worry? Lancet 2005, 366: definition and relationship to consumer satisfaction.
336-40. Med Care 1981, 19: 127-40. C.G. McLaughlin & L
18. C. Korsgaard. The Reasons We Can Share. Social philosophy Wyszewianski. Access to care: remembering old lessons.
and policy 1993, 10: 24-51. Health Sery Res 2002, 37: 1441-3.
19. G.B. Tangwa Moral agency, :moral worth and the question 39. M. Marmot. 2004. The Status Syndrome How Social
of double standards in medical research in developing Standing Affects Our Health and Longevity. London:
countries. Developing World Bioeth 2001, 1: 156-62. Bloomsbury Publishings.
40. N. Daniels. 1985. Just Health Care: Cambridge University 45. L. Belsky & H.S. Richardson. Medical researchers' ancillary
Press. clinical care responsibilities. BMJ 2004, 328: 1494-6.
41. Levinas op. cit. note 2. 46. L.B. Murphy. 2000. Moral Demands in Nonideal Theory.
42. D.C. Thomasma. The Vulnerability of the Sick. Bioethics Oxford, New York: Oxford University Press.
Forum 2000, 16: 5-12. 47. R. Macklin. Four forward-looking guidance points.
43. A. McIntyre Guilty Bystanders? On the Legitimacy of Duty Developing World Bioeth 2001, 1: 121-34.
to Rescue Statutes. Philosophy and Public Affairs 1994, 23: 48. D.W. Brock. 2002. Health Resource Allocation for
157-91. Vulnerable Populations. In Ethical Dimensions of Health
44. N. Daniels. Four unsolved rationing problems. A Policy. M. Danis, et al., eds. Oxford: Oxford University
challenge. Hastings Cent Rep 1994, 24: 27-9. Press: 283-309.
Best Practices for Health Research at the University of Montreal, conducted the research
and prepared the Best Practices. F. . .1
Involving Children and Adolescents:
Genetic, Pharmaceutical, and Guiding Principles
Longitudinal Studies The Best Practices use the three ethical principles intro-
duced in the TCPS2—respect for persons, concern for
D. Avard, L Black, J. Samuel, G. Griener, and welfare, and justice—as Guiding Principles because
B.M. Knoppers of their importance within the discourse of biomed-
With the collaboration of Institute for Human Development, ical ethics. These are essentially the same principles
Child and Youth Health (ciHR); and Ethics Office (CIHR) adopted in The Belmont Report and similar to the
principles used by Beauchamp and Childress in their
widely cited Principles of Biomedical Ethics.' This does
not mean that all ethical norms incorporate the defin-
Scope of Application
itions or terms used by the TCPS2 or that all commenta-
E.. .1 These Best Practices are designed to assist in the tors accept these principles, but the foundation of each
application of the TCPS2 in the paediatric population. principle is incorporated into the Best Practices to guide
They are complementary to, and do not supersede, the research decisions [. . .]
TCPS2, and are intended as voluntary guidance for the
Canadian health research community working with Respect for Persons in the Paediatric
children and adolescents... [...1 Research Context
It is clear that the morally important capacity to make
The Development of the Best Practices
decisions and determine the course of one's own life
does not instantaneously spring, fully formed, into
F.. .1 Considering the need for more explicit discussion existence. Rather, it develops over time and through
of paediatric health research, the National Council on exercise. Moreover, it may also be diminished or im-
Ethics in Human Research (NCEHR) proposed in 2008 paired to varying degrees by illness or other conditions.
to undertake a two-year project to develop a guidance One must carefully consider how to demonstrate re-
document for paediatric health researchers, research spect for the person with diminished capacity, which
ethics boards (REB), and institutions. This project was includes minors. Moreover, legal restrictions often exist
initiated and orchestrated through NCEHR's Emerging that limit the decision-making capacities of minors.
Issues Analysis Committee, and in collaboration with Most provinces promote an individual determination
the Canadian Institutes of Health Research (CIHR) Insti- of capacity, but not all.
tute of Human Development, Child and Youth Health In those instances where the capacity is devel-
(IHDCYH), the CIHR Ethics Office, Health Canada, and oping, but not yet sufficiently developed, one must
other key organizations such as the Maternal, Infant, respect the nascent autonomy and foster its further
Child, Youth, Research Network (MICYRN). The Centre development, while at the same time protecting the
of Genomics and Policy at McGill University, formerly interests of the individual. Ethically robust respect
part of the Centre de recherche en droit public (CRDP) for the developing autonomy of a child or adolescent
Source: Centre of Genomics and Policy (cGP), Maternal Infant Child and Youth Research Network (micYRN), 2012. Reprinted with
permission.
uu I ai I I ill 11111d
is shown by the procedures of informing the minor have different standards for when participation is per-
about the proposed research; doing so in a manner missible, illustrating the practical difference between
tailored to the child's abili:y to understand; solicit- adult and paediatric research. For instance, there
ing his or her assent for participation; and respecting must be sufficient justification for the research to be
dissent. In some instances, respecting the child's po- performed on minors instead of on adults. Addition-
tential for the development of autonomy may provide ally, the risk generally should be minimal.' (The dif-
a very strong reason to delay recruitment until that ficulties in defining minimal risk will be discussed in
autonomy blossoms and the mature child can decide greater detail in Chapter VI.) When the minor lacks
whether to participate. the capacity to consent, the TCPS2 requires either that
When initial recruitment cannot be delayed and the research pose no more than minimal risk or that
the individual's participation continues over a long participation is in the minor's best interests—that is,
period, the requirement that informed consent be research participation is the best option available to the
maintained generates a need to revisit assent and, minor. Notably, this requirement is somewhat more
subsequently, consent as autonomy advances. As the restrictive than that in the earlier version of the TCPS,
capacity of a child or adolescent to understand infor- and it is also more restrictive than regulations in the
mation matures, the sophistication of the informa- United States.°
tion provided to him or to her should increase, and The concept of "best interest" deserves additional
the ethical importance of assent increases. And at the discussion in the context of children. A competent adult
point the individual achieves autonomy (e.g., legal can be expected to judge his or her own best interests—
capacity), continued participation will require his or so long as the risk is within accepted bounds. A child or
her informed consent. adolescent, though, trusts others to make this judgment.
An additional and important consideration in Children and adolescents "are particularly vulnerable to
paediatric research is that minors are rarely, if ever, in a conflicting interests and values. The historical record
position to make decisions free from outside influence. demonstrates that those who make decisions on their
Even as children and adolescents gain greater capabil- behalf do not always serve the children's best inter-
ity to understand, they stir rely on their parents or est. At times, even their parents' interests conflict with
guardians for most decisions in their lives.' Moreover, theirs."' Researchers, parents, guardians, and research
physicians (or researchers) a-e likely to be in a position ethics boards should be careful that other considera-
of trust and, even without overt action on their part, tions do not impact a determination of what is in the
minors may do what is wanted of them merely because best interest of a child or adolescent. . .1
it is suggested by a trusted Ferson.3 In sum, the use in
the TCPS2 of "respect for persons"4—and the adoption Justice
of this principle in the Best Practices—goes beyond the [. ..1
traditional concept of autonomy to include protection
of those with diminished or no autonomy. Justice in the Paediatric Context
In sum, the use in the TCPS2 of "respect for per- Applying the principle of justice creates unavoidable
sons"—and the adoption of this principle in the Best tension in the context of research with children and
Practices—goes beyond the traditional concept of au- adolescents. At the broadest level of social policy, jus-
tonomy to include protection of those with diminished tice calls for research aimed at issues affecting children.
or no autonomy . . . In paeCiatric research, this exten- Such research will inevitably need to recruit minors
sion serves to improve the ethical standing of minors. as participants and in a great many instances minors
This principle is further strengthened by the second who are unable to make decisions for themselves. The
Guiding Principle of "concern for welfare." . .1 vulnerability of certain participants who can't decide
for themselves, such as children and adolescents, is
Concern for Welfare an important difference that must be taken into ac-
[. . .1 count so that they may enjoy equitable treatment in
research.8 Consequently, additional measures are re-
Concern for Welfare in the Paediatric Context quired to protect those who are not in a position to
Application of the principle of concern for welfare to protect themselves.
the context of paediatric research is similar to research Justice also requires researchers not to deliber-
involving adults: researchers still have an obligation ately avoid including children and adolescents in an
to minimize harms and to promote the benefit of the effort to reduce the perceived difficulty involved in
participant. Research involving minors does, however, conducting research ethically. It also requires those
Avard/Black/Samuel/Griener/Knoppers: Best Practices for Health Research Involving Children and Adolescents
who perform ethics review to carefully consider the existing international and national ethics. In response,
necessity of such inclusion and to work with research- several guidelines were established to ensure observ-
ers to avoid exploitation. The TCPS2 recognizes that it ance specifically of the rights of minors participating
would be unfair to exclude minors from research that in research.14 However, in an attempt to protect them,
might benefit them, especially as there are diseases minors were effectively excluded from research and
that are unique to children and adolescents or that are became "therapeutic orphans."13
expressed differently in children.9 This could particu- The unintended consequence of this exclusion was
larly concern specific cancers that affect the paediatric a lack of knowledge of paediatric development and of
population. Furthermore, inclusion of minors in areas appropriate medical treatments for minors in general,
such as pharmaceutical research could provide essen- thereby jeopardizing their health and well-being in
tial information of efficacy and dosage of medication the long term. It then became necessary for the inter-
when physicians are otherwise left without guidance.1° national community to re-evaluate their normative
However, particular care must be taken to ensure that guidance documents in order to promote a balance
minors in research are not shouldered with dispro- between the duty to protect vulnerable persons, such
portionate or unreasonable risk, especially in light of as children and adolescents, and the need to include
their vulnerability and general inability to consent to them in research. Though the need to conduct research
research on their own behalf. involving children and provide guidance for their in-
clusion is evident, it was not until the 1964 Declaration
Application of the Guiding Principles to the of Helsinki that inclusion criteria for minors were clearly
Best Practices laid out.16
These Guiding Principles serve as the backdrop to the Today, there is consensus in international and
Best Practices. However, as they are generally applied Canadian ethical norms regarding the need to include
in the context of adult research, their applicability to children and adolescents in research while offering
paediatric research is imperfect. Yet, perfection is not appropriate protection. Guidelines from the Council
necessary. Indeed, "principles are neither the start nor of International Organizations of Medical Sciences
the end of the process of ethical reflection."11 No single (ciomS) mention that the participation of minors is in-
principle has primacy when considering potential dispensable for research concerning diseases affecting
outcomes for research participants. The principles are infants, and the TCPS2 states that inequity is created
interdependent and complementary. They represent "when particular groups fail to receive fair benefits of
questions that should be asked and considerations that research or when groups, or their data or their bio-
should be taken into account and perhaps will lead to a logical materials, are excluded from research arbitrarily
deeper discussion of the ethical issues facing research- or for reasons unrelated to the research question."17
ers, REBs, and families working with the paediatric This reflects an important shift in norms related to re-
population. E. . .1 search involving minors.
The inclusion of minors in research is subject to
Chapter I: Inclusion of Minors in Research specific conditions to ensure their protection. Unan-
imity exists among international and national norms
Inclusion of Minors in Research: International on the following conditions: minors should only be
and Canadian Contexts involved when the research cannot be carried out on
The Nuremberg Code of Ethics of 1947 did not ad- adults; consent of the parents as well as the assent of
dress the inclusion of minors in research. This is un- the minor, when feasible, are required; research should
fortunate, as some research that took place during the involve no more than minimal risk if there is no pros-
Nazi period in Germany used children. Following the pect of direct benefit to the minor, and research should
Nuremberg Code, the twentieth century was further be approved by an REB and satisfy the legal require-
marked by a number of scandals involving paediatric ments of the jurisdiction. Additional elements may also
research, for example the Willowbrook case in the be considered, such as the importance of the research
1950s.12 While conducting a hepatitis study on healthy in validating adult data, its direct relation to a condi-
institutionalized children, researchers intentionally tion occurring in minors, or its legality.
infected them with hepatitis in order to understand With respect to the criterion of minimal risk, the
the disease and to develop a vaccine.13 Even without TCPS2 specifies that "REBs have special ethical obliga-
guidance that directly addressed research involving tions to individuals or groups whose circumstances
minors, it was clear that deliberately harming them make them vulnerable" (e.g., children) and that their
for the sake of paediatric research ran contrary to inclusion should not exacerbate their vulnerability.18
1111. II 11 II .11 IIILJJ
The TCPS2 also allows the inclusion of minors in re- Lack of Pharmaceutical Data
search that involves more than minimal risk (defined in There is currently a lack of pharmaceutical data on the
Chapter 2, section B of the TCPS2) if it has "the prospect paediatric population23 Most paediatric drugs are pre-
of direct benefits for them."19 Thus, it would be pos- scribed off-label due to the fact that they either have
sible for minors suffering from life-threatening diseases never been formally tested on children or adolescents
to participate in research that involves more than min- or data from paediatric clinical trials in other countries
imal risk if they can benefit f.:-om it2° (for additional dis- has not yet been submitted in Canada. Physicians may
cussion of minimal risk, see Chapter VI—Evaluation of also extrapolate paediatric dosage from adult data24
Risks and Benefits). even though "children are not small adults."25 Moreover,
Only a few ethical norms have addressed research there may be variations in maturation within the same
involving very vulnerable minors (e.g., impaired or age group of minors. The absence of age-appropriate
institutionalized minors). However, such inclusion is formulation of drugs and the lack of data on the ef-
limited to specific projects. For example, the Inter- ficacy and toxicity of these drugs may expose minors
national Conference on Harmonization (ICH) Clinical to serious harm. CIOMS underscores that, in the past,
Investigation of Medicinal Products in the Pediatric Popula- drugs that had not been tested on children were never-
tion El 1 provides that such research should be limited theless administered,26 exposing them to serious harm
to "diseases or conditions found principally or exclu- in the absence of sufficient knowledge on the safety and
sively in these groups."21 In addition, impaired or in- efficacy of such drugs. Even today, although drug labels
stitutionalized minors may b otherwise healthy, which may clearly state that the drug has not been approved
creates additional complexity when considering their for use in children and adolescents, clinicians might
participation in research: should protections for these still prescribe the drug because there is nothing else
populations be even greater than for healthy minors available.27 This is problematic because developmental
who are not impaired or institutionalized? The inclu- stages may influence the efficacy or toxicity of the
sion of healthy minors in research is generally limited drug28—prescribed drugs are metabolized, extracted,
to studies on prevention or vaccine trials.22 or absorbed differently.29 For example, Health Canada
Finally, international and Canadian ethical norms decided in 2008 that some over-the-counter cough and
do not generally provide an order of preference in the cold medicines (e.g., antitussives and expectorants)
selection of different groups of children or adolescents should not be labelled for use in children under 6 years
for inclusion. However, CIOMS, ICH, and the European old.3° Over-the-counter cough and cold medicines
Commission address the issue of involving older minors have a long history of use in children; however, there
first in research, if possible. The rationale for this age- is limited evidence available to support the efficacy of
based stratification of child participants is based on the these products in this population. In addition, reports
decreasing vulnerability of minors as they mature. of misuse, overdose, and rare side-effects have raised
concerns about the use of these medicines in children
General Statement on the Inclusion of under six.
Minors in Research Some countries and organizations have begun
initiatives to encourage clinical research involving the
The inclusion of minors M research promotes their paediatric population.31 For example, the United States
safety and well-being. adopted legislative provisions to increase the numbers
of drug trials involving minors by both offering incen-
Children and adolescents differ significantly from tives to manufacturers and by publishing requests for
adults, physiologically and psychologically as well as proposals to third parties in case of lack of interest
developmentally. Their developmental stages influence by manufacturers.32 The European Commission also
the limitations and potential benefits of research. Some adopted legislative provisions to facilitate and harmon-
diseases are found only in tie paediatric population. ize the conduct of paediatric clinical trials.33 In Canada,
However, given that minors are a vulnerable popula- two important initiatives were undertaken to encour-
tion, research should be subject to a rigorous govern- age paediatric research. First, Health Canada adopted
ance framework complying with national ethical and a guidance document called Guidance for Industry:
professional norms and legislation and informed by Clinical Investigation of Medicinal Products in the Pediatric
international ones. This framework respects the funda- Population, ICH Topic El 1 and prepared a document
mental principles of research involving human subjects. called Health Canada Addendum to ICH Guidance Docu-
[...I ment El 1: Clinical Investigation of Medicinal Products in
Conditions for Inclusion of Minors in Research

• participation of minors in research is justifiable when the research cannot be carried out with adults;
• when research requires the participation of minors, least vulnerable minors (e.g., older or more developed
children or adolescents) should be included first in the project, if possible and scientifically appropriate;
• minors should derive a direct or indirect benefit from their participation in research;
• minors should not be exposed to more than minimal risk when research does not hold the prospect of
direct benefit;
• a minor with a life-threatening disease may be included in research with hope of direct benefit only if:
1. the risks of participation are commensurate with the benefits; and
2. there is no treatment from which the minor can hope to benefit;
• a minor with a life-threatening disease may be included in research which does not offer hope of direct bene-
fit to her/him only if:
1. the research does not expose him/her to more than minimal risk; and
2. the research may benefit minors with the same life-threatening disease;
• impaired, abused or institutionalized minors should only be included in research if:
1. the research relates directly to their disease o r condition; and
2. the research offers hope of potential benefit to the minor concerned or the research represents no more
than minimal risk;
some research might require the participation of healthy minors in order to determine, for example, the effect
of diet or environmental factors on a genetic predisposition, or the efficacy of a paediatric vaccine, but the
research should not expose healthy minors to more than minimal risk.
the Pediatric Population to assist the industry and the a case-by-case basis. However, some organisations have
researchers conducting research on medicinal prod- adopted specific guidelines on the participation of dying
ucts for paediatric use.34 Second, the Food and Drug minors in research.37 Thus, they seem to create a special
Regulations were amended in 2006 to extend the data category for children and adolescents who present with
protection period by six months for certain paediatric a life-limiting illness and for whom there is no scientific-
drugs. To obtain this extension, manufacturers must, ally proven curative therapy available. For example, the
within the first five years of the protection period, American Academy of Pediatrics recognized specifically
submit the results of paediatric clinical trials, designed that minors at the end of life can participate in research
and conducted for the purpose of increasing know- when they suffer from "a life-threatening condition that
ledge of the use of the drug in paediatric populations. does not respond to all standard therapies, and the
Extending the term of data protection in this manner [child's] illness is such that death is imminent."38 The
is intended to encourage the submission of paediatric following conditions apply to such research:
research results to provide health benefits to children
and adolescents.35 [. . .] 1. The question being addressed is extremely
important.
Inclusion of Minors in Research at the End of Life 2. The therapy [experimental treatment] being pro-
The inclusion in research of minors who are dying and posed is well founded in animal and clinical re-
are receiving palliative care raises many ethical ques- search and/or there is a good expectation that the
tions because of their double vulnerability: 1) they are therapy may be beneficial.
minors; and 2) they are dying.38 As mentioned in Sec- 3. Physicians who are not involved in the research
tion 1.1, the need to include minors in research is widely must document that the clinical condition of the
recognized. There are not always specifications with patient is such that death appears inevitable and
regard to the categories of children or adolescents who standard therapy has not improved the patient's
should be included in research (e.g., healthy minors or prognosis.
minors with life-threatening illnesses). Each research 4. The potential benefits outweigh the potential
project involving minors should thus be evaluated on risks.39
An ethical argument often raised against the inclu- exhaustive. Depending on the type of research con-
sion of dying minors in research is the potential burden ducted, additional elements may need to be included
of participation in research_ 4° Some authors assert in the consent form. For example, in genetic research,
that these children or adolescents might be asked to it may be necessary to state the policy on the disclosure
take on a bigger burden than other minors because of of results of genetic tests and their familial implica-
their "compromised health status and limited remain- tions to the participant and family. In pharmaceutical
ing time."'" Moreover, they may prefer to spend their research, the availability of the drug after the comple-
remaining time in other ways than participating in tion of a trial or the lack of information on the drug
research.'" However, the data collected to date on the being studied should be included in the consent form.
potential burden on children and adolescents arising In the context of longitudinal studies using biobanks,
from their participation does not support this find- the future research uses (and the requirement for REB
ing.43 Therefore, it would be unjust to exclude them approval for those studies), storage and destruction
from research on this basis. There may however be a of data and/or samples, who can access the biobank,
need to conduct research to determine the perceptions and who has governance and control over the biobank
of the minors and parents regarding their participation should also be stated.
in research. E. . .1 International norms as well as Canadian norms
state that information should be given in understand-
Chapter II: Consent to Research able language. Researchers should therefore adapt the
language used to the abilities of the person consenting.
Consent should also be written except when the person
consenting cannot read or write, if a written consent
Free and Informed Consent Requirements is contrary to local custom, or when another reason
The informed consent of the competent adolescent, is judged acceptable by a REB. If this situation occurs,
or that of the parents if the minor is not competent verbal consent is possible but should be documented.
to consent,44 is subject to a number of requirements CIOMS and icPS2 address the possibility of an implied
according to international and Canadian norms on consent, meaning a consent implied by voluntary ac-
ethical research involving human participants. These tions (e.g., return of a questionnaire by mail). F. . .1
requirements—which are generally applicable to all re- Similarly, international and Canadian ethical
search participants and not just minors—mainly focus norms strongly recommend that cultural background
on the quality of the consent (e.g., free and informed), be considered (for additional information, refer to
the elements to include in the consent form, the cap- icPS2, Chapter 9, "Research Involving the First Na-
acity to understand the information provided, and the tions, Inuit and Metis Peoples of Canada"). For
actual process of obtaining informed consent. It should example, in some communities, a handshake may con-
be noted that these Best Practices do not provide guid- stitute evidence of trust sufficient to express consent,
ance on how to assess competency. while in other communities the giving and receiving of
International and Canadian ethical norms agree on gifts constitutes consent.45
the need to obtain the consent of the competent adoles-
cent, or that of the parents if the minor is not compe- Parental Consent for an Early Phase Clinical Trial
tent to consent, before inclusion in research. They are Pharmaceutical research also raises the question of
also in agreement on the characteristics of that consent. whether parents can consent to the participation of
Indeed, all of the norms analyzed provide that consent their child in an early phase clinical tria1.46 Usually,
must be free (e.g., obtained without manipulation or Phase I examines the "metabolism and pharmacologic
undue influence) and informed (e.g., researchers must actions of the drug in humans" while Phase II examines
provide all pertinent information). the "effectiveness of the drug for a particular indication
There is agreement on a list of core elements to [. . .1 in patients with the disease or condition under
be discussed and included in the consent process, study [. . .]"47 Thus, there is a low chance that anyone
which attempts to ensure that all pertinent informa- participating in a Phase I trial would directly benefit
tion is provided. These core elements include: the aims since the aim is not to test the efficacy of the new drug
of the research, research procedures, potential risks but rather its toxicity. Furthermore, Phase I or II stud-
and benefits, participant or third party access to the ies generally expose minors to more than minimal risk.
information collected, compensation of the participant For these reasons, it is unlikely that healthy minors
and/or family, right of withdrawal and a description would be permitted to participate in the early phases of
of alternative treatments. However, this list is not clinical trials due to the increase in risk. Evidently, the
development of vaccines and other preventative treat- include important information about the proposed
ments with healthy minors is legitimate prima facie.48 research project. Yet, unlike with consent, neither
Nonetheless, the inclusion of terminally ill minors may international nor Canadian ethical norms detail the
be considered for early phase studies.49 REBs should elements to include in obtaining assent, nor do they
consider that minors may benefit from the research advise on who should discuss assent with the minor
even if the probability of success is low. In any case, the (e.g., the researcher or someone else on the research
best interest of the child should prevail. [. . .] team). Some norms specify that information about
the project (e.g., nature and purpose of the research),
Chapter III: Assent of the Minor the right to decline, the right to withdraw, and in-
formation on potential risks and benefits should be
Assent of the Minor: International and provided to the minor. No further guidance is pro-
Canadian Contexts vided for researchers. Therefore, it may be useful to
[. . .] Minors are not presumed, as adults are, to have refer to the elements needed for an informed consent
the required competency to provide a free and informed and to adapt these to the particular context of assent.
consent to their participation in research. For minors This was suggested by the European Commission in
who are not considered to be legally competent or are its 2008 Ethical Considerations for Clinical Trials on
not legally emancipated, assent, rather than consent, Medicinal Products Conducted with the Paediatric Popu-
should be sought, together with parental consent. Most lation.'2 Although these elements of consent should
international and Canadian ethical norms acknowledge be considered, the assent process should not be as
the importance of including children and adolescents, legalistic as the consent process, considering that
who are capable of understanding, in the decision- minors have a limited comprehension of the research.
making process and of obtaining their assent. Assent However, since comprehension will evolve with age,
may be defined as the minor's willingness to participate an adolescent should receive more information than a
in the proposed research." 7-year-old, for instance.
The 1989 United Nations' Convention on the Rights Indeed, when seeking assent from a child or ado-
of the Child states that, when minors are able to ex- lescent not competent to consent, researchers need
press their own views, they have the right to express to take into consideration age, degree of maturity,
those views freely. Furthermore, "the views of the child developmental stage, and intellectual capacities (e.g.,
[should be] given due weight in accordance with the special needs or learning difficulties). Yet, minors of
age and maturity of the child."' International and the same age do not necessarily have the same degree
Canadian ethical norms require that researchers obtain of maturity and may not be at the same developmental
the assent of minors before involving them in research. stage. Thus, competency for assent should be deter-
However, assent might be impossible or impracticable mined on a case-by-case basis. The European Com-
to obtain in some circumstances. Only a few policy mission suggests that assent be obtained in a manner
documents address this matter by stating that not all appropriate to one of three different age groups: 1)
age groups can provide an assent (e.g., when the child children from birth to 3 years of age (where assent is
is too immature) and not all situations or types of re- impossible); 2) children from 3 years of age and up
search can foster this requirement (e.g., emergency re- (where children between ages 3-5 can understand
search, serious illness). However, when minors regain some expression of altruism; children from ages 6-7
capacity (e.g., following an emergency), assent should who have an emerging capacity to agree and under-
be sought to continue their participation in research. stand; and children from age 9 who can understand
Even if ethical norms do not frame such contexts, ex- the risks and benefits); and 3) adolescents (with an
ceptions to obtaining assent should also be extended to emerging capacity for independent decision-making
minors who cannot assent because of developmental and the capacity to make adult decisions in many
and cognitive disability and, obviously, to newborns areas of their life). The European Commission does
and the very young; that is, to all children who do not not draw a line as to when adolescence begins. Ac-
have the capacity to understand. In most Canadian cording to much of the literature, adolescence starts
jurisdictions, when the adolescent has a level of com- at 14 years of age," although there is not complete
prehension of research similar to a competent adult, agreement on this.54
consent should be sought. Information provided in the assent process should
The minor's assent alone is insufficient to be be disclosed to the extent allowed by the minor's ma-
included in research. It should be obtained in addi- turity and intelligence. Researchers should use a level
tion to the consent given by the parents and should of language and wording that is appropriate to the
SI II 6111 IIIlia
age and psychological and intellectual maturity of the the proposed research. Dissent requires of the minor
child or adolescent concerned and should also take the same level of capacity to understand as does assent.
into consideration any developmental or cognitive It may be expressed verbally or physically (e.g., crying,
challenges. The terms used should be understandable resistance). The European Commission recommends
and honest. documenting the dissent of the minor.
International and national norms do not neces- Most international and Canadian ethical norms
sarily require that assent be written. According to the do not provide detailed guidance on the dissent of
ICH and the European Commission, it is preferable that minors, except to state that it should be respected.
assent be written if the minor can read and write.55 The European Commission states that an "effort
However, TCPS2 acknowledges that assent may be ex- should be made to understand and respect differences
pressed verbally or physically.56 Since not all minors of opinion between the minor and his/her parents
can read and/or write, the assent process should be or legal representative."59 But if the minor expresses
documented to ensure that the rights of the minors strong and definitive objections, the dissent should
concerned have been respected. then prevaiL In Canada, the TCPS2 states that minor's
When assent is written, should the assent form be "expression of dissent or signs suggesting they do
distinct from the consent form? Most ethical norms are not wish to participate must be respected"6° with no
silent regarding this question ICH states that the assent provisions for overriding considerations except when
form may be separate from the consent form but does the minor does not have the ability to understand the
not clearly insist on this point. In contrast, the Euro- significance of the research. This inability to under-
pean Commission requires at the two documents be stand could be caused by age or lack of maturity (e.g.,
separate in order to ensure the use of age-appropriate newborns or very young children), or cognitive or
information. However, a common practice is to allow mental disorders. E. . .1
a space on the parental consent form for the assent of
the minor. Thus, researchers can explain the research Overriding Dissent
to the minor, verbally and in appropriate language, and It is worth noting that ethical norms used in some juris-
document the assent on the consent form. When the dictions suggest that overriding dissent is possible in
minor has provided a written assent, either on the same particular circumstances, although there is no unanim-
form as the consent or on a separate form, a copy of the ity regarding this question. For example, both CIOMS
assent form can be given to him/her where appropriate and ICH specify that the dissent of the minor may be
(primarily taking into account age). overridden when: 1) the minor is too immature or too
Finally, like consent, assent is a continuing pro- young; 2) there is no reasonable alternative other than
cess that should be renewed throughout the research what is available in research, and there are reasonable
project as the minor's capacities or the nature of the grounds to believe that it will offer benefit; or 3) the
research changes. For instance, in the context of longi- minor is suffering from a serious or life-threatening
tudinal studies, there is a need to continually reassess disease and dissent would jeopardize his/her welfare.6'
and renew assent throughout the duration of the re- When overriding dissisnt is possible, it is not clear
search project.57 Also, when adolescents develop the whether researchers need REB approval to do so. While
legal capacity to provide a fuly informed consent for CIOMS provides that such an approval is necessary to
themselves or reach the legal age of majority and are override the dissent of a minor who is "older and more
capable of making independent decisions during the nearly capable of independent informed consent,'
research, their informed consent should be sought ICH states that continued parental consent should be
when feasible—as a condition to their continued par- sufficient to maintain the participation of the minor in
ticipation in the research project.58 research.63 1. . .1
Chapter IV: Dissent of the Minor

The dissent of the minor, who is capable of understand-
ing, should be respected.
[. • .]
Dissent may be verbal or behavioural (e.g., body move-
Respecting Dissent ments) and may be expressed at any time during the
Just as refusal is the opposite of consent, dissent is the research. It should be respected if the minor is capable
opposite of assent and may be defined as the opposition of understanding, even if the parents consented to their
of the minor incompetent to consent to participate in minor's participation in the research project. F. . .1
Chapter V: Departures from Consent "(a) the research involves no more than minimal risk to
the participants;
Departures from Consent: International and (b) the lack of the participant's consent is unlikely to
Canadian Contexts adversely affect the welfare of the participant;
Consent of the competent adolescent, or that of the (c) it is impossible or impracticable to carry out the
parents if the minor is not competent to consent, is a research and to answer the research question prop-
fundamental requirement for participation in research. erly, given the research design, if the prior consent
However, this obligation may be abrogated under very of the participant is required;
specific and limited conditions. [. . .1 (d) whenever possible and appropriate, after participa-
In all cases, to ensure appropriate protection of tion, or at a later time during the study, participants
the population included in the proposed research, a will be debriefed and provided with additional pertin-
departure from consent is subject to REB approval at ent information in accordance with Articles 3.2 and
the very least. In Canada, with the exception of sec- 3.4, at which point they will have the opportunity to
ondary uses, the TCPS2 goes further by requiring that refuse consent in accordance with Article 3.1; and
all of the 5 following elements be satisfied before any (e) the research does not involve a therapeutic inter-
departure from the general process of consent may vention, or other clinical or diagnostic interven-
be approved: tions."64 [. . .1
Notes
1. Tom L. Beauchamp & James Childress, Principles of 11. J Harris, "In Praise of Unprincipled Ethics" (2003) 29(5) J
Biomedical Ethics, 6th ed. (New York: Oxford University Med Ethics 303.
Press, 2008) at 90. 12. HK Beecher, "Ethics and Clinical Research" (1966)
2. DF Merlo, LE Knudsen, K Matusiewicz, L NiebrOj & 274(24) N Engl Journal of Med 1354 [Beecher, "Ethics"].
KH Valtkangas, "Ethics in Studies in Children and 13. L Friedman Ross, Children in Medical Research: Access
Environmental Health" (2007) 33 J Med Ethics 408. versus Protection (New York: Clarendon Press, 2006) 155.
3. Ibid. 14. MA Grodin & LH Glantz, Children as Research Subjects:
4. Canadian Institutes of Health Research (CIHR), Natural Science, Ethics, and Law (New York: Oxford University
Sciences and Engineering Research Council of Canada, Press, 1994).
Social Sciences and Humanities Research Council of 15. E Kodish, Ethics and Research with Children: A Case-Based
Canada, Tri-Council Policy Statement: Ethical Conduct for Approach (New York: Oxford University Press, 2005).
Research Involving Humans (2010) Art. 1.1 [CIHR, Tri- 16. World Medical Association (wMA), Declaration of Helsinki:
Council Policy Statement]. Ethical Principles for Medical Research Involving Human
5. Ibid at Art. 4.6; JG Hoop, AC Smyth, LW Roberts, "Ethical Subjects (Finland: 1964).
Issues in Psychiatric Research on Children" (2008) 17 17. Council for International Organization of Medical
Child Adolesc Psych Clin N Am. 127. Sciences (CIOMS), International Ethical Guidelines for
6. The 1998 edition of TCPS required only that the Biomedical Research Involving Human Subjects (Geneva:
research have potential to benefit the participants or 2002) Guideline 14 looms, Guidelines 2002]; Canadian
"the group that they represent" (Article 5.3). The US Institutes of Health Research (CIHR), Natural Sciences and
regulations allow children to be exposed to greater than Engineering Research Council of Canada, Social Sciences
minimal risk if the research is of greater social benefit and Humanities Research Council of Canada, Tri-Council
and certain additional review procedures are followed. Policy Statement: Ethical Conduct for Research Involving
Canadian Institutes of Health Research (CIHR), Natural Humans (2010) at 10 ICIHR, Tri-Council Policy Statement].
Sciences and Engineering Research Council of Canada, 18. CIHR, Tri-Council Policy Statement, ibid. at Ch. 2B, p. 23.
Social Sciences and Humanities Research Council of 19. Ibid. at s. 4.6(b).
Canada, Tri-Council Policy Statement: Ethical Conduct 20. Ibid.
for Research Involving Humans (1998); 45 C.F.R. § 21. International Conference o n Harmonisation o f Technical
46.102(i) (2011). Requirements for Registration of Pharmaceuticals for
7. V H Sharav, "Children in Clinical Research: A Conflict of Human Use (I0-1), Clinical Investigation of Medicinal
Moral Values" (2003) 3 Am J Bioeth 12. Products in the Pediatric Population E 11 (20th July 2000),
8. Beauchamp and Childress, supra at 1 s. 2.6.3 [Ioi, Clinical Investigation E11].
9. Ibid. Chapter 4B. 22. European Medicines Agency (EMEA), Ethical
10. R Rhodes, "Rethinking Research Ethics" (2005) 5 Am J Considerations for Clinical Trials on Medicinal
Bioeth 7. Products Conducted with the Paediatric Population
11111 I II II .11 11111J
(Recommendations of the Ad Hoc Group for the with Life Threatening Conditions" (2001), online National
Development of Implementing Guidelines for Directive Hospice and Palliative Care Organization.
2001/20/EC Relating to Good Clinical Practice in 38. American Academy of Pediatrics, "Guidelines," ibid. at 292.
the Conduct of Clinical Trials on Medicinal Products 39. Ibid.
for Human Use) (2008), online s. 15 [EMEA, Ethical 40. D Tomlinson et al., Challenges to Participation in
Considerations for Clinical Tr als1. Paediatric Palliative Care Research: A Review of the
23. TB Ernest, DP Elder, LG Man ni, M Roberts & JL Ford, Literature (2007) 21 Palliative Medicine 435.
"Developing Paediatric Medicines: Identifying the 41. CM Ulrich et al., "Respondent Burden in Clinical Research:
Needs and Recognizing the Challenges" (2007) 59 When Are We Asking Too Much of Subjects?" (2005) 27
Journal of Pharmacy and Pharmacology 1043 [Ernest, 1RB 17.
"Paediatric Medicines"]; Saint A Raymond & D Brasseur, 42. DJ Casarett & JH Karlawish, Are Special Ethical Guidelines
"Development of Medicines for Children in Europe: Needed for Palliative Care Research? (2000) 20 J Pain Symptom
Ethical Implications" (2005) 6 Paediatric Respiratory Manage 130; Davies, "Palliative Care," supra note 36 at 191.
Reviews 45; Marilyn J Field & Richard E Behrman 43. D Tomlinson et al., "Challenges to Participation in Paediatric
eds., (Institute of Medicine of the National Academies, Palliative Care Research: A Review of the Literature," (2007)
Committee on Clinical Research Involving Children- 21 Palliative Medicine 435 [Tomlinson "Challenges"].
Board on Health Sciences Policy), Ethical Conduct of 44. To ease the reading of this document, the term "parents" also
Clinical Research Involving Children (Washington refers to legal representative or legal guardian of the child.
The National Academies Press., 2004). 45. C1HR, Tri-Council Policy Statement, supra note 4, art. 3.12.
24. R Truog, "Increasing the Participation of Children in 46. RC Barfield & C Church, Informed Consent in Pediatric
Clinical Research" (2005) 31 Intensive Care Med Clinical Trials (2005) 17(1) Current Opinion in Pediatrics 20.
760. 156. 47. 21 C.F.R § 312.21.
25. TP Klassen, L Harding, JC Craig & M Orrfinga, "Children 48. Health Canada, Guidance for Industry: Health Canada
Are Not Just Small Adults: The Urgent Need for High Addendum to ICH Guidance Document Ell: Clinical
Quality Trial Evidence in Children" (2008) 5 (8) PLOS 0001. Investigation of Medicinal Products in the Pediatric
26. aoms, Guidelines 2002, supra note 17. Population, Ottawa (2003), art 6.1.
27. DG Waller, "Off-Label and Unlicensed Prescribing for 49. Barfield, "Informed Consent," supra note 46.
Children: Have We Made Any Progress?' (2007) 64 Brit J 50. European Medicines Agency (EMEA), Ethical Considerations
Clin Pharmacology 1. for Clinical Trials on Medicinal Products Conducted with
28. Ernest, "Paediatric Medicines,' supra note 23. the Paediatric Population (Recommendations of the Ad Hoc
29. Ibid. Group for the Development of Implementing Guidelines
30. Health Canada, Health Canada's Decision on for Directive 2001/20/EC Relating to Good Clinical
Cough and Cold Medicines (Ottawa: 2008), online: Practice in the Conduct of Clinical Trials on Medicinal
<http://www.hc-sc.gc.ca/ahc-asclmedia/advisories- Products for Human Use), (2008), s. 5.7, online: <hap://
avis/_2008/2008_184info 1-erig.php>. ec.europa.eu/enterprise/pharmaceuticals/eudralex/vol-10/
31. Canada, "A Canada Fit for Children: Canada's Plan ethical_considerations.pdf>.
of Action in Response to the May 2002 United 51. United Nations, Convention on the Rights of the Child,
Nations Special Session on CF ildren," (2004), online GA Res. 44/25, UN GAOR, 44th Sess., UN Doc. A/
<http://www.hrsdc.gc.ca/eng/cs/sp/sdc/socpo]/ RES/44/25 (1989) s. 12.
publications/2002-002483/canadafite.pdf>. 52. EMEA, Ethical Considerations for Clinical Trials, ss. 7, 27.
32. Food and Drug Administratio a Modernization Act 53. K Senecal, J Samuel & D Avard, Research and the Assent
of 1997, Pub. L. No. 105-115, 2296 Stat. 111; Best of the Child: Towards Harmonization? in D Avard, J
Pharmaceuticals for Children Act of 2002, Pub. L. No. Samuel & BM Knoppers, eds., Paediatric Research in
107-109, 115 Stat. 1408. Canada (Montreal: Les Editions Themis, 2009) at 33; D
33. EMEA, Ethical Considerations for Clinical Trials, supra note 11. Wendler, Assent in Paediatric Research: Theoretical and
34. ICH, Clinical Investigation Ell, supra note 21. Practical Considerations, (2006) 32 J Med Ethics 230;
35. Regulations Amending the Food and Drug Regulations K Toner & R Schwartz, Why A Teenager over Age 14
(Data Protection), C. Gaz. 1996. I. Should Be Able to Consent, Rather Than Merely Assent, to
36. D Davies, "Palliative Care Research Involving Children and Participation as a Human Subject of Research" (2003) 3(4)
Adolescents," in D Avard, J Samuel & BM Knoppers, eds., The American Journal of Bioethics 39. Also see Canadian
Paediatric Research in Canada (Montreal: Les Editions Paediatric Society, Ethical Issues in Health Research in
Themis, 2009) at 191 [Davies, "Palliative Care"]. Children (2008) 13(8) Paediatr Child Health 707.
37. American Academy of Pediatrics, "Guidelines for the 54. International Conference on Harmonisation of Technical
Ethical Conduct of Studies to Evaluate Drugs in Pediatric Requirements for Registration of Pharmaceuticals for
Populations" (1995) 95(2) Pediatrics 286 [American Human Use (tcH), Clinical Investigation of Medicinal
Academy of Pediatrics, "Guidelines"]; Children's Project on Products in the Pediatric Population E 11 (20th July 2000).
Palliative/Hospice Services (CaiPPs), "A Call for Change: 55. International Conference on Harmonisation of Technical
Recommendation to Improve the Care of Children Living Requirements for Registration of Pharmaceuticals for
Human Use (10-1), Good Clinical Practice: Consolidated 2001/20/EC Relating To Good Clinical Practice in
Guidelines E6(R1), (10th June 1996), s. 4.8.12 [ICH, the Conduct of Clinical Trials on Medicinal Products
Guidelines E61; EMEA, Ethical Considerations for Clinical for Human Use) (2008), online: s. 7.2 [EMEA, Ethical
Trials, supra note 1, s. 7.1.2. Considerations for Clinical Trials].
56. Canadian Institutes of Health Research (CIHR), Natural 60. Canadian Institutes of Health Research (CIHR), Natural
Sciences and Engineering Research Council of Canada, Sciences and Engineering Research Council of Canada,
Social Sciences and Humanities Research Council of Social Sciences and Humanities Research Council of
Canada, Tri-Council Policy Statement: Ethical Conduct for Canada, Tri-Council Policy Statement: Ethical Conduct for
Research Involving Humans (2010), art. 3.10 [am, Tri- Research Involving Humans (2010), art. 3.10 [CIHR, Tri-
Council Policy Statement]. Council Policy Statement].
57. Council for International Organization of Medical Sciences 61. Council for International Organization of Medical
(cioms), International Ethical Guidelines for Biomedical Sciences (CIOMS), International Ethical Guidelines
Research Involving Human Subjects (Geneva: 2002), for Biomedical Research Involving Human Subjects
Guideline 14 [ooms, Guidelines 2002]; Council for (Geneva: 2002), Guideline 14 [ooms, Guidelines 2002];
International Organization of Medical Sciences (CIOMS), Council for International Organization of Medical
International Ethical Guidelines for Epidemiological Sciences (CIOMS), International Ethical Guidelines for
Studies (Geneva: 2008), Guideline 14 looms, Guidelines Epidemiological Studies (Geneva: 2008), Guideline 14
20081; EMEA, Ethical Considerations for Clinical Trials, [ooms, Guidelines 2008]; International Conference on
supra note 1, ss. 6.4, 7. Harmonisation of Technical Requirements for Registration
58. CIHR, Tri-Council Policy Statement, supra note 4, art. of Pharmaceuticals for Human Use (10-1), Clinical
3.9(e). Investigation of Medicinal Products in the Pediatric
59. European Medicines Agency (EMEA), Ethical Population E 11 (20th July 2000) s. 2.6.3 licit Clinical
Considerations for Clinical Trials on Medicinal Investigation E 11] .
Products Conducted with the Paediatric Population 62. Ibid.
(Recommendations of the Ad Hoc Group for the 63. ICH, Clinical Investigation Ell, supra note 61.
Development of Implementing Guidelines for Directive 64. CIHR, Tri-Council Policy Statement, supra note 4, art. 3.7.
9.5 Cases
Case 1 and one contained false data which he sent to be used

in trial results.'
Dr Roger Poisson: Fraud in
Dr Poisson defended his actions on the following
Breast Cancer Study grounds: (1) fraud in enrolling ineligible participants
in clinical trials is justified to permit participants access
In 1994, University of Montreal professor of surgery, to beneficial treatments that might otherwise be un-
Dr Roger Poisson, was found guilty by the US Office available to them and because the close follow-up they
of Research Integrity of enrolling ineligible participants receive is often better than clinical care. (2) Eligibility
and falsifying and fabricating data between 1977 and criteria for randomized clinical trials are too restrictive.
1990 in what was then the largest randomized clinical (3) Randomization of research participants to different
study of the safety and efficacy of surgical interventions study arms would prevent bias from enrolling ineligible
in breast cancer patients. The study compared lumpec- participants.'
tomy (with and without radiation treatment) to mastec- It is doubtful that these arguments are sufficient,
tomy. It confirmed what is now generally accepted, but but they are worth thinking about carefully. Enrolling
was contrary to "common sense" and the expectations ineligible participants may sometimes give more access
of many at the time, that lumpectomy-plus-radiation to better treatments and follow-up care, but this may be
was preferable to mastectomy for some types of breast at the expense of slowing scientific progress, adversely
cancer. Dr Poisson recruited more participants into this affecting many more patients than those enrolled in the
study than any other researcher, about 16 per cent of study. We also saw in the introduction to this chapter
all participants in the study. He accomplished this level that it is statistically unusual to receive access to un-
of recruitment in part by falsifying participants' eligi- available beneficial experimental treatments. Similarly,
bility data. He kept a double set of records. One set Dr Poisson's claims that randomization to different
contained the correct data, which he kept for himself, study arms would prevent biased results from enrolling
II IL
ineligible participants cannot be taken seriously. His to this chapter. It can be a psychological challenge
suggestion that eligibility for trials is too restrictive for physician-researchers to run randomized trials
raises a more interesting question. Entry criteria for when they strongly believe one arm to be superior to
studies must not be arbitrary, and there is some evi- another because of the duty of care physicians owe
dence that study trials contain unnecessary exclusion patients. The physician-researcher conflict raises
criteria. However, if studies are opened up too broadly the further question as to whether there is adequate
to give more patients a chance to access potentially scientific and ethical oversight of the conduct of
beneficial therapies, therapeutic misconception is argu- clinical trials. Today, many years after this scandal,
ably motivating research at the potential expense of sci- there remains no direct, on-site independent ethical
ence and future patients. oversight of trials after they receive initial review by
Dr Poisson's behaviour arguably reflects more gen- a research ethics board. Scientific monitoring is also
eral problems in combining the role of researcher with limited, typically to on-site review of collected data,
caregiver, which were discussed in the introduction which can be falsified.
Notes
1. Arthur Schafer, "The Moral Anatomy and Ethical 2. Charles Weijer, "The Breast Cancer Research Scandal,"
Pathology of the Randomized Clinical Trial," in P.P.D. Canadian Medical Association Journal 152 (1995) no. 8,
Deyn, ed., The Ethics of Animal and Human Experimentation 1195-7.
(John Libby & Company, 1994).
Case 2 unconscious for four days and was discharged 10 days

Mr Halushka: Human Research and later, having suffered brain damage and reduction of
his mental abilities.
Harm to Participants Mr Halushka brought a lawsuit against the re-
searchers and the university for trespass to the person
In August 1961, Walter Halushka participated in a and negligence. A jury found in Mr Halushka's favour,
clinical trial for a new drug at the University of Sas- but the university and researchers appealed. The Sas-
katchewan where he was a student. During his initial katchewan Court of Appeal found that inadequate
visit, Dr Wyant told Mr Halushka that the trial for this disclosure of information had occurred when the
drug had been conducted many times before and that doctor stated the test was safe and failed to state both
the drug was "perfectly safe." Dr Wyant explained that that fluoromar had not been used before and that the
electrodes would be put into Mr Halushka's head and catheter would be inserted directly into the heart and
limbs and a catheter would be inserted through an inci- not just into his arm vein. This negated Mr Halushka's
sion made in his arm; Mr Haluska would be paid $50 consent. In his statement, which has since been used
for participating in the study. Mr Halushka agreed and as the Canadian standard for information disclosure
signed a consent form, which waived liability for the to ensure adequate consent to human research, Justice
researchers and the University of Saskatchewan if there Hall made clear:
were any "untoward effects or accidents."
Two days later, while undergoing the procedure, There can be no exceptions to the ordinary re-
Mr Halushka was given the anaesthetic drug fluoro- quirements of disclosure in the case of research as
mar, which had not been previously tested or used, there may well be in ordinary medical practice. . . .
and the catheter was inserted into his heart cham- The subject of medical experimentation is entitled
bers. However the anaesthetic was too strong and to full and frank disclosure of all the facts, prob-
caused cardiac arrest. Resuscitation had to be done abilities, and opinions which a reasonable man
by making a large incision in Mr Halushka's chest might be expected to consider before giving his
and manually massaging the heart. Mr Halushka was consent.'
Note
1. Halushka v University o f Saskatchewan et al., [1965] 53
DLR (2d)436 (Sask. C.A.).
Cases 1 549
Case 3 assessing, on a case-by-case basis, whether research

Luka Magnotta and the Legal confidentiality is protected from breaches when a re-
search participant's communications may be relevant to
Recognition of Research a criminal trial. The Wigmore criteria are:
Confidentiality
1. The communications must originate in a con-
In May 2012, Luka Magnotta was arrested and charged fidence that they will not be disclosed.
with the gruesome, videotaped murder of a Concordia
University student in Montreal. Five years earlier, 2. This element of confidentiality must be essen-
Magnotta had been a research participant in a Univer- tial to the full and satisfactory maintenance of
sity of Ottawa study of sex workers (Magnotta had been the relation between the parties.
a male escort and pornography actor). A research assist-
ant who had interviewed him told police that Magnotta 3. The relation must be one which in the opin-
had been interviewed for the study. Police then arranged ion of the community ought to be sedulously
to seize the transcript and audiotape of the interview to fostered.
use it in evidence at Magnotta's murder trial, but they
held it in a sealed package while the lead researchers on 4. The injury that would inure to the relation by
the study, professors Chris Bruckert and Colette Parent, the disclosure of the communications must be
applied to the court to rule that the interview was in- greater than the benefit thereby gained for the
admissible. Bruckert and Parent argued that without the correct disposal of litigation.'
court's recognition that the confidentiality of such inter-
views was protected, it would be impossible to conduct The court held that these criteria were met and the
research on sex workers or other individuals living at interview should not be disclosed, observing that its rel-
the margins of the law or beyond. evance was "minimal and at most marginal" to the trial.
The hearing into the admissibility of the interview The decision recognizes researcher—participant
was held in April 2013 by Quebec Superior Court Jus- confidentiality as a legally protected relationship in
tice Sophie Bourque. In January 2014, she ruled that Canadian law However, it remains an interesting ques-
the interview could not be disclosed to the court on the tion whether the case-by-case framework using the
grounds that the public interest in protecting research Wigmore principles will sufficiently protect researcher—
confidentiality outweighed the value of the interview participant relationships in ways that will not threaten
to the trial. Justice Bourque's decision followed the or undermine some types of socially important research
"Wigmore criteria" to assess whether communications affecting human health—for example, research into use
that form part of research are admissible in criminal of illegal recreational drugs, child and elder abuse, and
proceedings. These criteria establish a framework for risky or illegal sexual behaviour.
Note
1. Parent & Bruchert v the Queen re Magnotta, Quebec

Superior Court, 21 January 2014, http://citoyens.soquij
.qc.ca/ID=7E4DBC4D25FC81DCECFB65F7FCA1BFE
Case 4 at the age of 18, Jesse participated in a clinical trial

Jesse Gelsinger: Research Conflicts at the University of Pennsylvania aimed at develop-
ing treatment for infants born with a severe form of
and Ethical Review the disease. To test the safety of the procedure, on
13 September Jesse was injected with an adenoviral
At the age of two, Jesse Gelsinger was diagnosed vector carrying a corrected version of the gene. He
with an X-chromosome-linked genetic disease af- died four days later on 17 September, apparently of
fecting his liver's ability to metabolize ammonia, a a massive immune response triggered by the viral
by-product of protein breakdown. Jesse was able vector leading to multiple organ failure and brain
to survive the normally fatal disease on a restricted death. Jesse was the first person to have died in a
diet and special medications. In September of 1999 clinical trial for gene therapy.
II III III JJ
A subsequent FDA investigation concluded that the • failure to disclose a possible conflict of interest
scientists involved in the study broke several rules of insofar as both the university and the primary re-
conduct, including: searcher had financial stakes in the research.'
• failure to disclose the deaths of monkeys given a Paul Gelsinger, Jesses father, had harsh words also
similar treatment; for the ethicist involved, Arthur Caplan. Dr Caplan
• failure to disclose that two previous patients had worked in the main researcher's department "effect-
experienced serious side effects from the gene ively making the researcher his boss, another serious
therapy, conflict."' The senior Gelsinger was disturbed by Dr
• inclusion of Gelsinger as a substitute for a volun- Caplan's claim that "all the controversy created by
teer who opted out in spite of Gelsinger having Jesse's death was good for the ethics train and that,
elevated ammonia levels, which would typically `we (bioethicists) thrive on scandal,'" claiming that this
have excluded his participation; and statement demonstrated a lack of good judgment.'
Note
1. Paul Gelsinger, "Jesse's Intent," wwwjesse-gelsinger.com,
20 September 2011.
9.6 Study Questions
1. What is therapeutic misconception, and how can it be reflected in research partici-

pants' and researchers' attitudes toward health research? Why is it ethically problem-
atic? What measures should be taken to avoid it?
2. In what ways is obtaining informed consent often more challenging in health research
as compared to the provision of clinical care? You might want to divide your answers
by considering how consent in health research puts burdens on each of the main
components of a valid informed consent: decision-making capacity (competence),
information disclosure, and voluntariness.
3. What conditions must be satisfied for there to be clinical equipoise? Must all research-
ers with relevant expertise be indifferent among the study arms or only the majority of
them? Or is a significant minority enough; if so, what would that be?
4. Does the notion of clinical equipoise properly describe when a multi-arm study is
ethically acceptable? As part of answering this question, consider whether researchers
owe research participants the same duty of care owed to patients.
5. Identify as many examples as possible of conflicts of interest that appear to be com-
monly present in experimental clinical research. What can be done to eliminate or
manage those conflicts?
6. In what circumstances should community consent be sought for research? What are
some ethical pitfalls associated with community consent?
7. What are the ethical issues raised by doing research on vulnerable populations?
Research Ethics Codes and Regulations

Canadian Institutes of Health Research, Natural Sciences and Engineering Research Coun-
cil of Canada, and Social Sciences and Humanities Research Council. 2014. Tri-Council
Policy Statement: Ethical Conduct for Research Involving Humans. http://www.pre.ethics
.gc.ca/eng/policy-politique/initiatives/tcps2-eptc2/Default.
• Note: The TCPS2 Core Tutorial is at http://www.pre.ethics.gc.ca/eng/education/
tutorial-didacticiel.
Council for the International Organizations of Medical Services. 2002. International Eth-
ical Guidelines for Biomedical Research Involving Human Subjects. http://www.cioms.chi
publications/layout_guide2002.pdf.
International Conference on Harmonization. 1996. Guideline for Good Clinical Practice.
http://www.ich.org/fileadmin/Public_Web_Site/ICH_Products/Guidelines/Efficacy/E6/
E6_Rl_Guideline.pdf.
Nuremberg Military Tribunal. 1947. The Nuremberg Code, from U.S. v Karl Brandt et al.
https://history.nih.gov/research/downloads/nuremberg.pdf.
US Department of Health and Human Services, National Institutes of Health, and Office
of Human Research Protections. 2009 (revised). The Common Rule, Title 45 (Public
Welfare), Code of Federal Regulations, Part 46 (Protection of Human Subjects). http://
www.hhs.gov/ohrp/regulations-and-policy/regulations/45-cfr-46.
World Medical Association. 2013. Declaration of Helsinki. http://www.wma.net/en/30
publications/lOpolicies/b3.
Research Ethics Textbooks

Coleman, Carl H., Jerry A. Menikoff, Jess A. Goldner, and Nancy N. Dubler. 2015. The Ethics
and Regulation of Research with Human Subjects. 2nd edn. New Providence, NJ: LexisNexis.
Emanuel, Ezekiel J., Christine C. Grady, Robert A. Crouch, Reidar K. Lie, Franklin G.
Miller, and David D. Wendler, eds. 2008. The Oxford Textbook of Clinical Research Ethics.
New York: Oxford University Press.
Emanuel, Ezekiel J., Robert A. Crouch, John D. Arras, Jonathan D. Moreno, and Christine
Grady. 2003. Ethical and Regulatory Aspects of Clinical Research. Baltimore: Johns Hop-
kins University Press.
Lavery, James V., Christine Grady, Elizabeth Wahl, and Ezekiel Emanuel, eds. 2007. Ethical
Issues in International Biomedical Research: A Casebook. Oxford University Press.
Past Abuses in Human Research

Beecher, Henry K. 1966. "Ethics and Clinical Research." New England Journal of Medicine
274: 1354-60.
Faden, Ruth R., ed. 1996. The Human Radiation Experiments: Final Report of the Advisory
Committee. New York: Oxford University Press.
Goodman, Jordan, Anthony McElligott, and Lara Marks, eds. 2008. Useful Bodies: Humans
in the Service of Medical Science in the Twentieth Century. Baltimore: Johns Hopkins
University Press.
Harris, Sheldon H. 1994. Factories of Death: Japanese Biological Warfare 1932-45 and the
American Cover Up. New York: Routledge.
Hodge, Felicia Schanche. 2012. "No Meaningful Apology for American Indian Unethical
Research Abuses." Ethics & Behaviour 22 (6): 431-44.
Jones, David S., Christine Grady, and Susan E. Lederer. 2016. "Ethics and Clinical
Research—The 50th Anniversary of Beecher's Bombshell." New England Journal of
Medicine 374: 2393-8.
Mosbey, Ian. 2013. "Administering Colonial Science: Nutrition Research and Human
Biomedical Experimentation in Aboriginal Communities and Residential Schools,
1942-1952. Social History 46 (91): 145-72.
O'Mathnna, DOnal P 2006. "Human Dignity in the Nazi Era: Implications for Contempor-
ary Bioethics." BMC Medical Ethics 7 (2).
Pence, Gregory E. 1995. "The Tuskegee Study," in Classic Cases in Medical Ethics, 2nd edn.
New York: McGraw Hill.
Shushman, Miriam. 2005. The Drug Trial: Nancy Olivieri and the Science Scandal that Rocked
the Hospital for Sick Children. Toronto: Random House Canada.
II Ail 111.thl
Weijer, Charles. 1995. "The Breast Cancer Research Scandal." Canadian Medical Association
Journal 152 (8): 1195-7.
Welsome, Eileen. 1999. The Plutonium Files: America's Secret Medical Experiments in the Cold
War. New York: Dial Press.
Topics in Human Research Ethics

Appelbaum, Paul S., Loren H. Roth, and Charles Lidz. 1982. "The Therapeutic Miscon-
ception: Informed Consent in Psychiatric Research." International Journal of Law and
Psychiatry 5 (3-4): 319-29.
Friedman, Benjamin. 1987. "Equipoise and the Ethics of Clinical Research." New England
Journal of Medicine 317(3): 141-5.
Henderson, Gail, et al. 2007. "Clinical Trials and Medical Care: Defining the Therapeutic Mis-
conception." PLoS Med 4 (11). http://journals.plos.org/plosmedicine/article?id=10.1371/
journal.pmed.0040324.
Heyd, David. 1996. "Experimentation on Trial: Why Should One Take Part in Medical
Research?" Jahrbuk fur Recht and Ethik 4: 189-204.
Kopelman, Loretta M. 2004. "Minimal Risk as an International Ethical Standard in Re-
search." Journal of Medicine and Philosophy 29 (3): 351-78.
Lewis, Steven, Patricia Baird, Robert G. Evans, William A. Ghali, Charles J. Wright, Elaine
Gibson, and Francoise Baylis. 2001. "Dancing with the Porcupine: Rules for Governing
University-Industry Relationship." Canadian Medical Association Journal 165 (5): 783-5.
McGuire, Amy, Timothy Caulfield, and Mildred K. Cho. 2008. "Research Ethics and the
Challenge of Whole Genome Sequencing." Nature Reviews Genetics 9: 152-6.
Miller, Franklin D., and Howard Brody. 2002. "What Makes Placebo-Controlled Trials Un-
ethical? American Journal of Bioethics 2 (2): 3-9.
Palys, Ted, and David MacAlister. 2016. "Protecting Research Confidentiality via the Wig-
more Criteria: Some Implications of Parent and Bruckert v The Queen and Luca Rocco
Magnotta." Canadian Journal of Law and Society 31 (3): 473-93.
Palys, Ted, and John Lowman 2014. Protecting Research Confidentiality: What Happens
When Law and Ethics Collide. Toronto: James Lorimer and Company.
Ross, Lainie Freidman. 2006. Children in Medical Research: Access versus Protection. Oxford
University Press.
Shalowitz, David I., and Franklin G. Miller. 2008. "Communicating the Results of Clinical
Research to Participants: Attitudes, Practices, and Future Directions. PLoS Medicine.
http://journals.plos.org/plosmedicine/article?id=10.1371/journal. pmed.0050091.
Weijer, Charles. 1999. "Placebo-Controlled Trials in Schizophrenia: Are They Ethical? Are
They Necessary? Schizophrenia Research 35 (3): 211-18.
Wendler, David. 2006. "Assent in Paediatric Research: Theoretical and Practical Considera-
tions." Journal of Medical Ethics 32: 229-34.
, Leah Belsky, Kimberly M. Thompson, and Ezekiel J. Emanuel. 2005. "Quantify-
ing the Federal Minimal Risk Standard Implications for Pediatric Research without a
Prospect of a Benefit." Journal of the American Medical Association 294 (7): 826-31.
Wendler, David, and Franklin Miller. 2004. "Deception in the Pursuit of Science. Archives
of Internal Medicine 164 (6): 597-600.
Animal Research
DeGrazia, David. 2002. Animal Rights: A Very Short Introduction. New York: Oxford
University Press.
Holms, Helen Bequaert. 1989. "Can Clinical Research Be Both Ethical and Scientific?"
Hypatia 4 (Summer): 154-65.
GDossary
abortion termination of pregnancy. clinical equipoise the state of genuine uncertainty within
the expert community about which study arm is therapeut-
active euthanasia administering a lethal agent to another ically superior in a multi-arm study of an experimental or
person in order to kill the person. existing therapy.
advance directive a legal document made by persons with club good a good that is not rival (in consumption) but is
decision-making capacity to guide medical decisions when excludable.
they lack decision-making capacity.
collective action problem any failure of a group of indi-
age-based rationing rationing based on age and age alone. viduals to achieve a result (for example, the production of a
As used in the literature, it means that if an elderly person good) that everyone values at least as much as the result of
and a non-elderly person each has essentially the same everyone acting in their own self-interest.
medical problem, needs life-preserving treatment, and has
the same prognosis, the treatment can be denied the elderly commercial surrogacy payment to a woman beyond re-
person on the basis of age. imbursement of expenses for agreeing to carry a pregnancy
for another person or couple who will become the child's
applied ethics the application of moral theories, parent(s).
principles, and ideas to specific moral problems.
common-pool resource good a good that is rival (in con-
assent a right that incompetent persons can have to sumption) but is not excludable.
accept or refuse treatment or participation in research.
It is to be contrasted with a competent person's right conflicts of interest situations in which a person has an
to consent. obligation to a person or an institution that is in opposition
to another interest or obligation, which could corrupt the
assisted suicide providing a person with the means of decision-making of that person.
death.
conscientious objection/refusal when health care provid-
beneficence prevent harm, reduce harm, and do good. ers refuse on moral grounds to provide a treatment that is
permitted by the standards of their profession.
best-interest judgment the judgment that the reasonable
person in the patient's position would make. consequentialism rightness is a function of promoting
good consequences.
care ethics morality is fundamentally about caring, fos-
tering positive relationships, and related concepts. Conservative Theory the fetus has a right to life from the
point of conception onwards.
categorical imperative or the supreme principle of the
categorical imperative the central concept in Kant's ethics; cultural relativism what is morally right is determined by
given several formulations. whatever moral standards a culture endorses.
■m I 11 .111 111111
Glossary
decision-making capacity individuals can understand fetus (as used in biology) a pre-natal organism in which
the nature and consequences of their decisions and are not all organ systems are in place and functioning (approx.
subject to distorting mental illness or mental states. Individ- 12-36 weeks old).
uals who have these abilities are also referred to as capable
or competent. free rider a person who benefits from or consumes a good
without contributing sufficiently to its production or con-
deontology rightness is not exclusively, and perhaps not at tinued existence.
all, a function of promoting good consequences.
gatekeeping the role played by health care providers in
determinism every event is brought about and neces- determining who has access to health care services.
sitated by prior causes that occur according to the laws of
nature. genetic exceptionalism the view that genetic information
is a special type of medical information and must therefore
distributive justice the fair allocation of scarce resources. be treated with greater care—for example, with special
laws—than other health care information.
direct or proximate cause the cause that is closest to the
event to be explained. harm principle interference in the conduct of competent
persons is only justified to prevent harm to others.
doctrine of double effect (DDE) the principle that a pro-
posed action that will have benefits but will also cause some Hawthorne Effect the alteration of research participants'
harm is permissible if the action itself is morally appropriate behaviour as a result of knowing that they are being
and the foreseen harm is not intended. observed by researchers, typically by improving their
behaviour.
egalitarian theory of justice every person is equally valu-
able, and society has an obligation to provide them with the health region a geographical area defined for administra-
basic resources for a good life. tive purposes by a provincial or territorial ministry of
health.
euthanasia bringing about the death of a person in that
person's interest. Hippocratic Oath this ancient Greek text requires a new
physician to swear to the healing gods to follow specific
excludable good a good that people can be prevented from ethical standards.
benefitting from without great cost. A non-excludable good
is one where people can be stopped from consuming it only indirect or remote cause a cause that causes the proximate
at great cost. cause.
experimental (or interventional) research attempts to infanticide killing an infant.

advance knowledge by studying the effects of research inter-
ventions on research participants. informed decision a decision made when a person knows
everything he or she would want to know before making the
false negative A test result that is negative but incorrect. A decision.
test result is negative when the test says a patient does not
have the illness being tested for or a predictor for it. instructional advance directive an advance directive
in which persons designate what medical decisions they
false positive a test result that is positive but incorrect. would or would not want made when they lack decision-
A test result is positive when the test says a patient has the making capacity.
illness being tested for or a predictor for it.
Liberal Theory the fetus lacks a right to life at all stages of
feticide killing the fetus. fetal development.
fetus (as generally used in philosophic discussions liberal theory of justice a social view that combines egali-
of abortion) a human organism at any stage of fetal de- tarian and libertarian elements. In the context of resource
velopment from zygote to (but not including) infant. allocation, it means two-tier health care.
Glossary
libertarian theory of justice every person has a right normative ethics the inquiry into the standards that deter-
to private property and to enter into any agreements they mine how to act morally and lead a moral life.
want to. Government can interfere only to prevent force
or fraud. observational research attempts to advance knowledge by
collecting data from research participants without making
luck egalitarianism individuals are responsible for conse- any change to their care, treatment, or lives in relation to
quences if and only if they are the result of their voluntary what is being studied.
choices.
overdiagnosis a correct diagnosis of illness (for example,
macroallocation the fair allocation of a society's resources cancer) where the illness will do no harm to the person
for health care under conditions of scarcity. during that person's lifetime.
MAID assisted suicide or active euthanasia performed by a overtreatment medical treatment for an illness given
physician or a nurse practitioner, as per amendments to the when the illness will do no harm to the person during that
Criminal Code of Canada in 2016. person's lifetime.
maxim Kant's name for a personal rule for acting in a cer- PAD physician-assisted suicide or active euthanasia per-
tain way. On some ideas about what Kant means, this must formed by a physician.
include a statement of the goal of acting and the kind of
action that will achieve the goal. passive euthanasia withholding or withdrawing life-
sustaining treatment (WEST).
mesoallocation the fair allocation of scarce health care
resources within a specific health care region or institution. paternalism interfering with someone for their own good.
metaethics the inquiry into the nature and status of morality. physician-assisted suicide a physician provides a patient
with the means of death.
microallocation the fair allocation of scarce resources to
specific individuals. placebo a treatment all of whose effects are placebo effects;
placebo effects are a good or bad effect of a treatment that
Middle Theory the fetus lacks a right to life up to some occurs because the recipient believes the treatment will have
point or threshold stage in fetal development and has a right the effect.
to life after that time.
positive right a right to be provided with something (a
moral realism moral facts exist independently of the good or benefit) by some institution or person or group. It
evidence for them and about which we can have at least corresponds to a positive duty on the part of that institution
approximate knowledge. or person or group to provide the good or benefit to those
with the right.
natural rights rights that all human beings have just in
virtue of being human and which are not subject to political primary goods general all-purpose means, such as liberty,
control. opportunities, education, and health, necessary for the
pursuit of life goals.
negative right A right not to be stopped by some institu-
tion or person or group from getting something (a good or principles of biomedical ethics non-maleficence, benefi-
benefit). It corresponds to a negative duty on the part of that cence, autonomy, and justice.
institution or person or group not to interfere with those
with the right from getting the good or benefit. professional autonomy roughly, health care providers'
right to say "no" when asked to act in ways that conflict with
non-maleficence do not cause harm. their professional obligations.
non-moral facts states of affairs that can be wholly or proxy directive an advance directive in which persons
adequately described separately from moral values. For designate who is to make decisions for them when they lack
example, biological, psychological, social, physical facts. decision-making capacity.
II .111 III 1.6.1
556 Glossary
randomized, double-blind, placebo-controlled trial a substitute decision-maker a person who makes decisions
study in which research participants are randomly assigned for a patient when the patient lacks decision-making cap-
to either an experimental treatment or a placebo and neither acity. A substitute decision-maker is sometimes also called
they nor the researchers know what they are receiving. a surrogate decision-maker or (when the person has been
legally designated as the decision-maker by the patient when
rationing addressing scarcity of health care resources by capable) a proxy decision-maker.
eliminating services or treatments offered, or excluding
classes of individuals from health services or treatments, or substituted judgment the judgment that a patient would
prioritizing access. make for himself or herself.
rights moral claims with special normative, apparently terminal sedation the use of high doses of pain medication
non-consequentialist, force. to treat a patients suffering at the end of life, with the aware-
ness that such high dosages will hasten the patient's death.
rival (or rivalrous) good a good where one person's bene-
fit reduces or eliminates a benefit to others. In a non-rival therapeutic misconception the failure to understand that
good, one persons benefit or consumption does not affect health research is aimed primarily at producing knowledge
another person's. and is not conducted specifically for the sake of research
participants' health.
screening test a medical test that looks for predictors
or the early stages of illness when a person shows no therapeutic privilege the idea that a medical professional
symptoms. may deceive or withhold the truth from a patient when
providing the truth will produce harm.
sentience the ability to have experiences.
tragedy of the commons when each individual seeks to
slippery slope a situation that is likely to lead to an undesir- maximize personal benefit from a scarce resource so that
able consequence, whether or not one tries to prevent that. demand overwhelms supply and the resource is no longer
available to some or all.
slippery slope argument an argument against allowing a
situation to occur because it is likely to lead to an undesir- triage the assignment of degrees of urgency to wounds,
able situation. injuries, or illnesses to decide the order of treatment of a
large number of patients under conditions of scarcity of
social determinants of health the social and economic medical resources.
conditions in which people are born, grow, live, work, and
age that have effects on their health. utilitarianism right actions are those that maximize overall
happiness or well-being.
strong paternalism it is permissible to interfere with a
competent or incompetent person's decision about treatment virtue ethics morality is principally a matter of realizing
if it is in their best interest. in action character traits (virtues) that express being a good
person.
study arm participants in clinical trials are randomly
assigned to different groups (arms) receiving different inter- voluntary decision a decision made when a person is not
ventions to compare the effectiveness and safety of those subject to undue influences.
interventions.
weak paternalism it not permissible to interfere with a
subjectivism morality is fundamentally about the expres- competent person's informed decision to accept or refuse an
sion of feeling or emotion. offered treatment.
Index
A Third Way (Sumner), 262 Brock, D.W., Standards of Competence, 32; Voluntary
Abortion of a Fetus Due to Diagnosis of Down Syndrome Active Euthanasia, 297
(case), 284 Brody, H., Transparency: Informed Consent in Primary
Abortion of Suspected Female Fetus (case), 283 Care, 51; Critique of Clinical Equipoise: Therapeutic
Abortion through a Feminist Ethics Lens (Sherwin), 268 Misconception in the Ethics of Clinical Trials, 509
Active and Passive Euthanasia (Rachels), 309 Browne, A., Defining Death, 339
Advance Directives for Resuscitation and Other Browne, K., Measles and Free Riders: California's
Life-Saving or Life-Sustaining Measures Mandatory Vaccination Law, 427
(Canadian Medical Association), 76 Buchanan, A.E., Standards of Competence, 32; Right to a
Allocation of Exotic Medical Life-Saving Therapy Decent Minimum of Health Care, 367
(Rescher), 391 Buying a Kidney in India but Requesting Canadian
American Medical Association, Deciding to Forego Life- After-Care (case), 413
Sustaining Treatment, 82
Arndt v Smith (case), 154 Callahan, D., When Self-Determination Runs Amok, 312
Avard, D., Best Practices for Health Research Involving Canada's Highest Court Unchains Injection Drug Users;
Children and Adolescents: Genetic, Pharmaceutical, Implications for Harm Reduction as Standard of
and Longitudinal Studies, 537 Healthcare (Small), 443
Canadian Medical Association, Advance Directives for
BC Physicians "Cherry-Picking" Patients for Ease of Care, Resuscitation and Other Life-Saving or Life-Sustaining
Refusing Those in Need (case), 207 Measures, 76
Beauchamp, T.L., The Concept of Informed Consent, 47 Canadian Medical Association, Caring in a Crisis: The
Best Practices for Health Research Involving Children and Ethical Obligations of Physicians and Society during a
Adolescents: Genetic, Pharmaceutical, and Longitudinal Pandemic, 188
Studies (Avard et al.), 537 Canadian Paediatric Society, Position Statement Treatment
Beyerstein, D., The Functions and Limitations of Decisions Regarding Infants, Children, and Adolescents, 80
Professional Codes of Ethics, 201 Caring in a Crisis: The Ethical Obligations of Physicians
Bioethics Committee, Position Statement: and Society during a Pandemic (Canadian Medical
Treatment Decisions Regarding Infants, Children, Association), 188
and Adolescents, 80 Caulfield, T., Direct-to-Consumer Genetics and Health
Biomedical Conflicts of Interest: A Defence of the Policy: A Worst-Case Scenario?, 434
Sequestration Thesis—Learning from the Cases of Chadwick, R., Ethics of Screening: Is "Screeningitis" an
Nancy Olivieri and David Healy (Schafer), 489 Incurable Disease?, 436
Black, L., Best Practices for Health Research Involving Christie, T., Virtue Ethics as an Alternative to Deontological
Children and Adolescents: Genetic, Pharmaceutical, and Consequential Reasoning in the
and Longitudinal Studies, 537 Harm Reduction Debate, 451
Boyle, J.M., Medical Ethics and Double Effect: The Case of Concept of Informed Consent (Faden & Beauchamp), 47
Terminal Sedation, 316 Consent and Public Health (O'Neill), 60
Breaching Confidentiality (Unger), 128 Controlling the Risks of PAD (British Columbia Supreme
Breast Cancer Screening (case), 473 Court), 321
Brett, A.S., Problem with Futility, 176 Crash of Germanwings Flight 9525 (case), 155
British Columbia Supreme Court, Controlling the Risks of Critique of Clinical Equipoise: Therapeutic Misconception
PAD, 321 in the Ethics of Clinical Trials (Miller & Brody), 509
II III 1111 l
Index
Culture, Power, and Informed Consent: The Impact of Griener, G., Best Practices for Health Research Involving
Aboriginal Health Interpreters on Decision-Making Children and Adolescents: Genetic, Pharmaceutical, and
(Kaufert & O'Neil), 56 Longitudinal Studies, 537
Groarke, L., Virtue Ethics as an Alternative to Deontological
Dalhousie University, Withholding and Withdrawal of and Consequential Reasoning in the Harm Reduction
Potentially Life-Sustaining Treatment, 295 Debate, 451
Daniels, N., When Are Health Inequalities Unjust?—The
Social Determinants of Health, 458 Harrison, C., Involving Children in Medical Decisions,
Deciding to Forego Life-Sustaining Treatment (Judicial 76; Position Statement: Treatment Decisions Regarding
Council & American Medical Association), 82 Infants, Children, and Adolescents, 80
Defence of Abortion (Thomson), 253 Hassan Rasouli and Medical Futility (case), 205
Defense of Unqualified Medical Confidentiality (Kipnis), 117 Having an Estate Sale of One's Organs (case), 413
Defining Death (Browne), 339 Health Care Workers and Flu Shots (case), 471
Department of Justice, Legislative Background: Medical Health Law Institute, Withholding and Withdrawal of
Assistance in Dying (Bill C-14), 323 Potentially Life-Sustaining Treatment, 295
Direct-to-Consumer Genetics and Health Policy: A Worst- Helping Residents to Live at Risk (case), 208
Case Scenario? (Caulfield), 434 Hope, T., Why Undervaluing "Statistical" People Costs
Dissent on Judicial Intervention in Pregnancy and Birth Lives, 377
(Royal Commission on New Reproductive Technologies) How Should Health Data Be Used? Privacy, Secondary Use,
(Scorsone), 281 and Big Data Sales (Kaplan), 134
Do Everything for Mom: Advance Directives and a Human Organs, Scarcities, and Sale: Morality Revisited
Surrogate's Right to Demand Treatment (case), 93 (Kishore), 407
Does Clifford Olson Deserve Medical Treatment? (case), 414 Hurst, S.A., Vulnerability in Research and Health Care;
Dr Ewen Cameron (Ross), 485 Describing the Elephant in the Room?, 526
Dr Nancy Morrison: Nonvoluntary Active Euthanasia of an
Adult (case), 342 Involving Children in Medical Decisions (Harrison et al.), 76
Dr Roger Poisson: Fraud in Breast Cancer Study (case), 547 Is It Ever Right to Violate Rules of Rationing? (case), 207
Dworkin, R., Justice and the High Cost of Health, 372
Jesse Gelsinger: Research Conflicts and Ethical Review
Elizabeth and Eric MacDonald: Assisted Suicide (case,) 345 (case), 549
Emanuel, E.J., An Ethical Framework for Biomedical Judicial Council, Deciding to Forego Life-Sustaining
Research, 495; Protecting Communities in Biomedical Treatment, 82
Research, 518 Judicial Intervention in Pregnancy and Birth (Royal
Endangering Behaviour in a Pregnant Woman (case), 284 Commission on New Reproductive Technologies), 277
Enough: The Failure of the Living Will (Fagerlin & Just Caring: In Defense of Limited Age-Based Healthcare
Schneider), 63 Rationing (Fleck), 399
Ethical Framework for Biomedical Research (Emanuel, Justice and the High Cost of Health (Dworkin), 372
Wendler, & Grady), 495
Ethical Relativism in a Multicultural Society (Macklin), 83 Kant, I., On the Supposed Right to Lie from Benevolent
Ethics of Screening: Is "Screeningitis" an Incurable Disease? Motives, 106
(Shickle & Chadwick), 436 Kaplan, B., How Should Health Data Be Used? Privacy,
Secondary Use, and Big Data Sales, 134
Faden, R.R., The Concept of Informed Consent, 47 Kaufert, J., Culture, Power, and Informed Consent:
Fagerlin, A., Enough: The Failure of the Living Will, 63 The Impact of Aboriginal Health Interpreters on
Fleck, L.M., Just Caring: In Defense of Limited Age-Based Decision-Making, 56
Healthcare Rationing, 399 Kenny, N.E. Involving Children in Medical Decisions, 76
Frader, J., Problem with Futility, 176 Kipnis, K., A Defense of Unqualified Medical
Free Riding and Organ Donation (Glannon), 431 Confidentiality, 117
Functions and Limitations of Professional Codes of Ethics Kishore, R. R., Human Organs, Scarcities, and Sale:
(Beyerstein), 201 Morality Revisited, 407
Knoppers, B.M., Best Practices for Health Research
Genetic Exceptionalism and "Future Diaries": Is Genetic Involving Children and Adolescents: Genetic,
Information Different from Other Medical Information? Pharmaceutical, and Longitudinal Studies, 537
(Murray), 147
George VIs Cancer (case), 153 Legislative Background: Medical Assistance in Dying (Bill
Glannon, W, Free Riding and Organ Donation, 431 C-14) (Department of Justice), 323
Grady, C., An Ethical Framework for Biomedical Luka Magnotta and the Legal Recognition of Research
Research, 495 Confidentiality (case), 549
Index 559
Prenatal Diagnosis and Abortion or Infanticide through

Macklin, R., Ethical Relativism in a Multicultural Declining Treatment (case), 282
Society, 83 Problem with Futility (Truog, Brett, & Frader), 176
Markayla Sault: Caring for an Aboriginal Patient (case), 95 Protecting Communities in Biomedical Research
Marquis, D., Why Abortion Is Immora,1238 (Weijer & Emanuel), 518
Mature Minors, Mental Illness, Advance Directives, Provincial-Territorial Expert Advisory Group on Physician-
and Conscientious Objection (Provincial-Territorial Assisted Dying and the Special Joint Committee on
Expert Advisory Group on Physician-Assisted Dying Physician-Assisted Dying, Mature Minors, Mental Illness,
and the Special Joint Committee on Physician-Assisted Advance Directives, and Conscientious Objection, 328
Dying), 328
Measles and Free Riders: California's Mandatory Vaccination QALYs vs DALYs vs LYs Gained: What Are the Differences,
Law (Browne), 427 and What Difference Do They Make for Health Care
Medical Ethics and Double Effect: The Case of Terminal Priority Setting? (Robberstad), 382
Sedation (Boyle), 316
Medical Secrecy: Patients' Right of Access to Medical Rachels, J., Active and Passive Euthanasia, 309
Records (Schafer), 113 Rationing Services to an Elder Who Is Responsible for His
Miller, F.G., A Critique of Clinical Equipoise: Therapeutic Medical Condition (case), 413
Misconception in the Ethics of Clinical Trials, 509 Relational Approach to Autonomy in Health Care
Mr Halushka: Human Research and Harm to Participants (Sherwin), 35
(case), 548 Repairing Research Integrity (Titus, Wells, & Rhoades), 523
Mr McCullough: Recommending Voluntary Passive Reproductive Freedom, Autonomy, and Reproductive Rights
Euthanasia (case), 344 (Overall), 218
Ms G and Refusal of Treatment while Pregnant (case), 283 Rescher, N.P., Allocation of Exotic Medical Life-Saving
Murray, T.H., Genetic Exceptionalism and "Future Diaries": Therapy, 391
Is Genetic Information Different from Other Medical Rhoades, 14, Repairing Research Integrity, 523
Information?, 147 Right to a Decent Minimum of Health Care (Buchanan), 367
Robberstad, B., QALYs vs DALYs vs LYs Gained: What Are
Nancy B.: Withdrawing Life-Sustaining Treatment the Differences, and What Difference Do They Make for
(case), 341 Health Care Priority Setting?, 382
No Chemotherapy for Anael: Surrogate Refusal of Treatment Robertson, J.A., Preconception Gender Selection, 226
for a Minor Child (case), 92 Romanow, R., Sustaining Medicare: The Commission on the
Nuts and Bolts of Obtaining Consent to Treatment Future of Health Care in Canada, 357
(Rozovsky & Rozovsky ), 45 Ross, C.A., Dr Ewen Cameron, 485
Rowell, M., Involving Children in Medical Decisions, 76
O'Neil, J., Culture, Power, and Informed Consent: The Royal Commission on New Reproductive Technologies,
Impact of Aboriginal Health Interpreters on Decision- Judicial Intervention in Pregnancy
Making, 56 and Birth, 277
O'Neill, 0., Consent and Public Health, 60 Royal College of Paediatrics and Child Health,
On the Moral and Legal Status of Abortion (Warren), 248 Withholding or Withdrawing Life-Sustaining Treatment
On the Supposed Right to Lie from Benevolent Motives in Children, 81
(Kant), 106 Royal Commission on New Reproductive Technologies,
Optimal Care versus Cost Containment: What Is a Doctor to Preconception Arrangements, 233
Do? (case), 412 Rozovsky, F.A., The Nuts and Bolts of Obtaining Consent to
Overall, C., Reproductive Freedom, Autonomy, and Treatment, 45
Reproductive Rights, 218 Rozovsky, L.E., The Nuts and Bolts of Obtaining Consent
to Treatment, 45
Patient and Physician Autonomy: Conflicting Rights and
Obligations in the Physician—Patient Relationship Samuel, J., Best Practices for Health Research Involving
(Pellegrino), 165 Children and Adolescents: Genetic, Pharmaceutical, and
Pellegrino, E.D., Patient and Physician Autonomy: Longitudinal Studies, 537
Conflicting Rights and Obligations in the Physician— SARS in Toronto: A Duty to Care (case), 206
Patient Relationship, 165 Schafer, A., Medical Secrecy: Patients' Right of Access
Personal Genome Mapping (case), 156 to Medical Records, 113; Biomedical Conflicts of
Position Statement: Treatment Decisions Regarding Infants, Interest: A Defence of the Sequestration Thesis—
Children, and Adolescents (Harrison et al.), 80 Learning from the Cases of Nancy Olivieri and
Preconception Arrangements (Royal Commission on New David Healy, 489
Reproductive Technologies), 233 Schneider, C.E., Enough: The Failure of the
Preconception Gender Selection (Robertson), 226 Living Will, 63
II III III
Index
Schuklnek, U., Why Medical Professionals Have No Moral Transparency: Informed Consent in Primary Care
Claim to Conscientious Objection Accommodation in (Brody), 51
Liberal Democracies, 191 Treatment for Jehovah's Witnesses: Adults and Children
Scorsone, S.R., Dissent on Judicial Intervention in (case), 94
Pregnancy and Birth (Royal Commission on New Truog, R.D., Problem with Futility, 176
Reproductive Technologies), 281
Scott Starson: Refusing Treatment while Incompetent Unger, D., Breaching Confidentiality, 128
(case), 91
Sex Selection in Canada (case), 282 Virtue Ethics as an Alternative to Deontological and
Sherwin, S., A Relational Approach to Autonomy in Health Consequential Reasoning in the Harm Reduction Debate
Care 35, Abortion through a Feminist Ethics Lens, 268 (Christie, Groarke, & Sweet), 451
Shickle, D., Ethics of Screening: Is "Screeningitis" an Voluntary Active Euthanasia (Brock), 297
Incurable Disease?, 436 Vulnerability in Research and Health Care; Describing the
Should Physicians Be Gatekeepers of Medical Resources? Elephant in the Room? (Hurst), 526
(Weinstein), 181
Sidarous, M., Involving Children in Medical Decisions, 76 Warren, M.A., On the Moral and Legal Status of
Small, D., Canada's Highest Court Unchains Injection Drug Abortion, 248
Users; Implications for Harm Reduction as Standard of Water in First Nations Communities (case), 474
Healthcare, 443 Weijer, C., Protecting Communities in Biomedical Research,
Smalling, R., Why Medical Professionals Have No Moral 518
Claim to Conscientious Objection Accommodation in Weinstein, M.C., Should Physicians Be Gatekeepers of
Liberal Democracies, 191 Medical Resources?, 181
Standards of Competence (Buchanan & Brock), 32 Wells, J.A., Repairing Research Integrity, 523
Stewart, R.S., Telling Patients the Truth, 110 Wendler, D., An Ethical Framework for Biomedical
Students and Study-Enhancing Drugs (case), 472 Research, 495
Sumner, L.W., A Third Way, 262 When Are Health Inequalities Unjust?—The Social
Sustaining Medicare: The Commission on the Future of Determinants of Health (Daniels), 458
Health Care in Canada (Romanow), 357 When Is Treatment Futile? (case), 209
Sweet. W., Virtue Ethics as an Alternative to Deontological When Self-Determination Runs Amok (Callahan), 312
and Consequential Reasoning in the Harm Reduction Why Abortion Is Immoral (Marquis), 238
Debate, 451 Why Medical Professionals Have No Moral Claim to
Conscientious Objection Accommodation in Liberal
Telling Patients the Truth (Stewart), 110 Democracies (Schuklnek & Smalling), 191
Telling the Truth to Patients: A Clinical Ethics Exploration Why Undervaluing "Statistical" People Costs Lives
(Thomasma), 107 (Hope), 377
Thomasma, D.C., Telling the Truth to Patients: A Clinical Withholding and Withdrawal of Potentially Life-Sustaining
Ethics Exploration, 107 Treatment (Health Law Institute & Dalhousie
Thomson, J.J., A Defence of Abortion, 253 University), 295
Titus, S.L., Repairing Research Integrity, 523 Withholding or Withdrawing Life-Sustaining Treatment
Tracy Latimer: Nonvoluntary Active Euthanasia of a Minor in Children (Royal College of Paediatrics and Child
(case), 343 Health), 81
"A very useful collection of important classic and contemporary readings on a variety
of important biomedical ethics issues that are relevant to Canadian students."
—Jeff McLaughlin, Thompson Rivers University
"The theoretical foundation this text provides is essential to teaching a good course in
medical ethics."
—Melany Banks, Wilfrid Laurier University
HIGHLIGHTS
• Completely revised chapter introductions give students more robust,
updated discussions of the concepts and issues arising in the readings.
• Fully updated, this third edition provides students with current information
and coverage of recent developments in the field, including a discussion of
the new Canadian legislation on physician-assisted dying.
• New chapters on public health and professional ethics introduce students

to these two important areas of biomedical ethics study.
• An ideal mix of classic and contemporary readings—with forty-four readings

new to this edition—introduces students to foundational readings alongside
the most recent developments in biomedical ethics.
• Engaging case studies, including nineteen new to this edition, start

students thinking about the possibilities and ramifications of bioethical
decision-making.
About the Editors

The late JOHNNA FISHER was an instructor in the Department of Philosophy at the University of
British Columbia and Langara College. J.S. RUSSELL is a member and past chair of the Department
of Philosophy at Langara College. ALISTER BROW NE is a faculty emeritus instructor and former chair
of the Philosophy Department at Langara College. LESLIE BURKHOLDER is a senior instructor at the
University of British Columbia.
www.oupcanada.com
SBN-13: 978-0- 9-902228-
OXFORD
UNIVERSITY PRESS MI 0
9 780199'022281

Biomedical Ethics A Canadian Focus Third Edition 9780199022281 0199022283

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Biomedical Ethics A Canadian Focus Third Edition 9780199022281 0199022283

Uploaded by

Copyright:

Available Formats

OXFORD THIRD EDITION

JOHNNA FISHER I J.S. RUSSELL

Oxford University Press is a department of the University of Oxford.

Copyright 0 Oxford University Press Canada 2018

All rights reserved. No part of this publication may be reproduced, stored in

Library and Archives Canada Cataloguing in Publication

Biomedical ethics : a Canadian focus / edited by Johnna Fisher,

Includes bibliographical references and index.

1. Medical ethics Canada. 2. Bioethics—Canada. 3. Medical

R724.8565 2018 174.20971 C2017-906106-2

Cover image: Assembly/Iconica/Getty Images

Oxford University Press is committed to our environment.

Printed and bound in the United States of America

1 Morality and Moral Decision-Making: A Brief Introduction 1

2 Medical Decision-Making: Self-Determination and

A Relational Approach to Autonomy in Health Care 35

2.3 Informed Consent 45

The Concept of Informed Consent 47

Transparency: Informed Consent in Primary Care 51

Culture, Power, and Informed Consent:The Impact of Aboriginal

Informed Consent and Public Health 60

2.4 Substituted Judgments 63

Advance Directives for Resuscitation and Other Life-Saving or

2.5 Best-Interest Judgments 76

Position Statement: Treatment Decisions Regarding Infants, Children,

Withholding or Withdrawing Life-Sustaining Treatment in Children 81

Deciding to Forego Life-Sustaining Treatment 82

Ethical Relativism in a Multicultural Society 83

3 Management of Medical Information 98

Telling the Truth to Patients: A Clinical Ethics Exploration 107

Telling Patients the Truth 110

Medical Secrecy: Patients' Right of Access to Medical Records 113

3.3 Confidentiality of Information 117

Breaching Confidentiality 128

3.4 Genetic Information 147

3.5 Cases 153

3.6 Study Questions 157

4 Professional Ethics 159

The Problem with Futility 176

Should Physicians Be Gatekeepers of Medical Resources? 181

Caring in a Crisis:The Ethical Obligations of Physicians and Society during a

Why Medical Professionals Have No Moral Claim to Conscientious Objection

The Functions and Limitations of Professional Codes of Ethics 201

4.3 Cases 205

Case 1 Hassan Rasouli and Medical Futility 205

4.4 Study Questions 209

5 Reproductive Ethics 211

Preconception Gender Selection 226

Preconception Arrangements 233

A Defence of Abortion 253

A Third Way 262

5.4 Completing Reproduction 277

5.6 Study Questions 285

6 End-of-Life Decision-Making 287

Voluntary Active Euthanasia 297

Active and Passive Euthanasia 309

When Self-Determination Runs Amok 312

Medical Ethics and Double Effect:The Case of Terminal Sedation 316

6.3 Physician-Assisted Dying in Canada 321

Legislative Background: Medical Assistance in Dying (Bill C-14) 323

Mature Minors, Mental Illness, Advance Directives, and

6.4 The Concept of Death and Its Practical Implications 338

6.5 Cases 341