Health Care Anal (2011) 19:269–281

DOI 10.1007/s10728-011-0194-8

ORIGINAL ARTICLE

Withdrawing from Research: A Rethink in the Context

of Research Biobanks

Søren Holm

Published online: 15 September 2011

 Springer Science+Business Media, LLC 2011

Abstract It is generally assumed in research ethics that research participants have

an unconditional right to withdraw from research without any detriment or reprisal.
This paper analyses this right in the context of biobank research and argues that the
traditional shape of the right in clinical research can be modified in biobank research
without incurring significant ethical cost. The paper falls in three parts. The first part
is a brief explication of the philosophical justification of the right to withdraw. The
second part presents a number of extant criticisms of the right. And the third and
final part argues that although a right to withdraw is crucial in relation to biobank
research, such a right has to be specified in a different way to the similar right in
relation to clinical research.

Keywords Biobank
Helsinki declaration
Research ethics
Right to withdraw

Unconditional right to withdraw

Introduction

It is generally assumed that research participants in biomedical research have

an unconditional right to withdraw from ongoing research projects they have
previously consented to participate in. This right is stated in various ways in all of
the central international research ethics documents.
The right first appears as principle 9 in the Nuremberg Code:

S. Holm (&)
Centre for Social Ethics and Policy, School of Law, University of Manchester,
Manchester M13 9PL, UK
e-mail: soren.holm@manchester.ac.uk

S. Holm
Center for Medical Ethics, Faculty of Medicine, University of Oslo, Oslo, Norway

123
270 Health Care Anal (2011) 19:269–281

9. During the course of the experiment the human subject should be at liberty
to bring the experiment to an end if he has reached the physical or mental state
where continuation of the experiment seems to him to be impossible.
It is explicit in the first Helsinki Declaration of 1964 where paragraph III.4b
states:
4b. At any time during the course of clinical research the subject or the
subject’s guardian should be free to withdraw permission for research to be
continued.
The investigator or the investigation team should discontinue the research if in
their judgment, it may, if continued be harmful to the individual.
It is restated in the current version of the Declaration in the central paragraph 22
about consent:
22. In any research on human beings, each potential subject must be
adequately informed of the aims, methods, sources of funding, any possible
conflicts of interest, institutional affiliations of the researcher, the anticipated
benefits and potential risks of the study and the discomfort it may entail. The
subject should be informed of the right to abstain from participation in the
study or to withdraw consent to participate at any time without reprisal.
After ensuring that the subject has understood the information, the physician
should then obtain the subject’s freely-given informed consent, preferably in
writing. If the consent cannot be obtained in writing, the non-written consent
must be formally documented and witnessed.1 (my emphasis)
It occurs in the Council of Europe’s Bioethics Convention in article 5, second
part:
The person concerned may freely withdraw consent at any time.
And the European Community Directive implementing the ICH Good Clinical
Practice rules in Europe states in Article 3.2.1:
2. (e) the subject may without any resulting detriment withdraw from the
clinical trial at any time by revoking his informed consent2;
Finally the US ‘Common Rule’ requires consent procedures to include:

1
World Medical Association Declaration of Helsinki—Ethical Principles for Medical Research
Involving Human Subjects
Adopted by the 18th WMA General Assembly Helsinki, Finland, June 1964 and amended by the 29th
WMA General Assembly, Tokyo, Japan, October 1975, 35th WMA General Assembly, Venice, Italy,
October 1983, 41st WMA General Assembly, Hong Kong, September 1989, 48th WMA General
Assembly, Somerset West, Republic of South Africa, October 1996 and the 52nd WMA General
Assembly, Edinburgh, Scotland, October 2000. Note of Clarification on Paragraph 29 added by the WMA
General Assembly, Washington 2002.
2
DIRECTIVE 2001/20/EC OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL of 4 April
2001 on the approximation of the laws, regulations and administrative provisions of the Member States
relating to the implementation of good clinical practice in the conduct of clinical trials on medicinal
products for human use.

123
Health Care Anal (2011) 19:269–281 271

(8) A statement that participation is voluntary, refusal to participate will

involve no penalty or loss of benefits to which the subject is otherwise entitled,
and the subject may discontinue participation at any time without penalty or
loss of benefits to which the subject is otherwise entitled. (CFR45 §46.116.8)
Many more examples could be given, but these ought to suffice. From these
sources we can gather that the right to withdraw can be characterised by 5
components. The right is:
• Absolute
• Unconditional, and requires no explanation or justification3
• Immediate
• Complete
• Non-tradeable or inalienable
The right has primarily been developed in the context of clinical research, as has
most of our current biomedical research ethics principles; and it was already firmly
established long before large-scale database or biobank based research became
prevalent. It was therefore transferred essentially unchanged to this new area of
research ethics.
The purpose of this paper is to consider whether the traditional right to withdraw
can be transferred to biobank research in this way, or whether we need to rethink the
components of the right.
The paper consists of 3 parts. The first part is a brief explication of the
philosophical justification of the right to withdraw. The second part presents a
number of extant criticisms of the right. And the third and final part argues that
although a right to withdraw is crucial in relation to biobank research, such a right
has to be specified in a different way to the similar right in relation to clinical
research.

The Justification of the Right to Withdraw

In the clinical research situation the justification of a right to withdraw with the
components outlined above is straightforward, although it relies on several distinct
lines of argument and on specific aspects of the clinical research situation. Some of
the lines of argument developed here have already been taken up by critics of the
right to withdraw who have exploited the fact that each line of argument only
supports part of the right as traditionally conceived (see below).
The following is an attempt to develop a justification of the right to withdraw as a
moral right in its own right and not merely as a component in an institutional
framework designed to protect other, presumably more fundamental moral rights. In
a recent article Schaefer and Wertheimer pursue a line of argument justifying a right

3
If there is a moral duty to participate in research as many have argued [5, 7] there may also be a moral
duty not to withdraw from research without good reason. Such a duty would potentially conflict with
unconditionality, since we normally find it acceptable to ask people to explain or justify their prima facie
breaches of moral duties. I will not pursue this line of argument here although I believe it has some merit.

123
272 Health Care Anal (2011) 19:269–281

to withdraw from research as an institutional right that should be afforded to

research participants ‘‘… because of the information asymmetry, inherent uncer-
tainty, and inability to hedge surrounding agreements that involve a potential
violation of their bodily integrity’’ [14, p. 349]. Applying their line of argument and
their conclusions in the biobank context would probably lead to conclusions and
recommendations fairly similar to those reached here.
Arguments from self-determination or autonomy straightforwardly lead to a right
to withdraw. If I do not want to participate in the research any more then it is an
infringement of my autonomy if I am forced to continue participation. It is,
however, unclear exactly how strong such a right will be. Unless we believe that
autonomy is always overriding other moral considerations, there will be instances
where other considerations take precedence. Withdrawing may cause harm to the
researchers, may waste communal resources and may impede important research. If
autonomy was the only consideration involved it is unlikely that we could sustain an
absolute and unconditional right to withdraw. And a justification relying only on
autonomy could not support complete inalienability either. In many circumstances
my autonomy is best served if I can bind myself to a future course of action. A
justification based only in autonomy could therefore justify something like the right
expressed in the Nuremberg Code, but not the more recent and more expansive
right.
But autonomy is not the only relevant moral consideration. Most clinical
research involves bodily interventions, either directly or through the ingestion of
pharmaceuticals, and there is thus a further line of argument for a right to
withdrawal based on the more general right to bodily integrity. This is the line of
argument that best supports the immediacy component. If you are about to stick a
needle into me, it is important that I can stop you now. Or if the colonoscope
suddenly appears much larger than I thought it would be when I consented to
colonoscopy in the research context I need to be able to withdraw my consent with
immediate effect.
However, neither autonomy nor bodily integrity justifies a right to complete
withdrawal if completeness is taken to include withdrawal of data or samples
already collected before the moment of withdrawal. Unless something has
happened that undermines the validity of my original consent it will be the
case that I autonomously consented to the collection of data or samples and
that this consent was valid at the time the collection happened. A right to,
retrospectively withdraw data or samples must therefore have a different
justification.
In order to get from a strong right, to an absolute and unconditional right we then
finally have to add considerations of the differences in power between the
researchers and the research participants that usually obtain. In order to protect
weak research participants from powerful researchers and allow them truly free
choice, the participants have to be able to withdraw without the researchers forcing
them to justify their withdrawal, or in any other way question their decision, even
implicitly. This is especially important in the clinical context because clinical
researchers or their close associates are often also involved in the ordinary clinical

123
Health Care Anal (2011) 19:269–281 273

care of the patient participants, either concurrently with the research or before and
after.4
We also need to recognise that the right to withdraw creates an important and
ethically justifiable incentive for researchers to develop and maintain good research
practice in order to keep withdrawal levels low.
Finally the right to withdraw cannot be seen in isolation. If there was no right to
withdraw we would have to go to much greater lengths to ensure that the consent to
participate in research was fully informed and completely voluntary, and we would
need to control much more strictly and diligently that researchers never deviated
from the research plan that participants had agreed to. Initial consent and a right to
withdraw protect the participant’s interests as one integrated package.
Given that the right to withdraw in its current form must draw on a number of
different lines of supporting argument it may be better to consider it as a family of
slightly different but related ‘rights to withdraw’, or alternatively as a bundle of rights.

Extant Criticisms of the Right to Withdraw

The literature on informed consent is huge, but the literature on withdrawal

relatively sparse. This makes it possible briefly to review the criticisms that have
been made of the right to withdraw, by reviewing the work of a few major critical
contributors to the debate.
Sarah Edwards argues against the view that the right to withdraw is absolute and
unconditional in cases where a competent person has given consent to participate in
a research project. Based on an analysis of autonomy that follows O’Neill and
Korsgaard she argues that [10, 13]:
Consent must conceptually involve some commitment to the project on which
consent is based and, along with it, certain responsibilities to oneself and to
others. Assuming for the moment that the agent is fully aware of the project
and is not coerced into accepting its aims and objectives, and that there are no
changes in the internal or external environment to bring the value of this
choice into question, then it would be irrational not to be committed to it to
some degree (Edwards [3], p. 120, emphasis in original).
It is therefore not wrong to hold participants to this commitment, and for instance
require an explanation of why they now want to withdraw; and furthermore not
wrong for researchers to negotiate with participants about withdrawal.
In a 2008 paper Eric Chwang argues that the right to withdraw is alienable. He
asks us to consider the following scenario:
Suppose a researcher wants to test a new drug that increases hemoglobin
oxygen saturation. The drug is known to be safe, and the researcher just wants
4
Apart from in the justification of the right to withdrawal, this power imbalance plays a major role in the
rule that rejecting participation in research should not affect the patient’s access to standard treatment, for
instance in paragraph 31 of the Helsinki Declaration: ‘‘31. The physician should fully inform the patient
which aspects of the care are related to the research. The refusal of a patient to participate in a study must
never interfere with the patient-physician relationship’’.

123
274 Health Care Anal (2011) 19:269–281

to test its efficacy. The proposed test is also safe: subjects take the drug, and
then their pulse oximetry is measured via finger-clip for the next five minutes.
However, suppose that the drug is so expensive that the researcher cannot
afford to have any subjects remove the clip from their finger within this five-
minute window, if statistical power is to be preserved. Thus, without a
guarantee that subjects will not withdraw, the researcher will not be able to
perform the study. In such a situation, I think that researchers and subjects are
morally permitted to come to an (informed) agreement whereby the subjects
autonomously choose to waive their right to withdraw (Chwang [2], p. 370).
Apart from the persuasive force of the scenario Chwang provides a meticulous
analysis of features of a right that might make that right inalienable or non-
tradeable, and argues that none of these features characterise the right to withdraw
from research in all the circumstances where it can currently be invoked (for a direct
counterargument see [11] and for a more restricted argument in the context of
follow up after xenotransplantation [16]).
Based on his analysis Chwang therefore concludes that:
… we should sometimes allow people to waive their right to withdraw. The
reason we should sometimes allow this is that it respects our autonomy and
allows us to make beneficial agreements, together with the important fact that
the right to withdraw doesn’t deserve special exemption to the general
principle that rights are typically alienable (Chwang [2], p. 378).
Even if we accept this conclusion in general, and I am inclined to accept it, we do
need to note that something like the Nuremberg right is probably inalienable. Even
in Chwang’s scenario there must be a right to withdraw if the pulse oximeter
malfunctions and the probe start to fry the finger of the research subject.
A slightly different argument to the same conclusion is developed in a paper by
Hansson and Hakama who argue that non-exploitative ‘Ulysses contracts’
restricting withdrawal should be allowed between researchers and research
participants, because this promotes the general interests of research participants
in the success of research [6]. This justification raises the interesting question of
whether part of such a Ulysses contract could be a requirement that the researchers
or the sponsors of the research committed to completing the research project and not
withdraw from it. This issue is beyond the scope of the present paper, but it is worth
remembering that we generally accept that a sponsor of a research project can
terminate it at any point without providing a reason to the research participants and
without taking any account of the public good.
Finally Gert Helgesson and colleagues consider the specific issue of withdrawal
from biobank research in two papers that, at least on the face of it seem to be
inconsistent. One paper argues that because the potential harms in biobank research
are smaller than in clinical research withdrawal should only be possible if the
participant can give sufficient reason [4]. The paper then gives the following
explication of this view:
To reflect the different context, the previously cited Nuremberg clause could
be rewritten as follows:

123
Health Care Anal (2011) 19:269–281 275

During the course of the research, a participant should be at liberty to

withdraw his consent if he can present sufficient reasons why it is no longer
reasonable to ask for his continued participation. When such reasons are
presented his sample(s) should be returned, destroyed or de-identified,
according to the wish of the participant. If sufficient reasons are lacking,
continued research is permissible.
Whether or not the reasons to withdraw are sufficient should be judged
primarily by the researchers or biobank holders. However, if they are inclined
to believe that the reasons are insufficient, then consultation with a research
ethics committee is recommended. Certain variations between countries due to
cultural differences can be expected in these kinds of decisions. However, we
feel that genuine, deeply felt concerns (i.e those in line with the person’s
character), not based on misconceptions, should generally be accepted as
sufficient reason. If the concerns are based on misconceptions, then these
misconceptions should be removed, if possible, by informing and explaining
to those concerned (Eriksson et al. [4], p. 1075).
In another paper also published in 2005 Helgesson and Johnson reach a
somewhat different conclusion based on considerations of public trust as a
necessary foundation for successful biobanking:
We maintain that the right to withdraw consent is important for the public trust in
research, and consequently to people’s willingness to participate in research and
to consent to research on their biobank samples. An analysis of the concept of
personal integrity, and its relation to autonomy, supports the view that the right to
withdraw consent to research on biobank samples should be maintained. Similar
considerations support permitting research participants to waive the right to
withdraw consent in a contract. However, for the sake of public trust in medical
research, such waivers must be handled with care (Helgesson et al. [8], p. 321).
It is probably not helpful to speculate concerning which of these conclusions that
reflect the fully considered view of Helgesson and his co-authors. It is, however
worth noting that the first suggestion that withdrawal should only be allowed if the
participant can present ‘sufficient reason’ and that the researchers should be the
judges of whether the reasons given are sufficient seems either naive or dangerous
(or possibly both).

Why Research Biobanks5 are Different

Research biobanks may arise and be structured in many ways (see [15]). The kind of
biobank which is of concern here is the large, non-specific, and open-ended research
biobank that is becoming increasingly common. Biological samples and other data

5
There are many kinds of biobanks (e.g. structured collections of human biological material). This paper
is only concerned with research biobanks that fit the description given in the following paragraphs,
not with therapeutic biobanks or with stem cell banks, or other cell banks where the individual cell line is
the central area of concern.

123
276 Health Care Anal (2011) 19:269–281

are collected from a large number of participants (patients and/or healthy

volunteers) with the intention of storing and analysing these for an indefinite
period of time, and for a number of purposes that cannot be completely specified in
advance (partly because the research questions will develop as knowledge develops,
partly because new analytic methods will become available leading to new
possibilities). The aim of the research is the linking of biological variation and/or
variations in environmental exposures to health outcomes. The individual record
consisting of sample(s) and associated data (data either collected or generated from
the samples) is of interest only as part of the bank, as the input to statistical analysis.
Although the collection of biological samples usually takes place only at entry into
the biobank there may be continuous collection of and/or linkage with health
information from other sources.
The relationship between researchers and participants is different in the biobank
context. Although the typical power differences may be found at the point of the
collection of the original sample, it is usually absent at later points where the
biobank researchers have no possibility of influencing the treatment or conditions of
the participant. The biobank researchers still have social power advantages, but
these are not nearly as important as the power advantages of clinical researchers.
In the remainder of the paper I will assume that all of the samples in the biobank
in question has been originally entered with adequate informed consent, that the
biobank is ethically constituted,6 that it has adequate ethical oversight of its
function,7 and that the research conducted is scientifically valid and has
scientifically and morally important goals, both with regard to the individual
projects and the research programme seen as a whole.

Alternative Conceptions of the Right to Withdraw

What makes the traditional right to withdraw problematic in the biobank context,
and what are the available alternatives to this right?
What makes the right of withdrawal problematic from a pragmatic and scientific
point of view is that it can significantly diminish the scientific value of the biobank,
with the attendant wastage of resources invested. Withdrawal is a cumulative
process, and insofar as new recruitment to the biobank is impossible (for instance in

6
I will leave out discussion of whether it is preferable that biobanks are public, private or constituted
along fiduciary or charitable trust lines [17], as well as discussion about benefit sharing arrangements.
Although the mode of constitution has some linkage with ethics I can see no reason to believe that one of
these forms of constitution is necessarily better. They will for instance all be subject to economic
incentives, simply in order to generate the amount of finance necessary to sustain the functioning of the
bank.
7
Specifying what is ‘adequate’ ethical oversight is not a simple matter, but it will at least include
consideration of each specific use of material and data including consideration of whether that specific use
is ethically justifiable in light of the original consent and the interests of the research participants.
I have previously written about the use of biobanks where materials were originally entered without
consent and will not further analyse these issues here but see [1].
Biobanks with material obtained from children following proxy consent also raise specific issues that I
will not analyse here but see [9].

123
Health Care Anal (2011) 19:269–281 277

time limited cohort biobanks), withdrawals will over time make the biobank smaller
and smaller, and in cohort biobanks less and less representative of the original
cohort of donors, because those who withdraw are very unlikely to be a random
sample of the participants.
The wastage of resources caused by withdrawal is not only a question of
economic waste (and the initial analysis of samples and the sustaining of the
biobank (storage costs etc.) may be very large), but also a question of the effect of
withdrawal on the scientific value of the remaining material in the bank. If someone
withdraws, he not only diminishes the scientific value of the bank, he also
diminishes the value of the contribution of each of those participants who do not
withdraw. In so far as the person withdrawing still wants to benefit from the fruits of
biobank research he is a Rawlsian ‘free rider’, and his free riding is perhaps even
more problematic than standard free riding because he has initially contributed to
the practice of biobanking, but now wants no longer to do his share.8
If some of the components of the right to withdraw can be modified to ameliorate
the problems caused by withdrawal, without incurring important moral costs, there
is good reason to make such modifications. Let us therefore look at the five
components of the traditional right to withdraw, i.e. that it is Absolute,
Unconditional, Immediate, Complete and Inalienable.
Let us first look at the immediacy component of the right. In the context of
biobank research immediacy may not be as important as in the clinical research
context. Research on stored samples does not involve breaches of bodily integrity,
and requiring immediate withdrawal may in some cases significantly sacrifice
further autonomous choice.
Consider the following hypothetical scenario which illustrates the problems that
the immediacy component may create:
A national, populist newspaper publishes a horror story about abuses of data in
a large national biobank, it is picked up by the electronic media and within one
week 20,000 people exercise their right of withdrawal from the biobank. It is
then revealed that the story was completely untrue and that the source was
maliciously intending to wreck an important research project. A lot of the
people who withdrew their data now regret their decision and want to
participate again, but this is impossible because the samples and data have
been destroyed.
In this example the right to immediate withdrawal leads to a result that is worse
for everyone, researchers and research participants alike. It is important to note that
withdrawal in this case has more profound consequences than withdrawal from a
usual clinical research project, by withdrawing from the biobank a person not only
withdraws from one project, but from an indefinite number of future projects,
including possibly some (many) that the person might want to participate in (this is
discussed more below in the section on completeness).

8
A person withdrawing could try to avoid the free rider problem by joining another biobank, but given
the irreversible loss to the first biobank, and the initial costs to be borne by the second biobank he will still
impose costs on others, even though this will only be in the short term.

123
278 Health Care Anal (2011) 19:269–281

But does withdrawal from a biobank have to have immediate and complete
effect? Not necessarily, what is presumably important is that the researchers stop
using my samples and data, not that they destroy them. We can therefore dissociate
completeness from immediacy when considering biobanks. This gives us the
opportunity to consider the option of a cooling off period, i.e. a system where
withdrawal has immediate effects concerning the researchers’ use of my samples
and data, but no immediate effect concerning the samples and data themselves.
The withdrawn samples and data could be parked in ‘limbo’, or be dead-locked
for a period of time (say 3 months) and only destroyed/erased at the end of that
period if the person withdrawing has not changed her mind. Here one could either
employ a system requiring confirmation of the decision to withdraw at the end of the
cooling off period, or a system automatically destroying/erasing if the patient has
not indicated a change of mind.
It is important to note that imposing a cooling off period could be communicated
to prospective participants during the original consent process, and that it
would only minimally infringe autonomy, and would not infringe privacy at all.
Completely blocking access to samples and data protects privacy as well as
destruction/erasure does!
What about the completeness component? Completeness of withdrawal is usually
taken to imply that data about the participant cannot be used further by the
researchers and that they have to stop collecting new data. The participant simply
has to be listed as ‘withdrawn’ or ‘lost to follow up’. In the biobank case this has, as
noted above been taken to imply that samples must be destroyed (or given back to
participants), and data erased.
Let us consider the following alternative options:
• Anonymisation
• With no further generation of data from existing samples
• With further generation of data from existing samples
• Consent to individual projects
• Donor’s committee/donor representation on REC
One possibility which is worth considering as an alternative to destruction is
anonymisation of data and samples, either without permission to generate further
data from the existing biological samples, or with permission to generate further
data.9 This will allow the researchers to include the record in all analyses for which
it contains a complete set of data.10 There are many examples of extremely valuable
research based on asking new questions of old datasets without further data
collection.

9
If there is complete anonymisation it will no longer be possible to link to information outside of the
biobank, or to collect new information about the person who has now withdrawn.
10
Many statistical methods will allow for the analysis of incomplete datasets which may further add to
the continued usefulness of incomplete records.

123
Health Care Anal (2011) 19:269–281 279

If anonymisation is complete the privacy rights of the person withdrawing will be

fully protected, and he will have no further involvement in the research, although
samples and data related to the, but now detached from, the person will still be used.
It is possible to derive potentially identifying information from tissue samples,
for instance by complete DNA sequencing, but this does not in itself show that
anonymity cannot be guaranteed within a specific biobank. Anonymity can be
guaranteed by restricting the information that the biobank is allowed to generate
from the samples.
Another option is to allow participants to withdraw their general consent, but
agree to be asked about consent to specific research projects. This would allow
researchers to re-contact withdrawn participants in situations where their specific
participation would be especially important.
And all of these options could be combined with measures aimed at giving
donors a greater say in the evaluation of the ethics and sensitivities of the research
done by the biobank.
Let us finally consider whether the right to withdraw samples from biobanks
should be absolute and unconditional.
One implication of the traditional view of unconditionality is that it not only
wrong to ask participants to explain why they want to withdraw, but that it is also
wrong to explain to them, or give them information about the wider implications of
their withdrawal. It would traditionally be seen as seriously wrong to give the
withdrawers a leaflet explaining how their withdrawal will impede the research in
question and ask them to reconsider their withdrawal.
I can see no reason why the same principle should apply generally in the biobank
context, and at least one reason why it should not. As noted above the strong power
differentials that justify the unconditionality component in clinical research is
typically absent in biobank research, and can therefore not justify unconditionality
in this new context. Second, in the clinical case there will be a strong temptation for
researchers to highlight the personal implications for the research participant in
terms of worse outcomes and/or treatment (whether or not we have rules prohibiting
such individual framing of the information). In the biobank case there will often be
no personal relationship between researchers and participant (withdrawal may be by
letter or e-mail), and often no personal implications of withdrawal. The risks created
by allowing the giving of information about the general implications of withdrawal
are therefore minimal.
The positive argument in favour of allowing such communication is that we are
usually justified in explaining the consequences of their actions to people if we think
their actions have problematic consequences, and we believe that they are ignorant
of these consequences. Even John Stuart Mill accepted this in the latter part of the
oft-quoted paragraph in ‘‘On Liberty’’:
The object of this Essay is to assert one very simple principle, as entitled to
govern absolutely the dealings of society with the individual in the way of
compulsion and control, whether the means used be physical force in the form
of legal penalties, or the moral coercion of public opinion. That principle is,
that the sole end for which mankind are warranted, individually or collectively

123
280 Health Care Anal (2011) 19:269–281

in interfering with the liberty of action of any of their number, is self-

protection. That the only purpose for which power can be rightfully exercised
over any member of a civilized community, against his will, is to prevent harm
to others. His own good, either physical or moral, is not a sufficient warrant.
He cannot rightfully be compelled to do or forbear because it will be better for
him to do so, because it will make him happier, because, in the opinions of
others, to do so would be wise, or even right. These are good reasons for
remonstrating with him, or reasoning with him, or persuading him, or
entreating him, but not for compelling him, or visiting him with any evil, in
case he do otherwise (Mill [12], p. 9 my emphasis).
Perhaps more contentiously we might also consider whether there should be a
possibility to consent to non-withdrawability for biobank research. In the clinical
research context consenting to non-withdrawability at the point of entry into the
research is highly problematic because it would involve the research participants in
a form of quasi-slavery. They would give away control over their bodily integrity,
and would essentially commit themselves to the specific performance of the
research ‘contract’. But these considerations do not apply in the biobank case. Non-
withdrawability could take several forms:
• Complete non-withdrawability, with further collection of data about the
participant and generation of data from existing samples
• Incomplete non-withdrawability, with no further collection of data about the
participant but either with or without further generation of data from existing samples
Complete non-withdrawability would lock participants into significant continued,
open-ended engagement with the biobank. In many circumstances the most valuable
contribution that participants give to the research enterprise is not their initial
samples, but the perpetual access to their health and other data. If participants have,
for instance come to disagree fundamentally with the research pursued by a biobank
it seems problematic that they were put in a situation where they could not end
their continued contributions of data to that biobank. Allowing consent to complete
non-withdrawability therefore raises very significant ethical issues. In relation to
incomplete non-withdrawability the issues are less acute. Participants can
effectively break off any future engagement with the biobank. This seems
analogous to many other kinds of social cooperation where participants can stop
future engagement without having a right to retrieve their past contributions.

Conclusion

The arguments in this paper have shown that although there must undoubtedly be a
right to withdraw from research that extends to biobank research, the exact contents
of that right does not necessarily have to be the same as in the context of clinical
research.
The arguments further suggest that there are good reasons to rethink the right to
withdraw in biobank research, to achieve a better balance between current and

123
Health Care Anal (2011) 19:269–281 281

future acts of self-determination, as well as a better balance between the protection

of research participants and the common interest in high quality biobank research.
This will necessarily require rewriting of central, international and national research
ethics documents.

References

1. Bennett, R., & Holm, S. (2001). Genetic research on tissues stored in tissue banks’ ISUMA.
Canadian Journal of Policy Research, 2, 106–112.
2. Chwang, E. (2008). Against the inalienable right to withdraw from research. Bioethics, 22, 370–378.
3. Edwards, S. J. L. (2005). Research participation and the right to withdraw. Bioethics, 19, 112–130.
4. Eriksson, S., & Helgesson, G. (2005). Potential harms, anonymization, and the right to withdraw
consent to biobank research. European Journal of Human Genetics, 13, 1071–1076.
5. Evans, H. M. (2004). Should patients be allowed to veto their participation in clinical research.
Journal of Medical Ethics, 30, 198–203.
6. Hansson, M. G., & Hakama, M. (2010). Ulysses contracts for the doctor and the patient. Contem-
porary Clinical Trials, 31, 202–206.
7. Harris, J. (2005). Scientific research is a moral duty. Journal of Medical Ethics, 31, 242–248.
8. Helgesson, G., & Johnsson, L. (2005). The right to withdraw consent to research on biobank samples.
Medicine, Health Care and Philosophy, 8, 315–321.
9. Holm, S. (2005). Informed consent and the bio-banking of material from children. Genomics, Society
and Policy, 1, 16–26.
10. Korsgaard, C. M. (1996). Creating the kingdom of ends. Cambridge: Cambridge University Press.
11. McConnell, T. (2010). The inalienable right to withdraw from research. The Journal of Law, Med-
icine & Ethics, 38, 840–846.
12. Mill, J. S. (1978). In E. Rapaport (Ed.), On liberty. Indianapolis: Hackett.
13. O’Neill, O. (2002). Autonomy and trust in bioethics. Cambridge: Cambridge University Press.
14. Schaefer, G. O., & Wertheimer, A. (2010). The right to withdraw from research. Kennedy Institute of
Ethics Journal, 20, 329–352.
15. Solbakk, J. H., Holm, S., & Hofmann, B. (Eds.). (2009). The ethics of research biobanking. New
York: Springer.
16. Spillman, M. A., & Sade, R. M. (2007). Clinical trials of xenotransplantation: Waiver of the right to
withdraw from a clinical trial should be required. Journal of Law, Medicine & Ethics, 35, 265–272.
17. Winickoff, D. E., & Winickoff, R. N. (2003). The charitable trust as a model for genomic biobanks.
New England Journal of Medicine, 349, 1180–1184.

123