You are on page 1of 822

Handbook of

HUMAN FACTORS in
MEDICAL DEVICE
DESIGN

TAF-ER562X_WINGR-09-0401-C000.indd i 11/9/10 1:53:35 AM


TAF-ER562X_WINGR-09-0401-C000.indd ii 11/9/10 1:53:35 AM
Handbook of
HUMAN FACTORS in
MEDICAL DEVICE
DESIGN
Edited by
Matthew B. Weinger
Michael E. Wiklund
Daryle J. Gardner-Bonneau

Assistant Editor
Lori M. Kelly

Boca Raton London New York

CRC Press is an imprint of the


Taylor & Francis Group, an informa business

TAF-ER562X_WINGR-09-0401-C000.indd iii 11/9/10 1:53:35 AM


CRC Press
Taylor & Francis Group
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742

© 2011 by Taylor and Francis Group, LLC


CRC Press is an imprint of Taylor & Francis Group, an Informa business

No claim to original U.S. Government works

Printed in the United States of America on acid-free paper


10 9 8 7 6 5 4 3 2 1

International Standard Book Number-13: 978-1-4200-6351-6 (Ebook-PDF)

This book contains information obtained from authentic and highly regarded sources. Reasonable efforts have been made to
publish reliable data and information, but the author and publisher cannot assume responsibility for the validity of all materials
or the consequences of their use. The authors and publishers have attempted to trace the copyright holders of all material repro-
duced in this publication and apologize to copyright holders if permission to publish in this form has not been obtained. If any
copyright material has not been acknowledged please write and let us know so we may rectify in any future reprint.

Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmitted, or utilized in any
form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying, microfilming,
and recording, or in any information storage or retrieval system, without written permission from the publishers.

For permission to photocopy or use material electronically from this work, please access www.copyright.com (http://www.copy-
right.com/) or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400.
CCC is a not-for-profit organization that provides licenses and registration for a variety of users. For organizations that have been
granted a photocopy license by the CCC, a separate system of payment has been arranged.

Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identifica-
tion and explanation without intent to infringe.
Visit the Taylor & Francis Web site at
http://www.taylorandfrancis.com

and the CRC Press Web site at


http://www.crcpress.com
Dedication
The editors express their gratitude to Peter Carstensen for making this Handbook possible.

After a distinguished career in aerospace, Peter spent over 30 years with the U.S. Food and
Drug Administration (FDA), where he introduced the agency to human factors engineering
and ultimately served as its human factors team leader. His intelligence, patience, and wit
served him well in his role as a human factors advocate within the government and as a
convener of and participant in international human factors standards committees. Having
retired from the FDA in 2008, Peter is still improving medical device safety and usability
through his efforts as a human factors engineering and regulatory compliance consultant.
On joining the FDA in the mid-1970s, Peter and his colleagues recognized that medical
device mishaps were frequently related to user-interface design shortcomings. Beginning
with initiatives to make anesthesia equipment safer through the application of human fac-
tors design principles and more recently leading the international community to establish
standards for the application of human factors in medical device design, Peter’s tireless
work helped give human factors a “seat at the table” in settings ranging from engineering
meetings to regulatory reviews. As a result, with support from his FDA colleagues (notably
Dick Sawyer, Robert Cangelosi, Ron Kaye, and Michael Mendelson), Peter made a substan-
tial difference in the world, improving the safety of medical devices and likely saving an
untold number of patient lives.

TAF-ER562X_WINGR-09-0401-C000.indd v 11/9/10 1:53:36 AM


TAF-ER562X_WINGR-09-0401-C000.indd vi 11/9/10 1:53:36 AM
Contents
Introduction ..................................................................................................................... ix
Michael E. Wiklund

Chapter 1 General Principles ...........................................................................................1


Michael E. Wiklund and Matthew B. Weinger

Chapter 2 Basic Human Abilities...................................................................................23


Edmond W. Israelski

Chapter 3 Environment of Use .......................................................................................63


Pascale Carayon, Ben-Tzion Karsh, Carla J. Alvarado,
Matthew B. Weinger, and Michael Wiklund

Chapter 4 Anthropometry and Biomechanics ................................................................97


W. Gary Allread and Edmond W. Israelski

Chapter 5 Documentation ............................................................................................153


John W. Gwynne III and David A. Kobus

Chapter 6 Testing and Evaluation ................................................................................201


Edmond W. Israelski

Chapter 7 Controls ......................................................................................................251


Stephen B. Wilcox

Chapter 8 Visual Displays ............................................................................................297


William H. Muto and Michael E. Maddox

Chapter 9 Connections and Connectors ....................................................................... 351


Joseph F. Dyro

Chapter 10 Alarm Design ..............................................................................................397


Stephen B. Wilcox

Chapter 11 Software User Interfaces ............................................................................425


Michael E. Wiklund

vii

TAF-ER562X_WINGR-09-0401-C000.indd vii 11/9/10 1:53:36 AM


viii Contents

Chapter 12 Workstations ................................................................................................471


Michael E. Wiklund

Chapter 13 Signs, Symbols, and Markings ....................................................................543


Michael J. Kalsher and Michael S. Wogalter

Chapter 14 Packaging ....................................................................................................595


Michael E. Maddox and Larry W. Avery

Chapter 15 Device Life Cycle ........................................................................................623


Michael E. Maddox and Larry W. Avery

Chapter 16 Hand Tool Design ........................................................................................645


Richard Botney, Mary Beth Privitera, Ramon Berguer, and
Robert G. Radwin

Chapter 17 Mobile Medical Devices..............................................................................715


Richard Stein and Michael E. Wiklund

Chapter 18 Home Health Care .......................................................................................747


Daryle J. Gardner-Bonneau

Chapter 19 Cross-National and Cross-Cultural Design of Medical Devices .................771


Uvo Hoelscher, Long Liu, Torsten Gruchmann, and Carl Pantiskas

Editors ............................................................................................................................795
Contributors ..................................................................................................................797
Index .............................................................................................................................. 805

TAF-ER562X_WINGR-09-0401-C000.indd viii 11/9/10 1:53:36 AM


Introduction
Michael E. Wiklund

The editors and contributing authors developed this Handbook to promote the design of
safe, effective, and usable medical devices. Broadly speaking, it is a response to the histori-
cally high rate of injuries and deaths that have been directly or indirectly caused by medi-
cal devices with user-interface design shortcomings*. Following the handbook’s guidance
should help medical device developers create user interfaces that are less prone to use error.
The resulting devices will be more likely to support the highest quality of medical care
while enhancing user productivity.

ABOUT HUMAN FACTORS


Readers who are new to human factors should be excited to learn how much the discipline
has to offer the medical device industry, health care providers, and patients. By apply-
ing knowledge about human characteristics to produce effective interfaces, human factors
practitioners help to create medical devices that are easier to learn and use. Specifically,
a well-designed user interface enables device users, such as physicians, nurses, therapists,
and technicians, to draw on their past experience to use a new device, avoid errors, quickly
detect and recover from errors when they occur, feel confident about their decisions and
actions, and be physically comfortable. As such, a good user interface can improve safety
and the quality of patient care. Conversely, a poorly designed user interface can respond in
unexpected ways to user actions, increase workload, induce critical errors that jeopardize
patient and clinician safety, and cause fatigue and discomfort.
Medical device developers who invest in human factors find that increased user interface
quality gives them a competitive advantage in the marketplace because customers prefer
their products to other offerings. The investment also meets a standard of care that has been
established over several decades as more companies incorporate human factors processes
and design practices into their device development process. Moreover, medical device
manufacturers are now compelled to invest in human factors by international standards†
and the expectations set forth by medical device regulators, such as the U.S. Food and Drug
Administration, Health Canada, and Germany’s Federal Institute for Drugs and Medical
Devices.

HANDBOOK CONTENT
As a prelude to detailed design guidance, Chapters 1 through 6 provide advice on funda-
mental topics, such as aligning the interactive nature of medical devices to the expected use

* Institute of Medicine. To err is human – Building a safer health system (Washington, DC: National Academies
Press, 2000) p. 1.
† International Electrotechnical Commission. IEC 60601-1-6: Medical Electrical Equipment – Part 1-6:

General Requirements for Safety – Collateral Standard: Usability. 2004.

ix

TAF-ER562X_WINGR-09-0401-C000.indd ix 11/9/10 1:53:36 AM


x Introduction

environments ranging from hospitals to ambulances to patients’ homes, drawing on anthro-


pometric and biometric data to ensure that designs match the intended users’ bodies and
physical abilities, and conducting usability tests and other evaluations to ensure that devices
perform as intended. Chapters 7 through 19 have an applied design focus, offering guid-
ance on the design of specific types of devices (e.g., Chapter 16, “Hand Tool Design”) as
well as designing devices for particular use environments (e.g., Chapter 18, “Home Health
Care”).
Adapted in part from established design standards and conventions, the design guid-
ance also reflects professional judgment extracted from many years of applied analysis and
design. Therefore, while much of the guidance is founded on empirical data, it is also based
on the contributing authors’ applied design and evaluation experience. Accordingly, read-
ers should exercise their own judgment when applying the design guidance and note that
published standards should always take precedence. Although this Handbook’s content is
the most current available as of early 2007, other resources should be consulted, as needed,
to ensure that the design guidelines remain valid.

RELATIONSHIP TO AAMI STANDARDS


The genesis for this Handbook was the work of the Association for the Advancement
of Medical Instrumentation’s (AAMI’s) Human Factors Engineering Committee, which
produced ANSI/AAMI HE-74-2001, Design Process for the Human Factors Engineering
of Medical Devices. This standard provides design process guidance but no specific device
design guidelines. The committee recognized the need for a comprehensive resource for
human factors design guidance that was specifically tailored to medical devices. One
of the committee’s first national standards, ANSI/AAMI HE-48-1993, Human Factors
Engineering Guidelines and Preferred Ppractices for the Design of Medical Devices,
contained a modest amount of design guidance, much of which was adopted from an
existing standard for military equipment (MIL-STD-1472—Human Factor Engineering).
The Committee concluded that a subgroup of its members, augmented by outside human
factors specialists, should independently produce a handbook that the committee could
then draw on to produce an updated national standard.
As planned, AAMI’s Human Factors Engineering Committee drew content from this
handbook to produce its own design standard for medical device user interfaces (ANSI/
AAMI HE-75-2007, a complement to ANSI/AAMI HE-74-2001). However, there are nota-
ble differences between the AAMI’s design standard and this Handbook:

• This Handbook includes expanded discussions of design issues, product design


case studies, and supporting illustrations.
• AAMI’s standard has a terse writing style and explicit organizational structure befit-
ting a standard.
• Readers can expect minor differences in the guidance found in the two
documents.
• This Handbook’s content was not vetted by the AAMI committee, which includes
representatives from numerous medical device manufacturers as well as indepen-
dent medical product design professionals.

TAF-ER562X_WINGR-09-0401-C000.indd x 11/9/10 1:53:36 AM


Introduction xi

INTENDED AUDIENCE
This Handbook should be a valuable resource for the following people involved in the
myriad aspects of medical device development, marketing, and support:

• Clinical engineers • Patient safety and medical device


• Design strategists/managers researchers
• Electrical engineers • Product liability litigators
• Graphic designers • Product planners/managers
• Human factors engineers/ • Regulatory affairs specialists
ergonomists • Quality assurance specialists
• Industrial designers • Risk analysts/managers
• Instructional media developers/ • Safety engineers
writers • Software developers
• Marketing specialists • User-interface designers
• Mechanical engineers
• Medical practitioners

HOW TO USE THIS HANDBOOK


The Handbook can serve several purposes:
1. Medical device developers can convert pertinent design guidelines into product
requirements.
2. Medical device evaluators, including test and evaluation personnel and regulators,
can set performance criteria for design evaluations (e.g., inspections and usability
tests) based on the design guidelines.
3. Human factors advocates can cite the Handbook and specific content to raise their
organizations’ user-interface quality standards and secure resources for human fac-
tors programs.
4. Students and interested professionals can learn about good human factors design
practices.

Each chapter is written to stand alone, assuming that people will refer to the subject of interest
rather than read the Handbook from start to finish. Therefore, certain topics are intentionally
addressed by multiple chapters, ensuring the completeness of technical discussion but lead-
ing to some inconsistencies because of varying author perspectives and opinions on design.
While each chapter presents its own approach to promoting user-centered design, most
share a common organization scheme. Most begin with a general introduction to the
selected topic, followed by the presentation of general and special design considerations
and then specific, numbered design guidelines. Some chapters include one or more cases
studies to instantiate the guidance, and conclude with a listing of resources, literature, and
Web site references.
Guidelines are numbered sequentially and prefixed with the number of the chapter in
which they appear (e.g., the guidelines presented in Chapter 11, “Software User Interface,”
are numbered from 11.1 to 11.105). Most guidelines employ the term “should” to promote
a desirable design characteristic or performance level. In contrast, design standards usually

TAF-ER562X_WINGR-09-0401-C000.indd xi 11/9/10 1:53:36 AM


xii Introduction

employ the term “shall.” By deliberately using the term “should,” the authors leave room for
designer judgment and acknowledge that this Handbook serves an advisory purpose rather
than establishing mandates.
Chapters include numerous cross references to related guidance found in other chap-
ters. This is especially true of closely related chapters, such as Chapter 7, “Controls,” and
Chapter 8, “Visual Displays.” Chapters also include numerous tables and illustrations to
further elucidate technical discussions and improve readability.

CLOSING REMARKS
Undoubtedly, this Handbook could cover many more topics in the manner of a multivolume
encyclopedia. Also, each chapter could expand to cover a wider range of design issues.
However, the editors are satisfied that the 19 chapters focus on some of the most important
human factors issues facing medical device developers while providing numerous refer-
ences to other valuable resources.
We hope that our readers will appreciate the importance of designing medical devices
compatible with human needs and preferences. Enhanced medical device user interfaces
are certain to make a positive difference in the lives of caregivers and patients alike.

TAF-ER562X_WINGR-09-0401-C000.indd xii 11/9/10 1:53:36 AM


1 General Principles
Michael E. Wiklund, MS, CHFP; Matthew B. Weinger, MD

CONTENTS
1.1 Seek User Input............................................................................................................2
1.1.1 Involve Users Early and Often ..........................................................................2
1.1.2 Refine Designs through Usability Testing ........................................................3
1.2 Establish Design Priorities...........................................................................................4
1.2.1 Err on the Side of Design Simplicity ................................................................4
1.2.2 Ensure Safe Use ................................................................................................4
1.2.3 Ensure Essential Communication .....................................................................5
1.2.4 Anticipate Device Failures................................................................................5
1.2.5 Facilitate Workflow ...........................................................................................5
1.3 Accommodate User Characteristics and Capabilities ..................................................6
1.3.1 Do Not Expect Users to Become Masters.........................................................6
1.3.2 Expect Use Errors .............................................................................................6
1.3.3 Accommodate Diverse Users ............................................................................7
1.3.4 Maximize Accessibility ....................................................................................7
1.3.5 Consider External Factors That Influence Task Performance...........................8
1.4 Accommodate Users’ Needs and Preferences .............................................................8
1.4.1 Accommodate User Preferences up to a Point ..................................................8
1.4.2 Do Not Rely Exclusively on “Thought Leaders” ..............................................9
1.4.3 Enable Users to Set the Pace .............................................................................9
1.5 Establish Realistic Expectations of Users ..................................................................10
1.5.1 Do Not Rely on Training ................................................................................10
1.5.2 Do Not Rely on Instructions for Use...............................................................10
1.5.3 Do Not Rely on Warnings ...............................................................................11
1.5.4 Do Not Rely on Memory ................................................................................11
1.5.5 Avoid Information Overload ...........................................................................11
1.5.6 Do Not Assign Users Tasks That Are Better Suited to Other Users
or Devices .......................................................................................................12
1.6 Consider Real-World Demands..................................................................................12
1.6.1 Consider the Context of Use ...........................................................................12
1.6.2 Consider Worst-Case Scenarios ......................................................................13
1.6.3 Make Devices as Rugged as Necessary ..........................................................13
1.6.4 Limit User Workload ......................................................................................14
1.6.5 Consider the Potential for Device Migration into Additional Uses or
Use Environments ...........................................................................................15

TAF-ER562X_WINGR-09-0401-C001.indd 1 11/8/10 11:39:53 PM


2 Handbook of Human Factors in Medical Device Design

1.7 Develop Compatible Designs .....................................................................................15


1.7.1 Accommodate Mental Models ......................................................................15
1.7.2 Establish Natural or Conventional Mappings ...............................................15
1.7.3 Follow Industry Conventions and Consensus Standards ..............................17
1.8 Optimize User Interactions ........................................................................................17
1.8.1 Make Devices Error Tolerant ........................................................................17
1.8.2 Fail in a Safe Manner ....................................................................................17
1.8.3 Avoid Physical Strain, Repetitive Motions, and Cumulative Trauma ...........18
1.8.4 Enable Users to Anticipate Future Events .....................................................18
1.8.5 Confirm Important Actions ...........................................................................18
1.8.6 Make Critical Controls Robust and Guard Them .........................................19
1.8.7 Clarify Operational Modes ...........................................................................19
1.8.8 Employ Redundant Coding ...........................................................................20
1.8.9 Design to Prevent User Confusion ................................................................20
1.8.10 Don’t Neglect Device Appeal .......................................................................21
1.9 Summary ...................................................................................................................22
Resources ...........................................................................................................................22
References ..........................................................................................................................22

Although an understanding of detailed human factors guidelines is helpful when designing


a medical device, a command of the general design principles of human factors engineer-
ing is critical. After all, clinicians and caregivers are usually able to cope with devices that
have specific design shortcomings as long as the flaws do not lead to serious use errors or
pose insurmountable obstacles to accomplishing a task. In fact, few medical devices are
perfect from a user-interface design standpoint. They usually violate one specific guideline
or another. It’s another story altogether if a medical device violates a general human factors
design principle.
Serious violations, such as presenting information at an excessive pace or expecting the
user to carefully read an instruction manual before using a device, can render a medical
device unsafe or unusable. Accordingly, designers should focus on meeting the high-level
design principles before they perfect the details. After all, there is no sense in refining a
fundamentally flawed device. In contrast, great products arise from fundamentally correct
solutions that are subsequently honed to a state of excellence. This chapter provides an
overview of these human factors. Many of these principles will be echoed or built on in
subsequent chapters.
This chapter presents several high-level design principles intended to help designers pro-
duce fundamentally correct user interfaces. For those readers unfamiliar with the human
factors design process, reference to ANSI/AAMI HE-74 (2001) may be of value.

1.1 SEEK USER INPUT


1.1.1 INVOLVE USERS EARLY AND OFTEN
Medical device users can offer invaluable guidance at each stage of user-interface develop-
ment. Early in the design process, they can critique existing devices, explain contextual
factors that must be accommodated in the design, offer a vision of user interactions with

TAF-ER562X_WINGR-09-0401-C001.indd 2 11/8/10 11:39:54 PM


General Principles 3

the device, and help set usability objectives. As a user-interface design evolves, users can
comment on features they like and dislike, describe design attributes that give them trouble,
and participate in more formal usability tests (see Chapter 6, “Testing and Evaluation”).
Toward the end of the design process, users can help to verify the quality of a near-final
design by participating in a usability test of a working prototype. A high level of diverse
user involvement throughout the design process helps to ensure that the final device is
well suited to the intended users. It also avoids last-minute design modifications to resolve
usability problems.

1.1.2 REFINE DESIGNS THROUGH USABILITY TESTING


Usability testing is a critical component of the human factors engineering process.
Usability testing is a reliable method by which to discern user-interface design issues that
could affect safety, efficacy, and satisfaction. Progressive manufacturers often extend test-
ing beyond primary tasks (e.g., using a defibrillator to shock a patient in cardiac arrest) to
include setup, storage, maintenance, and even repair tasks. In a typical test session, rep-
resentative device users perform typical or critical tasks in an appropriate environment,
which may range from a conference room to a sophisticated, high-fidelity simulation
of the intended clinical care environment (Figure 1.1). The level of test fidelity usually
increases as the device progresses from a concept to a refined prototype. Testing early in
the design process and then several more times as the design evolves is an effective way
to prevent user interaction problems from persisting into the later stages of the design
process, a time when effective solutions to problems are more limited and expensive to
implement. It is important to find a sample of test participants who accurately reflect
the range of user characteristics rather than choosing “thought leaders” who bring spe-
cial knowledge and motivation to the test. User characteristics that should be considered
include physical attributes (i.e., traditional ergonomics), abilities and skills, needs, and
psychological attributes. When a device will be used by several distinctly different user
groups (e.g., both physicians and patients), tests should be conducted separately with rep-
resentative participants from each group.

FIGURE 1.1 Clinicians participating in a medical simulation that incorporates several medi-
cal devices and a sophisticated mannequin. (From the Center for Experimental Learning and
Assessment, Vanderbilt University, Nashville, TN. With permission.)

TAF-ER562X_WINGR-09-0401-C001.indd 3 11/8/10 11:39:54 PM


4 Handbook of Human Factors in Medical Device Design

1.2 ESTABLISH DESIGN PRIORITIES


1.2.1 ERR ON THE SIDE OF DESIGN SIMPLICITY
In medical device design, simpler is usually better. Most medical device users dislike devices
equipped with all the “bells and whistles,” especially if the “extras” get in the way of per-
forming basic tasks. Indeed, some medical devices are loaded with features intended to give
them a competitive advantage over competing devices. Yet features aimed at enhancing
sales can cost a company in terms of customer goodwill if they complicate device operation.
Accordingly, developers are well advised to produce devices that focus on the basics and
exclude features offering little day-to-day value. The added complexity of “bells and whis-
tles” can interfere with initial ease of use and is usually not worth it. That said, designers
have to be careful about eliminating advanced features that offer real value to sophisticated
users even if such users represent a small percentage of the overall user population. In such
cases, faced with divergent market needs, a manufacturer should consider developing two
devices rather than a single, compromised version (Figure 1.2). Similarly, for maintenance
tasks, simple is again better. Designers should seek ways to limit the level of skill required to
maintain and repair a device as well as the number of steps and the need for special tools.

1.2.2 ENSURE SAFE USE


Medical devices should minimize the risk of injury to both users and patients, includ-
ing physical and psychological injury, during normal and emergency device operation.
Applying this principle to a CT scanner, designers would promote design solutions that
reduce users’ risk of trauma due to moving parts and of patients feeling claustrophobic
as a result of being enclosed in a tight space. Applying the principle to a portable patient

FIGURE 1.2 Two ultrasound scanners (GE Logiq 9 and SonoSite 180Plus System), developed
by separate manufacturers, are targeted toward different user populations and use environments.
(From http://www.stormoff.com and http://images.google.com. With permission.)

TAF-ER562X_WINGR-09-0401-C001.indd 4 11/8/10 11:39:54 PM


General Principles 5

monitor, designers would avoid placing a heavy instrument on a wheeled pole that is vulner-
able to tipping over, which could cause injury to both clinicians and patients, cause property
damage, and disrupt the care delivery process. On the other hand, user-interface designers
must consider the consequences of dynamic user interactions. For example, in designing a
portable glucose monitor for home use, the control–display relationships must be designed
to minimize the risk of the diabetic user (who may have a preexisting visual impairment
and an acutely abnormal blood sugar) collecting the sample incorrectly or misreading the
resulting displayed value. Thus, potential device-induced user harm can either be due to
static design characteristics (e.g., mechanical consequences of the physical design) or to use
errors during device interaction.

1.2.3 ENSURE ESSENTIAL COMMUNICATION


During busy and stressful times, people often must work harder to communicate with each
other, and this might lead them to raise their voices, repeat themselves to make sure they
were heard, or even grasp someone’s arm to get their attention. Similarly, a well-designed
medical device should be capable of reliably communicating critical information, such as
a change in a patient’s vital signs that could be life threatening, during busy and stress-
ful moments. Accordingly, designers should employ redundant methods to communicate
vital information. Also, where possible, they should provide users with a clear and concise
explanation of any problem (including the source) and how to correct it. For example, the
device may employ both a sufficiently loud audible alarm and a complementary visual
alarm, thereby using two sensory channels to increase the chance of detection. Moreover,
the visual alarm might flash to draw attention, and the audible alarm might be set at an
attention-getting frequency that some would regard as noxious—a means to motivate users
to attend to it. (Note: Standards have taken such needs into account, leading to alarms that
some users do find annoying when they are not contextually appropriate; see Chapter 10,
“Alarm Design.”) Finally, all designs should be evaluated in the context of the overall use
environment, including other devices commonly in use, to ensure that the design solu-
tion does not result in unintended consequences, including impaired clinician–patient,
clinician– clinician, or clinician–device communication.

1.2.4 ANTICIPATE DEVICE FAILURES


Devices will fail. When they do, it is important to communicate the failure to users and, where
possible, indicate the cause and recommend appropriate remedial action. This is especially
important when a device failure, such as the failure of an air-in-blood detector, places a patient
at immediate risk. Ideally, devices will fail safely, but sometimes user intervention is needed
to ensure a safe outcome. Therefore, designers should consider the full range of failure modes
and develop strategies and detailed user-interface or other solutions for coping with them. As
with communicating critical information (see above), it is helpful to communicate both the
cause of the device failure and proper remedial or coping actions in clear and concise terms.

1.2.5 FACILITATE WORKFLOW


Humans are loathe to change to accommodate new medical devices unless there is a clear
benefit in terms of work efficiency or effectiveness. Therefore, designers should take care to

TAF-ER562X_WINGR-09-0401-C001.indd 5 11/8/10 11:39:54 PM


6 Handbook of Human Factors in Medical Device Design

understand the impact that their device might have on task flow, obviously avoiding nega-
tive effects. Additionally, designers should analyze how people will use their device, mak-
ing sure that the user interface is organized to facilitate urgent, frequent, and critical tasks.
For example, in a software interface of a physiological monitor, a dedicated, surface-level
control for recording a patient parameter waveform would be a better design than relegating
this control to a lower-level screen. Potential device uses should be analyzed formally using
techniques such as contextual inquiry, task analysis, or usability testing.

1.3 ACCOMMODATE USER CHARACTERISTICS AND CAPABILITIES


1.3.1 DO NOT EXPECT USERS TO BECOME MASTERS
While well-designed devices should have high learnability, do not overestimate the ability
of users to master a device’s functions. In reality, most users master just the critical (from
their perspective) portions of a device’s functions even if the device is only modestly com-
plex (see Table 1.1). In other words, being practical-minded and often pressed for time,
users master just the functions they use frequently. They tend to disregard other device
functions until they are forced to deal with them, expecting at that time to draw on their
intuition and peer support to operate the device correctly. Accordingly, designers should
make infrequently used tasks maximally intuitive (especially if the task is life-critical)
because most users will approach them in the same manner as novices.

1.3.2 EXPECT USE ERRORS


Many factors can contribute to device use error. Therefore, while maintaining a respectful
view of device users, designers should assume that users will err frequently (Beydon et al.,
2001; Samore et al., 2004). They should not assume that all users will operate a device with
equivalent levels of preparation, attitude, vigilance, and motivation. Rather, a disconcerting
proportion of users may have insufficient training, have forgotten their training since the
last time they used the device, be fatigued from working long hours (Gaba and Howard,
2002; Weinger and Ancoli-Israel, 2002), have insufficient technical aptitude to interact with
complex devices, or be rushed, overworked, or distracted. If anything, designers should
overestimate rather than underestimate the chances of a use error. Thus, designers should
make the errors obvious, provide the means for rapid recovery, and guide users through the
recovery process.

TABLE 1.1
Comparison of the Varying Levels of Mastery of Infusion Pump Setup
and Operation Tasks
Level of Mastery at Performing Specific Tasks
Determine the Total Set up a “Piggyback” Change the
Sample Users Volume of IV Fluid Infusion Battery
Nurse X High Medium Low
Physician Y Medium Low Low
Biomedical engineer Z Medium Low High

TAF-ER562X_WINGR-09-0401-C001.indd 6 11/8/10 11:39:54 PM


General Principles 7

1.3.3 ACCOMMODATE DIVERSE USERS


It is perilous to assume homogeneity in the user population. It is even more perilous for
designers to assume that users are just like them. This is why many human factors research-
ers choose to conduct fieldwork leading to the formulation of personas (also called user pro-
files) of typical users to guide a design effort. Depending on the device, users may indeed
be a comparatively small and specialized population, such as highly trained interventional
cardiologists who operate catheterization laboratory equipment. However, over-the-counter
devices, such as glucose meters, blood pressure monitors, metered dose inhalers, and infant
apnea monitors, will be used by diverse individuals, including the young and old and people
with disabilities. In either case, designers should accommodate the needs of users who have
different sizes, shapes, physical abilities, intellectual capabilities, experiences, and so on (see
Chapter 4, “Anthropometry and Biomechanics,” and Chapter 2, “Basic Human Abilities”).
A simple example of accommodating user diversity is designing a surgical tool that can be
used comfortably by individuals with either small or large hands (see Chapter 16, “Hand
Tool Design”). Other examples include designing a computerized patient data entry screen
for people who have extensive computer experience as well as those with relatively little
experience or a mammography machine that is accessible to both ambulatory individuals
and those in wheelchairs.

1.3.4 MAXIMIZE ACCESSIBILITY


The term accessibility has traditionally been associated with architectural features, such as
sidewalks, building entrances, and restrooms. Features such as curb cuts, automatic doors,
and large restroom stalls equipped with assist bars are products of regulations and politi-
cal activism that have improved accessibility to public spaces. In recent years, consumer
electronic and information technology products have incorporated features to make them
more accessible to users with physical or sensory impairments. For example, every federal
Web site now describes all figures in the text to accommodate people with visual impair-
ments who use screen readers. Similar improvements can be made to medical devices,
making them more usable by people with disabilities. As cited above, imaging devices can
be made more accessible to people in wheelchairs and those with limited range of motion
or sensory-motor control by addressing their needs during device design (Figure 1.3). The

FIGURE 1.3 Mammography machines can accommodate a seated patient. (From http://qap.sdsu.
edu/education/breastcancerreview/Bc_diag/Bccore/photo4d.gif. With permission.)

TAF-ER562X_WINGR-09-0401-C001.indd 7 11/8/10 11:39:54 PM


8 Handbook of Human Factors in Medical Device Design

FIGURE 1.4 First responders might have difficulty reading a defibrillator display if space con-
straints on a commuter train require them to place the unit at an acute viewing angle. (From http://
www.sanarena.ch/icons/fotos/Nothilfetag2004/reanimation1.jpg. With permission.)

same may be said of many other types of medical devices, including examination tables and
all sorts of diagnostic and therapeutic devices. An example of the latter is a glucose meter
that speaks user instructions and meter readings to facilitate use by individuals with visual
impairments.

1.3.5 CONSIDER EXTERNAL FACTORS THAT INFLUENCE TASK PERFORMANCE


Sometimes, people use medical devices in a relatively isolated manner, such as reprogram-
ming an insulin pump while seated at a desk in a quiet room. However, medical devices
are often used in a more dynamic and potentially distracting setting (see Chapter 3,
“Environment of Use”). It might be quite noisy. It may be particularly hot (e.g., using devices
in the course of a rescue conducted outdoors on a 100°F day). Other people or devices may
be vying for the user’s attention. Users may be wearing protective gear (e.g., glasses and
gloves) to prevent injury or contamination. When considering the many potential external
factors, a given design might be found to be incompatible with some uses, such as a para-
medic who is wearing thick gloves and trying to press the right button or to read a display
at an acute angle (see Figure 1.4). Analyzing the resulting design trade-offs (e.g., different
control element choices might be better under different use conditions) is a core part of
effective human factors engineering.

1.4 ACCOMMODATE USERS’ NEEDS AND PREFERENCES


1.4.1 ACCOMMODATE USER PREFERENCES UP TO A POINT
Medical device consumers—particularly large hospitals that purchase large lots—
often ask manufacturers to customize devices according to their institutions’ needs.
Such requests often motivate designers to add configuration options into devices. This

TAF-ER562X_WINGR-09-0401-C001.indd 8 11/8/10 11:39:55 PM


General Principles 9

supports the appropriate goal of making devices adapt to the users rather than the other
way around. However, such adaptability brings with it the risk of user interfaces becom-
ing less stable or less predictable or even compromising performance. Someone accus-
tomed to one device’s setup may be confused when encountering what appears to be the
same type of device but with a completely different setup. For example, a nurse who
works shifts at several different hospitals each week may encounter infusion pumps
that look the same but work quite differently, setting the stage for a use error due to
negative transference. One solution is to limit user-interface variability where possible
by developing optimal solutions for all users and then make the remaining setup differ-
ences obvious.
Consistency across device models regardless of manufacturer (e.g., all parenteral
infusion pumps would have the same controls just like all cars have a steering wheel)
could substantially reduce use errors. This approach may be resisted, however, because
it asks manufacturers to sacrifice brand identity and potential competitive advantages.
Also, manufacturer compliance with de facto industry standards may be compromised
by existing patents and licensing agreements. An alternative is to design a device that can
be readily set to a particular institution’s or individual’s preferences, thereby accommo-
dating market niches while ensuring that such preferences do not impede typical users.
Designers should keep in mind that users are not designers. Thus, although users might
express specific needs and preferences or even suggest a detailed design, this input may
prove unreliable, undesirable, or unworkable. Accordingly, designers should view them-
selves as interpreters, taking and prioritizing user input while also applying their own
expertise and creativity to produce the best possible designs. Selective user input can be
particularly dangerous when care is not taken to rigorously study and evaluate potential
design alternatives.

1.4.2 DO NOT RELY EXCLUSIVELY ON “THOUGHT LEADERS”


Manufacturers are prone to rely on guidance from accomplished and interested clini-
cians, individuals also referred to as “thought leaders.” Manufacturers also tend to rely
more on clinicians who represent large accounts, the goal being to give extra emphasis to
the particular institution’s needs in order to keep their business. Indeed, such individuals
can be an excellent source of design guidance, particularly with regard to identifying use
needs and preferences. However, their relatively sophisticated viewpoint and capabilities
as well as their expanded knowledge of specific design issues and trade-offs lead them
to develop biases. Also, such individuals might push a particular design solution harder
than is appropriate because of their emotional investment in it or even for the sake of ego
gratification. Accordingly, design teams should seek users who are more representative of
the typical user to define user needs and preferences as well as to get reactions to designs
in progress.

1.4.3 ENABLE USERS TO SET THE PACE


Human beings become annoyed when machines set the work pace. Often the pace will be
too slow or too fast because of individual performance differences. Moreover, machine-
paced tasks do not readily accommodate work stoppages due to interruptions, including

TAF-ER562X_WINGR-09-0401-C001.indd 9 11/8/10 11:39:55 PM


10 Handbook of Human Factors in Medical Device Design

FIGURE 1.5 These software user interfaces give users control over the pace of work by present-
ing controls to continue and start tasks. (From http://www.visionchips.com/images/s3.gif and http://
alaris.choc.org/graphics/display_wsk_gdl_lido_new_1b.jpg. With permission.)

emergency situations. Thus, designers should let device users set the work pace by, for
example, requiring them to provide input before proceeding to the next step in a procedure
(Figure 1.5).

1.5 ESTABLISH REALISTIC EXPECTATIONS OF USERS


1.5.1 DO NOT RELY ON TRAINING
Caregivers do not always receive proper training before using a given device. As discussed
earlier, clinicians’ work demands leave little time for training and for reading instruction
manuals. Also, a new employee or a traveling nurse who is filling a temporary gap in staff-
ing may not have received training before he or she uses a device, institutional policies
prohibiting such untrained use notwithstanding. Even when users receive proper training,
they may forget what they learned by the time they use the device, especially if the device is
used infrequently (e.g., a few times per year). Or caregivers who are familiar with a device
might simply forget or be confused about how to perform an uncommon task. Accordingly,
medical devices should be designed for intuitive operation even when users are expected to
be highly trained. Furthermore, if designers anticipate that a medical device may migrate
from clinical use to off-label use, for example, from hospitals to use in people’s homes, it
should be designed for intuitive use by laypersons even if it is not initially intended for use
by such individuals.

1.5.2 DO NOT RELY ON INSTRUCTIONS FOR USE


While some manufacturers produce excellent instructions for use, users might still disre-
gard them in favor of a hands-on demonstration (i.e., an in-service) from a manufacturer’s
representative, staff educator, or knowledgeable peer. The instruction manual will often
be difficult or impossible for users to access while also using the given device (Figure
1.6). Therefore, designers should not count on users reviewing and absorbing information
found only in the instructions for use. While the instructions may describe a device’s theory
of operation in a helpful manner or may even be essential to understanding the device’s
performance, it is unlikely that many users will read them. In most cases, the caregiver’s
workday is too hectic to spend the time to thoroughly read or even skim device instructions.
Users may even neglect or overlook instructions for use placed on the device, perhaps in
the form of a label or online help system. Many users simply ignore instructions in favor of

TAF-ER562X_WINGR-09-0401-C001.indd 10 11/8/10 11:39:55 PM


General Principles 11

FIGURE 1.6 User manuals are often stored in a central location rather than with the medical
device.

other learning methods. Therefore, designers should account for the reduced opportunity
to inform device users through documentation by making devices as intuitive to operate as
possible.

1.5.3 DO NOT RELY ON WARNINGS


The presence of many warnings and on-device instructions often indicates user-interface
shortcomings (see also Chapter 10, “Alarm Design,” and Chapter 13, “Signs, Symbols, and
Markings”). The best way to address a hazard is to eliminate it or design a user interface
to guard against it. As such, warnings should be regarded as the last but nonetheless useful
step toward preventing a problem, especially problems that can lead to property damage
or personal injury. Unfortunately, devices become “papered” with warnings when develop-
ment teams fail to implement the design changes necessary to correct fundamental design
problems. This solution is problematic not only because it leaves fundamental flaws intact
but also because the presence of multiple warnings may lead to “warning fatigue,” causing
users to be less attentive to any warnings on the device.

1.5.4 DO NOT RELY ON MEMORY


People can be forgetful and are often easily distracted. Also, daily life can place excessive
burdens on short-term memory. Therefore, designers should not rely on users to remember
information, such as a test result or numerical code, to perform a task. Moreover, opera-
tional sequences should not require users to remember next steps. It is far better to present
users with the crucial information they need to perform a task correctly. It is also helpful to
bring required tasks to the user’s attention.

1.5.5 AVOID INFORMATION OVERLOAD


Medical devices often flood caregivers with more information than they could possibly
use to accomplish associated tasks. The result can be information overload, a condition
in which the caregiver cannot receive and process the information fast enough for it to
be useful (see Chapter 2, “Basic Human Abilities”). One solution is to provide users with

TAF-ER562X_WINGR-09-0401-C001.indd 11 11/8/10 11:39:55 PM


12 Handbook of Human Factors in Medical Device Design

information of primary interest at the moment it is needed while allowing them to easily
obtain secondary information at a later time. For example, an at-home dialysis patient may
usually want to know that “everything is okay” and determine how long before his or her
treatment is complete but occasionally opt to check his or her blood pressure, the amount of
fluid “taken off,” and other dialysis parameters. Another solution is to preprocess informa-
tion, relieving the caregiver of the task. For example, a device might graph the relationship
between two parameters and indicate minimum and maximum values, thereby performing
the work that users otherwise must do in their heads. The key is to simplify information
while still ensuring that important contextual information and subtle nuances are retained.
Another example is emphasizing values in a list that exceed set limits rather than making
users recall the limits and search the entire list for excursions from those limits.

1.5.6 DO NOT ASSIGN USERS TASKS THAT ARE BETTER


SUITED TO OTHER USERS OR DEVICES
Assign users and machines the tasks they are best qualified to perform and make sure that
the distribution ensures safety and user satisfaction. One effective way to reduce workload is
to let devices perform the functions they do best rather than give people the extra work (see
Chapter 2, “Basic Human Abilities”). For example, a device is usually better at monitoring
a steady-state process for unusual events, a task that is eventually fatiguing for most people
and leads to reduced vigilance. Computer-based devices easily retain information that users
may forget. A robotic device can hold an instrument steadily in a precise position for long
periods of time. On the other hand, some functions (e.g., complex problem solving or pattern
recognition) might be better performed by people and should not be quickly delegated to
technology. Designers should automate tasks skillfully to avoid unintended consequences
on overall user performance. Even if devices might better perform some tasks, designers
should be cautious about shifting so many functions to the device that the users lose their
awareness of the current situation as well as their ability to respond to emergencies. People
generally prefer to be actors rather than observers in a process, except when the required
actions are tedious or fatiguing, divert their attention from more important tasks, or are
clearly performed better by machines.

1.6 CONSIDER REAL-WORLD DEMANDS


1.6.1 CONSIDER THE CONTEXT OF USE
There is a tendency to create designs that work well for trained users who focus their full
attention on operating the device in a quiet environment, such as the previously cited exam-
ple of reprogramming an insulin pump at a desk in a quiet room. But medical devices are
commonly used by diverse users in several different use environments. In addition, medical
devices are sometimes used by distracted, fatigued, and/or marginally trained individuals.
Clinicians often work in chaotic environments and need to focus more attention on their
patients than their equipment. These real-world conditions can have a significant effect on
a user’s interactions with a medical device, masking audible signals and interfering with
concentration, for example. Designers need to consider realistic use conditions. The first
step is to learn about users through field research conducted in a manner that does not
appreciably influence the way people are working. Researchers often discover that medical

TAF-ER562X_WINGR-09-0401-C001.indd 12 11/8/10 11:39:56 PM


General Principles 13

devices are in the periphery of the users’ minds, particularly devices that function in a
relatively autonomous manner, requiring only occasional checking. Later during device
design, the impact of a new device on the ability of users to deliver quality care as well as
its ability to work well with other devices should be tested. Well-designed medical devices
tend to fit naturally into the home or workplace, garnering favorable reviews from users.
Flawed devices tend to draw end users’ dissatisfaction, generating complaints about how
they interfere with the normal workflow, and require too much attention.

1.6.2 CONSIDER WORST-CASE SCENARIOS


In the normal course of device development, mechanical engineers purposely drop devices
to the floor or shake them for hours in a test chamber to see if and how they break. User-
interface designers need to perform equivalent tests of their designs. Thus, user interfaces
must be subjected to worst-case scenarios to see if and how they fail. The goal is to see
how well an untrained or minimally trained user performs when asked to operate an unfa-
miliar device; to observe people using a device under harsh environmental conditions, such
as at night in a moving ambulance (Figure 1.7) or helicopter; and to see what happens
when people are under time pressure to perform a critical and difficult task. By stressing
the user interface, designers can learn how to make the device more effective in the real
world. Note that minimizing the risk of faulty design or use errors causing human injury
requires an integrated approach that includes rigorous design practices, usability testing,
and risk analysis.

1.6.3 MAKE DEVICES AS RUGGED AS NECESSARY


Some medical devices take a beating during their useful life, which may span 10 to 20
years of daily use, and, just like mechanical components, user-interface components should

FIGURE 1.7 Paramedics use myriad medical devices in a moving vehicle that may compli-
cate device-related tasks, such as reading a display and operating controls. (Shutterstock image
31634293.)

TAF-ER562X_WINGR-09-0401-C001.indd 13 11/8/10 11:39:56 PM


14 Handbook of Human Factors in Medical Device Design

FIGURE 1.8 Patient transport device has rugged design to be specifically used on stairs over many
years. (From http://www.ems.stryker.com/detail.jsp?id=6 and Strykker 2005 Catalog, ems_2005_
catalog.pdf.)

be designed to last this long (see Figure 1.8). Heavily used devices that are subject to jos-
tling and impacts require heavier-duty user-interface components. This means selecting
switches that are unlikely to break even if struck with considerable force, screens that will
not crack if elbowed or bumped, and labels that will not become unreadable after years
of scrubbing with antiseptic solutions. The designer will need to balance the need for rug-
gedness against other design goals, such as ease of switch actuation and avoiding display
parallax.

1.6.4 LIMIT USER WORKLOAD


Caregivers are often overworked, enduring vigorous 12-hour or longer shifts, sometimes
several days in a row. This explains why some caregivers actively or passively reject
medical devices that create too much mental or physical work, as discussed earlier.
Clinicians want to focus on their patients, not on distracting technology. Caregivers will
seek shortcuts and work-arounds if a device distracts them from more important tasks
even if the work-saving strategies are strongly discouraged by their institution. Patients
who use medical devices are similarly disinterested in investing undue time or effort
into device use given the many other demands and interests in their lives. Therefore,
designers should pursue opportunities to reduce the time required to learn how to use
and operate devices. For example, a high-quality video complemented by a quick refer-
ence card might shorten the time required to learn to operate a ventilator. As another
example, a point-of-care blood gas analyzer might allow a clinician to commence with
data entry tasks while the device completes a calibration check or prepares to analyze
a sample.

TAF-ER562X_WINGR-09-0401-C001.indd 14 11/8/10 11:39:57 PM


General Principles 15

1.6.5 CONSIDER THE POTENTIAL FOR DEVICE MIGRATION INTO


ADDITIONAL USES OR USE ENVIRONMENTS
It is common for medical devices to migrate from sophisticated medical settings, such as an
intensive care unit, to less sophisticated settings, such as inpatient wards, outpatient clinics,
and patients’ homes (see Chapter 18, “Home Health Care”). Or a device intended for use on
pediatric patients might ultimately be used on adults, such as small women. In some cases,
the device manufacturer might not have anticipated such migration or considered the needs
of the new user population in the original design efforts. This can lead to problems for
both the manufacturer and new users. Consider the case of the infusion pump designed for
use in the hospital but used by the parent of a sick child at home, a so-called off-label use.
Lacking sufficient medical knowledge or training on how to use the device properly, the
parent might experience a use error, leading to a tragic loss of life. In turn, this might lead
to a costly lawsuit against the manufacturer. Or consider the case in which a pulse oximeter
is used as a respiratory monitor in patients receiving narcotic pain medications at home,
but in the presence of supplemental oxygen, the monitor does not alarm until the patient
has stopped breathing, leading to an adverse outcome. Such scenarios underscore the value
of anticipating alternative uses of medical devices. After all, manufacturers are probably
better off protecting users against hazards associated with predictable, unintended device
uses rather than reacting to a product liability or personal injury claim.

1.7 DEVELOP COMPATIBLE DESIGNS


1.7.1 ACCOMMODATE MENTAL MODELS
People frequently have an established mental model or “big picture” in mind when they use
a new device. Usually, their mental model is based on previous experience using similar
devices. For example, users may expect certain controls to function in a particular man-
ner and be surprised by a control that functions differently. Or they may be accustomed to
configuring the device the way they were taught in nursing school, only to find that a new
device requires an alternative approach. Such incompatibilities can make a device harder to
learn to use and can induce errors even among experienced users who may unconsciously
fall into an old use pattern (sometimes called negative transfer of training). Therefore,
designers need to take care to identify established mental models and accommodate them
where possible. When the need for change exists, designers are often better off making
major rather than minor changes so that the difference is more readily apparent. Also,
designers should provide “affordances” that help users form an accurate mental model of
how a device works. Affordances might include organizing the user interface according to
a simple “metaphor,” effective use of labels, clear and redundant feedback in response to
user inputs, helpful warning messages, and a quick reference card that emphasizes how a
device is different from other similar devices.

1.7.2 ESTABLISH NATURAL OR CONVENTIONAL MAPPINGS


When people associate an action with a design element, such as turning a knob clockwise,
they are mapping. An example of natural mapping is squeezing the bag on an anesthesia
machine to fill a patient’s lungs with air; the action and outcome are largely self-evident

TAF-ER562X_WINGR-09-0401-C001.indd 15 11/8/10 11:39:57 PM


16 Handbook of Human Factors in Medical Device Design

FIGURE 1.9 The axial and left–right alignment of this power injector’s piston controls (top of
photo) maps naturally to the movement of the pistons (bottom of photo).

(see Figures 1.9 and 1.10 for other examples of mapping). Turning a knob on an anesthesia
machine clockwise to increase the rate of gas flow is an example of a conventional mapping;
while it may not be as self-evident to a naive individual, experienced anesthesia providers
will perform the task in an automatic, subconscious manner. When mappings are natural or
conventional, users will find the associated devices more intuitive to operate. Incorrect or

Head
Neck

Chest

Abdomen Arm

Hand
Thigh

Calf

Misc Foot

FIGURE 1.10 This power injector’s software screen (unrelated to the power injector shown in
Figure 1.9) provides users with an intuitive means of configuring the device for scanning specific
body parts by mapping control elements to corresponding body parts. (From EZEM [design] and
Wiklund, 1995 [photo].)

TAF-ER562X_WINGR-09-0401-C001.indd 16 11/8/10 11:39:57 PM


General Principles 17

unconventional mappings may cause users to take longer to learn to use a device or increase
use errors. The challenge for designers is to establish effective mappings when the natural
or conventional one that has emerged over time departs from a given company’s established
practice. Or the designer might learn that conventional mappings differ among user popula-
tions (e.g., France, Japan, and the United States). In such cases, it might be appropriate to
customize devices to the given market or develop optimal mapping through a rigorous pro-
cess of design iteration and usability testing. More information on how to design interface
elements to be more intuitive is provided in other sections of this handbook (e.g., Chapter 7,
“Controls,” Chapter 8, “Visual Displays,” and Chapter 12, “Workstations”).

1.7.3 FOLLOW INDUSTRY CONVENTIONS AND CONSENSUS STANDARDS


Manufacturers often seek ways to make their devices stand out from competitors’ devices,
thereby fortifying their brand identity and competitiveness. On the other hand, device users
value consistency, particularly with regard to operational characteristics. When a new
device works like similar devices, including consumer products such as cellular phones and
word processing software, users have less to learn about the new device and can apply past
experience more readily. Human factors specialists call this positive transfer. Therefore,
designers should not diverge substantially from conventional design practice or industry
standards unless there is a compelling reason to do so. There can be very good reasons to
deviate, such as a demonstrable increase in design intuitiveness, task efficiency, or error
prevention. In fact, to foster innovation and continual design improvements, designers
are encouraged to challenge de facto conventions and incorporate new designs if there is
good evidence that the alternative will lead to better user performance. On the other hand,
making a device seem different just for the sake of being different is a poor practice. For
example, designers should not diverge from the standard color codes for alarm signals
simply because they think that a magenta-colored alarm indicator is more conspicuous
than a red one or the company’s branding scheme dictates use of the nonstandard color.
American users have learned that high-priority visual alarms are red, for example, and
expect all devices to adopt this convention.

1.8 OPTIMIZE USER INTERACTIONS


1.8.1 MAKE DEVICES ERROR TOLERANT
As stated previously, devices fail, and users make errors. Consistent with modern principles
of resilience engineering, device designs should be tolerant of error to minimize harm
to users or patients. Design approaches to accomplish error tolerance include failing safe
(see below), considering the device in the overall context of use, providing more informa-
tion about the implications of specific use actions, making errors or unwanted deviations
more visible to users, making potential risks more visible to users, and facilitating error
recovery.

1.8.2 FAIL IN A SAFE MANNER


A basic engineering design principle is to fail safely. Consider the household example of
an iron that shuts itself off if the user fails to do so, possibly preventing a fire. For medical

TAF-ER562X_WINGR-09-0401-C001.indd 17 11/8/10 11:39:58 PM


18 Handbook of Human Factors in Medical Device Design

devices, failing in a safe manner is important because patient lives are at stake. The concept
of failing safely can be extended from electromechanical failures to use errors. For exam-
ple, an infusion device might be designed to prevent users from setting an unsafe level of
drug delivery. A laser treatment device should not be able to be activated if the emergency
stop control is inoperative.

1.8.3 AVOID PHYSICAL STRAIN, REPETITIVE MOTIONS, AND CUMULATIVE TRAUMA


The repetitious nature of many medical procedures, such as firmly squeezing and releas-
ing a surgical stapler, puts caregivers at risk for repetitive motion or cumulative trauma
disorders (see Chapter 16, “Hand Tool Design”). Therefore, designers should seek oppor-
tunities to reduce the number of repetitive actions required to operate a given device (see
Chapter 4, “Anthropometry and Biomechanics”). They should also keep manually applied
forces to a minimum, eliminate pressure points between devices and users, and enable
users to maintain neutral joint positions. Moreover, they should limit the amount of time
users are required to apply a constant force (e.g., continuously squeezing the handles on a
grasping tool) even if the force is relatively small. Often, a relatively simple design change
will achieve these goals, protecting users from injury.

1.8.4 ENABLE USERS TO ANTICIPATE FUTURE EVENTS


To provide the best patient care, caregivers sometimes need to predict the most likely future
course of disease manifestations and therapeutic interventions. In other words, caregivers
try to figure out what is about to happen rather than simply reacting to what has happened
or is happening at the moment. This is especially true in situations in which a caregiver is
delivering a therapy that can have a dramatic effect on the patient’s physiological state, such
as the intravenous delivery of a blood pressure medication. Where possible, designs should
enable caregivers to “see ahead.” For example, a monitoring device might present historical
parametric values as well as values forecasted for the next 5, 10, and 30 minutes.

1.8.5 CONFIRM IMPORTANT ACTIONS


In the medical arena, confirmation messages may serve an important or even critical purpose,
considering that some actions are irreversible and could lead to patient injury (Figure 1.11).

FIGURE 1.11 Sample confirmation message.

TAF-ER562X_WINGR-09-0401-C001.indd 18 11/8/10 11:39:58 PM


General Principles 19

Therefore, even though some users may regard confirmation messages as a hindrance—a
wasted extra step—the benefit of such messages often outweighs the annoyance they cause.
However, the need for users to perform tasks efficiently should not be underestimated when
weighing decisions about which actions require confirmation. Thus, the benefits of confir-
mation messages should be demonstrated through user testing, ensuring that they do not
replace one problem with another, such as users confirming their actions without thinking
about it (i.e., performing tasks in a rote manner).

1.8.6 MAKE CRITICAL CONTROLS ROBUST AND GUARD THEM


Certain medical devices may be exposed to rough handling, particularly when they are
used outdoors or during emergency responses (e.g., patient resuscitation procedures or
“codes”). Medical devices are subject to being dropped or bumped. Therefore, the user
interface needs to be designed to prevent accidental actuation of critical controls. Incidental
contact with a device’s front panel should not, for example, deactivate the device or alter a
critical control setting. This is why many medical devices have physically guarded power
buttons (Figure 1.12) and require the user to confirm critical adjustments (see Chapter 7,
“Controls”). If a critical control fails, an alternative means of control should be provided.
For example, in the case of a damaged pump stop switch, a mechanical means of ceasing
pump action should be provided (and be readily identifiable by the user).

1.8.7 CLARIFY OPERATIONAL MODES


One way that designers seek to simplify medical devices is to incorporate multiple opera-
tional “modes.” In principle, this approach is a sensible way to facilitate context-specific
tasks and to limit users’ exposure to non-relevant device features. However, problems can
arise when the user enters the wrong mode and does not realize it (often called mode error).
It would be problematic, for example, if an anesthesiologist were monitoring an adult patient

FIGURE 1.12 (See color insert following page 564.) An emergency stop button on a scanner is
recessed to prevent inadvertent actuation. Large size, red color, symbolic label, shape, and recessing
also provide redundant means of differentiating this control from others.

TAF-ER562X_WINGR-09-0401-C001.indd 19 11/8/10 11:39:58 PM


20 Handbook of Human Factors in Medical Device Design

using a monitor placed in “demonstration” mode (in which the numbers displayed are nor-
mal and unchanging) or had the anesthesia workstation’s software user interface set for a
pediatric case. Accordingly, designers should make operational modes and their character-
istics readily apparent.

1.8.8 EMPLOY REDUNDANT CODING


Redundant coding of displays and controls can be a powerful way to ensure reliable device
operation (see Chapter 7, “Controls,” and Chapter 8, “Visual Displays”). The concern is
that a user who may be fatigued or distracted might actuate the wrong control or mistake
one display value for another (e.g., a “1” for a “7”). These kinds of use errors are less likely
if displays and controls employ more than one means of coding. Coding options include
varying the user-interface element’s size, shape, color, texture, or placement. For example,
anesthesia machines use redundant coding (knob color, shape, and texture) to ensure that
caregivers turn the correct knob to increase the flow of 100% oxygen rather than air (see
Chapter 12, “Workstations”).

1.8.9 DESIGN TO PREVENT USER CONFUSION


While considering the need for making devices compatible, as discussed elsewhere, also
consider when it is appropriate to make devices distinct. For example, it is critical to dis-
tinguish power cable receptacles from sensor cable receptacles, thereby avoiding circum-
stances in which a user might plug a patient sensor lead into an AC outlet and shock the
patient (Figure 1.13). Devices and elements thereof may be distinguished using the coding
methods described above. In the case of plugs and receptacles, size and shape coding is
particularly appropriate, making it impossible to fit a given lead into the wrong port. This
principle of using a physical constraint to dissuade undesirable user actions applies to con-
trols, connectors, and other design elements.

FIGURE 1.13 Monitor’s sensor leads are incompatible with AC power receptacles. (From Wiklund,
M., Medical Device and Equipment Design: Usability Engineering and Ergonomics, Interpharm
Press, Boca Raton, FL, 1995.)

TAF-ER562X_WINGR-09-0401-C001.indd 20 11/8/10 11:39:58 PM


General Principles 21

1.8.10 DON’T NEGLECT DEVICE APPEAL


It is important to recognize that human factors in medical device design are not just about
achieving safe and effective task performance. They are also about satisfying user needs,
which includes making medical devices pleasing to use. Devices that are easy to use as well
as appealing to view and touch will engender greater user satisfaction. One payoff from
making devices appealing is that patients—and particularly children—may find them less
frightening. Moreover, users may be more motivated to learn how to use appealing devices
properly. Added appeal may also lead to increased user vigilance and job satisfaction. For
example, a user may pay closer attention to a display with a pleasing appearance that also
draws attention to important information as opposed to one that has a garish appearance that
draws attention to less important information (see Chapter 11, “Software User Interfaces”).
Devices designed to provide an impression of quality can inspire greater user confidence
(and even pride of ownership). A user may be drawn to a portable patient monitor because
of design qualities (e.g., an enclosure that looks attractive and easy to handle) that extend
beyond functionality to boost appeal. The same can be said of tools that look comfortable to
hold (Figure 1.14). However, medical devices intended for use in the home should not look
like toys; otherwise, they might attract children to play with them.

FIGURE 1.14 Close attention to visual and tactile design considerations, such as rounded surfaces,
distinguishable controls, and other styling cues, contribute to a device’s usability and appeal and
earned these sinus cavity surgery and eye examination devices Medical Design Excellence Awards.
(From http://www.gyrus-ent.com/health/diego/rhinologyDiego.htm and http://www.welchallyn.
com/medical/products/catalog/detail.asp?ID=29365. With permission.)

TAF-ER562X_WINGR-09-0401-C001.indd 21 11/8/10 11:39:59 PM


22 Handbook of Human Factors in Medical Device Design

1.9 SUMMARY
The general considerations (i.e., principles) discussed above are just a fraction of the broad-
level factors, albeit some of the more important ones, to consider when designing a medical
device’s user interface. Accordingly, readers should regard these as a starting point and
supplement them with additional considerations presented in the following sections of this
handbook as well as other reference documents.

RESOURCES
Jacko, J. A. and Sears, A. (2003). The Human-Computer Interaction Handbook: Fundamentals, Evolv-
ing Technologies and Emerging Applications. Mahwah, NJ: Lawrence Erlbaum Associates.
National Aeronautics and Space Administration. (1989). Man-Systems Integration Standards.
NASA-STD-3000a. Houston: Lyndon B. Johnson Space Center.
Nielson, J. (1993). Usability Engineering. Boston: Academic Press.
Norman, D. (1988). The Design of Everyday Things. New York: Basic Books.
Rouse, W. B. (1991). Design for Success: A Human-Centered Approach to Designing Successful
Products and Systems. New York: Wiley-Interscience.
Salvendy, G. (Ed.). (2006). Handbook of Human Factors and Ergonomics (3rd ed.). New York: John
Wiley & Sons.
Sanders, M. S. and McCormick, E. J. (1993). Human Factors in Engineering and Design (8th ed.).
New York: McGraw-Hill.
Sawyer, D. (1996). Do it by Design: An Introduction to Human Factors in Medical Devices.
Washington, DC: U.S. Department of Health and Human Services, Food and Drug
Administration.
U.S. Department of Defense. (1996). Human Engineering Design Criteria for Military Systems,
Equipment, and Facilities. MIL-STD-1472F. Washington, DC: U.S. Department of Defense.
Wickens, C. and Holland, J. (2000). Engineering Psychology and Human Performance (3rd ed.).
New York: Prentice Hall.
Wiklund, M. (Ed.). (1995). Medical Device and Equipment Design: Usability Engineering and
Ergonomics. Boca Raton, FL: Interpharm Press.
Wiklund, M. and Wilcox, S. (2005). Designing Usability into Medical Devices. Boca Raton, FL:
Interpharm Press.
Woodson, W. E., Tilman, B., and Tilman, P. (1992). Human Factors Design Handbook: Information
and Guidelines for the Design of Systems, Facilities, Equipment, and Products for Human Use
(2nd ed.). New York: McGraw-Hill.

REFERENCES
American National Standards Institute/Association for the Advancement of Medical Instrumentation
(ANSI/AAMI). (2001). Human Factors Design Process for Medical Devices. ANSI/AAMI
HE-74-2001. Arlington, VA: Association for the Advancement of Medical Instrumentation.
Beydon, L., Conreux, F., Le Gall, R., Safran, D., Cazalaa, J. B., et al. (2001). Analysis of the French
health ministry’s national register of incidents involving medical devices in anaesthesia and
intensive care. British Journal of Anaesthesia 86:382–87.
Gaba, D. M. and Howard, S. K. (2002). Patient safety: Fatigue among clinicians and the safety of
patients. New England Journal of Medicine 347:1249–55.
Samore, M. H., Evans, R. S., Lassen, A., Gould, P., Lloyd J., Gardner, R. M., et al. (2004). Surveillance
of medical device-related hazards and adverse events in hospitalized patients. Journal of the
American Medical Association 291:325–34.
Weinger, M. B. and Ancoli-Israel, S. (2002). Sleep deprivation and clinical performance. Journal of
the American Medical Association 287:955–57.

TAF-ER562X_WINGR-09-0401-C001.indd 22 11/8/10 11:39:59 PM


2 Basic Human Abilities
Edmond W. Israelski, PhD

CONTENTS
2.1 Sensory and Perceptual Abilities ...............................................................................25
2.1.1 Vision ..............................................................................................................25
2.1.1.1 Threshold of Seeing ........................................................................25
2.1.1.2 Visual Acuity ..................................................................................26
2.1.1.3 Visual Angle....................................................................................27
2.1.1.4 Minimum Visual Angle ..................................................................27
2.1.1.5 Dynamic versus Static Acuity .........................................................29
2.1.1.6 Accommodation (Focusing Abilities) .............................................29
2.1.1.7 Visual Field .....................................................................................29
2.1.1.8 Color Vision ....................................................................................29
2.1.1.9 Color Vision Deficiencies ................................................................32
2.1.1.10 Color Discrimination Recommendations ........................................33
2.1.1.11 Recommendations for Printed Colors .............................................34
2.1.1.12 Recommendations for Colored Lights.............................................34
2.1.1.13 Recommendations for Color Combinations (Legibility
and Visibility) ..................................................................................35
2.1.1.14 Dark Adaptation ..............................................................................36
2.1.2 Visual Perception ............................................................................................37
2.1.2.1 Distance and Perceived Size ...........................................................37
2.1.2.2 True Object Size ..............................................................................38
2.1.2.3 Common Visual Illusions ................................................................38
2.1.2.4 Perception of Motion .......................................................................39
2.1.2.5 Flickering Lights .............................................................................39
2.1.2.6 Photosensitive Epilepsy ...................................................................39
2.1.3 Auditory Perception ........................................................................................40
2.1.3.1 Loudness Measurements .................................................................40
2.1.3.2 Relationship of Phones to Sones .....................................................41
2.1.3.3 Loudness (Sones) Calculation for Complex Sounds ........................42
2.1.3.4 Pitch Measurement ..........................................................................42
2.1.3.5 Differential Hearing Thresholds .....................................................42
2.1.3.6 Effects of Aging on Hearing Sensitivity .........................................43
2.1.4 Other Sensory Modalities ...............................................................................45
2.1.4.1 Skin (Somesthetic) Senses ...............................................................45
2.1.4.2 Muscle Sense ...................................................................................46

23

TAF-ER562X_WINGR-09-0401-C002.indd 23 11/8/10 11:40:31 PM


24 Handbook of Human Factors in Medical Device Design

2.1.4.3 Sense of Balance .............................................................................46


2.1.4.4 Chemical Senses .............................................................................46
2.1.5 Other Perceptual Abilities...............................................................................46
2.1.5.1 Estimation of Time Intervals...........................................................46
2.1.5.2 Estimation of Other Physical Quantities .........................................46
2.2 Human Information Processing .................................................................................48
2.2.1 Limitations on Information Processing Abilities ............................................48
2.2.1.1 Channel Capacity ............................................................................48
2.2.1.2 Attention..........................................................................................49
2.2.1.3 Vigilance (Sustained Attention) ......................................................49
2.2.2 Speed of Information Processing ....................................................................49
2.2.2.1 Reaction Time .................................................................................49
2.2.2.2 Speed versus Accuracy....................................................................51
2.2.2.3 Human Memory ..............................................................................51
2.2.2.4 Working Memory ............................................................................51
2.2.2.5 Long-Term Memory ........................................................................53
2.2.2.6 Estimation and Decision-Making Abilities .....................................53
2.3 Human Response Capabilities ...................................................................................54
2.3.1 Speed of Movement ........................................................................................54
2.3.2 Principles of Motion Economy .......................................................................56
2.3.3 Speech Attributes ............................................................................................57
2.3.3.1 Loudness Levels of Speech .............................................................57
2.3.3.2 Frequency Characteristics of Speech ..............................................58
2.4 Human versus Machine Capabilities .........................................................................59
Resources ...........................................................................................................................60
References ..........................................................................................................................60

This chapter presents a brief overview of basic human skills and abilities that will aid
the designer of medical devices to better understand the guidance provided in subsequent
chapters in this book. These basic human skills and abilities and their interrelationships
are shown in Figure 2.1. The organization of this chapter follows the flow of how humans
sense, perceive, process, and respond to the world around them, as shown in the figure that
portrays the basic inputs and outputs of the human information processing system.
Specifically, the following areas of basic human capabilities and corresponding limita-
tions are covered: vision, visual perception, audition (or hearing), sensation, information
processing, human response capabilities, and human versus machine trade-offs. Many spe-
cific design recommendations result from knowledge of basic human skills and abilities,
and these recommendations are covered in specific chapters in this book (see Chapter 8,

Stimuli Response

Sensory Information Response


Input Perception Output
processing processing processing

FIGURE 2.1 Organization of basic human skills and abilities.

TAF-ER562X_WINGR-09-0401-C002.indd 24 11/8/10 11:40:32 PM


Basic Human Abilities 25

“Visual Displays”; Chapter 10, “Alarm Design”; and Chapter 16, “Hand Tool Design”).
Design guidance from this chapter is noted by numbered guidelines.

2.1 SENSORY AND PERCEPTUAL ABILITIES


This section covers basic human sensory and perceptual abilities in the areas of vision, hear-
ing, touch, balance, and perceptual estimation. Individual differences in perceptual abilities
are quite large and are due not only to inherent physiological factors but also differences
in experience, motivation, and preconceived ideas about incoming sensory information,
sometimes called one’s “psychological set” or “expectations.” It should be noted that unless
specifically mentioned, much of the following data are based on relatively young humans
without major disabilities. Designers must be careful when applying data from this chapter
if their intended users include special populations, such as the elderly or disabled. In that
case, information provided in Chapter 18, “Home Health Care,” may be more useful.

2.1.1 VISION
The human visual sensory system is quite complex. In this section, information is presented
on visual thresholds for seeing, visual acuity (both static and dynamic), visual angles,
accommodation or focusing, visual field, color vision, and dark adaptation.

2.1.1.1 Threshold of Seeing


The sensitivity of the human visual system covers a wide range, as does the visual threshold
(i.e., the minimum light level in which an object can be visually identified) under various
ambient lighting conditions (Figure 2.2; Van Cott and Kinkade, 1972). Rod vision comes
from visual receptors found on the back of the eye, the retina, that are most sensitive under
100,000
Upper limit of visual tolerance
10,000 Cone vision
Fresh snow on a clear day
only
1,000 Average earth on a clear day

100 Average earth on a cloudy day


Approximate luminance level
(brightness) in ft. lamberts

10 White paper in good reading light

1 White paper 1 ft. from std. candle

0.1
Rod and cone
Snow in full moon vision
0.01
Average earth in full moon
0.001

0.0001 Snow in starlight


Rod vision
Grass in starlight only
0.00001

0.000001 Absolute threshold of seeing

FIGURE 2.2 Threshold of seeing and luminance levels of various objects.

TAF-ER562X_WINGR-09-0401-C002.indd 25 11/8/10 11:40:32 PM


26 Handbook of Human Factors in Medical Device Design

low light conditions to shades of black and white (e.g., a patient room during sleeping
hours). Cone vision comes from visual receptors that are sensitive to color and operate best
under higher light levels (e.g., in an operating room).
2.1.1.2 Visual Acuity
A number of measures of visual acuity exist:

• Minimum distinguishable (detection of detail in an arbitrary test target)


• Minimum perceptible (detection of a spot, e.g., on a radiograph)
• Minimum separable (detection of a gap between parts of a target)
• Stereoscopic acuity (detection of depth for a three-dimensional target)
• Vernier acuity (detection of lateral displacement of one line from another)

These measurements of acuity all apply to static or stationary objects. In addition, there is
dynamic visual acuity of the smallest detail that can be detected for a moving target. Visual
acuity can be affected by a variety of factors (Table 2.1).

TABLE 2.1
Factors That Affect Visual Acuity
Factor Positive Example Negative Example
Amount and kind of Bright operating room light Glare from outside sunlight on IV
illumination pump screen
Viewing time Momentary occlusion message on IV Long time period for viewing pulse
pump screen oximeter readings
Object contrast with Light-colored numbers showing pulse Yellow trace lines of respiratory rate
background rate on a dark-background patient on a white-background patient
monitor monitor
Object size (visual angle Large font indicating on/off for a Small print on a catheter package
subtended by the object at ventilator power switch label indicating French size
the eye)
Object color Bright red flashing alarm on an Lettering in pale pastel colors
enteral pump indicating it is empty indicating length of a nasogastric tube
Direction of viewing Patient monitor screen placed at a PCA pump placed on the bottom of
(position of the image on 45-degree angle above a patient’s bed an IV pole at knee level
the retina)
Movement of the object or Stationary IV pump screen Heart monitor vibrating from the
viewer motion in an ambulance
Accommodation or Large-screen monitor (over 17 Fatigued operator of a small-screen
focusing abilities of the inches) for an ultrasound machine portable patient monitor
viewer’s visual system
Optical alignment of both Surgeon being able to accurately Surgeon attempting to judge distance
eyes or convergence judge the depth of cutting with a from a laparoscopic pincer by
abilities scalpel looking at a video monitor
Dark adaptation Ambulance driver reading red gauges Trying to find a central line port in a
while driving at night darkened patient room after entering
from a brightly lit hallway

TAF-ER562X_WINGR-09-0401-C002.indd 26 11/8/10 11:40:32 PM


Basic Human Abilities 27

7 mm
Size of the object So
Visual angle A

So
Distance of object do tan A
do
(57.3) 60 So if A 10°
or A (minutes of arc)
do

FIGURE 2.3 Calculation of visual angle A in minutes of arc when object size and distance are known.

2.1.1.3 Visual Angle


The visual angle subtended on the retina at the back of the eye can be calculated using the
formula and related diagram in Figure 2.3 (Cornsweet, 1970). In the diagram, the apex of
the triangle is assumed to be 7 mm behind the foremost point of the cornea.
Examples of the visual angle A cast by some common objects at a given distance do are
shown in Table 2.2 (Cornsweet, 1970).

2.1.1.4 Minimum Visual Angle


Minimum visual angle is the value of the visual angle cast on the retina for the following
types of limiting conditions (Dreyfuss, 1966):

• Minimum perceptible visual angle is approximately 1 second of a degree for


a thin wire against bright sky. (Visible stars may subtend an angle as low as
0.056 second.)
• Preferred angle for reading English text is 20 to 22 minutes of arc. Marginally
acceptable angles range from 16 to 18 minutes of arc, with 12 minutes considered
the threshold of readability.

Guideline 2.1: Minimum Type Size


Type size should not be less than 3 points when read at 14 inches under the most favorable
lighting conditions (1 point = 1/72, or 0.01384 inch). See Table 2.3 for details.

TABLE 2.2
Visual Angle for Common Objects
Object Distance (do) Visual Angle (A)
Sun 93,000,000 miles 30 minutes
Moon 240,000 miles 30 minutes
Quarter Arm’s length (70 cm) 2 degrees
Quarter 90 yards 1 minute
Quarter 3 miles 1 second
Lowercase pica-type letter Reading distance (40 cm) 13 minutes

1 degree = 60 minutes of arc; 1 minute = 60 seconds of arc; tan 1 second = 0.0000048;


tan 1 minute = 0.00029. For small angles, the tangent of an angle varies linearly with the
size of the angle (e.g., tan 10 minutes = 10 × tan 1 minute).

TAF-ER562X_WINGR-09-0401-C002.indd 27 11/8/10 11:40:32 PM


28 Handbook of Human Factors in Medical Device Design

TABLE 2.3
Recommended Character Sizes and Corresponding Font Sizes for
Various Reading Distances
Character Height vs. Reading Distance
Character Reading Visual Angle
Heighta (in.) Distance (in.) (minutes of arc) Font Sizeb

Preferred (Upper Bound)


0.102 16 22 Actual font size 7.5
0.154 24 22 Actual font size 11
0.230 36 22 Actual size 16
0.768 120 22 Font 55
1.152 180 22 Font 72

Preferred (Lower Bound)


0.093 16 20 Actual font size 7
0.140 24 20 Actual font size 10
0.209 36 20 Actual font size 15
0.698 120 20 Font 50
1.047 180 20 Font 75

Adequate (Upper Bound)


0.084 16 18 Actual font size 6

0.126 24 18 Actual font size 9


0.188 36 18 Actual font size 13.5
0.628 120 18 Font 45
0.942 180 18 Font 68

Adequate (Lower Bound)


0.074 16 16 Actual font size 5.5

0.112 24 16 Actual font size 8


0.168 36 16 Actual font size 12
0.558 120 16 Font 40
0.838 180 16 Font 60

Minimum Threshold
0.056 16 12 Actual font size 4

0.084 24 12 Actual font size 6

0.126 36 12 Actual font size 9


0.419 120 12 Font 30
0.628 180 12 Font 45
a Smallest lowercase letter height.
b Font size is the distance from the highest ascender to the lowest descender of any character
in the font set.
Assumptions: Contrast ratio >7:1; luminance >35 cd/m2.

TAF-ER562X_WINGR-09-0401-C002.indd 28 11/8/10 11:40:32 PM


Basic Human Abilities 29

2.1.1.5 Dynamic versus Static Acuity


An individual’s dynamic visual acuity of moving targets is not strongly related to static acu-
ity. Thresholds for dynamic acuity increase rapidly for rates of motion exceeding 60 inches
of visual field per second and are considerably higher with shorter viewing times or longer
target travel distance. Both dynamic and static acuity decrease with age (Burg, 1966).

2.1.1.6 Accommodation (Focusing Abilities)


Guideline 2.2: Typical Visual Deficiencies
Designers need to be aware of typical visual deficiencies of their users, including nearsighted-
ness, farsightedness, astigmatism, and aging eyes. This would require making text and images
larger and with higher contrast on labels, displays, and documentation.
Accommodation is the adjustment of the lens of the eye to focus light rays properly on the
receptor cells of the retina. Normal accommodation and common types of inadequate accom-
modation abilities are shown in Figure 2.4. The figure shows what causes the common visual
impairments of nearsightedness (myopia) and farsightedness (hyperopia) and describes the
effect of age on the ability to focus on near objects (presbyopia). Presbyopia is the inability
of the eye to focus sharply on nearby objects, resulting from loss of elasticity of the crystal-
line lens with advancing age. The average age of onset is 40 years. Astigmatism, which also
decreases the ability to focus, is a defect in which the unequal curvature of one or more refrac-
tive surfaces of the eye, usually the cornea, prevents light rays from focusing clearly at one
point on the retina, resulting in blurred vision.

2.1.1.7 Visual Field


Figure 2.5 shows the binocular (for both eyes) visual field measured in degrees. The normal
line of site can be as small as 10 degrees from the horizontal plane as shown in the figure.
Figure 2.6 shows the monocular visual field (for the right eye). The field for the left eye
would be the reverse, or mirror image, of the right eye (Woodson and Conover, 1964).

2.1.1.8 Color Vision


There are two types of receptor cells in the retina: rods and cones. The rods function under
dim light and do not respond to color. They are located on the periphery of the retina away
from the fovea, which is in the center of the retina, where vision is the sharpest. The cones

Retina Lens normal Normal far vision – Light focused on retina


Object
Normal near vision – Light focused on retina
Lens curved Age
d 20 40 60
d = nearest point
of focus (average 4.0 8.8 40.0
inches)
Nearsightedness (elongated eyeball and/or excess
curvature of lens) – Light focused in front of
retina for far objects

Farsightedness (shortened eyeball and/or inability


to increase lens curvature sufficiently as in aging)
– Light focused in back of retina for near objects

FIGURE 2.4 Common visual impairments and focus point distances for different ages.

TAF-ER562X_WINGR-09-0401-C002.indd 29 11/8/10 11:40:33 PM


30 Handbook of Human Factors in Medical Device Design

50°

15°
Color
e 70°
vision al lin
limit Norm t
h
of sig
30° 30° 94°
94°

FIGURE 2.5 Normal visual fields in vertical and horizontal directions for both eyes.
function under relatively higher-intensity light and respond to color. They are located pri-
marily in the area of the fovea. Factors that influence color vision are brightness (i.e., light
intensity), hue (dominant wavelength), and saturation (pureness of color). As illumination
decreases, so does color vision sensitivity.
Not all zones of the retina are equally sensitive to color. Toward the periphery, objects
can still be distinguished even though their color cannot. Some colors are recognized at
greater angles away from the fovea than others. Figure 2.7 shows the limits of the retina
zones in which the various colors can, under normal illumination, be correctly recognized
(Woodson and Conover, 1964).
Figure 2.8 shows the sensitivity of the human visual system to different colors (wave-
lengths of light). One curve shows the sensitivity of the cone receptors, and the second
Nasal Temporal

55° Occluded by eyebrow


30°
No color (under
normal levels of illumination)
60° 34° 65°
Occluded Blind spot
by nose 40°
Cornea 70° Color area (cones)
Pupil Occluded by cheek
opening

Lens
Eye
Retina

Fovea
(Sharpest vision at fovea
with rapidly decreasing acuity toward
retina periphery)

FIGURE 2.6 Monocular visual field of the right eye showing occlusions and color areas. (From
Woodson, W. E. and Conover, D. W., Human Engineering Guide, University of California Press,
Berkeley, CA, 1964. With permission.)

TAF-ER562X_WINGR-09-0401-C002.indd 30 11/8/10 11:40:33 PM


Basic Human Abilities 31

Temporal Nasal

White
Blue
Yellow
Red

Green

90°

65°
60°

50°

33°
25°
15° 10°
Fovea

FIGURE 2.7 Retinal zones that are primarily sensitive to different colors.

curve shows the sensitivity of the rod receptors. The rod system takes over primarily after
the eye is adapted to very low light levels, which is also known as dark adaptation (Glazer
and Hammell, 1970).

Guideline 2.3: Use of Color in Low Light Conditions


Data from the human visual system have led to the following design recommendations:
1. Red and orange are poorly visible under low light conditions and should be avoided.
2. Blue, green, and yellow are equally visible under both low and higher light conditions
and are good color choices,
3. Under low light conditions, blues and cyan colors are more visible. There is a shift in color
sensitivity under these lower light conditions toward the blue end of the color spectrum.

48
100
Rod vision Cone vision
(shades of gray) (color)
80
Relative visibility

60

40
Orange
Yellow
Violet

Green
Blue

Red

20

0
400 450 500 550 600 650 700
Wavelength (mμ)

FIGURE 2.8 Relative visibility of different colors (wavelengths) for rod vision (dim light) and
cone vision (brighter light). Curves are individually normalized and not normalized to each other.

TAF-ER562X_WINGR-09-0401-C002.indd 31 11/8/10 11:40:33 PM


32 Handbook of Human Factors in Medical Device Design

TABLE 2.4
Types of Deficient Color Vision and Their Frequencies
Percentage of
Population Affected
Type Effect Males Females

Trichromacy (three colors) Normal color vision 92.0 98.0


Dichromacy (two colors) (Sees only two colors plus shades of gray)
Protanopia (red blindness) Inability to see hues other than yellow and 1.0 0.02
blue with red and bluish-green seen as same
shade of gray
Deuteranopia Inability to see hues other than yellow and 1.1 0.01
(green blindness) blue with green and bluish-red seen as same
shade of gray
Tritanopia Inability to see hues other than red and green 0.0001 Very rare
with yellow-green and purplish-blue seen as
same shade of gray
Tetartanopia Inability to see hues other than red and green Very rare Very rare
where color spectrum appears red at long
wavelengths, green in the center, and red
again at short wavelengths
Anomalous trichromacy (Sees all colors but mismatches them,
especially under dim light or small light
sources)
Protanomaly There is a foreshortened red end of the 1.0 0.02
(red weak) spectrum that requires more red than normal
to match pure yellow; the spectrum is
shifted, and normal yellow is seen as
greenish
Deuteranomaly There is no foreshortening of the red end of 4.9 0.38
(green weak) the spectrum, and more green is required to
match pure yellow; the spectrum is shifted,
and normal yellow is seen as orange
Tritanomaly More blue than normal is required to match Very rare Very rare
cyan or blue-green
Monochromatism Complete loss of color discrimination and 0.003 0.002
poor acuity

2.1.1.9 Color Vision Deficiencies


Approximately 8% of males and 2% of females have some degree of deficient color vision.
Table 2.4 describes the various types of deficient color vision and their relative incidence in
the U.S. population (Israelski, 1978).
An example of the effects of color vision impairments is shown in Figure 2.9, which
shows a patient monitor as seen by users with normal color vision, tritanopia, red blindness
(protanopia), and green blindness (deuteranopia).

TAF-ER562X_WINGR-09-0401-C002.indd 32 11/8/10 11:40:33 PM


Basic Human Abilities 33

Normal color vision Tritanopia

Red blind (protanopia) Green blind (deuteranopia)

FIGURE 2.9 (See color insert following page 564.) Comparison of colors seen by various users
of a patient monitor with normal and deficient color vision, including tritanopia, protanopia, and
deuteranopia.

2.1.1.10 Color Discrimination Recommendations


Up to eight saturated surface colors (excluding black and white) can be used for color cod-
ing with practically error-free discrimination for color normal people. Color coding with
more than eight colors produces higher error rates. Fewer colors would place less demand
on memory. Under strictly controlled conditions, with a high level of training and the use
of various combinations of hue and saturation, up to 50 colors can be identified with high
accuracy. In any color-coding scheme, colors should subtend a visual angle of at least
15 minutes of arc.

Guideline 2.4: Color Only One Form of Coding


In any design, color coding should be a redundant information source and never stand as the
only means of coding. If the population for which equipment is being designed is known to
include significant numbers of color-deficient users, color coding should be avoided.

However, if a significant number of users may be color-deficient and color coding is still
desired, only three colors may be safely used according to military aviation standards for
the Army and Navy (Army-Navy Aeronautical Specification AN-C-56). These colors are
not pure colors and therefore allow better discrimination for color-deficient users:

• Aviation red (MIL-C-2505A Red)


• Aviation green (MIL-C-2505A Green)
• Aviation blue (MIL-C-2505A Blue)

Other red, green, and blue hues may cause confusion. Colors are not recommended for use
at larger distances because blue and green are more likely to be confused. White or yellow
should not be added to the code because of probable red–yellow and green–white confusion
for color-deficient individuals.

TAF-ER562X_WINGR-09-0401-C002.indd 33 11/8/10 11:40:33 PM


34 Handbook of Human Factors in Medical Device Design

TABLE 2.5
Recommended Printed Color Codes Using the Munsell Color System
8-Color Code 7-Color Code 6-Color Code 5-Color Code 4-Color Code
n p n p n p n p n p
1R 999 5R 1008 1R 999 1R 999 1R 999
9R 892 3YR 890 3YR 890 7YR 884 1Y 946
1Y 946 5Y 1128 9Y 1131 7GY 960 9G 1099
7GY 960 1G 1103 5G 1101 1B 1093 1P 1135
9G 1099 7BG 1095 5B 1087 5P 1007
5B 1087 7PB 1133 9P 1005
1P 1135 3RP 1003
3RP 1003

n, book notation of Munsell color system; p, Munsell production number; R, red; Y, yellow; G, green; B, blue;
P, purple.

2.1.1.11 Recommendations for Printed Colors


Guideline 2.5: Printed Color Recommendations
For printed materials, colors should be chosen from the Munsell color systems (Table 2.5)
(Cleland, 2004; Conover and Kraft, 1958; McCormick, 1970).

The Munsell color system is the system of color notation developed by A. H. Munsell in
1905 and identifies color in terms of three attributes—HUE, VALUE, and CHROMA—
which are described symbolically.
The HUE (H) notation of a color indicates its relation to a visually equally spaced scale
of 100 hues. There are five principal and five intermediate positioned hue steps within
this scale. The hue notation in general use is based on the 10 major hue names: Red (5R),
Yellow-Red (5YR), Yellow (5Y), Green-Yellow (5GY), Green (5G), Blue-Green (5BG), Blue
(5B), Purple-Blue (5PB), Purple (5P), and Red-Purple (5RP).
The VALUE (V) notation indicates the lightness or darkness of a color in relation to a
neutral gray scale, which extends from absolute black (value symbol 0/) to absolute white
(value symbol 10/). The symbol 5/ is used for the middle gray and for all chromatic colors
that appear halfway in value between absolute black and absolute white.
The CHROMA (C) notation indicates the degree of divergence of a given hue from a
neutral gray of the same value. The scale of chroma extends from /0 for a neutral gray to /10,
/12, /14, or further, depending on the strength (saturation) of the sample to be evaluated.

2.1.1.12 Recommendations for Colored Lights


Guideline 2.6: Recommendations for Colored Lights
Colored lights need to be chosen using available data to have good recognizability and the
least amount of confusion. Table 2.6 recommends colors and describes their effect on color
recognition of small-point light sources near the threshold of visibility (Dreyfuss, 1966).
Table 2.7 shows 10 colored light choices that were shown to reduce confusion error; that
is, these 10 wavelengths had a less than 2% misidentification error in experimental studies

TAF-ER562X_WINGR-09-0401-C002.indd 34 11/8/10 11:40:34 PM


Basic Human Abilities 35

TABLE 2.6
Recommendations for Colored Lights
Color Discrimination Colors Recommended
Easiesta Red and green lights are easiest to recognize for color-normal individuals.
Easier White light is the next easiest to recognize.
Difficult Yellow (or orange) is more difficult to recognize.
More difficult Blue and green lights are very difficult to differentiate at a distance greater than10 feet.
More difficult Yellow, white, and orange lights are difficult to differentiate at distances greater
than 10 feet.

Source: Based on Dreyfuss, H., Measure of Man (2nd ed.), Watson Guptill, New York, NY, 1966.
a The best set of three colored lights is red, green, and white.

(Chapanis and Halsey, 1956). Use of the values in Table 2.7 increases the number of color
choices that are easily recognizable and less likely to be misidentified.

2.1.1.13 Recommendations for Color Combinations (Legibility and Visibility)


Guideline 2.7: Recommendations for Color Combinations
Research has shown that the following lists are the best combinations of colors for providing
good legibility, discrimination, and visibility.

The most legible color combinations for text are listed in order of legibility (Dreyfuss, 1966):

1. Black on white (most legible).


2. Black on yellow (most attention gained).
3. Green on white.

TABLE 2.7
Ten Colored Light Choices That Have Less Than a 2% Misidentification Error Rate
RGB Values
Color Description Wavelength (nm) R G B
Red 642 255 12 0
Burnt orange 610 255 137 0
Orange 596 255 192 0
Yellow 582 255 247 0
Yellow-green 556 168 255 0
Green 515 18 255 0
Light green 504 0 255 76
Cyan 494 0 255 204
Pale blue 476 0 184 255
Medium blue 430 28 0 255

Source: Based on Chapanis, A. and Halsey, R., J Psychol, 42, 99, 1956.
RGB, red, green, blue.

TAF-ER562X_WINGR-09-0401-C002.indd 35 11/8/10 11:40:34 PM


36 Handbook of Human Factors in Medical Device Design

4. Red on white.
5. White on blue.
6. Combinations of pure red and green or red and blue should not be used.
7. White on black may cause problems of smearing or irradiation of white on the
black background if printing or electronic displays are not carefully controlled.

For visibility of opaque colors under typical light conditions, the following colors and
combinations are recommended:

• Yellow is the most luminous and visible.


• Orange and red-orange hold maximum attention value.
• Blue is likely to be out of focus and indistinct.
• Red on blue or blue on red should not be used since each focuses differently on the
retina and creates an induced three-dimensional effect called chromosteriopsis.
2.1.1.14 Dark Adaptation
The rod system takes time to function efficiently (becoming dark adapted) after the eyes
experience a change from bright to dark illumination conditions. There are both physiologi-
cal and neurochemical changes that occur on dark adaptation. The pupils dilate, and the rod
receptors become more sensitive.
Guideline 2.8: Red Maintains Dark Adaptation
The use of red goggles or red lighting does not affect rod vision and is useful in maintaining
dark adaptation.

Figure 2.10 shows the lowering of visual thresholds as time in the dark progresses (Glazer
and Hammel, 1970; Nutting, 1916). Full dark adaptation can take as long as 30 minutes,

5
Centrally fixated fields
Log of threshold brightness (μml)

Visual angle
4


3


2
10°

20°

1
0 10 20 30
Minutes in the dark

FIGURE 2.10 Dark adaptation. Threshold of seeing decreases as a function of time in darkness
and width of visual field or angle.

TAF-ER562X_WINGR-09-0401-C002.indd 36 11/8/10 11:40:34 PM


Basic Human Abilities 37

B A

FIGURE 2.11 Distance and perceived size of objects. Although arrows A and B subtend the same
visual angle at the observer’s eye and therefore produce images of the same size on his retina, arrow
A is seen to be farther away and hence appears to be larger.

but significant adaptation can take place in 8 to 10 minutes. The human eye also adapts to
bright light chiefly by constricting the pupil. On subsequent exposure to low light levels,
the pupils will normally dilate, after which the rod system starts the chemical process of
increasing sensitivity to these low light levels.

2.1.2 VISUAL PERCEPTION


There are known limits to human visual perception and its processing. Some of
these visual limitations and resulting common errors are summarized in this section.
Awareness of these common limitations will reduce the display of ambiguous informa-
tion. Chapter 8, “Visual Displays,” provides practical recommendations for the design of
visual displays.

2.1.2.1 Distance and Perceived Size


The principle of perspective states that objects appear to be smaller when they are farther
away. Distance and perceived size are related in the manner shown in Figure 2.11.
We learn in childhood the distance–size relationship, and therefore it is automatically
taken into account when we observe distant objects. If two objects that are actually the
same size are perceived as being at different distances, the one that seems to be farther
away will look larger. However, designers may unintentionally create misperceptions in
size perception, as in Figure 2.12.

FIGURE 2.12 Perceived size of distant objects. The cone on the right is perceived as being larger
than the identical cone at the left only because it seems to be farther away. (From Kaufman, L. and
Rock, I., Scientific American, July 1972. With permission.)

TAF-ER562X_WINGR-09-0401-C002.indd 37 11/8/10 11:40:34 PM


38 Handbook of Human Factors in Medical Device Design

FIGURE 2.13 Object size misperception. The horizon moon appears to be farther away, although
it is not. The viewer automatically takes the apparent distance into account. The viewer then uncon-
sciously applies the rule that, of two objects forming images of equal size, the more distant must be
the larger. (From Kaufman, L. and Rock, I., Scientific American, July 1972. With permission)

2.1.2.2 True Object Size


The same concept applies to perceived object size in the presence of misperceived visual
reference information. The apparent-distance theory holds that this is what happens in the
case of the moon illusion and is illustrated in Figure 2.13.

Guideline 2.9: Object Size and Distance Misperceptions


Designers should not create objects that may be misperceived by users because of well-known
visual processing limitations and illusions.

2.1.2.3 Common Visual Illusions


The human visual system is easily fooled by visual illusions. Designers need to be aware
of these visual illusions and of course avoid any graphic treatments that might trigger these
problems of unreliable visual interpretations.
Parallax error is one form of visual illusion that is commonly encountered and must be
avoided. The error is seen as an apparent change in the position of an object, such as a medi-
cal device meter reading caused by a change of the observer’s line of sight, as illustrated in
Figures 2.14 and 2.15.

Viewing from the left Viewing from the right

FIGURE 2.14 Parallax errors. Note that the apparent reading of the meter needle position changes
depending on the observer’s position from either the left or the right viewing angle. The image of
the needle on the background display mirror is the true reading. Mirrors such as these are an aid to
reduce parallax error.

TAF-ER562X_WINGR-09-0401-C002.indd 38 11/8/10 11:40:34 PM


Basic Human Abilities 39

FIGURE 2.15 Straight-line illusions. Note how the observer sees lines that are not straight even
when they are.

2.1.2.4 Perception of Motion


Humans are often confused by slow relative motion between us and another object when
there is no dependable frame of reference. Apparent motion can be induced, as is com-
monly observed in motion pictures (24 still frames per second with each shown twice) as
well as television and cathode ray tube (CRT) displays (typically 50 or 60 frames per sec-
ond). Another example of induced apparent motion is the phi-phenomenon, in which rapid
successive flashes of individual lights arranged in a row or circle give the appearance of
individual light-source motion.

Guideline 2.10: Perception of Motion


The designer should be aware of methods to create apparent motion in visual displays either
to avoid confusion or to take advantage of them as potential information sources.

2.1.2.5 Flickering Lights


Prolonged perception of flicker (over 20 minutes) in a light being flashed on and off causes
visual fatigue and annoyance. A related concept is when flickering lights are perceived
as being steady. The frequency at which a flashing light is perceived as having a continu-
ous intensity level is called the critical fusion frequency (CFF). The CFF increases with
increasing average light intensity and with decreasing proportion of the light–dark cycle
occupied by the flash (percent modulation or duty cycle). CFF varies from 2 Hz up to 50 to
60 Hz for high-intensity light sources.

2.1.2.6 Photosensitive Epilepsy


Flashing lights or the flicker of a computer monitor at certain speeds can trigger a seizure
in susceptible individuals. This problem is called photosensitive epilepsy, photic epilepsy,
or photogenic epilepsy (Harding and Jeavons, 1995):
• Approximately 1 in 200 people have epilepsy, and of these, only 3% to 5% have
seizures induced by flashing lights.
• Photosensitivity is more common in children and adolescents and becomes less
common after the early 20s.

TAF-ER562X_WINGR-09-0401-C002.indd 39 11/8/10 11:40:35 PM


40 Handbook of Human Factors in Medical Device Design

• The flicker frequencies that can trigger seizures vary greatly by individual and are
in the range of 5 to 60 Hz.
• Only 50% of photosensitive people are sensitive to 50 Hz, but 75% are sensitive to
25 Hz. Rates below 5 Hz are considered relatively safe.
Guideline 2.11: Display Flicker
Liquid crystal displays have no flicker and are preferable to CRTs in situations where flicker should
be avoided, such as in medical display devices for use in the hospital rooms of seizure patients.

2.1.3 AUDITORY PERCEPTION


This section describes the basics of human hearing abilities (see Chapter 10, “Alarm Design,”
and Chapter 3, “Environment of Use,” for more detail and recommendations on environ-
mental effects of sound). Loudness or sound volume is the subjective measure equivalent to
sound intensity. The pitch or tone of a sound is the subjective measure equivalent to sound
frequency. The loudness of a sound is perceived differently at various frequencies. For
example, infusion pump alarm signals will be perceived as louder at higher frequencies.
Thresholds for hearing sounds and feeling pain at different sound intensity levels are also a
function of frequency (see Figure 2.16).

2.1.3.1 Loudness Measurements


Two commonly used subjective loudness measures of sound intensity are phones and sones:
• Loudness level (phones). Phones are the subjective measure of any tone intensity
that is numerically equal to the sound pressure level (SPL) in decibels (dB) of a
140
Threshold of pain

120
Threshold of feeling

100

Prolonged exposure
Intensity level (dB)

causes hearing loss 80

60

40
Words combine to
make meaning Threshold of hearing
tones
20
Sounds become words
Sounds can be heard 0

–20

FIGURE 2.16 Thresholds of hearing and pain. The curves show the intensity levels of sound
as a function of frequency for two sets of thresholds: minimal threshold to hear pure tones and
the threshold of pain. (From Woodson, W. E. and Conover, D. W., Human Engineering Guide,
University of California Press, Berkeley, CA, 1964. With permission.)

TAF-ER562X_WINGR-09-0401-C002.indd 40 11/8/10 11:40:35 PM


Basic Human Abilities 41

TABLE 2.8
Loudness for Common Noise Sources, Including Health Care Settings
Noise Source SPL (dB) Loudness (sones)
Patient room at night (EPA recommended) 35 <0.5
Residential inside, quiet 42 1
Patient room during day (EPA recommended) 45 1.3
Household ventilating fan 56 7
Automobile at 50 feet 68 14
Typical patient room—peaks 70 24
Anesthesia equipment—peaks 76 54
ICU—peak sound levels 80 74
IV pump alarm 85 99
Hospital beeper 89 127
Operating room—peaks 90 164
Inside MRI machine 95 222
Punch press, 3 feet 103 350
Nail-making machine, 6 feet 111 800
Pneumatic riveter 128 3,000

standard 1,000-Hz tone. This measure indicates subjective equality of any tone
compared to a 1,000-Hz standard.
• Loudness (sones). Sones are another measure of relative subjective sound intensity.
One sone is defined as the loudness of a 1,000-Hz tone at 40-dB SPL. A sound twice
as loud is 2 sones, a sound half as loud is half a sone, and so on. Common noise
sources and typical loudness values in sones are listed in Table 2.8 (McCormick,
1970; Weinger and Englund, 1990).
2.1.3.2 Relationship of Phones to Sones
Sones are measured on a logarithmic scale, while phones are measured on a linear scale.
This relationship is subject to many influences, but it can be shown reasonably well by the
line in Figure 2.17. Basically, loudness in sones doubles for every 10-phone increase in
loudness level (Stevens, 1955).
200
100

50
Loudness (sones)

20

10
5

2
1
40 50 60 70 80 90 100 110
Loudness level (phones)

FIGURE 2.17 The relationship between two common measures of subjective loudness. Sones
(loudness) and phones (loudness level).

TAF-ER562X_WINGR-09-0401-C002.indd 41 11/8/10 11:40:35 PM


42 Handbook of Human Factors in Medical Device Design

TABLE 2.9
Loudness (Sones) Calculation for Complex Sounds
(SPL) Band Loudness Index (S) for Octave Bands
Band Midpoint of Octave Band (Hz) Loudness Level
(dB) 31.5 63 125 250 500 1,000 2,000 4,000 8,000 (sones) (phones)
20 0.18 0.30 0.45 0.61 0.25 20
30 0.16 0.49 0.67 0.87 1.10 1.35 0.50 30
40 0.07 0.37 0.77 1.18 1.44 1.75 2.11 2.53 1.00 40
50 0.26 0.62 1.13 1.82 2.24 2.68 3.2 3.8 4.6 2.00 50
60 0.94 1.56 2.44 3.4 4.1 4.9 5.8 7.0 8.3 4.00 60
70 2.11 3.2 5.0 6.2 7.4 8.8 10.5 12.6 15.3 8.00 70
80 4.3 6.7 9.3 11.1 13.5 16.4 20.0 24.7 30.5 16.0 80
90 8.8 13.6 17.5 21.4 26.5 32.9 41.0 52.0 66.0 32.0 90
100 18.7 28.5 35.3 44.0 56.0 71.0 90.0 113.0 139.0 64.0 100
110 44.0 61.0 77.0 97.0 121.0 149.0 184.0 226.0 278.0 128.0 110
120 105.0 130.0 160.0 197.0 242.0 298.0 367.0 256.0 120

2.1.3.3 Loudness (Sones) Calculation for Complex Sounds


Complex sounds, such as the drone of a laboratory diagnostic system for blood testing, have
many combinations of sounds at various frequencies. Sones are calculated for complex
sounds by using the following procedure. The SPL is measured for each of the nine octave
bands with center frequencies listed in Table 2.9 (McCormick, 1970; Peterson and Gross,
1967). After obtaining these values, the procedure is as follows:
1. From Table 2.9, find the proper loudness index for each band level (S).
2. Add all the loudness indexes (∑ S).
3. Multiply this sum by 0.3.
4. Add this product to 0.7 times the index that has the largest value (Smax). The total
loudness in sones is (0.3 ∑ S + 0.7 Smax).
5. This total loudness (sones) can be converted to loudness level (phones) by using the
two columns at the right of the Table 2.9.

2.1.3.4 Pitch Measurement


Pitch is a subjective attribute of sound quality and is determined primarily by frequency but
also by intensity and the complexity of a sound’s spectrum (see Figure 2.18). The scale unit
is the mel, which is defined as the pitch of a 1,000-Hz tone at 40-dB SPL.

2.1.3.5 Differential Hearing Thresholds


Humans can discriminate sounds on the basis of sound frequency and intensity (Figures
2.19 and 2.20). The term just noticeable difference (JND) is used by psychophysicists to
mean a difference in some stimulus attributes that is detectable 75% of the time on the aver-
age (Foley and Moray, 1987).

TAF-ER562X_WINGR-09-0401-C002.indd 42 11/8/10 11:40:35 PM


Basic Human Abilities 43

4,000

3,000

Pitch (mels)
2,000

1,000

0
20 50 200 500 2,000 5,000 20,000
Frequency (Hz)

FIGURE 2.18 Pitch in the subjective units of mels as a function of frequency. (From Stevens, S. S.
and Vollunan, J., Am J Psychol, 53, 329, 1940.)

2.1.3.6 Effects of Aging on Hearing Sensitivity


Guideline 2.12: Adjustable Volume
Sounds for medical devices should have volume adjustments to control for ambient noise lev-
els and to accommodate older users with hearing loss.

As we age, our hearing sensitivity decreases, particularly for higher frequencies.


Figure 2.21 shows the decrease in hearing sensitivity with increasing age across the normal

50
Just noticeable difference in frequency (Hz)

20

10 5 dB

10 dB
5 15 dB
20 dB
40 dB
70 dB
2
50 100 200 500 1,000 2,000 5,000 10,000
Frequency (Hz)

FIGURE 2.19 Just noticeable differences in frequency changes to pure tones of various frequen-
cies as a function of sound pressure levels. Smaller-frequency differences are more noticeable at
higher sound intensity levels. (From Shower, E. G. and Biddulph, R., J Acoust Soc Am, 3, 275, 1931.
With permission.)

TAF-ER562X_WINGR-09-0401-C002.indd 43 11/8/10 11:40:35 PM


44 Handbook of Human Factors in Medical Device Design

10

Just noticeable difference in


5

sound pressure level (dB)


2

1,000 Hz
0.5 70 Hz White
noise
4,000 Hz 1,000 Hz
0.2
0 20 40 60 80 100 120
Level above threshold (dB)

FIGURE 2.20 Just noticeable differences in sound pressure level for pure tones of various fre-
quencies and white noise. Differences in sound intensity are easier to hear at higher frequencies.
(From Miller, G. A., J Acoust Soc Am, 19, 609, 1947 and Riesz, R. R., Physiol Rev, 31, 867, 1928.
With permission.)

spectrum of human audition with separate curves for males and females. There are many
causes for age-related hearing loss (also called presbycusis), including the following:

• Physiological changes to the inner ear


• Cumulative effects of exposure to loud noises
• Effects of stress
• Indirectly due to some slowing of the cognitive processing of auditory information

Commonly, as hearing sensitivity decreases for frequencies above 2,000 Hz, it becomes
increasingly more difficult to hear speech. Speech consonants, such as “b,” “c,” “f,” and “t”
become harder to discern.
Women Men
–5
0
500
5 and 500 Hz
1,000 Hz
10
Hearing loss (dB)

1,000 Hz
15
2,000 Hz
20 3,000 Hz 4,000 Hz
2,000 Hz
25
30
3,000
35 Hz
4,000 Hz
40
20 30 40 50 60 70 20 30 40 50 60 70
Age (yr)

FIGURE 2.21 The loss of hearing with age for males and females as a function of frequency. (From
American Standards Association, The Relations of Hearing Loss to Noise Exposure, American
Standards Association, New York, 1954. With permission.)

TAF-ER562X_WINGR-09-0401-C002.indd 44 11/8/10 11:40:36 PM


Basic Human Abilities 45

2.1.4 OTHER SENSORY MODALITIES


Other sensory modalities are generally poorer information input systems relative to the
visual and auditory systems. Touch, vibration, temperature, pain, and other skin sensa-
tions are not as fully developed neurologically in humans as our sight and hearing. A brief
description of these other sensory modalities follows.

2.1.4.1 Skin (Somesthetic) Senses


Senses related to the skin are multifaceted (Woodson and Conover, 1964; Corso, 1967).

• Touch—The stimulus for this sense is deformation of the skin due to pressure being
applied. Sensitivity varies with location on the body to which pressure is applied;
for example, JND thresholds range from pressures of 3 g/mm2 on the fingertips to
12 g/mm2 on the back of the hand and from 0.2 g/mm2 on the cornea of the eye to
250 g/mm2 on the thick parts of the sole of the foot. More rapid dynamic applica-
tion of pressure results in lower thresholds. Once a constant pressure is applied,
the sense will adapt, and awareness of contact will cease, as in the case of wearing
surgical gloves. Laparoscopic surgical instruments include touch as one source of
feedback (i.e., tactile feedback) to the surgeon.
• Vibration—Sensitivity to local skin vibrations is another dimension of touch and
varies with location on the body and the frequency of the vibrations (Figure 2.22).
Medical devices such as implantable heart rhythm monitors use vibrations as warn-
ings to patients of an incipient electrical shock.
• Temperature—Sensitivity to hot and cold stimuli also varies with body location.
Thinner, softer skin areas, such as the inner thighs, are more sensitive than rough,
thick skin areas, such as the soles of the feet. Subjective judgments of absolute
temperature of objects or fluids in contact with the skin are unreliable. Even rela-
tive temperature judgments are often erroneous. Normal skin temperature is in the
range 90.5°F to 92.3°F. Skin temperatures below 32°F and above 125°F become
painful.
• Pain—This important sense provides a clear signal of a hazardous condition.
Common stimuli include thermal, mechanical, chemical, and electrical sources.
Pain localization can be precise but is sometimes referred to distant locations.
• Other skin sensations—Other skin sensations are generally combinations or varia-
tions of the primary sensory modalities, for example, the sensation of moisture is a
combination of pressure and thermal stimuli.
(thousandths of an inch)

0.3
Threshold amplitudes

0.2

0.1

0.03
0
10 100 250 1,000 4,000
Frequency (Hz)

FIGURE 2.22 Vibration sensitivity thresholds as a function of frequency.

TAF-ER562X_WINGR-09-0401-C002.indd 45 11/8/10 11:40:36 PM


46 Handbook of Human Factors in Medical Device Design

2.1.4.2 Muscle Sense


Proprioception and kinesthesis are the sensory feedback mechanisms for motor muscle con-
trol and body posture. Together these senses provide important information for the coordina-
tion of complex physical acts. Proprioception is the usually unconscious perception of spatial
orientation arising from stimuli within the body itself and would include knowledge of where
a joint is positioned. Kinesthesis is the sensation that informs our brains about joint motion
and acceleration. For example, a medical device user can discriminate differences in control
knob shape, position, and degree of rotation through a combination of these senses (Carlson,
2000). Sensors in the muscles provide information about muscle stretch (tension) and tone.

2.1.4.3 Sense of Balance


Organs located in the inner ear provide sensory information necessary to maintain bal-
ance or orientation and to detect motion of the body. These organs are part of the vestibu-
lar system and include the semicircular canals. The semicircular canals are a three-part
fluid-filled network, detecting acceleration in the three perpendicular planes. The absolute
threshold for detection of angular acceleration of the body is 0.1°/s2 (Carlson, 2000).

2.1.4.4 Chemical Senses


The taste sense receptors (gustation) are located on the tongue and provide four basic taste
qualities: salty, sour, bitter, and sweet. Other taste sensations are combinations of these four.
The smell sense receptors (olfaction) are located in the upper part of the nasal cavity and
are estimated to be 10,000 times more sensitive than taste for absolute detection thresholds
but much poorer for difference detection thresholds. The fundamental odors are considered
to be spicy, fragrant, resinous, burnt, and putrid. Stimulus thresholds vary; 5.83 mg/L of air
for detection of ethyl ether is required, whereas the threshold for musk oil is 1.00004 mg/L
of air. The commonly accepted conversion to parts per million (ppm) is 1 mg/L = 1 ppm.
Both chemical senses are unsuitable for reliable information transmission primarily
because they quickly adapt to incoming stimuli so that awareness of these stimuli ceases.
Neither of these sources should ever be depended on as a primary information source even
though they often do have some practical information value, as in the detection of hazard-
ous gases or burning insulation (Carlson, 2000).

2.1.5 OTHER PERCEPTUAL ABILITIES


There are a number of other human perceptual attributes that may be of relevance to device
designers. This section briefly describes the extent and limitations of these human abilities.

2.1.5.1 Estimation of Time Intervals


Even with extensive training, human judgment of time passage is generally inaccurate and
unreliable. This statement applies whether a person is asked to produce a stated interval
of time or to estimate time that has passed. Research has shown that we tend to underesti-
mate elapsed time when we are active and overestimate elapsed time when we are passive.
Table 2.10 illustrates the large range of error in these judgments (Fraise, 1963).

2.1.5.2 Estimation of Other Physical Quantities


Humans have inherent biases in their ability to estimate many other physical quantities
(Table 2.11). Designers need to take into account the imperfect estimation skills of device users.

TAF-ER562X_WINGR-09-0401-C002.indd 46 11/8/10 11:40:36 PM


Basic Human Abilities 47

TABLE 2.10
Errors in Judgments of Elapsed Time Estimation
Activity During a Standard Average Estimation Average Percentage Range of Error
200-Second Interval (in seconds) Error (standard deviation)

Passive
Resting and trying to sleep 241.7 12 107.8
Holding the arms outstretched 228.4 14 96.2
Listening to a metronome
66 beats per minute 223.7 12 92.4
184 beats per minute 214.1 7 85.2
Pressing a point on the skin 210.2 5 78.4

Active
Reading a passage in a mirror 181.8 14 77.6
Taking dictation 174.6 13 77.6
Doing numerical division 168.9 11 70.2

TABLE 2.11
Human Estimation of Physical Quantities
Physical Quantity Human Estimation Tendency Comments and References
Horizontal distance Underestimate Thirty percent of the population has
some depth perception deficiencies
(Richards, 1973)
Vertical height Overestimate when looking up Tendency to overestimate is greatest
Underestimate when looking down for pilots at night (Weintrab and
Virsu, 1975)

Speed of an object Overestimate if object is (Hatayama and Tada, 1972)


accelerating

Constancy of speed Speed perceived as fluctuating (Runeson, 1974)

Geometric angle Underestimate acute angles (Weintrab and Virsu, 1975)


Overestimate obtuse angles

Temperature (ambient) Overestimate when hot Depends greatly on adaptation level,


Underestimate when cold humidity, and air movement
(Geldard, 1972)

Weight of object Overestimate if bulky (Geldard, 1972)


Underestimate if compact

Number of items (without Consistently underestimate (Bevan et al., 1963)


counting)
Volume, area, object temperature, Unreliable estimates with no (Geldard, 1972)
acceleration, and compass bearing general tendency

TAF-ER562X_WINGR-09-0401-C002.indd 47 11/8/10 11:40:36 PM


48 Handbook of Human Factors in Medical Device Design

2.2 HUMAN INFORMATION PROCESSING


This section provides an overview of basic human abilities in processing sensory and per-
ceptual input information. The point at which the perceptual process ends and higher-level
information processing begins cannot be clearly defined.
A bit is defined by the field of information theory as the amount of information obtained
when one of two equally likely alternatives is specified.

2.2.1 LIMITATIONS ON INFORMATION PROCESSING ABILITIES


2.2.1.1 Channel Capacity
Channel capacity is remarkably similar among the sensory modalities in terms of the maxi-
mum number of levels of input stimuli that can be reliably discriminated on an absolute
basis. The maximum number of levels for a single dimension of a sensory channel clusters
around 7.0, which represents 2.8 bits of information for each single presentation of the input
stimulus. Humans are better at making relative judgments than absolute judgments of input
stimuli (Tables 2.12 and 2.13). For example, the average person can reliably identify about
five different tone frequencies when each is presented in isolation from the others, while the
number of reliable discriminations increases to over 1,800 if only a relative discrimination
between two tones presented in sequence is required (adapted from Miller, 1956; Van Cott
and Kinkade, 1972; Woodson and Conover, 1964).

TABLE 2.12
Channel Capacities for Various Human Senses with One Stimulus at a Time
Number of Discriminable Levels
On an Absolute Basis
Sense Stimulus Dimensions (H = bits/stimulus) On a Relative Basis
Vision Pointer position on a linear scale 15 (3.9)
Size of squares 5 (2.2)
Areas 6 (2.6)
Hue 9 (3.1) 128 (medium
brightness)
Brightness 5 (2.2) 570 (white light)
Flashing rate of white light 3 (1.7) 460 (duty cycle = 0.5)
Line length 8 (3.0)
Direction of line inclination 11 (3.3)
Line curvature 5 (2.2)
Audition Loudness 5 (2.2) 325 (at 2,000 Hz)
Pitch 5 (2.2) 1,800 (at 60-dB SPL)
Touch (vibration) Intensity 4 (2.0) 1.5
Duration 5 (2.2)
Location on chest 7 (2.8)
Frequency 5 (2.2) 180
Taste Intensity (salty or sweet) 5 (2.2)
Smell Intensity 4 (2.0)

TAF-ER562X_WINGR-09-0401-C002.indd 48 11/8/10 11:40:36 PM


Basic Human Abilities 49

TABLE 2.13
Channel Capacities for Various Human Senses with Multiple Stimuli Being
Presented at the Same Time
Number of Discriminable Levels
On an Absolute Basis On a Relative
Sense Stimulus Dimensions (H = bits/stimulus) Basis
Vision Size, brightness, and hue (varied together) 18 (4.1) >570
Position of points in a square (no grid) 24 (4.6)
Hue and saturation 13 (3.6) >570
Audition Loudness and pitch 9 (3.1) >1,800
Loudness, pitch, cadence, duty cycle, total 150 (7.4)
duration, and specific location

Guideline 2.13: Improving Information Discrimination


Information discrimination can be increased by allowing relative judgments to be made, pre-
senting multidimensional stimuli, and increasing the rate of sequential presentation of stimu-
lus material.

2.2.1.2 Attention
People usually attend to only one source of sensory information at a time. This is true
because we are basically single-channel processors, although we may do some time shar-
ing of attention similar to the way a single-processor computer multiplexes or multitasks.
Information from other unattended input channels is not totally blocked but rather is pro-
cessed only incompletely with the result that sometimes certain kinds of information do get
through. For example, individuals at a cocktail party can attend to a single voice amidst a
babble of competing voices (the so-called cocktail party effect) yet still hear and recognize
their own name being spoken in another conversation. A possible design implication is that
a simple alternating tone or flashing light could precede important information presented
in a channel likely to be otherwise unattended.

2.2.1.3 Vigilance (Sustained Attention)


Psychologists are beginning to understand more about human performance during long
periods of sustained attention (usually called vigilance activity) such as during a search
for infrequent trouble signals on a display (e.g., missed heartbeats on a patient monitor).
Table 2.14 lists various task conditions that affect human performance during prolonged
vigilance (Van Cott and Kincaid, 1972; Wickens and Carswell, 1997).

2.2.2 SPEED OF INFORMATION PROCESSING


2.2.2.1 Reaction Time
The time it takes for a person to react to an input stimulus and initiate a response is called
reaction time. Simple reaction time (RT) involves only one response to a single stimulus.

TAF-ER562X_WINGR-09-0401-C002.indd 49 11/8/10 11:40:36 PM


50 Handbook of Human Factors in Medical Device Design

TABLE 2.14
Task Conditions That Affect Vigilance Performance
Improved Probability of Signal Detection
• Simultaneous presentation of signals to dual channels
• Observers monitoring display in pairs; members of pairs permitted to speak with one another; 10 minutes
of rest each 30 minutes of work; random schedule inspection by supervisor
• Introduction of artificial signals during vigilance period to which a response is required
• Introduction of knowledge of results of artificial signals
Decreased Probability of Correct Detection
• Introduction of artificial signals for which a response is not required
• Higher or lower task load
• Introduction of a secondary display monitoring task
• Users report only signals of which they are sure
Change in Probability of Detection with Time
• A short pretest followed by infrequently appearing signals during vigilance
• High initial probability of detection, decreasing rapidly
• A few pretest signals before vigilance period
• No CR reduces decrement in probability of detection with time
• Prolonged continuous vigilance
• No CR decreases probability of correct signal detection

Simple RTs for the different senses are provided in Table 2.15 (Brebner and Welford, 1980;
Pierce and Karlin, 1957). Response time, a related concept, is the sum of reaction time, the
cognitive processing time to an input stimulus, and the time to generate a response. The
processing time is sometimes called think time.
Choice RTs require a choice to be made among a number of stimuli and responses and
typically are longer than simple RTs. The rate at which information is processed, such as
during choice RT, is dependent on many factors (e.g., sensory channel, the kind and inten-
sity of stimulus) (Pierce and Karlin, 1957). RT can decrease with training (up to 10%), the
use of an alerting signal, an increase in intensity or duration of the stimulus, and optimized
compatibility between delivered stimulus and expected response. RT increases (deterio-
rates) generally with age, fatigue, and the use of central nervous system– depressing drugs,
such as alcohol.

TABLE 2.15
Simple Reaction Times for Various Sensory Stimuli
Stimulus Type Reaction Time (ms)
Visual 150–225
Auditory 120–185
Tactual (haptic) 115–190
Pain 400–1,000
Cold 150
Warm 180
Movement (body rotation) 520

TAF-ER562X_WINGR-09-0401-C002.indd 50 11/8/10 11:40:36 PM


Basic Human Abilities 51

50

40
Fast speeds
(8 decisions

Errors per 100 moves


per minute)
30

20

10 Slow speeds
(3 decisions
per minute)

0
0 5 10 20 30 40 50
Load (number of columns displayed)

FIGURE 2.23 Speed versus accuracy trade-off (error rate versus speed of decisions). These data
come from a study using a panel of varying numbers of columns of comparison numbers. The men-
tal load increased with the number of columns displayed. The fast and slow speeds were, respec-
tively, six and three decisions per minute. (From Mackworth, J. F. and Mackworth, N. H., J Opt Soc
Am, 52, 713–716, 1958 and McCormick, E. J., Human Factors Engineering, McGraw Hill, New
York, 1970. With permission.)

2.2.2.2 Speed versus Accuracy


The so-called “speed-versus-accuracy trade-off” describes an approximately linear rela-
tionship with accuracy decreasing with either increased demands/faster response or addi-
tional workload. This relationship is shown in Figure 2.23 as the effects of increasing
mental load on errors.

2.2.2.3 Human Memory


Experimental cognitive psychologists refer to three basic types of human memory: sensory
information memory, working short-term memory (STM), and long-term memory (LTM).
Sensory memory has less bearing on device design and thus is not discussed further. Long-
term memory is further categorized into two main areas: procedural (memory for processes
and how to do things) and declarative (memory for facts and what to do). Table 2.16 sum-
marizes the most important and distinguishing characteristics of working, declarative, and
procedural memory (Kyllonen and Alluisi, 1987).

2.2.2.4 Working Memory


This kind of STM transiently holds new information from the senses or other mental
processes as might a computer data buffer. STM can be characterized by having fast access
and retrieval time, limited capacity, and rapid loss of content unless actively attended to.
The fast-access advantage of STM is countered by its limited capacity. The capacity is
limited to five to nine “chunks” of information. This concept of working memory size
limitation is also known as the “magic number seven plus or minus two” (Miller, 1956). The
unit “chunk” is not defined precisely but can be considered a psychologically meaningful
unit of information for material to be placed in STM. A large but limited number of bits of
information may be contained in a chunk. This depends on the schemes used to recode bits

TAF-ER562X_WINGR-09-0401-C002.indd 51 11/8/10 11:40:37 PM


52 Handbook of Human Factors in Medical Device Design

TABLE 2.16
Main Characteristics of Memory Systems
Working Memory Long-Term Memory
Characteristic (Short Term) Declarative Procedural
Primary function Center of all thought/learning Stores meaning of inputs Permanent storage of
how-to knowledge

Subset of declarative Storage of facts

Temporary storage in flux

Capacity Highly limited, 7 ± 2 chunks Unlimited Unlimited

Contents Primarily acoustical codes Semantic codes (primary) Same as declarative

Secondarily visual/spatial Spatial codes


Acoustic codes
Motor codes (physical
movement skills)
Temporal codes

Information units Same as declarative Concepts Production rules from


Schemata/frames/scripts very specific to general
(if–then rules)

Organization Same as declarative Hierarchical with multiple Flat


levels of complexity

Learn/forget Decays with time (73 seconds Learning by being told Generalizations (inductive
processes for one item, 7 seconds for (passive advice taking) and deductive)
three items)

Increased decay time with Encoding Learn by doing (active


rehearsal practice)

Interference from similar Limited by retrieval paths Strengthening and


stimuli and associations reinforcement

Displacement (3–7 slots) Very slow decay Discrimination


Analogies
Problem solving

of input information into psychologically meaningful units. For example, if an individual


had to memorize randomly generated alphabet letters, each letter would be a chunk. But if
the task involved the memorization of randomly chosen words, then each word would be
a chunk. A chunk of medical information could be patient name, room, and bed position.
Information can be held in STM or working memory for as long as it is actively given
attention, which usually involves rehearsing or reviewing the material over and over again.
There is a human tendency to store information in working memory in an acoustical form
whenever possible. The practical implication is that devices should be designed to minimize
listening and talking when information is being held in working memory because these
activities interfere more with STM.

TAF-ER562X_WINGR-09-0401-C002.indd 52 11/8/10 11:40:37 PM


Basic Human Abilities 53

2.2.2.5 Long-Term Memory


LTM has capacity ranging from 109 to 1015 or more bits of information. Retrieval from
declarative LTM (facts) is slower than from working memory. Information gets transferred
into LTM (either from STM or directly from the senses) only if appropriate links or associa-
tions can be established with psychologically meaningful material already in the long-term
store LTM. Procedural memory (how to do things) is best learned by actively practicing a
skill and appears to be very slow to decay (e.g., riding a bicycle, programming a computer,
or remembering how to perform a laryngoscopy). This is because declarative LTM is orga-
nized much like a thesaurus with information having close meaningful associations being
grouped near each other in the brain. People see and hear verbal messages more quickly
and accurately if words with associated meanings are grouped close together, especially
when competing signals are present.

2.2.2.6 Estimation and Decision-Making Abilities


Humans are limited in both estimation and decision-making abilities. Among the highest-
level mathematical operations that people can perform “in their heads” are first-order inte-
gration and differentiation; these are performed crudely at best. Even simple arithmetic
operations are performed poorly as soon as a person is stressed by demands for higher
speed or accuracy. Our guessing behavior or probabilistic estimation skills show the fol-
lowing human tendencies (Kahneman and Tversky, 2000):

• Overestimation of true probability for low-probability events and underestimation of


true probability for high-probability events (especially overestimation of the proba-
bility of chains of unlikely events and underestimation of cumulative risks of events
over a long period of time, such as the relationship between smoking and cancer).
• Overestimation of true probability for events viewed as favorable and under-
estimation of true probability for events viewed as unfavorable. An example of
overestimation is playing the lottery with enormously high odds. An example of
underestimation is feeling threatened by a neurosurgery with an 80% success rate.
• Unwillingness to believe constant probabilities for outcomes of successive indepen-
dent events (also called the gambler’s fallacy, e.g., disbelief that p = .50 for a tail on
the next toss after a fair coin has been tossed 10 times, each time coming up tails).
• Human decision making is often not logical and, depending on circumstances or
on how a problem is framed, we can be risk aversive or risk seeking.
• Humans want to avoid false alarms when it comes to safety-related events. Signal
detection theory (Swets, 1964) makes certain predictions about our decision-making
criteria levels depending on the relative costs for false alarms versus the benefits of
hits or true correct decisions. These trade-offs between decision criteria level and
outcome consequences are dynamic and sometimes rapidly change. For example,
when we are answering yes/no questions as part of a dangerous disease-screening
questionnaire, we answer “yes” more frequently with more lenient criteria for deci-
sion making if the benefits of early valid detection of the disease are very positive
and the therapy is simple, and we answer “yes” less frequently with stricter criteria
if the costs of a false alarm are high (e.g., a very painful therapy for the disease).
• Humans rationalize and rethink decisions in ways that are not always obvious or
predictable. At the root of these problems is the concept of cognitive dissonance as

TAF-ER562X_WINGR-09-0401-C002.indd 53 11/8/10 11:40:37 PM


54 Handbook of Human Factors in Medical Device Design

studied by Festinger (1957). When faced with two options, we need to resolve any
conflict that each choice presents in terms of differences in positive and negative
consequences. Dissonance or conflict may exist both before and after we make deci-
sions. Among common mechanisms for resolving dissonance are the following:
• Rationalization—After making a highly conflicted choice, we reduce disso-
nance by altering our beliefs about something for which we formerly had strong
convictions. For example, we choose to spend time participating in a long, bor-
ing memory experiment for very little reward or pay. We later rationalize our
choice by saying that the experiment really was not that boring but was in fact
very interesting.
• Selective exposure or filtering—We begin to notice only positive attributes or
features of our conflicted choice and ignore or filter any subsequent negative
attributes that reveal themselves. For example, after being prescribed a certain
drug, you begin to notice with increased frequency many other people taking it
and also notice the same drug appearing in television ads.
• Commitment—We increase our confidence about the quality of our conflicted
choice and become steadfast in this belief. This behavioral tendency is related
to sunk-cost bias, whereby we are not willing to give up or cut our losses on a
bad investment that keeps falling in value.
• Defensiveness—We resist any challenges to the soundness of our conflicted
choices and categorize all feedback on our decision as wrong or biased.
• Regret—We admit that we made the wrong conflicted decision and wish we
had made a different choice. An example is the so-called buyer’s remorse we
experience after making a large expenditure for an item of questionable quality
(Malle, 2001). Regret has been shown to have different long-term and short-
term strengths.
− Short-term regret is stronger for actions or commitments taken, such as
buying a house.
− Long-term regret is stronger for inactions, such as the trip you never took.

2.3 HUMAN RESPONSE CAPABILITIES


In addition to limits in the ability to sense, perceive, and process information, humans
have significant limitations in response capabilities after initial cognitive processing is fin-
ished. Only highlights of this broad topic are covered in this section, and the reader is
referred to more in-depth treatments (see Eastman Kodak Company, 1989; Salvendy, 1997).
Physical response data, such as strength, reach, and endurance, are covered in Chapter 4,
“Anthropometry and Biomechanics.”

2.3.1 SPEED OF MOVEMENT


Speed of hand and arm movements are dependent on a number of factors (Brown, 1949;
Fitts, 1954). Hand movement time increases nonlinearly as a function of distance moved
(Figure 2.24):

• Maximum hand velocity for distances less than 3.2 feet is about 10 feet/sec.
• Maximum hand velocity is about 20 feet/sec for longer distances.

TAF-ER562X_WINGR-09-0401-C002.indd 54 11/8/10 11:40:37 PM


Basic Human Abilities 55

1.0
0.8
Total time
0.6

0.4
Primary movement time

Time (sec)
Reaction time

0.2
Secondary
movement
time

0.1
0.08
2 4 6 8 10 20 40
Distance moved (cm)

FIGURE 2.24 Hand movement time as a function of distance moved. Movement time is for non-
repetitive movements of an object by the right hand from one position to another with complete
visual feedback. Primary movement time is the time taken to make the major movement toward a
target after the reaction time delay. Secondary movement time is the time taken to make small final
adjustments while reaching the target.

Figure 2.25 shows the relationship of hand movement distance as a function of time.
Fitts’s law describes the relationship between the speed of a control movement versus its
difficulty (Fitts, 1954). The law can be used to predict a wide variety of user movements,
including those involving surgical tools, computer mouse movements, and foot controls.
Movement time increases proportionately with distance to a target and decreases with

16
Pick up
Movement of right hand (inches)

15” object
12

10”
8

5”
Return
4

0
0 0.20 0.40 0.60 0.80 1.00
Time (sec)

FIGURE 2.25 The times shown are for the movement of one hand from an initial position to some
other position to pick up an object and to return to the original position. Reaction time delay is not
included. The operation of a control panel switch involves this type of movement. The curves illus-
trate that long movements take proportionately less time in relation to length than short movements.
(From Barnes, R. M., Motion and Time Study, John Wiley & Sons, New York, 1963 and Glazer, S.
and Hammell, R., Physical Design of Electronic Systems, Prentice Hall, Englewood Cliffs, NJ,
1970. With permission.)

TAF-ER562X_WINGR-09-0401-C002.indd 55 11/8/10 11:40:37 PM


56 Handbook of Human Factors in Medical Device Design

larger target sizes. Fitts’s law implies that constant ratios of movement accuracy and move-
ment distances result in constant movement times. It is beyond the scope of this chapter to
describe the details. An excellent description of Fitts’s law and its practical applications for
product design is found in Knight (1987).

2.3.2 PRINCIPLES OF MOTION ECONOMY


Industrial engineers have completed large numbers of time and motion studies that sug-
gest that the following principles may be used to increase the speed, accuracy, and ease
of manual operations. Such tasks are not satisfying to most workers and may lead to
errors as well as musculoskeletal disorders. See Chapter 16, “Hand Tool Design,” for
more detailed guidance on design to accommodate human limitations in body movement
and Chapter 12, “Workstations,” for guidance on the design of workspaces to include
manual tasks.

Guideline 2.14: Principles to Achieve Motion Economy


The following principles, adapted from Barnes (1963), should be considered:
1. Avoid repetitive manual tasks. Normally, machines are better for repetitive tasks.
2. Both hands should begin and complete their motion at the same instant.
3. The hands should not be idle except during rest periods.
4. Motions of the arms should be made simultaneously and in opposite and symmetrical
directions.
5. The motion sequence that uses the fewest steps is the best for performing a given task.
6. Horizontal hand movements are faster than vertical.
7. Hands should be relieved of all work that can be performed more advantageously by the
feet or other parts of the body.
8. Where possible, work should be held by jigs or vises so that hands are more free to
operate.
9. Tools, materials, and controls should be located in an arc around the workplace and as
near the worker as possible.
10. Tools and materials should be pre-positioned to eliminate searching and selecting.
11. Two or more tools should be combined whenever possible.
12. The height of the workplace and the chair should preferably be arranged so that alter-
nate sitting and standing at work is easily possible.
13. Continuous, curved motions are preferable to straight-line motions involving sudden
and sharp changes in direction.
14. Ballistic movements are faster, easier, and more accurate than restricted or controlled
movements.
15. Rhythm is essential to the smooth and automatic performance of an operation, and the
work should be arranged to permit easy and natural rhythm whenever possible.
16. Successive movements should be so interrelated that one movement passes eas-
ily into the next, each ending in a position favorable for the beginning of the next
movement.
17. A movement is less fatiguing if it occurs in the direction that takes the greatest possible
advantage of gravity.
18. When a forcible stroke is required, the movements and the material of the worker
must be arranged so that when the stroke is delivered, it has reached its greatest
momentum.

TAF-ER562X_WINGR-09-0401-C002.indd 56 11/8/10 11:40:37 PM


Basic Human Abilities 57

19. Momentum should be reduced to a minimum if it must be overcome by muscular


effort.
20. Hesitation or the temporary and often minute cessation of motion should be analyzed;
its cause should be accounted for and, if possible, eliminated.
21. If a specific combination of movements has been determined as most suitable, empha-
size form rather than accuracy even if this results in poor performance at the beginning
of the learning period.
22. Arm movements that are mostly a pivoting of the elbow with small shoulder and upper-
arm action are faster and more accurate than those with a greater amount of shoulder
and upper-arm action.
23. Manual limb movements terminated by mechanical devices take shorter periods of time
compared to movements terminated solely by visual cues.
24. Single-hand visual positioning movements are faster and more accurate for short dis-
tances on a line 60 degrees from straight ahead on the same side of the body.
25. Two-handed visual positioning movements are most accurate straight ahead and fastest
30 degrees right or left of straight ahead.
26. In blind positioning movements, humans tend to undershoot long distances and over-
shoot short distances. Straight-ahead movements tend to be the most accurate.
27. Continuous movements in a horizontal plane are more accurate in certain angular direc-
tions from the midline of the body. For example, if 0 degrees is straight ahead, the most
accurate movements for right-handed people would be 45 and 225 degrees and for left-
handed people 135 and 315 degrees.
28. Tremor or small vibrations of parts of the body degrade precision work and can be
controlled by providing a visual reference, providing support of the body in general
and any body part in particular, providing placement of the hand within 8 inches (20.3
cm) above or below the heart, or providing a limited amount of mechanical friction to
absorb vibration energy.

2.3.3 SPEECH ATTRIBUTES


2.3.3.1 Loudness Levels of Speech
This section describes some quantitative aspects of the human speech response and produc-
tion capabilities. These attributes will become more important in device design as technol-
ogy advances to allow more reliable and robust speech recognition systems. Table 2.17
shows average talker speech output in terms of sound pressure levels in dB (Morgan, 1963).
Table 2.18 shows the distribution of talker loudness levels (Fletcher, 1953).

TABLE 2.17
Sound Pressure Levels of Speech 1 Meter from the Talker
Normal Level (dB)
Measure of Sound Pressure Whisper (dB) Minimum Average Maximum Shout
Peak instantaneous pressure 70 79 89 99 110
Speech peaks 58 67 79 87 98
Long-time RMS pressures 46 55 65 75 86
Speech minimum 30 39 49 59 70

TAF-ER562X_WINGR-09-0401-C002.indd 57 11/8/10 11:40:37 PM


58 Handbook of Human Factors in Medical Device Design

TABLE 2.18
Distributions of Speaking Volume for Persons
Using the Telephone
Volume Level Range (dB)a Percentage of Speakers
<54 7
54–57 9
57–60 1
4
60–63 18
63–66 22
66–69 17
69–72 9
72–75 4
>75 ~0
a Above sound pressure of 0.0002 millibars at a point 1 m
from the talker’s lips.

2.3.3.2 Frequency Characteristics of Speech


Because speech is a complex time varying quantity, its measurement is complex. Usually,
speech is divided into a number of frequency bands and into a number of time segments
(Figure 2.26). The average frequency is 128 Hz for males and 256 Hz for females. Most of
the energy is below 1,000 Hz, with very little above 5 KHz (Morgan, 1963).

Overall level (dB*)


60 90 Peak
instantaneous
50 A. Instantaneous pressure 80 pressures
exceeded Speech peaks
Speech-spectrum level (dB*)

40 1% of the
B. RMS pressure time 70
exceeded 10% Long time RMS
of the time pressures
30 60
Speech minima
20 50

C. RMS pressure
10 exceeded 80%
of the time
0

–10
100 200 500 1,000 2,000 5,000 10,000
Frequency (Hz)
*Re 0.0002 μ bar
Male speech

FIGURE 2.26 The curves show how the intensity of male speech varies as a function of frequency
and by different measurement criteria. At the upper right of the figure are corresponding overall
levels for unanalyzed, unfiltered speech.

TAF-ER562X_WINGR-09-0401-C002.indd 58 11/8/10 11:40:37 PM


Basic Human Abilities 59

2.4 HUMAN VERSUS MACHINE CAPABILITIES


One way to summarize the basic elements of this introduction to basic human skills and
abilities is to compare the relative advantages and disadvantages of humans versus machines
or mechanical systems (Table 2.19). For example, we know that some human limitations
when compared to machines include limited strength, low endurance, slower processing
speed, less accuracy, emotionally impaired decision making, and severe STM and LTM
limitations. On the other hand, machines do not fatigue, are faster, are more accurate, can
more easily do parallel processing, and are much better at repetitive tasks. Many trade-offs

TABLE 2.19
Human versus Machine Capabilities
Humans
Limitations Advantages
Force—Limited strength Visual acuity and range are very good.
Endurance—Fatigues easily Visual information processing system extremely
logical and flexible.
Speed—Significant time needed for decision making Range of detection is extremely wide with good
and movement time. sensitivity for audition and vision.
Accuracy—Unreliable; makes constant and variable Perception—Ability to make order out of
errors. complex situations; detection possible under high
noise.
Computing—Slow and error prone. Can reason inductively; can follow up intuition.
Decision making—Best strategy not always adapted; Very flexible; can easily change rules of operation
emotions interfere. with changes in situation.
Information processing—Basically a single-channel Attention is easily shifted; only essential
processor, which is easily overloaded; performance information is selected for processing.
greatly dependent on motivation.
Limited short-term working memory; long-term When highly motivated can perform under adverse
memory, although large, has unreliable and slow conditions with parts out of order (injuries).
access.
Machines
Limitations Advantages
Decision making is limited. Great forces are possible.
Inductive reasoning not possible. Does not fatigue easily.
Must be monitored. High speed.
All activities must be thoroughly planned and Great accuracy attainable.
preprogrammed thoroughly.
Must get careful maintenance. Large short-term working memory.
May not operate at all if some parts are broken. For narrow applications long-term memory is
superior.
Complex problems can be handled deductively.
Excellent for repetitive work; unaffected by
emotions and motivational needs.
Can perform simultaneous operations easily.

TAF-ER562X_WINGR-09-0401-C002.indd 59 11/8/10 11:40:38 PM


60 Handbook of Human Factors in Medical Device Design

are required in assigning tasks to humans versus machines. The designer must fully under-
stand the intended tasks, users, and use environment to be able to effectively select and
design device functional attributes. The remainder of this book provides detailed guide-
lines on specific design topics.

RESOURCES
Cobb, P. W. and Moss, F. K. (1928). The four variables in the visual threshold. Franklin Institute
Journal, 205, 831.
Hill, N. E. G. (1947). The recognition of colored light signals which are near the limit of visibility.
Proceedings of the Physiological Society, 59, 560.
Kahneman, D., Slovic, P., and Tversky, A. (1982). Judgment Under Uncertainty: Heuristics and
Biases. Cambridge, MA: Cambridge University Press.
Kling, J. W. and Riggs, L. A. (1971). Experimental Psychology. New York: Holt, Rinehart and
Winston.
McFarling, L. H. and Heimstra, N. W. (1975). Pacing, product complexity, and task perception in
simulated inspection. Human Factors, 17(4), 361–67.
Moray, N. (1972). Listening and Attention. Baltimore: Penguin.
Stevens, S. S. (1947). Sensitivity to changes in the intensity of white noise and its relations to mask-
ing and loudness. Journal of the Acoustical Society of America, 19, 609.
Stevens, S. S. (1951). Handbook of Experimental Psychology. New York: John Wiley & Sons.
Stroh, C. M. (1971). Vigilance: The Problem of Sustained Attention. Oxford: Pergamon.

REFERENCES
American Standards Association. (1954). The Relations of Hearing Loss to Noise Exposure. New
York: American Standards Association.
Barnes, R. M. (1960). Motion and Time Study. New York: John Wiley & Sons.
Bevan, W., Maier, R. A., and Helson, H. (1963). The influence of context upon the estimation of
number. American Journal of Psychology, 76, 464–69.
Brebner, J. T. and Welford, A. T. (1980). Introduction: An historical background sketch. In A. T.
Welford (Ed.), Reaction Times (pp. 1–23). New York: Academic Press.
Brown, J. S. and Slater-Hammel, A. T. (1949). Discrete movements in the horizontal plane as a func-
tion of their length and direction. Journal of Experimental Psychology, 39, 84–95.
Burg, A. (1966). Visual acuity as measured by static and dynamic tests: A comparative evaluation.
Journal of Applied Psychology, 50, 460–66.
Carlson, N. R. (2000). Foundations of Physiological Psychology (5th ed.). New York: Allyn &
Bacon.
Chapanis, A. and Halsey, R. (1956). Absolute judgments of spectrum colors. Journal of Psychology,
42, 99.
Cleland, T. M. (2004). A Practical Description of the Munsell Color System and Suggestions for Its
Use 1937. Whitefish, MT: Kessinger Publishing.
Conover, D. W. and Kraft, C. L. (1958, October). The Use of Color in Coding Displays (USAF
WADC-TR-55-471). Dayton, OH: Wright-Patterson Air Force Base.
Cornsweet, T. N. (1970). Visual Perception. New York: Academic Press.
Corso, J. F. (1967). The Experimental Psychology of Sensory Behavior. New York: Holt, Rinehart
and Winston.
Dreyfuss, H. (1966). Measure of Man (2nd ed.). New York: Watson Cuptill.
Drury, C. G. (1975). Inspection of sheet materials—Model and data. Human Factors, 17(3),
257–65.

TAF-ER562X_WINGR-09-0401-C002.indd 60 11/8/10 11:40:38 PM


Basic Human Abilities 61

Eastman Kodak Company. (1989). Ergonomic Design for People at Work: Vol. II. The Design of
Jobs, Including Work Patterns, Hours of Work, Manual Materials Handling Tasks, Methods
to Evaluate Job Demands, and the Physiological Basis of Work. New York: Van Nostrand
Reinhold.
Festinger, L. (1957). Theories of Cognitive Dissonance. Palo Alto, CA: Stanford University Press.
Fitts, P. M. (1954). The information capacity of the human motor system in controlling the ampli-
tude of movement. Journal of Experimental Psychology 47:381–91.
Fletcher, H. and Munson, W. A. (1933). Loudness: Its definition, measurement, and calculation.
Journal of the Acoustical Society of America, 5, 82–108.
Foley, P. and Moray, N. (1987). Sensation, perception and systems design. In G. Salvendy (Ed.),
Handbook of Human Factors. New York: John Wiley & Sons.
Fraisse, P. (1963). The Psychology of Time. New York: Harper & Row.
Geldard, F. A. (1972). The Human Senses. New York: John Wiley & Sons.
Glazer, S. and Hammell, R. (1970). Man-machine interaction. Physical Design of Electronic
Systems: Vol. 1. Englewood Cliffs, NJ: Prentice Hall.
Harding G. F. and Jeavons, P. M. (1995). Photosensitive Epilepsy. Cambridge, UK: Mac Keith Press.
Hatayama, T. and Tada, H. (1972). The experimental study of speed perception of the car on the road
III: Speed judgements of drivers in the night time. Tohoku Psychologia Folia, 31(1), 4.
Israelski, E. W. (1978). Commonplace human factors problems experienced by the colorblind: A
pilot questionnaire survey. Proceedings of the 22nd Annual Meeting of the Human Factors
Society, 347–51. Human Factors and Ergonomics Society, Santa Monica, CA.
Kahneman, D. and Tversky, A. (Eds.) (2000). Choices, Values and Frames. New York: Cambridge
University Press.
Kaufman, L. and Rock, I. (1972). The moon illusion. Scientific American, July.
Knight, J. L. (1987). Manual control and tracking. In G. Salvendy (Ed.), Handbook of Human
Factors. New York: John Wiley & Sons.
Kyllonen, P. C. and Alluisi, E. H. (1987). Learning and forgetting: Facts and skills. In G. Salvendy
(Ed.), Handbook of Human Factors. New York: John Wiley & Sons.
Lindsay, P., II. and Norman, D. A. (1973). Human Information Processing. New York: Academic
Press.
Mackworth, J. F. and Mackworth, N. H. (1958). Eye fixations recorded on changing visual scenes by
television eye-marker. Journal of Optical Society of America, 52, 713–16.
Malle B. F. (2001). Intentions and intentionality. In L. J. Moses and D. A. Baldwin (Eds.), Foundations
of Social Cognition. Cambridge, MA: MIT Press.
McCormick, E. J. (1970). Human Factors Engineering. New York: McGraw-Hill.
Miller, G. A. (1956). The magic number seven, plus or minus two: Some limits on our capacity for
processing information. Psychological Review, 63, 81–97.
Miller, G. A. (1947). Sensitivity to changes in the intensity of white noise and its relations to mask-
ing and loudness. Journal of the Acoustical Society of America, 19, 609.
Morgan, C. T. (Ed.) (1963). Human Engineering Guide to Equipment Design. New York:
McGraw Hill.
Norman, D. A. (1969). Memory and Attention. New York: John Wiley & Sons.
Nutting, P. G. (1916). Effects of brightness and contrast in vision. Transcriptions of the Illuminating
Engineering Society, 11, 939.
Peterson, A. P. G. and Gross, E. E., Jr. (1967). Handbook of Noise Measurement. New Concord,
MA: General Radio Corp.
Pierce, J. R. and Karlin, J. E. (1957). Reading rates and the information rate of a human channel.
Bell System Technical Journal, 36, 497–516.
Richards, W. (1973). Factors Affecting Depth Perception: Final Report (Government Report
AD-75926 1). Washington, DC: U.S. Government Printing Office.
Riesz, R. R. (1928). Differential intensity sensitivity of the ear for pure tones. Physiological Review,
31, 867.

TAF-ER562X_WINGR-09-0401-C002.indd 61 11/8/10 11:40:38 PM


62 Handbook of Human Factors in Medical Device Design

Runeson, S. (1974). Constant velocity: Not perceived as such. Psychological Research, 37, 3–23.
Salvendy, G. (1997). Handbook of Human Factors and Ergonomics. New York: John Wiley &
Sons.
Shower, E. G. and Biddulph, R. (1931). Differential pitch sensitivity of the ear. Journal of the
Acoustical Society of America, 3, 275.
Stevens, S. S. (1955). The Measurement of loudness. Journal of the Acoustical Society of America,
27, 815.
Stevens, S. S. and Vollunan, J. (1940). The relation of pitch to frequency: A revised scale. American
Journal of Psychology, 53, 329.
Swets, J. A. (1964). Signal Detection and Recognition by Human Observers. New York: John Wiley
& Sons.
Van Cott, H. and Kinkaide, R. (1972). Human Engineering Guide to Equipment Design. Washington
DC: U.S. Government Printing Office.
Weinger, M. B. and Englund, C. E. (1990). Ergonomic and human-factors affecting anesthetic vigi-
lance and monitoring performance in the operating-room environment. Anesthesiology, 73,
995–1021.
Weintrab, D. J. and Virsu, V. (1972). Estimating the vertex of converging lines: Angle mispercep-
tion. Perception and Psychophysics, 11, 277–83.
Wickens, D. C. and Carswell, C. M. (1997). Information processing. In G. Salvendy (Ed.), Handbook
of Human Factors and Ergonomics. New York: John Wiley & Sons.
Woodson, W. E. and Conover D. W. (1964). Human Engineering Guide. Berkeley, CA: University
of California Press.

TAF-ER562X_WINGR-09-0401-C002.indd 62 11/8/10 11:40:38 PM


3 Environment of Use
Pascale Carayon, PhD; Ben-Tzion Karsh, PhD;
Carla J. Alvarado, PhD; Matthew B. Weinger, MD;
Michael Wiklund, MS, CHFP

CONTENTS
3.1 Space and Physical Constraints .................................................................................65
3.1.1 General Principles...........................................................................................65
3.1.2 Design Guidelines ...........................................................................................66
3.1.2.1 Device Form and Configuration .......................................................67
3.1.2.2 Security and Privacy .........................................................................69
3.1.2.3 Device Labeling................................................................................69
3.2 Lighting ....................................................................................................................70
3.2.1 General Principles...........................................................................................70
3.2.2 Design Guidelines ...........................................................................................71
3.2.2.1 Illumination ......................................................................................71
3.2.2.2 Special Medical Environments and Applications .............................73
3.3 Noise ..........................................................................................................................74
3.3.1 General Principles...........................................................................................74
3.3.2 Design Guidelines ...........................................................................................76
3.3.2.1 Critical Communications ..................................................................79
3.4 Climate: Thermal Environment .................................................................................80
3.4.1 General Principles...........................................................................................80
3.4.2 Design Guidelines ...........................................................................................82
3.5 Climate: Humidity .....................................................................................................82
3.5.1 General Principles...........................................................................................82
3.5.2 Design Guidelines ...........................................................................................83
3.6 Climate: Airflow and Pressure...................................................................................83
3.6.1 General Principles...........................................................................................83
3.6.2 Design Guidelines ...........................................................................................84
3.7 Vibration ....................................................................................................................85
3.7.1 General Principles...........................................................................................85
3.7.2 Design Guidelines ...........................................................................................86
3.8 Energy Sources ..........................................................................................................87
3.8.1 General Principles...........................................................................................88
3.8.2 Design Guidelines ...........................................................................................88

63

TAF-ER562X_WINGR-09-0401-C003.indd 63 11/8/10 11:41:07 PM


64 Handbook of Human Factors in Medical Device Design

3.9 Design for Infection Control ......................................................................................89


3.9.1 General Principles...........................................................................................89
3.9.2 Design Guidelines ...........................................................................................90
3.10 Case Studies ...............................................................................................................91
3.10.1 Cleaning of Flexible Gastroscopes .................................................................91
3.10.2 Patient Use Insulin Infusion Pump .................................................................92
Resources ...........................................................................................................................93
References ..........................................................................................................................93

This chapter describes the environmental issues that need to be considered in the pro-
cess of designing effective and safe medical devices. It is important to recognize that the
physical environment is only one element of the work system in which devices are used
(Vincent, Taylor-Adams, and Stanhope, 1998; Weinger and Englund, 1990; Wiklund, 1995).
According to Smith and Carayon (Carayon and Smith, 2000; Smith and Carayon-Sainfort,
1989), the work system is comprised of five elements: (1) the individual (end user), (2) tasks,
(3) tools and technologies (including the medical devices), (4) physical environment, and (5)
organizational conditions. In this chapter, we focus on the effects of the physical environ-
ment on end users and the interaction between the physical environment and a particular
type of technology: medical devices.
The ergonomic aspects of the environment include factors such as space and physi-
cal constraints, lighting, noise, climate, vibration, and electromagnetic radiation (Parsons,
2000). In this chapter, we examine each of these factors individually, addressing each envi-
ronmental characteristic in the context of medical device use. People interact continuously
with their environment in a dynamic manner, experiencing the environment as a whole
(Parsons, 2000), and, therefore, typically do not respond to a single environmental factor
in isolation. Moreover, various environmental factors can interact to produce synergistic or
compound effects on human performance. Thus, medical device designers must consider
the entire integrated physical environment. Both the main effects of environmental factors
and their interactions can affect end users and their ability to use medical devices safely
and effectively (Figure 3.1).
Medical devices should be designed to be both easy to use and effective (i.e., do what they
are supposed to do) in the context of expected use. This means that medical devices must
be designed with consideration of the environment of use, the people who will use them,
the other devices likely to be used at the same time, and the way in which the device will
be used. Medical device designers must study the anticipated environments of use and test
devices within them if they are to design devices that will succeed in those environments
(Bruckart, Licina, and Quattlebaum, 1993; International Electrotechnical Commission
[IEC], 2004; Wickens et al., 2004).
This chapter covers medical devices and medical device use environments in the United
States. Unusual or special environments, such as mobile (e.g., ambulance, helicopter; see
Chapter 17, “Mobile Medical Devices”), public or outdoor (e.g., playgrounds, roadways),
or home (see Chapter 17 and Chapter 18, “Home Health Care”) environments, require spe-
cial design considerations. However, recommendations herein may be applicable to many
health care environments and should be used as appropriate. This guidance is not intended
to restrict medical device innovation or improvements in current design or use techniques,
noting that device design is regulated and advised by many federal agencies (e.g., the Food
and Drug Administration [FDA]) and national and international standards).

TAF-ER562X_WINGR-09-0401-C003.indd 64 11/8/10 11:41:07 PM


Environment of Use 65

FIGURE 3.1 The complex environment of hospital operating rooms (shown) or intensive care
units provides substantial challenges for the medical device designer who must consider myriad
factors, including lighting, temperature, noise levels, and the impact of other devices. (Photo
courtesy of Frank Painter.)

3.1 SPACE AND PHYSICAL CONSTRAINTS


The goal of human factors is to design systems that reduce human error, increase produc-
tivity, and enhance safety and comfort (Wickens et al., 2004). Designing ergonomic work-
spaces is one of the major ways to improve the fit between humans, medical devices, and
the patient care physical space. Health care presents many challenging venues and physical
design constraints for the device designer. Hospitals and clinics are forever adding patient
care equipment, and on-site storage is often limited or not considered at the time of pur-
chase. The designer needs to consider not only how the device will be used in the primary
use environment but also how the device will interact with people and other devices in the
available space and where the device will be stored when not in use. Additionally, devices
designed for clinical arenas are often difficult to modify for mobile or home care, where use
and storage space may be far more limited.

3.1.1 GENERAL PRINCIPLES


The physical space in which a device will be used must be sufficiently large to accom-
modate the functions, people, and devices for which it is intended. In the case of health
care, if the space is too constrained, not only could users be uncomfortable and device use
impaired, but there might also be a greater risk of use errors resulting in harm to patients
and caregivers. Devices that cannot be placed in a convenient location within a work area
are likely to be used less effectively (and may even be used less often). Therefore, designing
devices for use in a specific environment must consider the space available and the associ-
ated reach, line of sight, and related physical use requirements. Line of sight (i.e., ability to
see a critical device attribute from a typical use position), access (i.e., ability to reach and
manipulate the device), and clearance (i.e., space between use elements such as adjacent

TAF-ER562X_WINGR-09-0401-C003.indd 65 11/8/10 11:41:07 PM


66 Handbook of Human Factors in Medical Device Design

FIGURE 3.2 Helicopter interior showing substantial physical space limitations. (Photo courtesy
of Wiklund Research and Design.)

controls) must consider the entire use environment. For example, if a device is likely to be
placed atop an existing workstation (e.g., an anesthesia monitor placed on top of an anesthe-
sia gas machine), then its design should accommodate the visual field of view and reach of
both sitting and standing users. In some use environments, the presence of caregivers, the
patient, and medical devices leaves little room for the addition of new devices without com-
promising clinical care. For example, in the transport care environment (e.g., ambulances
and helicopters; see Figure 3.2), space is extremely limited, and clearance challenges can
be substantial.
Designers also need to consider typical task workflow and other environmental factors.
For example, devices that require power cords or other connections may present a sig-
nificant tripping hazard in a congested use environment occupied by fast-moving workers
(e.g., a busy emergency room). Trip and fall hazards are relatively common in health care
for both patients and workers. People could, for example, fall and sustain injuries by slip-
ping on wet surfaces created by the device, tripping over cords or tubing associated with
the device, or tripping over the device itself. If not disposable, the medical device may stay
in the use environment for prolonged periods. Power supplies, device cleaning, and main-
tenance should be considered during design if the device is likely to reside in a patient’s
room, for example, for many days.

3.1.2 DESIGN GUIDELINES


The following specific design guidelines apply to physical and space considerations.
Guideline 3.1: Device Visibility
Medical devices should be visible in their expected use environments, which might range
from outdoor locations to operating rooms. For example, a device commonly placed on top
of a patient covered by medium-blue sheets should be a contrasting color to make the device
easier to find. Similarly, devices frequently exposed to blood should not be colored red.

TAF-ER562X_WINGR-09-0401-C003.indd 66 11/8/10 11:41:08 PM


Environment of Use 67

Guideline 3.2: Adequate Space for Device Use


The physical space in which a device will be used should be sufficiently large to accommo-
date the functions, people, and devices for which it is intended.

Guideline 3.3: Space Considerations in Device Design


Device designers should consider the space available and the associated reach, line of sight,
and related physical use requirements.

Guideline 3.4: Reach Requirements for Smallest Users


Reach requirements of the device and environment should not exceed the reach of the small-
est users as they operate a hand-operated device and/or activate a foot control (Wickens et al.,
2004) (see Chapter 4, “Anthropometry and Biomechanics”).

Guideline 3.5: Line of Sight and Device Positioning


Device displays should be readily seen and easily read by the users (see Chapter 8, “Visual
Displays”) when placed in the expected range of locations and positions in the intended use
environment. This requires proper positioning and a clear line of sight with respect to the
device and other equipment in the area.

Guideline 3.6: Critical Controls Accessible


Critical controls should be accessible and not placed in tandem with or too close to other
critical controls or devices that may be activated/deactivated inadvertently (see Chapter 7,
“Controls”).

Guideline 3.7: Device Clearances


Device clearances should accommodate not only routine and nonroutine use but also cleaning
and maintenance requirements. When designing clearances, the expected positioning of the
device in the use environment and associated equipment and devices should be considered.

Guideline 3.8: Slipping and Tripping Hazards


Designers should identify and minimize or eliminate slipping and tripping hazards associated
with device use.

3.1.2.1 Device Form and Configuration


A medical device’s form in a particular use environment can have a strong influence on
its safety and usability. For example, an ultrasound scanner that works well in a hospital’s
surgical unit might not be able to be on a helicopter ambulance because of its contextually
large size and weight. In contrast, a new-generation handheld ultrasound scanner, which
is light and compact, could easily fit in a helicopter. Similarly, medical devices that have
a single, inflexible configuration may function well in some use environments but not in
others, where configurable devices would have a distinct advantage.

Guideline 3.9: Compactness for Use in Congested Areas


Medical devices intended for use in congested care environments, such as intensive care units,
should be as compact as possible to enable effective personnel movement and provide room
for other essential equipment (see Chapter 12, “Workstations”).

TAF-ER562X_WINGR-09-0401-C003.indd 67 11/8/10 11:41:09 PM


68 Handbook of Human Factors in Medical Device Design

FIGURE 3.3 Anesthesia workstations may be used on either the patient’s right or left side to
accommodate clinician preference, room configurations, and surgical procedures. (From Music
Therapy World, mmmagazine and University of Michigan Anesthesia.)

Guideline 3.10: Self-Contained Devices


Medical devices should be as self-contained as possible rather than spreading out unnecessar-
ily in ways that interfere with effective personnel movement.

Guideline 3.11: Flexible Placement Within Use Environment


Medical devices should enable various placements within the care environment. For example,
anesthesia workstations should be equally usable when they are placed on the right or left side
of a caregiver positioned in front of the patient’s head (see Figure 3.3).

Guideline 3.12: Cable, Tube, and Wire Management


Medical devices should incorporate the means to organize cables, tubes, and wires so they do
not become tangled or confused with one another or with those associated with other devices.

Guideline 3.13: Device Use In Small Workspaces


When necessary, medical devices should facilitate use in small workspaces. Environments
such as a helicopter ambulance might limit users to reaching controls placed on a front panel,
precluding access to a back panel (see Figure 3.4). In such cases, placing a device’s power
switch on the back panel would pose a major usability problem.

FIGURE 3.4 Ventilator mounting scheme (see right-hand wall) provides access only to the device’s
front panel in this helicopter ambulance. (Photos courtesy of Wiklund Research and Design.)

TAF-ER562X_WINGR-09-0401-C003.indd 68 11/8/10 11:41:09 PM


Environment of Use 69

Guideline 3.14: Damage Resistance


Medical devices used in rescue operations should be able to withstand frequent impacts and
rough handling by personnel focused on a given emergency rather than handling equipment
carefully. Therefore, a patient monitor used in an ambulance or during patient transfers might
be equipped with an oversized handle (encouraging proper carrying) and shock-absorbing
bumpers.

Guideline 3.15: Stable Platform to Prevent Tipping


Medical devices should be sufficiently stable to withstand the likely types of impacts in the
intended use environments. For example, a cart-mounted electrocardiograph should not be
subject to tipping over when quickly pushed out of the way in an emergency or struck by a
hospital bed being rolled down the hallway.

3.1.2.2 Security and Privacy


Medical devices should be protected against accidental actuations such as might occur
when bumping into a control panel. Also, some medical devices, such as infusion pumps
used to deliver narcotic drugs, might be subject to unauthorized use or even malicious tam-
pering (i.e., for drug diversion), suggesting the need for protective mechanisms. Finally, the
details of medical care are not a matter of public disclosure. Federal laws strictly regulate
the access and release of medical information.

Guideline 3.16: Protection Against Inadvertent Actuation


Controls on medical devices used in congested environments should be protected against
inadvertent actuation by moving equipment and personnel that might bump against them (see
Chapter 17, “Mobile Medical Devices”).

Guideline 3.17: Tamper Resistance


Medical devices used in uncontrolled or unsupervised environments should be protected
against tampering, including unauthorized use. For example, electronic controls might require
the user to enter a password to unlock them or use a physical key to open a cover.

Guideline 3.18: Protecting Patient Privacy


Where practicable, medical devices should be designed to prevent bystanders (e.g., hospital
visitors and other patients) from viewing confidential patient information appearing on dis-
plays. For example, a computer display might be equipped with an overlay (e.g., polarizing
film) that prevents viewing from the side. However, such overlays should not impede reading
by intended users from likely viewing angles.

3.1.2.3 Device Labeling


It is important for clinicians to rapidly locate, identify, and take note of any necessary pre-
cautions associated with a given medical device.

Guideline 3.19: Indicate Prohibited Uses


Medical devices should indicate pertinent, prohibited uses. For example, a medical device
containing ferrous metals that is likely to be used near an MRI scanner should warn users
about exposing it to an intense magnetic field.

TAF-ER562X_WINGR-09-0401-C003.indd 69 11/8/10 11:41:11 PM


70 Handbook of Human Factors in Medical Device Design

Guideline 3.20: Bold Labeling


Medical devices should be labeled boldly to facilitate rapid identification, particularly when
the device will be used or stored among with many other devices and supplies (see Chapter
13, “Signs, Symbols, and Markings”).

Guideline 3.21: Warning Label Placement


Warning labels should be placed where intended users are most likely to see them. In some
cases, multiple labels with the same warning may be necessary to ensure reliable viewing in
various use environments and use positions.

Guideline 3.22: Consistent Color Coding


Color coding should be consistent with local and national conventions. For example, controls
on medical air valves used in the United States should be colored yellow, while those used in
several European countries should be checkered black and white.

3.2 LIGHTING
Characteristics of lighting in the use environment that are relevant to device design include
illumination, luminance, contrast, glare, and shadow (see also Chapter 2, “Basic Human
Abilities,” and Chapter 8, “Visual Displays”). Illumination is the amount of light falling
onto a surface, commonly measured in units of lux, an SI unit (one foot-candle equals 10.76
lux). Luminance is the light generated by a surface and is commonly measured in units of
candelas per square meter (cd/m2), an SI unit (one foot-lambert equals 3.426 cd/m2). The
ratio of the amount of light reflected by a surface (luminance) to the amount of light strik-
ing the surface (illuminance) is called the reflectance. The formula for reflectance is

Reflectance = π × luminance (cd/m2)/illuminance (lux)

The contrast or luminance ratio is the ratio of luminance of any two surfaces or areas in
the visual field. If there are large differences in luminance in the environment, the eye must
adapt to these different levels as it moves from one area of the environment to another.
Large differences in luminance in the environment can be sources of glare and cause visual
discomfort, annoyance, or decreased visual performance (Sanders and McCormick, 1993).
Glare is the reflectance of bright light off a surface that reduces visual contrast and impairs
visibility. Glare can be direct if it is caused by a bright area in the visual field or indirect
if it is caused by light being reflected by a surface in the visual field. Glare depends on the
range of luminances in the visual field, on the reflectance of surfaces in the visual field, and
on the position of the light source relative to the line of sight of the user. As the glare source
becomes closer to the line of sight, discomfort increases (Sanders and McCormick, 1993).

3.2.1 GENERAL PRINCIPLES


A poorly designed lighting environment will provide excessive or insufficient light for the
intended tasks or activities. The effects of lighting on performance depend on the visual
stimulus, individual characteristics (e.g., age), visual demands of the task, and physical
environment (Boyce, 1997). Lighting parameters mainly affect the visual and perceptual

TAF-ER562X_WINGR-09-0401-C003.indd 70 11/8/10 11:41:11 PM


Environment of Use 71

TABLE 3.1
Recommended Illumination Levels by Location
Location Illumination (lux)
Bed, observation 400
Bed, examination 1,000
Operating table (directed locally) 50,000–100,000
Operating room where other work is being performed 400–500
X-ray processing room 50

Source: From The Chartered Institution of Building Services Engineers, Lighting


Guide—Hospitals and Health Care Buildings (LG2: 1989 ed.), The Chartered
Institution of Building Services Engineers, London, 1989. With permission.

aspects of a task. For instance, light produced by medical devices and other equipment (e.g.,
bright lighting in an operating room) can create direct glare. Other lighting attributes affect-
ing visual performance include illumination, luminance contrast (e.g., contrast between an
object and its background), and reflectance. Display luminance can affect visual search
performance when viewing radiographs (Krupinski, Roehrig, and Furukawa, 1999).
Lighting levels vary widely depending on the health care use environment and the asso-
ciated medical procedure or therapy. The visual requirements of clinical tasks performed
with medical devices tend to be very high because of the precision required. For exam-
ple, illumination levels of 10,000 to 20,000 lux are recommended for surgical procedures
(Sanders and McCormick, 1993). Brighter illumination can improve the depth of field and
therefore increase visual acuity. On the other hand, high illumination increases the likeli-
hood of glare and shadows and may overburden the visual system. Even in the same use
environment, lighting levels may vary considerably depending on the specific task being
performed. For example, while operating rooms are normally brightly lit by general room
lights as well as special spotlights (i.e., operating room lights), they might be dimly lit
to facilitate a minimally invasive surgical procedure, such as arthroscopy, during which
procedural guidance is derived from images on large displays. Therefore, devices used
in the operating room, such as a ventilator, must be designed for use in variable lighting
conditions. Lighting conditions will be even more variable for medical devices designed for
patients to use in homes, public spaces, and outdoors.
Recommendations for illumination levels in various areas of hospitals and health care
buildings can be found for the United Kingdom in the Lighting Guide of The Chartered
Institution of Building Services Engineers (1989). For North America, illumination recom-
mendations can be found in the IES Lighting Handbook (Illuminating Engineering Society
of North America, 1981) (see Tables 3.1 and 3.2).

3.2.2 DESIGN GUIDELINES


3.2.2.1 Illumination
Designers should determine the visual demands of the tasks performed with the medical
devices, such as the need for precision work and the characteristics of the visual envi-
ronment. This information should be used to determine the level of illumination that

TAF-ER562X_WINGR-09-0401-C003.indd 71 11/8/10 11:41:11 PM


72 Handbook of Human Factors in Medical Device Design

TABLE 3.2
Recommended Illumination Levels by Visual Task Demands
Visual Demands Illumination (lux)
Visual demands are not high 100
Performance of visual tasks—high-contrast items or large size 300
Performance of visual tasks—medium-contrast items or small size 500
Performance of visual tasks—low-contrast items or very small size 1,000
Performance of exacting visual tasks (e.g., surgery) 3,000

Source: From S. N. Chengalur, S. J. Rodgers, and T. E. Bernard, Kodak’s Ergonomic Design for
People at Work (2nd ed.), John Wiley & Sons, Hoboken, NJ, 2004. With permission.

should be provided for the tasks and thereby dictate the design of displays, controls, and
device labeling.

Guideline 3.23: Intended Use Environment Illumination


The illumination levels necessary to safely and effectively use a device should be determined
by knowledge of the intended use environments and user tasks.

Guideline 3.24: Adjustable Lighting Levels


Light sources (e.g., a display backlight) should be adjustable (including the option to turn them
off) to facilitate use in a variety of expected use environments.

Guideline 3.25: Luminance Ratios


Visual performance is clearly affected not only by the type of display equipment (e.g., view-
ing box versus cathode ray tube monitor) but also by its luminance characteristics (Krupinski
et al., 1999). The luminance ratios of device displays should be as follows: about 1 between
tasks and adjacent darker surroundings, 0.33 between tasks and adjacent lighter surroundings,
and 10 between tasks and more remote surfaces (Chengalur, Rodgers, and Bernard, 2004).

Guideline 3.26: Display Reflectance to Prevent Glare


The reflectance of the device display should be such that it produces as little glare as possible.
In general, surfaces should diffuse light (e.g., matte finishes) (Chengalur et al., 2004). If the
medical device produces light, the light source should be shielded and not produce direct
glare.

Guideline 3.27: Illuminated Displays and Controls


Medical devices used in low lighting conditions (e.g., outdoors at night or in a diagnostic
procedure room with its lights dimmed to facilitate minimally invasive procedures) should
be illuminated as required to ensure proper device operation (see Figure 3.5). For example,
keyboards and controls can be spotlighted, while displays can be backlighted.

Guideline 3.28: Attentuation of Transmitted Light


If a device incorporates see-through materials (glass/Plexiglas between the user and what
he or she is to view, then the design should consider the attenuation of transmitted light.

TAF-ER562X_WINGR-09-0401-C003.indd 72 11/8/10 11:41:11 PM


Environment of Use 73

FIGURE 3.5 Medic takes a soldier’s vital signs while traveling in a vibrating and dimly lit mili-
tary transport aircraft. (Courtesy of Air Force Photo Gallery, image 081107-F-062E-477.)

For instance, under normal lighting conditions, a reduction of 6% to 16% of the illumination
was observed during the illumination’s passage through the walls or roof of infant incubators
(Sjors et al., 1992). Incubators with double walls produce a greater decrease in illumination
than incubators with single walls.

Guideline 3.29: Display Legibility in Bright Light


As required, displays should remain legible when exposed to direct sunlight and intense arti-
ficial light sources (e.g., operating room lights).

Guideline 3.30: Display Cleanability


Device displays should be designed so that they are easy to clean, thereby maintaining
intended brightness (McCarthy and Brennan, 2003).

3.2.2.2 Special Medical Environments and Applications


3.2.2.2.1 Radiographic Imaging Environments
Interpretation of medical/clinical images is dependent on image viewing conditions, includ-
ing the lighting characteristics of the display equipment (e.g., viewing boxes and visual
display). A range of guidelines has been developed that define appropriate radiological
viewing conditions (see Table 3.3).

3.2.2.2.2 Endoscopic Devices


The presence or absence of shadow can affect visual performance. The presence of shad-
ows can provide depth cues that facilitate the reconstruction of three-dimensional images.
Hanna, Cresswell, and Cuschieri (2002) demonstrated that the use of a shadow-inducing
system facilitated mental processing of an endoscopic image reconstructed as a three-
dimensional picture of the operative field. On the other hand, the presence of shadows
may affect visual performance as well as working postures (e.g., the surgeon has to adopt

TAF-ER562X_WINGR-09-0401-C003.indd 73 11/8/10 11:41:11 PM


74 Handbook of Human Factors in Medical Device Design

TABLE 3.3
Guidelines on Radiological Image Viewing Conditions
Brightness of Viewing Uniformity of
Source Box (cd/m2) Viewing Box (%)a Illumination (lux)
World Health Organization (1982) 1,500–3,000 <15 <100
CEC (1996) 2,000–4,000 Not available Not available
CEC (1997) >1,700 <30 <50

Source: Adapted from E. McCarthy and P. C. Brennan, British Journal of Radiology, 76, 94–97, 2003.
a Uniformity of viewing box is measured as the maximum deviation of brightness as a percentage of the sum

of the maximum and minimum brightness levels (McCarthy and Brennan, 2003).

awkward postures to avoid shadows in the visual field). Instruments equipped with coaxial
illumination provide illumination that is parallel to the line of sight and facilitate shadow-
free images, and can significantly improve working postures (Chang, 2002).

3.2.2.2.3 Use of Goggles or Other Eye Gear


Because of concerns about infection exposure, clinical personnel often wear face shields,
protective glasses, or goggles. Some situations (e.g., laser procedures) require use of colored
or other types of special glasses. When indicated, device displays should accommodate
the resulting glare, parallax, and altered vision. The use of night-vision goggles allows the
performance of various medical procedures, such as intubations at night under emergency
field conditions (Gillis and Miles, 2001). However, providing a small light source that pro-
jects a focused beam of light may provide safer care than the use of night-vision goggles,
for instance, in obtaining intravenous access (Schwartz and Charity, 2001).

Guideline 3.31: Device Use with Goggles


If the expectation is that a device will be used by people wearing goggles or protective eye
gear, the design should accommodate such use without decreases in user performance.

Guideline 3.32: Displays Viewed through Goggles


Displays that might be viewed by users wearing night-vision goggles (e.g., paramedics serving
in helicopters at night) or laser protection goggles (during laser procedures) should be opti-
mized for such use and provide readily accessible brightness controls.

3.3 NOISE
3.3.1 GENERAL PRINCIPLES
Noise is a common attribute of medical device use environments and must be considered
during device design (see also Chapter 2, “Basic Human Abilities,” and Chapter 10, “Alarm
Design”). Noise can contribute to a number of undesirable outcomes in health care deliv-
ery settings. For example, noise can increase length of hospital stay (Fife and Rappaport,
1976) and contribute to patients’ sleep–wake abnormalities in the intensive care unit

TAF-ER562X_WINGR-09-0401-C003.indd 74 11/8/10 11:41:12 PM


Environment of Use 75

(Freedman et al., 2001). Noise can increase heart rate, subjective stress, and annoyance
(Morrison et al., 2003), and possibly impair task performance, concentration, and com-
plex problem-solving (Morrison et al., 2003). Depending on the task conditions and the
type of sounds, noise may impair performances, have minimal impact, or can even improve
performance (Sanders and McCormick, 1993). Difficult tasks that place high demands on
perceptual-motor skills and/or information processing (e.g., team communication during
high-risk emergency procedures) may be most affected by noise (Sanders and McCormick,
1993). Loud noise can impair reading comprehension. Even noise as low as 68 dBA (where
dBA is decibels on the A-weighted logarithmic scale for sound loudness)—about the same
volume as a person speaking normally from 3 to 5 feet away—can impair the retrieval of
information from memory (Crocker, 1997). Noise also influences response bias; people
tend to think an auditory signal is present only when they are absolutely sure they heard the
signal. Loud noises can cause a startle response that disrupts perceptual-motor performance.
However, with repeated exposure to noise, the magnitude of this startle response diminishes.
Noise appears to have little impact on motor performance or visual functions such as acuity,
contrast discrimination, dark vision, and accommodation (Sanders and McCormick, 1993).
There are two important ways that noise can affect medical device use. First, noise can
prevent health care personnel from hearing device-generated communications, informa-
tion signals, or alarm sounds. That is, noise (from speech, devices, procedures, and so on)
can mask sounds emitted from medical devices. Masking occurs when one aspect of the
sound environment reduces the sensitivity of the ear to other aspects of the sound environ-
ment. For example, paramedics, emergency medical technicians, emergency physicians,
and transport nurses could correctly identify 96% of breath sounds in a quiet environment
but only 54% when inside an ambulance (Brown et al., 1997). Masking of device sounds by
noise increases when device sounds are near the frequency of the masking tone and its har-
monic overtones and as noise intensity increases (Sanders and McCormick, 1993). Complex
sounds are more difficult to mask than simple sounds.
For sound levels in clinical facilities, the World Health Organization and the U.S.
Environmental Protection Agency recommend that hospital rooms should not be louder
than 30 to 45 dBA (Shankar et al., 2001; World Health Organization, 1999). Table 3.4 shows
noise levels (dBA) of common sound-producing events. The noise guidelines for newly con-
structed or renovated hospital nurseries recommended that continuous sound in occupied

TABLE 3.4
Noise Levels
Quality dBA Source Effect
Just audible 10
Very quiet 20–30 Whisper <35 dbA desired for sleep
Quiet 50 Light traffic <50 dbA desired for work
Moderately loud 70 Vacuum cleaner Annoyance
Loud 90 Pneumatic drill Hearing loss with chronic exposure
Very loud 100 Power mower
Uncomfortably loud 120 Discotheque Pain and distress

Source: Reproduced from W. E. Morrison, E. C. Haas, D. H. Shaffner, E. S. Garrett, and J. C. Fackler,


Critical Care Medicine, 31, 113–119, 2003. With permission.

TAF-ER562X_WINGR-09-0401-C003.indd 75 11/8/10 11:41:12 PM


76 Handbook of Human Factors in Medical Device Design

TABLE 3.5
Noise Levels Measured in Hospital Environments
Study Setting Noise Levels (dBA)
Bentley et al. (1977) ICU Average 42–53
Peaks >70
Redding et al. (1977) ICU Average 74
Meyer et al. (1994) ICU Peaks 83.6
Meyer-Falcke et al. SICU Average >65
(1994) Peaks >110
McLaughlin et al. (1996) CSICU Average 58–77
Peak 101
Tsiou et al. (1998) Six-bed ICU Average 60–67
Peak 83–90
Keipert (1985) Peds ward Peaks 80–88
Couper et al. (1994) Peds ward 65

Source: Adapted from W. E. Morrison, E. C. Haas, D. H. Shaffner, E. S. Garrett, and J. C. Fackler,


Critical Care Medicine, 31, 113–119, 2003. With permission.
ICU, intensive care unit; SICU, surgical ICU; CSICU, cardiac surgical ICU; Peds, pediatrics.

bed spaces or patient care areas should not exceed an hourly equivalent continuous sound
level (Leq) of 50 dBA (Philbin, Robertson, and Hall, 1999). But, as shown in Table 3.5,
adapted from Morrison et al. (2003), noise averages from a variety of hospital studies regu-
larly exceed these guidelines, and peaks can exceed 100 dBA, a level considered dangerous
for hearing (Wickens et al., 2004).
The second way that noise can affect medical device use is through its effects on perfor-
mance and stress. Acute noise exposure causes a stress response, while chronic exposure
can adversely affect health (Weinger and Englund 1990; Wickens et al., 2004). People have
more difficulty hearing an auditory signal when they are simultaneously attending to or
performing another activity (Sust and Lazarus, 2003); this is typical in health care, as
clinicians are typically conducting a variety of patient care tasks while at the same time
attending to medical devices’ auditory signals. Blood pressure values obtained by medical
personnel in a quiet environment were significantly more accurate than those obtained in
an ambulance, although the results may have been affected by the motion of the ambulance
(Prasad et al., 1994).
Table 3.6 provides examples of a variety of noise sources in hospitals. Medical devices
are also used in transport vehicles such as ambulances, fixed-wing aircraft, and rotary-wing
aircraft where noise levels can range from 65 to more than 99 dBA (Campbell et al., 1984;
Macnab et al., 1995).

3.3.2 DESIGN GUIDELINES*


A key design principle is that if a medical device auditory signal occurs in the presence of
noise, the threshold of detectability of the signal will be increased; this auditory threshold

* See Chapter 10, “Alarm Design.”

TAF-ER562X_WINGR-09-0401-C003.indd 76 11/8/10 11:41:12 PM


Environment of Use 77

TABLE 3.6
Noise Levels Associated with Common Events and Activities in Hospitals
Noise Sources Sound Level (dBA) Study

Adult ICU item


Connecting/disconnecting gas supplies 88 Tsiou et al. (1998)
Oximeter alarm 81 Kahn et al. (1998)
Nebulizer 80 Kahn et al. (1998)
Telephone ringing 80 Kahn et al. (1998)
Ventilator alarm 79 Kahn et al. (1998)
Monitor alarm 79 Kahn et al. (1998)
Television (loud) 79 Kahn et al. (1998)
Ventilator 78 Kahn et al. (1998)
IV alarm 77 Kahn et al. (1998)
Air conditioner 75 Kahn et al. (1998)
Open suctions 70–82 Tsiou et al. (1998)
Open oxygen sources 70–77 Tsiou et al. (1998)
Dialysis alarm 63 Kam et al. (1994)
Patient transfer 60–66 Kam et al. (1994)
Moving bed 58 Kam et al. (1994)
Respirators 49–72 Tsiou et al. (1998)
Pediatric ICU item
Cleaning crew/equipment ≤96 Morrison et al. (2003)
Ventilator alarms ≤79 peak Morrison et al. (2003)
Infants crying 78 Morrison et al. (2003)
Conversations ≤73 peak Morrison et al. (2003)
Trauma phone 73 Morrison et al. (2003)
Monitor alarms 62–74 Morrison et al. (2003)
Overhead pages 59–84 Morrison et al. (2003)
Neonatal incubators 58 Kam et al. (1994)
Medication pump alarms 55–56 Morrison et al. (2003)
PACU
Patient crying 80–86 Kam et al. (1994)
ECG alarms 75–78 Kam et al. (1994)
Patient coughing 70 Kam et al. (1994)
Changing bed linen 56–66 Kam et al. (1994)
Staff normal conversation 56–60 Kam et al. (1994)
Opening plastic packets 55–72 Kam et al. (1994)
Cardiac monitors 44–78 Kam et al. (1994)
Operating room
Suction in use 75–80 Kam et al. (1994)
Suction not in use 73 Kam et al. (1994)
Surgeon instructions 66–72 Kam et al. (1994)
Ventilator 65 Kam et al. (1994)
Ward areas 58 Kam et al. (1994)
Operating theaters 57 Kam et al. (1994)
Intensive care 57 Kam et al. (1994)
continued

TAF-ER562X_WINGR-09-0401-C003.indd 77 11/8/10 11:41:12 PM


78 Handbook of Human Factors in Medical Device Design

TABLE 3.6 (CONTINUED)


Noise Levels Associated with Common Events and Activities in Hospitals
Noise Sources Sound Level (dBA) Study

Other hospital areas


Nonmedical care 65 Bayo et al. (1995)
Corridors 61 Bayo et al. (1995)
Clinics 59 Bayo et al. (1995)

ICU, intensive care unit; IV, intravenous; PACU, postanesthesia care unit; ECG, electro-
cardiogram.

needs to be exceeded if the signal is to be detected (Sanders and McCormick, 1993).


Effective design thus requires a thorough understanding of the noise levels in the environ-
ment of use. For example, one of the recommendations is for auditory signals to be at least
15 dB above the masking threshold.

Guideline 3.33: Quiet Device Except when Signaling


Medical devices should be as quiet as possible except when a certain amount of noise provides
useful audible feedback to the user or patient. For example, portable oxygen generators should
operate quietly so that they do not disturb patients, particularly when they are trying to sleep.
Conversely, it might be best for a hospital bed to make a whirring noise when moving as a
means to help protect against harm that could come from inadvertent bed motion.

Guideline 3.34: Auditory Signal Intensity above Masking Threshold


The sound level of an auditory signal should generally be at least 15 dBA but not more than
25 dBA above the threshold. However, Sust and Lazarus (2003) suggest that levels as high as
30 dBA may sometimes be required to attract attention during other patient care activities.
Avoid auditory alarms that are in the dangerous range for hearing (above 85 to 90 dBA) unless
that is the only intensity that the auditory signal can be heard. In this latter case, the signal
duration should be brief.

Guideline 3.35: Auditory Signal Frequency


Auditory signals should have frequencies between 500 and 3,000 Hz because the ear is most
sensitive to the middle ranges.

Guideline 3.36: Auditory Signal Duration


Auditory signals should be at least 500 ms in duration (preferably between 500 and 1,000 ms).
A shorter duration signal will not sound as loud, and the ear does not respond immediately to
sound. If auditory signals must be shorter, their intensity should be increased.

Guideline 3.37: Get Users’ Attention, then Provide


Useful Information
When presenting complex information with auditory signals, use an attention-demanding sig-
nal, followed by a designation (or information-emitting) signal that provides the meaning of
the signal.

TAF-ER562X_WINGR-09-0401-C003.indd 78 11/8/10 11:41:12 PM


Environment of Use 79

Guideline 3.38: Auditory Signal Consistency


The same signal should always be used to convey the same information. Similarly, avoid con-
flict with signals that are already being used to convey other types of information. This may
be especially important in noisy environments where some degree of masking is occurring
and some signals may mask others.

Guideline 3.39: Consistency of Auditory Signals with


Learned Meanings
When appropriate, auditory signals should be compatible with natural relationships (i.e., con-
sistent with human associations with other environmental and man-made sounds). For exam-
ple, higher-frequency wailing signals are generally interpreted as indicating an emergency.

Guideline 3.40: Auditory Signal Usability Testing


All auditory signals should undergo usability testing with representative users for detection
and discrimination in the intended use environments.

3.3.2.1 Critical Communications


When a medical device requires attention, perhaps because there is a problem or a consum-
able has run out, it must draw the user’s attention in a reliable manner. The most appropriate
means of communicating a critical message will depend on the use environment.
If task and environmental analyses indicate that the device will be used in a noisy envi-
ronment, nonauditory signals should be incorporated to convey critical device informa-
tion. Auditory displays are preferred over visual displays when the message is simple and
short, will not be referred to later, deals with events in time, and calls for immediate action
(Sanders and McCormick, 1993). Auditory displays may also be preferred when the visual
system is overburdened, illumination limits vision, the user moves from place to place,
or a verbal response is required. Overall, auditory alarms produce greater compliance
than visual alarms (Wickens et al., 2004). But, considering the noise in most health care
environments, especially transport vehicles, visual signals and displays should usually be
incorporated into the device as a complement to the auditory signals. Fromm, Campbell,
and Schlieter (1995) studied the ability of health care personnel to react to visual alarms
in medical helicopters. Reaction times to visual alarms were very long and variable, with
mean reaction times of 81 ± 78 seconds, with a range of 3 seconds to more than 5 minutes.
Therefore, careful analysis followed by rigorous testing in potential use environments
should guide design decisions in noisy environments.

Guideline 3.41: Reliable Detection of Alarm Signals


Where appropriate, alarms should be designed for reliable detection in visually congested
and/or noisy environments. In some cases, this might call for redundant annunciation in mul-
tiple locations, such as in a patient’s room, adjacent corridor, and central workstation.

Guideline 3.42: Audible Alarms


In visually congested environments, alarms should be annunciated by means of an audible
signal (see Chapter 10, “Alarm Design”).

TAF-ER562X_WINGR-09-0401-C003.indd 79 11/8/10 11:41:12 PM


80 Handbook of Human Factors in Medical Device Design

Guideline 3.43: Visual Alarms


In noisy environments, alarms should be annunciated by means of a visual signal, such as a
flashing light (see Chapter 10, “Alarm Design”).

Guideline 3.44: Tactile Alarms


In visually complex and noisy environments, alarms should be annunciated by means of a
tactile signal, such as making a device worn on the body (e.g., pager) vibrate (see Chapter 10,
“Alarm Design”).

Guideline 3.45: Voice Prompts


Voice prompting should only be used to communicate critical information (i.e., instructions
on how to operate an automated external defibrillator) when other information channels
(e.g., vision and touch) are already devoted to sensing other stimuli or when a voice is most
likely to produce the desired behavior (see Chapter 10, “Alarm Design”).

3.4 CLIMATE: THERMAL ENVIRONMENT


Many medical devices, such as stretchers and portable ventilators, venture outside of envi-
ronmentally controlled settings like hospitals and doctors offices. Clearly, such devices will
be exposed to a wide range of climatic conditions, placing special demands on the equip-
ment and users. The devices and users might also be subjected to physical stresses, such
as intense vibration or jarring. Also, the device might be used in a potentially contagious
environment. Although most clinical settings in the United States are highly controlled
thermal/humidity environments, medical devices are often used in much less controlled
environments, including outdoors. Even the most highly controlled environment, the oper-
ating room, can have temperatures ranging from 12°C (cardiac surgery) to 30°C (burn or
neonatal surgery).

3.4.1 GENERAL PRINCIPLES


Personal comfort varies with time of day, season, diet, health status, and clothing choices
as well as job or task stress and cultural variables and expectations (Chengalur et al., 2004).
Even under optimal comfort conditions, a small percentage (<3%) of the population is too
warm or too cold (Fanger, 1970). Six main factors should be considered in device design
and user responses to thermal environments:

• Air temperature
• Radiant temperature
• Air velocity (affected by air-handling systems and fans)
• Humidity
• Level of user activity
• Users’ clothing

If the user is subjected to thermal stress, then the thermoregulatory system responds by
changing its state to maintain core body temperature. This response has consequences
for the user’s health, comfort, and working efficiency (Parsons, 2000). It is difficult to

TAF-ER562X_WINGR-09-0401-C003.indd 80 11/8/10 11:41:12 PM


Environment of Use 81

accurately predict the effects of thermal environments on specific patient care task perfor-
mance because of the many variables involved. There are, however, numerous nonmedical
studies on the effects of thermal stresses on human performance (Parsons, 1993). When
humans are exposed to a hot environment, the effects on task performance will depend on
individual levels of arousal and motivation as well as individual differences, such as degree
of environmental acclimatization (Parsons, 2000). As heat stress increases, mental and
physical performance will be affected adversely. Performance on sustained vigilance tasks,
often required during patient care, can be impaired in slightly warm environments that pro-
duce soporific effects. For example, a nurse who is heavily garbed for patient isolation may
be more likely to miss early signs of an evolving patient crisis if performing monotonous,
repetitive monitoring tasks in a room warmed to a higher-than-normal temperature for the
health of a burn patient who loses heat more readily.
Cold stress is generally a less common health care occupational hazard than heat stress.
However, providing emergency care outdoors can expose the device user to the cold and
elements, including snow and ice (Figure 3.6). Additionally, because of the requirements of
the Occupational Safety and Health Administration’s Bloodborne Pathogens Rule to wear
moisture-resistant, personal protective barriers for surgical procedures, many operating
rooms have lowered ambient room temperature to accommodate medical staff wearing this
extra protective gear. Operating room nurses and anesthesia providers, who do not typically
wear as much protective clothing, must cope with these cold temperatures. The two pri-
mary physiological reactions to cold stress, vasoconstriction and shivering, reduce net heat
loss, thus maintaining core temperature (Sanders and McCormick, 1993). However, vaso-
constriction can cause decreased tactile sensitivity, increasing the risk of use error when
interacting with medical devices. Shivering can be severe enough to impair vision and the
execution of even simple fine motor tasks, such as activating a control or holding a portable
device. Additionally, clothing required to reduce heat loss, such as heavy gloves, can inter-
fere with tactile sensation or the ability to grasp. Hats can cause hearing loss. Jackets can

FIGURE 3.6 Ski patrollers prepare to attach a splint in cold weather. The use of medical devices
in such outdoor conditions will be affected not only by the temperature but also by the required
protective gloves and clothing. (From http://www.nsw-ski-patrol.org.au/perisher/pics/treat.jpg. With
permission.)

TAF-ER562X_WINGR-09-0401-C003.indd 81 11/8/10 11:41:12 PM


82 Handbook of Human Factors in Medical Device Design

reduce mobility and range of motion. Thus, user-interface designs need to account for users
wearing protective clothing and gear as well as any physical or mental impairments that a
user might be under in harsh conditions.

3.4.2 DESIGN GUIDELINES


Guideline 3.46: Minimize Effect on Ambient Temperature
Medical devices should not unduly affect the ambient temperature of likely use areas.

Guideline 3.47: Consider Users’ Clothing Requirements


If the temperature in the use environment will necessitate unusual clothing requirements, the
implications for device use should be incorporated into design and evaluation activities.

Guideline 3.48: Device Use Wearing Gloves


Controls and handles should be designed to provide sufficient tactile feedback and control to
persons wearing protective gloves in cold weather conditions.

Guideline 3.49: Use in Inclement Weather


Medical devices intended for use outdoors should be usable in all possible weather conditions,
such as extreme heat or cold, extremely low or high humidity, high wind, blowing sand, rain
or snowfall, and so on.

Guideline 3.50: Protective Eyewear


Information on displays that might be viewed by users wearing protective eyewear (e.g., gog-
gles protecting the eyes from extreme cold) should be oversized to facilitate reading, particu-
larly when the eyewear is frosted, fogged, smudged, or scratched.

Guideline 3.51: Handles and Grips


Handles and equivalent gripping surfaces should continue to provide a secure grip when wet
and icy. This goal might be achieved by allowing a closed grip around a handle (with the
thumb crossing over the fingers), making it less likely to slip out of the hand.

3.5 CLIMATE: HUMIDITY


3.5.1 GENERAL PRINCIPLES
Humidity is the amount of moisture in the air. Without active intervention, indoor humid-
ity tracks the prevailing outdoor humidity. Outdoor humidity is usually lower during the
winter months and higher in the summer. In the summer, air-conditioning systems reduce
indoor humidity. Home care environments have wider fluctuations in humidity than clinical
care environments. Home humidifiers are used in conjunction with home heating units or as
stand-alone humidifiers. Hospitals and clinics no longer use such devices because they can
introduce potential infectious agents (e.g., fungi and bacteria) into the environment.
Low humidity can dry user’s mucous membranes, skin, and lips (Sanders and McCormick,
1993). Eye irritation occurs when the relative humidity is at or below 30%, particularly for

TAF-ER562X_WINGR-09-0401-C003.indd 82 11/8/10 11:41:13 PM


Environment of Use 83

those who wear contact lenses. This effect becomes pronounced after 4 hours (Rohles and
Konz, 1987).
Areas used for device cleaning and reprocessing are often warm and humid because of
instrument reprocessing equipment; the effects on cleaning personnel’s performance are
amplified by the required personal protective equipment (PPE) (i.e., gloves, goggles, and
gowns). In addition to their thermal burden, PPE can interfere with job performance and
jeopardize worker safety. Medical devices should be designed so that maintenance/repro-
cessing personnel can do their jobs safely and efficiently. An ASTM F23 Committee for
Protective Clothing Human Factors Subcommittee is developing new standards for mea-
suring the overall thermal comfort and breathability (i.e., ability to release moisture, per-
haps while also preventing its ingress) of protective clothing. The standard will also specify
consistent methods to evaluate the effectiveness of cooling devices used with protective
clothing.

3.5.2 DESIGN GUIDELINES


Guideline 3.52: Minimize Effect on Ambient Humidity
As humidity increases, discomfort will be felt at the higher end of the thermal comfort zone
(79°F/26°C). Device use should not increase the work area’s humidity above 55% for a person
performing light work (Chengalur et al., 2004). If at all possible, the device should not con-
tribute to ambient humidity. Humidity levels above the desired range can cause condensation,
damaged equipment, or microbial contamination.

Guideline 3.53: Device Storage in High Humidity


In areas of high humidity and/or temperature, attention should be given to device stor-
age requirements and degradation of device components, such as adhesives and chemical
reagents.

3.6 CLIMATE: AIRFLOW AND PRESSURE


3.6.1 GENERAL PRINCIPLES
Many medical devices incorporate fans to maintain component stability, evacuate gener-
ated heat, or dispose of toxic substances (e.g., smoke). Devices, especially large devices
such as whole-body imaging systems, can significantly influence not only temperature but
also airflow and pressure within patient care environments.
Designers of medical devices that could affect or be affected by the climate and air
quality within their use environments need to be familiar with existing health care facility
environmental standards and guidelines. The construction and environmental guidelines
of the American Institute of Architects (2001) and the standards of the American Society
of Heating, Refrigerating and Air-Conditioning Engineers (2003) provides minimum stan-
dards for design and construction of health care facility ventilation systems. The envi-
ronment of care standards of The Joint Commission (formerly the Joint Commission on
Accreditation of Healthcare Organizations [JCAHO]) require hospitals to install systems
that maintain appropriate pressure relationships, air exchange rates, and filtration efficien-
cies systems to control airborne contaminants such as biological agents, gases, fumes, and
dust (JCAHO, 2004).

TAF-ER562X_WINGR-09-0401-C003.indd 83 11/8/10 11:41:13 PM


84 Handbook of Human Factors in Medical Device Design

3.6.2 DESIGN GUIDELINES


Guideline 3.54: Device Effects on Air Handling Systems
Medical device designs should minimize all potential interactions between the device and the
air-handling systems in the expected environments of use.

Guideline 3.55: Device Exhaust Direction


Fans, such as those used to cool electrically powered components (see Figure 3.7), should not
blow directly on medical personnel and patients.

Guideline 3.56: Device Exhaust Filtration


If a device generates an exhaust, filtration of the exhaust should be incorporated into the
design.

Guideline 3.57: No Airborne Contaminants


Medical devices should not generate any airborne contaminants that exceed prescribed con-
tainment standards and guidelines.

Guideline 3.58: Minimal Effect on Negative-Pressure Rooms


The exhaust from medical devices used in isolation environments should not be so forceful
that it appreciably changes the direction of air flow in the room, which presumably is set to
remove contaminated air through a low pressure inlet.

Guideline 3.59: Use in High- and Low-Pressure Environments


Medical devices may be used at high altitude (i.e., subatmospheric) or at greater than atmos-
pheric pressures (i.e., decompression chambers or in submarines). Devices to be used in
decompression chambers should function correctly and be usable at pressures up to 6 atmos-
pheres (600 kilopascals [kPa], or ∼85 psi [lbf/in2]). Subatmospheric pressures are attained at
high-altitude medical facilities (e.g., Everest base camp medical clinic) and in medical aircraft

FIGURE 3.7 Ultrasound scanner’s cooling vents point toward the side and back, avoiding direct-
ing heated air toward the technician or patient.

TAF-ER562X_WINGR-09-0401-C003.indd 84 11/8/10 11:41:13 PM


Environment of Use 85

(cabin pressurized typically at 62 kPA) or spacecraft. For devices intended for use at very high
altitude, the reduced gravity during use must also be considered.

3.7 VIBRATION
3.7.1 GENERAL PRINCIPLES
Design guidelines for vibration include how to design for device use in a vibrating envi-
ronment and how to reduce the vibration generated by a device. This section covers only
the former (see also Chapter 17, “Mobile Medical Devices”), while Chapter 16, “Hand Tool
Design,” covers the latter. Common medical device use scenarios with vibration include
moving vehicles (ambulance and helicopter) and device use while moving (e.g., use of a patient
monitor on a gurney while transporting the patient from one place to another in a hospi-
tal). Nonmedical devices (e.g., heating, ventilating, and air-conditioning systems) can induce
vibration in otherwise stationary use environments if improperly designed or installed.
Exposure to vibration can cause pain, shortness of breath, anxiety, and changes in blood
pressure (Macnab et al., 1995). Vibration can impair visual or motor performance (Griffin,
1997; Sanders and McCormick, 1993; Wickens et al., 2004). In other words, both input
and output processes can be affected by vibration. Table 3.7 shows the perceived comfort
reported by healthy adults experiencing different levels of vibration.
Visual performance is impaired by vibration because the continual movement of the
image on the retina causes the image to appear blurry. Vibration can disrupt eye–hand
coordination in the absence of effective stabilization. Thus, control input devices, such as
touch screens, might be unreliable (Wickens et al., 2004). Additionally, medical device
display vibration can reduce the ability to see fine detail. Performance can be impaired
when using isometric controls (depending on the control gain and control order) (Griffin,
1997). Control performance in vibrating environments will be affected by control shape
and orientation, task difficulty, and the axis of vibration (Griffin, 1997).
Patient transport by ambulance, fixed-wing, or rotary-wing aircraft can produce sub-
stantial vibration (see Table 3.8). Vibration acceleration magnitude in transport vehicles
varies from 0.4 to 5.6 m/s2, depending on the axis and the type of vehicle (Campbell et al.,
1984). Rotary-wing aircraft (i.e., helicopters) tend to produce the greatest vibration, while

TABLE 3.7
Vibration Comfort Zones for Healthy Adults
Comfort Level Acceleration (m/s2)
Still comfortable <0.31
A little uncomfortable 0.31–0.63
Fairly uncomfortable 0.5–1.0
Uncomfortable 0.8–1.6
Very uncomfortable 1.25–2.5
Extremely uncomfortable >2.0

Source: Adapted from A. Macnab, Y. Chen, F. Gagnon, B. Bora, B., and


C. Laszlo, Aviation Space and Environmental Medicine, 66,
212–219, 1995. With permission.

TAF-ER562X_WINGR-09-0401-C003.indd 85 11/8/10 11:41:14 PM


86 Handbook of Human Factors in Medical Device Design

TABLE 3.8
Vibration Experienced in Flying Vehicles
Vehicle Vibration Acceleration (m/s2) Study
Cruising fixed-wing aircraft 0.04 (while cruising) Campbell et al. (1984)
Fixed-wing aircraft 0.25–0.4 (average) Macnab et al. (1995)
Fixed-wing aircraft 0.86 (maximum obtained) Macnab et al. (1995)
Hovercraft 0.5 (average) Macnab et al. (1995)
Helicopter 1 0.75 (average) Macnab et al. (1995)
Helicopter 2 1.3 (average) Macnab et al. (1995)
Rotary-wing aircraft 2.35 (maximum obtained) Macnab et al. (1995)
Rotary-wing aircraft 5.6 (while cruising) Campbell et al. (1984)

Source: Reproduced from A. Macnab, Y. Chen, F. Gagnon, B. Bora, B., and C. Laszlo, Aviation
Space and Environmental Medicine, 66, 212–219, 1995. With permission.
Measurements will vary depending on make and model of vehicle as well as many other factors.

fixed-wing aircraft produce less. In general, propeller-powered aircraft produce more vibra-
tion than jet-powered aircraft. Other factors affecting vibration include vehicle load, road
surface and tires (for ambulances), altitude, and weather.

3.7.2 DESIGN GUIDELINES


Guideline 3.60: Minimal Device Vibration
Medical devices should produce as little vibration as possible except when vibration provides
useful tactile feedback to the user or patient.

Guideline 3.61: Vibrating Controls


Controls subject to vibration due to vehicular movement, for example, should allow adjust-
ment by means of relatively gross motor hand and finger movements rather than fine ones.
Alternatively, they should be isolated from sources of vibration.

Guideline 3.62: Vibrating Displays


Information on displays subject to vibration should be oversized to facilitate reading (see
Figure 3.8). Information should also contrast sharply against its background to counterbal-
ance the effect of vibration. The space in between rows of displayed information should be
similarly enlarged (Griffin, 1997; Wickens et al., 2004).

Guideline 3.63: Vibrating Quantitative Display


Quantitative displays should distinguish between values that can be made reliably given the
expected level of vibration. The length of scale units should be greater than the 1.3 to 1.8 mm
recommended for use in nonvibrating environments.

Guideline 3.64: Usability Testing with Vibration


In the face of known or expected vibration, the usability and effectiveness of controls and
displays should be rigorously tested.

TAF-ER562X_WINGR-09-0401-C003.indd 86 11/8/10 11:41:14 PM


Environment of Use 87

Blood
Blood pressure
pressure

Blood pressure

Blood pressure
FIGURE 3.8 Text should be made larger to account for reductions in legibility due to vibration
(simulated in this graphic).

3.8 ENERGY SOURCES


A range of energy sources exists in environments where medical devices are used, includ-
ing ultrasound, x-ray, infrared, gamma, laser, and magnetic (e.g., MRI) radiation. Medical
devices can generate energy sources themselves. These energy sources may interact with each
other and affect the use of other medical devices (e.g., electromagnetic interference). Many
energy sources can affect user or patient health and safety, necessitating the use of personal
protective equipment. The presence of energy sources affects facility design (e.g., shielding)
as well as the design of other devices used in the same environment. Table 3.9 provides a list
of energy sources and examples of medical devices and implications for use and safety.
The increasing use of laser light in medicine has implications for the design of other
devices used in the same clinical environment. For example, because laser energy is an

TABLE 3.9
Energy Sources, Associated Medical Devices, and Sample Design Implications
Examples of Potential Use or
Type of Energy Source Examples of Medical Device Safety Implications for Design
Electromagnetic energy
Gamma Gamma knife Need for shielding and PPE
X-ray Fluoroscopy Need for shielding and PPE
CT scan
Infrared Infant warmer Heating of surroundings, burns
Radio frequency Electrocautery Interference
MRI Other items in suite cannot be
ferrormagnetic
Laser Surgical laser Eye damage or fires

Other sources of energy


Ultrasound Cardiac ultrasound Need for skin contact by probe—
infection and sterility issues
Audible sounda Voice-activated robot control Affect hearing alarms of other
devices
a See Section 3.3 on noise.

TAF-ER562X_WINGR-09-0401-C003.indd 87 11/8/10 11:41:14 PM


88 Handbook of Human Factors in Medical Device Design

ignition source, other devices used in proximity (e.g., endotracheal tubes and drapes) should
not be a ready fuel source that might lead to fire.

3.8.1 GENERAL PRINCIPLES


Medical devices should function in all anticipated or reasonably foreseeable use environ-
ments, including environments containing electromagnetic energy sources. Bruckart et al.
(1993) tested 34 medical devices under extreme conditions of temperature, humidity, alti-
tude, and vibration as well as their electromagnetic compatibility. Ninety-one percent of the
tested devices failed at least one of the electromagnetic interference tests, and many of the
devices exceeded the electromagnetic emissions standard established for use in U.S. Army
helicopters (U.S. Department of Defense, 1983).
Electromagnetic interference can come from a variety of sources, including radio trans-
mitters, aircraft, power lines and other power sources, computer timing and control circuits,
electrical switching transients (e.g., lights turned on or off), fans or motors, and electro-
static discharges (e.g., lightning). Between 1979 and 1995, the FDA received more than 100
reports of malfunctioning electronic medical devices presumably due to electromagnetic
interference (Silberberg, 1996). For example, electromagnetic interference often affects
medical device displays (e.g., the ECG tracings shown in Figure 3.9). Myriad regulations,
standards, and publications are available to guide the design of devices with improved elec-
tromagnetic compatibility.
Medical devices may themselves produce harmful radiation and therefore require
shielding and/or the wearing of protective equipment by users and patients. For instance,
linear accelerators and cobalt emitters used in radiotherapy produce highly toxic gamma
radiation.

3.8.2 DESIGN GUIDELINES


Guideline 3.65: Electromagnetic Compatibility Testing
Designers should test a medical device’s electromagnetic compatibility in the environment(s)
in which the device will be used.

Normal ECG tracing ECG tracing during FM transmission

FIGURE 3.9 Example of electromagnetic interference in an ECG monitor on board an operat-


ing U.S. Army Medevac helicopter. (From J. E. Bruckart, J. R. Licina, and M. Quattlebaum, Air
Medical Journal, 1, 51–56, 1993. With permission.)

TAF-ER562X_WINGR-09-0401-C003.indd 88 11/8/10 11:41:14 PM


Environment of Use 89

Guideline 3.66: Display Reliability in Radiant Energy Environments


Designs should minimize the display of erroneous or misleading information when a device
is used as intended in radiant energy environments (e.g., near MRI machines).

Guideline 3.67: Labeled as Excluded Use


As appropriate, devices should be clearly labeled as unsuitable for use in environments with
radiant energy sources (e.g., banning ferromagnetic materials adjacent to MRI devices). The
rationale for the exclusion should be provided.

Guideline 3.68: Shielding Provided


Adequate shielding and other means should be provided to protect users and patients from
harmful energy sources produced by a device.

Guideline 3.69: Protective Precautions


If shielding and other designed protective measures are insufficient, the device instructions
for use (including on-device labeling) should clearly specify all necessary protection precau-
tions to be taken by users and patients.

Guideline 3.70: Accommodate Use of Protective Equipment


When PPE or adjuvant protective equipment (e.g., vertical lead shield on wheels) is required,
the device should accommodate its use through design. For example, a portable x-ray machine
should be designed to allow users to operate the device from a safe distance. Bulky protective
clothing may limit users’ range of motion and hamper work movements. Device shielding or
the use of protective goggles may increase visual demands and affect visual performance.
Note that other devices in the use environment with different operational requirements may
also necessitate PPE use.

Guideline 3.71: Storage of Protective Equipment


When protective equipment is required for use with a device, the device should provide means
to store this equipment.

3.9 DESIGN FOR INFECTION CONTROL


Adverse outcomes or illness among patients or health care workers can occur from medi-
cal device use due to, for example, inadvertent exposures to environmental pathogens (e.g.,
endoscopes and respiratory therapy equipment), improper ventilation and distribution of
airborne pathogens (e.g., Mycobacterium tuberculosis and varicella-zoster virus), or toxic
chemical exposure. Thus, the ability to sterilize, reliably clean, and maintain the integrity
of medical devices is a critical design feature.

3.9.1 GENERAL PRINCIPLES


A key to infection control is to design medical devices that are easy to clean or steril-
ize. Sterile devices are those that must be inserted into the body, typically via invasive

TAF-ER562X_WINGR-09-0401-C003.indd 89 11/8/10 11:41:14 PM


90 Handbook of Human Factors in Medical Device Design

procedures. Surgical procedures are almost always performed under sterile conditions.
Thus, surgical devices must be easily used by care personnel wearing surgical gowns,
gloves, hair coverings, and protective eyewear.
While one type of medical device might be cleaned multiple times a day, another might
be cleaned only rarely. Sterile medical devices may be used only once (single use or dis-
posable device) or be reusable, in which case they must be resterilized after each use. For
example, a stretcher might need a scrub down with a bleach disinfectant after each use,
while a tonometer, used to test for glaucoma, might receive infrequent cleaning (except for
the part that touches the eye, which is cleaned after each use). Accordingly, the use environ-
ment and the device’s use profile establish cleaning requirements. In an attempt to reduce
cleaning efforts and difficulties, some devices might be outfitted with protective, dispos-
able covers. It is important for a user to be able to correctly and quickly ascertain whether
a device is clean (or sterile).
Devices should not be a potential source of infections. For example, a device should not
generate high humidity or have a moisture reservoir that can become contaminated and
serve as a nidus for infectious agents.
For PPE for infection control, medical devices should be accessible and usable when
the user is in full PPE (i.e., mask, isolation gown, gloves, and protective eyewear). For
example, a patient in isolation for respiratory tuberculosis, a wound infection, or a blood-
borne pathogen may require the health care provider to wear a respirator, impervious
gown, face shield, and gloves. This provider’s range of motion, tactile senses, and vision
will be affected by PPE use and this should be considered in device design. Additionally,
a powered air-purifying respirator may be used (e.g., during total joint replacement sur-
gery). These systems include helmets, hoods, head covers, and face pieces and generate
significant noise.

3.9.2 DESIGN GUIDELINES


Guideline 3.72: Obvious Evidence of Contamination
Medical devices that must be clean (though not necessarily sterile) should reveal contamina-
tion. For instance, a light-colored finish will show grime better than a dark one.

Guideline 3.73: Impervious to Fluid Contamination


As appropriate, medical devices should be impervious to fluids that might be wiped on, spilled
over, or projected toward them, such as cleaning solutions, urine from a detached catheter,
saline from a burst IV bag, or blood spurting from a lacerated artery.

Guideline 3.74: Device Cleanability


Medical devices intended for use in clean or sterile environments should be designed for easy,
thorough cleaning.

Guideline 3.75: Instructions for Cleaning


Users should be provided with clear instructions for device cleaning. Appropriate (and inap-
propriate) cleaners and disinfectants should be identified. Devices to be used in environ-
ments where particular cleaning methods are typically used should support the same cleaning
methods.

TAF-ER562X_WINGR-09-0401-C003.indd 90 11/8/10 11:41:14 PM


Environment of Use 91

Guideline 3.76: Hand Hygiene


Medical device use should not interfere with proper hand hygiene. If the device itself can
become contaminated during patient care and use, the user must be able to properly decon-
taminate his or her hands.

Guideline 3.77: Sterility Support


As appropriate, devices used in sterile environments should accommodate protective covers,
such as those placed over operating room light handles. Displays and handles should be posi-
tioned for ready visual and physical access underneath such covers.

Guideline 3.78: Usable Wearing Protective Equipment


Medical devices should be accessible and usable when the user is in full personal protec-
tive equipment (i.e., mask, isolation gown, gloves, and protective eyewear). For example, a
patient in isolation for respiratory tuberculosis, a wound infection, or a blood-borne pathogen
might require the health care provider to wear a respirator, imperious gown, face shield, and
gloves.

3.10 CASE STUDIES


3.10.1 CLEANING OF FLEXIBLE GASTROSCOPES
To date, all published episodes of pathogen transmission related to gastrointestinal (GI)
endoscopy have been associated either with a failure to follow established cleaning and
disinfection/sterilization guidelines or with the use of defective equipment (Nelson et al.,
2004). All flexible GI endoscopes, used for diagnosis and intervention procedures in the
human GI system, require initial decontamination, thorough cleaning, and final high-level
disinfection or sterilization. On removal from the patient, the endoscope is initially cleaned
externally with a manufacturer-approved detergent (Figure 3.10a) and then transported to
another area for further cleaning and reprocessing. All internal channels of the device must
be physically cleaned (Figure 3.10b) prior to terminal reprocessing. Maintainers must wear

(a) (b)

FIGURE 3.10 Initial cleaning of a flexible endoscope. (a) External cleaning. (b) Internal cleaning.
(Photos courtesy of Carla J. Alvarado.)

TAF-ER562X_WINGR-09-0401-C003.indd 91 11/8/10 11:41:14 PM


92 Handbook of Human Factors in Medical Device Design

gear that fully protects them from both potentially infective human material (PPE as per
OSHA’s Bloodborne Pathogens Standard) and toxic chemicals used during cleaning and
reprocessing. Flexible endoscopes are prone to damage and are extremely difficult to clean.
Cleaning involves accessing near microscopic, difficult-to-navigate 90-degree angles in the
device’s internal channels. The entire cleaning process must be efficient, given the need
for rapid instrument turnaround time in busy endoscopy clinics or operating rooms. This
may result in significant production pressure and the risk of rushed, incomplete cleaning.
Improved device design that facilitates the cleaning process would be a major improvement
in safety and efficiency.

3.10.2 PATIENT USE INSULIN INFUSION PUMP


Insulin infusion minipumps deliver precise amounts of rapid-acting insulin throughout the
day via a catheter into the diabetic patient’s body, usually via a needle inserted under the
skin. The pump is worn by patients and gives them the option to deliver extra doses when
needed. In principle, the technology gives people with diabetes better control over their dis-
ease, enabling them to lead longer and healthier lives. To deliver these benefits, the insulin
pump needs to small, lightweight, and easy to use (Figure 3.11).
In addition, these pumps must be designed for use in a wide range of use environments.
People who use the pump typically clip it onto their belt, a close distance from the abdomi-
nal injection site (Figure 3.12). A disposable set, which includes a tube, adhesive patch, and
catheter, conveys insulin from the pump to the user’s body. The whole system must work
properly in a wide range of climate conditions and be able to withstand considerable physi-
cal abuse. For example, a user might wear the pump outdoors on an extremely cold winter
morning to catch a bus to work. On boarding the bus, he or she might accidentally bang
the pump against a stanchion. The device’s rugged construction and protective plastic case
defend it against damage, while its internal workings are engineered for use in extremes of
temperature.

FIGURE 3.12 The patient wears the pump


FIGURE 3.11 Insulin infusion pump worn by on his belt and can carry on activities of daily
a diabetic patient. living without impediment.

TAF-ER562X_WINGR-09-0401-C003.indd 92 11/8/10 11:41:15 PM


Environment of Use 93

The pump is moisture sealed, preventing precipitation and other fluids from contami-
nating it and possibly leading to a malfunction. Specifically, the pump shown in this case
(Figure 3.11) has elastomeric push buttons mounted on top of mechanical switches to keep
moisture out while also providing good tactile feedback. Thus, users do not need to worry
about using the device in inclement weather, such as administering a bolus dose while
watching a child play soccer in the rain. Patients who want to keep their medical condition
and course of treatment private can do so. The device is as compact as a cellular phone,
making it easy to hide under an overshirt or jacket.
Accounting for use in dim lighting or darkness, the pump has a built-in display back-
light. A software-based control allows users to increase message tone volume in noisy
environments (e.g., bus station) and minimize the volume or switch to vibratory alerts in
quieter settings (e.g., library).

RESOURCES
Salvendy, G. (Ed.). (2006). Handbook of Human Factors and Ergonomics (3rd. ed.). New York:
John Wiley & Sons.
Sanders, M. S. and McCormick, E. J. (1993). Human Factors in Engineering and Design. New York:
McGraw-Hill.

REFERENCES
American Institute of Architects. (2001). Guidelines for Design and Construction of Hospital and
Health Care Facilities. Washington, DC: American Institute of Architects.
American Society of Heating, Refrigerating and Air-Conditioning Engineers. (2003). HVAC Design
Manual for Hospitals and Clinics. American Society of Heating, Refrigerating and Air-
Conditioning Engineers. Atlanta, GA: ASHRAE.
Bentley, S., Murphy, F., and Dudley, H. (1977). Perceived noise in surgical wards and an intensive
care area: An objective analysis. British Medical Journal, 2, 1503–1506.
Bayo, M. V., Garcia, A. M., and Garcia, A. (1995). Noise levels in an urban hospital and workers’
subjective responses. Archives of Environmental Health, 50(3), 247–251.
Boyce, P. R. (1997). Illumination. In G. Salvendy (Ed.), Handbook of Human Factors and Ergonomics
(pp. 858–890). New York: John Wiley & Sons.
Brown, L. H., Gough, J. E., Bryan-Berg, D. M., and Hunt, R. C. (1997). Assessment of breath
sounds during ambulance transport. Annals of Emergency Medicine, 29(2), 228–231.
Bruckart, J. E., Licina, J. R., and Quattlebaum, M. (1993). Laboratory and flight tests of medical
equipment for use in U.S. Army Medevac helicopters. Air Medical Journal, 1(3), 51–56.
Campbell, A. N., Lightstone, A. D., Smith, J. M., Kirpalani, H., and Perlman, M. (1984). Mechanical
vibration and sound levels experienced in neonatal transport. American Journal of Diseases of
Children, 138(10), 967–970.
Carayon, P. and Smith, M. J. (2000). Work organization and ergonomics. Applied Ergonomics, 31,
649–662.
Chang, B. J. (2002). Ergonomic benefits of surgical telescope systems: Selection guidelines. Journal
of the California Dental Association, 30(2), 161–169.
Chengalur, S. N., Rodgers, S. J., and Bernard, T. E. (2004). Kodak’s Ergonomic Design for People
at Work (2nd ed.). Hoboken, NJ: John Wiley & Sons.
Commission of the European Communities. (1996). European Guidelines on Quality Criteria for
Diagnostic Radiographic Images (Vol. EUR 16260). Brussels: Commission of the European
Communities.

TAF-ER562X_WINGR-09-0401-C003.indd 93 11/8/10 11:41:17 PM


94 Handbook of Human Factors in Medical Device Design

Commission of the European Communities. (1997). Criteria for Acceptability of Radiological


(Including Radiotherapy) and Nuclear Installations (Radiation Protection No. 91).
Luxembourg: Commission of the European Communities.
Couper, R. T., Hendy, K., Lloyd, N., Gray. N., Williams, S., Bates, D. J., et al. (1994). Traffic and
noise in children’s wards. Medical Journal of Australia, 160, 338–341.
Crocker, M. J. (1997). Noise. In G. Salvendy (Ed.), Handbook of Human Factors and Ergonomics
(2nd ed., pp. 790–827). New York: John Wiley & Sons.
Fanger, P. O. (1970). Thermal Comfort, Analysis and Applications in Environmental Engineering.
Copenhagen: Danish Technical Press.
Fife, D. and Rappaport, E. (1976). Noise and hospital stay. American Journal of Public Health,
66(7), 680–681.
Freedman, N. S., Gazendam, J., Levan, L., Pack, A. I., and Schwab, R. J. (2001). Abnormal sleep/
wake cycles and the effect of environmental noise on sleep disruption in the intensive care unit.
American Journal of Respiratory and Critical Care Medicine, 163(2), 451–457.
Fromm, R. E., Jr., Campbell, E., and Schlieter, P. (1995). Inadequacy of visual alarms in helicopter
air medical transport. Aviation Space and Environmental Medicine, 66(8), 784–786.
Gillis, W. L. and Miles, R. J. (2001). Orotrachial intubation in darkness using night vision goggles.
Military Medicine, 166, 984–986.
Griffin, M. J. (1997). Vibration and motion. In G. Salvendy (Ed.), Handbook of Human Factors and
Ergonomics (2nd ed., pp. 828–857). New York: John Wiley and Sons.
Hanna, G. B., Cresswell, A. B., and Cuschieri, A. (2002). Shadow depth cues and endoscopic task
performance. Archives of Surgery, 137, 1166–1169.
Illuminating Engineering Society of North America. (1981). IES Lighting Handbook. New York:
Illuminating Engineering Society of North America.
International Electrotechnical Commission. (2004). IEC 60601-1-6. Medical Electrical Equipment—
Part 1-6: General Requirements for Safety—Collateral Standard: Usability. Geneva:
International Electrotechnical Commission.
Joint Commission on Accreditation of Healthcare Organizations. (2004). Environment of Care
Essentials for Health Care (4th ed.). Oak Brook, IL: Joint Commission Resources.
Kahn, D. M., Cook, T. E., Carlisle, C. C., Nelson, D. L., Kramer, N. R., and Millman, R. P. (1998).
Identification and modification of environmental noise in an ICU setting. Chest, 114(2),
535–540.
Kam, P. C., Kam, A. C., and Thompson, J. F. (1994). Noise pollution in the anaesthetic and intensive
care environment. Anaesthesia, 49(11), 982–986.
Keipert, J. A. (1985). The harmful effects of noise in a children’s ward. Australian Paediatric
Journal, 21, 101–103.
Krupinski, E., Roehrig, H., and Furukawa, T. (1999). Influence of film and monitor display lumi-
nance on observer performance and visual search. Academic Radiology, 6(7), 411–418.
Macnab, A., Chen, Y., Gagnon, F., Bora, B., and Laszlo, C. (1995). Vibration and noise in pedi-
atric emergency transport vehicles: a potential cause of morbidity? Aviation Space and
Environmental Medicine, 66(3), 212–219.
McCarthy, E. and Brennan, P. C. (2003). Viewing conditions for diagnostic images in three major
Dublin hospitals: A comparison with WHO and CEC recommendations. British Journal of
Radiology, 76, 94–97.
McLaughlin, A., McLaughlin, B., Elliott, J., Campalani, G, et al. (1996). Noise levels in a cardiac
surgical intensive care unit: A preliminary study conducted in secret. Intensive Critical Care
Nursing, 12, 226–230.
Meyer, T. J., Eveloff, S. E., Bauer, M. S., Schwartz, W. A., Hill, N. S., Millman, R. P., et al. (1994). Adverse
environmental conditions in the respiratory and medical ICU settings. Chest, 105, 1211–1216.
Meyer-Falcke, A., Rack, R., Eichwede, F., Jansing, P. J., et al. (1994). How noisy are anaesthe-
sia and intensive care medicine? Quantification of the patients’ stress. European Journal of
Anaesthesiology, 11, 407–411.

TAF-ER562X_WINGR-09-0401-C003.indd 94 11/8/10 11:41:17 PM


Environment of Use 95

Morrison, W. E., Haas, E. C., Shaffner, D. H., Garrett, E. S., and Fackler, J. C. (2003). Noise, stress,
and annoyance in a pediatric intensive care unit. Critical Care Medicine, 31(1), 113–119.
Nelson, D. B., Jarvis, W. R., Rutala, W. A., Foxx-Orenstein, A. E., Isenberg, G., Dash, G. P., et al.
(2004). Multi-society guideline for reprocessing flexible gastrointestinal endoscopes. Diseases
of the Colon and Rectum, 47(4), 413–421.
Parsons, K. C. (1993). Human Thermal Environments. London: Taylor & Francis.
Parsons, K. C. (2000). Environmental ergonomics: A review of principles, methods and models.
Applied Ergonomics, 31, 581–594.
Philbin, M. K., Robertson, A., and Hall, J. W., III. (1999). Recommended permissible noise criteria
for occupied, newly constructed or renovated hospital nurseries. Journal of Perinatology, 19(8,
Pt. 1), 559–563.
Prasad, N. H., Brown, L. H., Ausband, S. C., Cooper-Spruill, O., Carroll, R. G., and Whitley, T.
W. (1994). Prehospital blood pressures: Inaccuracies caused by ambulance noise? American
Journal of Emergency Medicine, 12(6), 617–620.
Redding, J. S., Hargest, T. S., and Minksy, S.H. (1977). How noisy is intensive care? Critical Care
Medicine, 5, 275–276.
Rohles, F. H. and Konz, S. A. (1987). Climate. New York: John Wiley & Sons.
Sanders, M. S. and McCormick, E. J. (1993). Human Factors in Engineering and Design (7th ed.).
New York: McGraw-Hill.
Schwartz, R. B. and Charity, B. M. (2001). Use of night vision goggles and low-level light source
in obtaining intravenous access in tactical conditions of darkness. Military Medicine, 166,
982–983.
Shankar, N., Malhotra, K. L., Ahuja, S., and Tandon, O. P. (2001). Noise pollution: A study of
noise levels in the operation theatres of a general hospital during various surgical procedures.
Journal of the Indian Medical Association, 99(5), 244, 246–247.
Silberberg, J. L. (1996). What can/should we learn from reports of medical device electromagnetic
interference. Compliance Engineering, 13(4), 41–57.
Silberberg, J. L. (2001). Achieving medical device EMC: The role of regulations, standards, guide-
lines and publications. IEEE International Symposium on Electromagnetic Compatibility, 2,
1298–1303.
Sjors, G., Hammarlund, K., Oberg, P.-A., and Sedin, G. (1992). An evaluation of environment and
climate control in seven infant incubators. Biomedical Instrumentation and Technology, 26(4),
294–301.
Smith, M. J. and Carayon-Sainfort, P. (1989). A balance theory of job design for stress reduction.
International Journal of Industrial Ergonomics, 4, 67–79.
Sust, C. A. and Lazarus, H. (2003). Signal perception during performance of an activity under the
influence of noise. Noise and Health, 6(21), 51–62.
The Chartered Institution of Building Services Engineers. (1989). Lighting Guide—Hospitals
and Health Care Buildings (LG2: 1989 ed.). London: The Chartered Institution of Building
Services Engineers.
Tsiou, C., Eftymiatos, D., Theodossopoulou, E., Notis, P., and Kiriakou, K. (1998). Noise sources
and levels in the Evgenidion Hospital intensive care unit. Intensive Care Medicine, 24(8),
845–847.
U.S. Department of Defense. (1983). Military Standard: Environmental Test Methods and
Engineering Guidelines (MIL-STD-810). Washington, DC: U.S. Department of Defense.
Vincent, C., Taylor-Adams, S., and Stanhope, N. (1998). Framework for analysing risk and safety in
clinical medicine. British Medical Journal, 316(7138), 1154–1157.
Weinger, M. B. and Englund, C. E. (1990). Ergonomic and human factors affecting anesthetic vigi-
lance and monitoring performance in the operating room environment. Anesthesiology, 73(5),
995–1021.
Wickens, C. D., Lee, J. D., Liu, Y., and Becker, S. E. G. (2004). An Introduction to Human Factors
Engineering (2nd ed.). Upper Saddle River, NJ: Prentice Hall.

TAF-ER562X_WINGR-09-0401-C003.indd 95 11/8/10 11:41:17 PM


96 Handbook of Human Factors in Medical Device Design

Wiklund, M. E. (1995). Medical Device and Equipment Design—Usability Engineering and


Ergonomics. Buffalo Grove, IL: Interpharm Press.
World Health Organization. (1982). Quality Assurance in Diagnostic Radiology. Geneva: World
Health Organization.
World Health Organization. (1999). Guidelines for Community Noise. Retrieved June 1, 2004, from
http://www.who.int/docstore/peh/noise/guidelines2.html

TAF-ER562X_WINGR-09-0401-C003.indd 96 11/8/10 11:41:17 PM


4 Anthropometry
Biomechanics
and

W. Gary Allread, PhD, CPE; Edmond W. Israelski, PhD

CONTENTS
4.1 General Principles of Good Anthropometric Design ................................................99
4.1.1 One-Dimensional Measurements ..................................................................100
4.1.1.1 Hand Data........................................................................................102
4.1.1.2 Foot Data .........................................................................................107
4.1.1.3 Data Regarding Children.................................................................107
4.1.1.4 Additional Data Sources.................................................................. 110
4.1.2 Mobility and Functional Measurements ....................................................... 111
4.1.2.1 Mobility ........................................................................................... 111
4.1.2.2 Functional Work .............................................................................. 111
4.1.2.3 Visual Work..................................................................................... 111
4.1.3 Strength ......................................................................................................... 112
4.1.3.1 Strength and Gender Differences .................................................... 116
4.1.3.2 Strength of the Upper Extremity ..................................................... 116
4.1.3.3 Strength of the Fingers and Hands .................................................. 117
4.1.3.4 Strength of the Feet ......................................................................... 117
4.1.4 Special Considerations ..................................................................................121
4.1.5 Design Guidelines for Body Dimension Data...............................................122
4.1.5.1 Designing for Population Extremes .................................................122
4.1.5.2 Designing for the Average User.......................................................124
4.1.5.3 Designing for Adjustability .............................................................124
4.1.5.4 Deriving Missing Data ....................................................................125
4.2 Case Studies in Anthropometric Design..................................................................127
4.2.1 Viewing Angle Determination for a Cardiac Output Monitor ......................127
4.2.1.1 Analysis ...........................................................................................127
4.2.1.2 High-Mount Position .......................................................................127
4.2.1.3 Calculated Viewing Angles for the High-Mount Position ...............127
4.2.1.4 Low-Mount Position ........................................................................128
4.2.1.5 Calculated Viewing Angle for the Low-Mount Position .................128
4.2.1.6 Viewing Angles, Summary, and Implications .................................129
4.2.2 Keyboard Height in a Diagnostic System Workstation .................................129
4.2.3 Finger Clearance Space Calculations ...........................................................130
4.2.3.1 Prototype Blood Analyzer—Sample Probe Hand Clearance .......... 131

97

TAF-ER562X_WINGR-09-0401-C004.indd 97 11/9/10 2:23:06 AM


98 Handbook of Human Factors in Medical Device Design

4.2.3.2 Portable Infusion Pump Access for Thumb Bolus Activation ......... 131
4.2.3.3 Clinical Chemistry Diagnostic System Access Opening ................132
4.3 General Principles of Good Biomechanical Design ................................................132
4.3.1 Critical Design Considerations .....................................................................133
4.3.1.1 Body Posture ...................................................................................133
4.3.1.2 Trade-Offs .......................................................................................133
4.3.1.3 Endurance........................................................................................134
4.3.1.4 Repetitive Motions ..........................................................................135
4.3.2 Special Populations .......................................................................................135
4.3.3 Design Guidelines for Tasks Involving Lifting .............................................136
4.3.3.1 NIOSH Revised Lifting Equation ...................................................136
4.3.3.2 ACGIH Lifting Threshold Limit Values .........................................137
4.3.3.3 Industrial Lumbar Motion Monitor Risk Assessment System ........138
4.3.3.4 Psychophysical Limits .....................................................................139
4.3.4 Design Guidelines for Tasks Involving Use of the Upper Extremity ............139
4.3.4.1 Strain Index .....................................................................................139
4.3.4.2 Rapid Upper Limb Assessment .......................................................140
4.3.5 Design Guidelines to Determine Strength Requirements .............................140
4.3.5.1 Three-Dimensional Static Strength Prediction Program.................140
4.4 Case Studies in Biomechanical Design ................................................................... 141
4.4.1 Knob Twisting Forces on Pole Clamps ......................................................... 141
4.4.1.1 Analysis ...........................................................................................142
4.4.2 Bending Forces on an Autoinjector Device .................................................. 143
4.4.2.1 Analysis ........................................................................................... 143
4.4.3 Snap-On Lid Removal Forces .......................................................................144
4.4.3.1 Analysis ...........................................................................................144
4.4.4 Pulling Forces on IV Tubing.........................................................................146
4.4.4.1 Analysis ...........................................................................................146
4.4.5 Pulling Force to Remove a Plunger from a Vial ...........................................146
4.4.5.1 Analysis ........................................................................................... 147
Anthropometry-Related Resources ..................................................................................148
Biomechanics-Related Resources ....................................................................................149
References ........................................................................................................................150

The understanding of human physical capabilities and limitations is fundamental to the


design of effective medical devices. Properly designed medical tools, equipment, and work-
stations will help to reduce errors, decrease injury risk, increase productivity, and improve
user satisfaction.
To produce an effective medical device, one must have at least a basic understanding of
the many scientific disciplines that are involved in proper design—engineering, psychol-
ogy, anatomy, and physiology. Although this chapter does not provide details of all these
areas, it acquaints the reader with the essential information and reference materials that
designers commonly use to solve human-related design problems. The citations provided
here are not meant to be an all-inclusive listing but were carefully chosen sources on topics
that relate to many design concerns in the medical device realm.
The two design issues addressed in this chapter are anthropometry and biomechanics.
Anthropometry is the science of measuring and quantifying various human physical traits,

TAF-ER562X_WINGR-09-0401-C004.indd 98 11/9/10 2:23:06 AM


Anthropometry and Biomechanics 99

such as size, weight, proportion, mobility, and strength. Biomechanics is the use of laws of
physics and engineering principles to study various body segments as they move and are
acted on by internal and external forces.
Most of the anthropometric and biomechanical information available today is not spe-
cific to health care professionals and patients or to medical equipment. However, much of
the published body size data are just as useful to the medical device designer as they are
to any other designer who must combine human physical dimensions with workplace tasks
and equipment. Human biomechanical principles apply to the design process of a device
as well. Therefore, this chapter contains many data tables, figures, and other resources that
will facilitate the application of anthropometric and biomechanical data to the design of
medical devices.
This chapter presents a broad assortment of anthropometric data because of the inherent,
wide range of human physical dimensions that exists in any population. This is particularly
an issue in ethnically diverse workplaces, as individuals may differ drastically in their
physical characteristics. In the United States, variability in the physical sizes of workers
is expected to increase in the future. The U.S. Bureau of Labor Statistics projects that, in
the upcoming decade, the percentage of newly hired African American, Hispanic, and
Asian employees will far exceed new Caucasian workers. This affects the design of medical
devices because these groups can differ in many physical dimensions. The resulting vari-
ability in size, strength, and mobility among device users requires that anthropometric and
biomechanical factors be taken into account to ensure the proper design of medical devices,
accessories, and work environments.
This chapter includes data on body size distributions of males and females. This is
important, as there are many medical devices for which knowledge of user anthropomet-
rics is critical to design. Some examples include (1) the keyboard and monitor height for
laboratory diagnostic (IVD) workstations, (2) the hand span for manipulating a suture clos-
ing device, and (3) the gaze height for the shortest and tallest users of an overhead vital
signs monitor. This chapter also contains reference information for tools used to assess the
biomechanical injury risk of work performed in medical environments.
Anthropometric and biomechanical design principles will assist the medical device
designer in producing tools, equipment, and workstations that will accommodate most of
the individuals for whom they were intended. These principles also will improve safety
and appropriate use as well as reduce the risk of musculoskeletal strain. However, they do
not guarantee that the device will fit every user, and they do not guarantee that injuries or
misuse will be eliminated among those interacting with these devices.
Finally, it is impossible to include in this chapter information about every possible user
group and work situation. Further, there are numerous types of physical limitations that
can arise from a multitude of disabilities. However, this chapter directs the medical device
designer to information not included here and illustrates how to determine needed data
from existing sources.

4.1 GENERAL PRINCIPLES OF GOOD ANTHROPOMETRIC DESIGN


In most cases, the design of medical tools, equipment, and workstations should accommo-
date adults ranging in size from a 5th-percentile female to a 95th-percentile male. A 5th-
percentile value for any particular body dimension indicates that 5% of the population will
be equal to or smaller than that value and that the sizes of the remaining 95% will be larger.

TAF-ER562X_WINGR-09-0401-C004.indd 99 11/9/10 2:23:06 AM


100 Handbook of Human Factors in Medical Device Design

Conversely, a 95th-percentile value indicates that 95% of the population will be equal to or
smaller than that value on a particular dimension and that 5% will be larger. There is con-
siderable internal variation among human body sizes. In other words, an individual who is
measured at the 95th percentile on one particular body dimension likely will not fall in this
same percentile for other dimensions. In addition, the relationship between two dimensions
may vary as a function of gender. That is, body segment proportions for males can differ
considerably from the proportions of the same segments in females. Medical device design-
ers should take these facts into account.
Designing a medical device in a range that will accommodate users from the 5th-percentile
female to the 95th-percentile male will theoretically provide coverage for 90% of the user
population for that dimension. While designers should accommodate the widest possible
range of users, higher costs and infeasibility often prevent the designer from fully accom-
modating the entire user population. Certain medical equipment devices and tasks involve
critical functions, when it is necessary to accommodate more than the central 90% of
potential users. In these instances, data representing larger ranges of the population could
be used, such as the first to 99th percentiles. However, there often are trade-offs involved
in designing to these ranges. Cost and functionality must be considered when attempting
to accommodate larger percentages of any given population. For example, designing an
adjustable armrest for the 5th through 95th percentile of U.S. women, based on their seated
elbow resting height, requires a range of 10.2 cm. However, this range increases nearly
50% (to 15.0 cm) when accommodating the 1st through 99th percentile of this popula-
tion. This simple example shows that the inclusion of a greater segment of the population
for a device’s design on any particular dimension may not be feasible or may significantly
increase material cost.
In summary, good anthropometric design of medical devices should do the following:

• At a minimum, accommodate adults who range from a 5th-percentile female to a


95th-percentile male in size
• Accommodate larger percentages of the population if (1) the device involves criti-
cal functions, (2) device usability and functionality are not compromised, or (3)
there are not excessive costs involved in doing so

4.1.1 ONE-DIMENSIONAL MEASUREMENTS


Individuals from ethnically diverse populations often operate the same medical devices.
As a result, designers must understand the size ranges of these users. This chapter pres-
ents a representative sample of available data. Table 4.1 defines body segment terms, and
Table 4.2 lists the measurements. These are based on values for a 40-year-old American
male and a 40-year-old Asian female projected to the year 2000, which is the most
recent international data available (National Aeronautics and Space Administration,
1989). Figure 4.1 graphically illustrates these dimensions. Device designers should use
Tables 4.1 and 4.2 and Figure 4.1 as a starting point from which to assess the adequacy
of their equipment designs.
Historically, most anthropometric data have come from military populations, and
some medical device designers may question the relevancy of this information. However,
Kroemer, Kroemer, and Kroemer-Elbert (1997, p. 17) reported that “with proper caution
and insight, we can use military anthropometric data to approximate size data for the general

TAF-ER562X_WINGR-09-0401-C004.indd 100 11/9/10 2:23:06 AM


Anthropometry and Biomechanics 101

TABLE 4.1
Descriptions of Male and Female Anthropometric Measurements
Ref. No.a Dimension Definition
64 Ankle height The height of the level of minimum circumference of the leg
100 Biacromial breadth The distance across the shoulders from the right to the left acromion
178 Buttock The circumference of the body measured at the level of the maximum
circumference posterior protuberance of the buttocks
194 Buttock-knee length The horizontal distance from the rearmost surface of the buttocks to the
front of the kneecaps
200 Buttock-popliteal The horizontal distance from the rearmost surface of the buttock to the
length back of the lower leg
215 Calf height The height of the level of the maximum circumference of the lower leg
236 Chest depth (male) The depth of the torso measured at nipple level
249 Crotch height The height of the midpoint of the crotch
308 Elbow height The height of the radiale
312 Elbow rest height The height of the bottom of the tip of the elbow above the sitting surface
330 Eye height, sitting The height of the inner corner of the eye above the sitting surface
378 Forearm-forearm The distance across the tissue mass of the forearms measured with the
breadth elbows flexed and resting lightly against the body
381 Forearm-hand length The distance from the tip of the elbow to the tip of the longest finger
411 Hand breadth The breadth of the hand as measured across the distal ends of the
metacarpal bones
416 Hand circumference The circumference of the hand measured around its knuckles
420 Hand length The distance from the base of the hand to the top of the middle finger
measured along the long axis of the hand
459 Hip breadth, sitting The breadth of the body as measured across the widest portion of the hips
506 Interscye The tape distance across the back between the posterior axillary folds at
the lower level of the armpits
529 Knee height, sitting The height, from the footrest surface, of the musculature just above the knee
612 Mid-shoulder The height of the point on the shoulder halfway between the neck and the
height, sitting acromion above the sitting surface
639 Neck circumference The maximum circumference of the neck, including the Adam’s apple (in males)
678 Popliteal height The height of the underside of the upper leg above the footrest surface
735 Scye circumference The circumference of the scye, measured in a vertical plane, as high as
possible in the armpit and passing over the acromion
754 Shoulder length The surface distance from the acromion, at the end of the shoulder blade,
to the junction of the shoulder and the neck
758 Sitting height The height, from the sitting surface, to the top of the head
761 Shoulder-elbow The vertical distance from the acromion to the bottom of the elbow,
length measured with the elbow bent 90 degrees and the lower arm held horizontal
805 Stature The height of the top of the head
856 Thigh clearance The height of the highest point of the thigh above the sitting surface
916 Vertical trunk The circumference of the torso measured with the tape passing diagonally
circumference across the front of the body from the midpoint of the shoulder to the
crotch, through the crotch, over the posterior protuberance of the buttock
and along the small of the back

continued

TAF-ER562X_WINGR-09-0401-C004.indd 101 11/9/10 2:23:06 AM


102 Handbook of Human Factors in Medical Device Design

TABLE 4.1 (CONTINUED)


Descriptions of Male and Female Anthropometric Measurements
Ref. No.a Dimension Definition
921 Waist back The surface distance, along the spine, from waist level to the cervicale (the
top of the seventh cervical vertebra)
946 Waist front The surface distance in the midsagittal plane from waist level to the
suprasternale
949 Waist height The height of waist level
973 Wrist height The height of the stylion

Source: National Aeronautics and Space Administration, Man-Systems Integration Standards. NASA-STD-
3000A, Lyndon B. Johnson Space Center, Houston, TX, 1989. With permission.
a Reference numbers refer to numbers shown in Figure 4.1, Table 4.2, and in the NASA standard document cited.

population. Dimensions of the head, hand, and foot are virtually the same in military and
civilian populations.” Therefore, the data referred to in this chapter are arguably the best
available. However, designers should learn about any unique physical characteristics of
their medical device’s user population (e.g., elderly individuals, pregnant women) that may
make it different from a military population so as to achieve a proper integration of these
factors and a suitable design.
A recent project has established a more comprehensive database of the civilian U.S.
and European population: the Civilian American and European Surface Anthropometry
Resource (CAESAR™) Project (Robinette, 2000). These anthropometric data are used in
computer-aided design software applications to model human movement. The CAESAR™
data, from a sample of thousands of individuals, can be filtered by age (18–65 years), gen-
der, weight, and ethnic group. These data are available for purchase from the Society of
Automotive Engineers (Warrendale, PA). Data samples can be obtained from the U.S. Air
Force’s Computerized Anthropometric Research and Design Laboratory (Wright-Patterson
Air Force Base, OH).
The choice of anthropometric database for use by medical device designers should do
the following:
• Be specific to the population of users of that device if such a database exists
• Use civilian-based information when possible
• Consider use of existing military data if it can be assumed that the dimensions of
interest are not different between military and civilian populations

Medical device designers often need even more specific anthropometric data. Obtaining
these data may require measurements of specific body parts (such as the hands or feet), per-
haps in specialized populations (such as children). Some of this information is presented
in the following sections.

4.1.1.1 Hand Data


Many medical devices are used specifically with the hands, making knowledge of measure-
ment ranges for this body part critical. Devices that require hand dimension data would
include any surgical instruments that are manipulated and need to accommodate specific

TAF-ER562X_WINGR-09-0401-C004.indd 102 11/9/10 2:23:07 AM


TABLE 4.2
Male and Female Anthropometric Measurements, in cm (in.)
40-Year-Old American Male, 40-Year-Old Japanese Female,
Projected for Year 2000 Projected for Year 2000
Percentile Percentile

TAF-ER562X_WINGR-09-0401-C004.indd 103
Ref. No. Dimension 5th 50th 95th 5th 50th 95th
64 Ankle height 12.0 (4.7) 13.9 (5.5) 15.8 (6.2) 5.2 (2.0) 6.1 (2.4) 7.0 (2.8)
100 Biacromial breadth 37.9 (14.9) 41.1 (16.2) 44.3 (17.5) 32.4 (12.8) 35.7 (14.1) 39.0 (15.4)
178 Buttock circumference 91.0 (35.8) 100.2 (39.4) 109.4 (43.1) 79.9 (31.5) 87.1 (34.3) 94.3 (37.1)
Anthropometry and Biomechanics

194 Buttock-knee length 56.8 (22.4) 61.3 (24.1) 65.8 (25.9) 48.9 (19.2) 53.3 (21.0) 57.8 (22.7)
200 Buttock-popliteal length 46.9 (18.5) 51.2 (20.2) 55.5 (21.9) 37.9 (14.9) 41.7 (16.4) 45.5 (17.9)
215 Calf height 32.5 (12.8) 36.2 (14.3) 40.0 (15.7) 25.5 (10.0) 28.9 (11.4) 32.3 (12.7)
236 Chest depth (male) 21.8 (8.6) 25.0 (9.8) 28.2 (11.1) na na na
249 Crotch height 79.4 (31.3) 86.4 (34.0) 93.3 (36.7) 65.2 (25.7) 70.6 (27.8) 76.1 (30.0)
308 Elbow height na na na 92.8 (36.5) 98.4 (38.8) 104.1 (41.0)
312 Elbow rest height 21.1 (8.3) 25.4 (10.0) 29.7 (11.7) 20.7 (8.2) 25.0 (9.9) 29.3 (11.5)
330 Eye height, sitting 76.8 (30.3) 81.9 (32.2) 86.9 (34.2) 68.1 (26.8) 73.8 (29.1) 79.6 (31.4)
378 Forearm-forearm breadth 48.8 (19.2) 55.1 (21.7) 61.5 (24.2) na na na
381 Forearm-hand length na na na 37.3 (14.7) 41.7 (16.4) 44.6 (17.6)
411 Hand breadth 8.2 (3.2) 8.9 (3.5) 9.6 (3.8) 6.9 (2.7) 7.8 (3.1) 8.6 (3.4)
416 Hand circumference 20.3 (8.0) 21.8 (8.6) 23.4 (9.2) 16.5 (6.5) 17.9 (7.0) 19.3 (7.6)
420 Hand length 17.9 (7.0) 19.3 (7.6) 20.6 (8.1) 15.8 (6.2) 17.2 (6.8) 18.7 (7.3)
459 Hip breadth, sitting 34.6 (13.6) 38.4 (15.1) 42.3 (16.6) 30.4 (12.0) 33.7 (13.3) 37.0 (14.6)
506 Interscye 32.9 (13.0) 39.2 (15.4) 45.4 (17.9) 32.4 (12.8) 35.7 (14.1) 39.0 (15.4)
529 Knee height, sitting 52.6 (20.7) 56.7 (22.3) 60.9 (24.0) 41.6 (16.4) 45.6 (17.9) 49.5 (19.5)
612 Midshoulder height, sitting 60.8 (23.9) 65.4 (25.7) 70.0 (27.5) na na na
639 Neck circumference 35.5 (14.0) 38.7 (15.2) 41.9 (16.5) 34.5 (13.6) 37.1 (14.6) 39.7 (15.6)
678 Popliteal height 40.6 (16.0) 44.4 (17.5) 48.1 (19.0) 34.7 (13.6) 38.3 (15.1) 41.9 (16.5)
735 Scye circumference 44.4 (17.5) 49.0 (19.3) 53.6 (21.1) 32.3 (12.7) 36.1 (14.2) 39.8 (15.7)
103

continued

11/9/10 2:23:07 AM
104

TABLE 4.2 (CONTINUED)


Male and Female Anthropometric Measurements, in cm (in.)
40-Year-Old American Male, 40-Year-Old Japanese Female,
Projected for Year 2000 Projected for Year 2000
Percentile Percentile

TAF-ER562X_WINGR-09-0401-C004.indd 104
Ref. No. Dimension 5th 50th 95th 5th 50th 95th
754 Shoulder length 14.8 (5.8) 16.9 (6.7) 19.0 (7.5) 11.3 (4.4) 13.1 (5.1) 14.8 (5.8)
758 Sitting height 88.9 (35.0) 94.2 (37.1) 99.5 (39.2) 78.3 (30.8) 84.8 (33.4) 91.2 (35.9)
761 Shoulder-elbow length 33.7 (13.3) 36.6 (14.4) 39.4 (15.5) 27.2 (10.7) 29.8 (11.7) 32.4 (12.8)
805 Stature 169.7 (66.8) 179.9 (70.8) 190.1 (74.8) 148.9 (58.6) 157.0 (61.8) 165.1 (65.0)
856 Thigh clearance 14.5 (5.7) 16.8 (6.6) 19.1 (7.5) 11.2 (4.4) 12.9 (5.1) 14.5 (5.7)
916 Vertical trunk circumference 158.7 (62.5) 170.7 (67.2) 182.6 (71.9) 136.9 (53.9) 146.0 (57.5) 155.2 (61.1)

921 Waist back 43.7 (17.2) 47.6 (18.8) 51.6 (20.3) 35.2 (13.9) 38.1 (15.0) 41.0 (16.1)
946 Waist front 37.2 (14.6) 40.9 (16.1) 44.6 (17.5) na na na
949 Waist height 100.4 (39.5) 108.3 (42.6) 116.2 (45.7) 90.1 (35.5) 96.7 (38.1) 103.4 (40.7)
973 Wrist height na na na 70.8 (27.9) 76.6 (30.2) 82.4 (32.4)

Source: National Aeronautics and Space Administration, Man-Systems Integration Standards. NASA-STD-3000A, Lyndon B. Johnson Space Center, Houston, TX, 1989.
With permission.
na, not available.
Handbook of Human Factors in Medical Device Design

11/9/10 2:23:07 AM
Anthropometry and Biomechanics 105

(a) 378
805
754 639

236 506 612


921
308
916

973 459

64

(b) 758
411
330
420

761

416

529
194

678

(c)

735 100
946

178 312 381

856
949
200

249
215

FIGURE 4.1 Graphical illustrations of the body dimension data provided in Table 4.2. The refer-
ence numbers shown in these drawings relate to those listed in the first column of Table 4.2 (From
National Aeronautics and Space Administration, Man-Systems Integration Standards, NASA-STD-
3000A, Lyndon B. Johnson Space Center, Houston, TX, 1989. With permission.)

ranges of hand dimensions, such as hand span. For example, an over-the-wire catheter con-
trol device needs to have a functional hand spread no larger than a 5th-percentile female’s
dimension, so that a majority of intended users can operate the device properly. Table 4.3
presents (and Figure 4.2 illustrates) anthropometric hand data for both males and females.
Pheasant (1996) compiled this information from a number of civilian and military sources

TAF-ER562X_WINGR-09-0401-C004.indd 105 11/9/10 2:23:07 AM


106 Handbook of Human Factors in Medical Device Design

TABLE 4.3
Male and Female Anthropometric Hand Measurements, in cm (in.)
Male Female

Ref. Percentile Percentile


No.a Dimension 5th 50th 95th 5th 50th 95th
1 Hand length 17.30 (6.81) 18.90 (7.44) 20.50 (8.07) 15.90 (6.26) 17.40 (6.85) 18.90 (7.44)
2 Palm length 9.80 (3.86) 10.70 (4.21) 11.60 (4.57) 8.90 (3.50) 9.70 (3.82) 10.50 (4.13)
3 Thumb length 4.40 (1.73) 5.10 (2.01) 5.80 (2.28) 4.00 (1.57) 4.70 (1.85) 5.30 (2.09)
4 Index finger length 6.40 (2.52) 7.20 (2.83) 7.90 (3.11) 6.00 (2.36) 6.70 (2.64) 7.40 (2.91)
5 Middle finger length 7.60 (2.99) 8.30 (3.27) 9.00 (3.54) 6.90 (2.72) 7.70 (3.03) 8.40 (3.31)
6 Ring finger length 6.50 (2.56) 7.20 (2.83) 8.00 (3.15) 5.90 (2.32) 6.60 (2.60) 7.30 (2.87)
7 Little finger length 4.80 (1.89) 5.50 (2.17) 6.30 (2.48) 4.30 (1.69) 5.00 (1.97) 5.70 (2.24)
8 Thumb breadth (at 2.00 (0.79) 2.30 (0.91) 2.60 (1.02) 1.70 (0.67) 1.90 (0.75) 2.10 (0.83)
interphalangeal joint)
9 Thumb thickness (at 1.90 (0.75) 2.20 (0.87) 2.40 (0.94) 1.50 (0.59) 1.80 (0.71) 2.00 (0.79)
interphalangeal joint)
10 Index finger breadth 1.90 (0.75) 2.10 (0.83) 2.30 (0.91) 1.60 (0.63) 1.80 (0.71) 2.00 (0.79)
(at the proximal
interphalangeal joint)
11 Index finger thickness 1.70 (0.67) 1.90 (0.75) 2.10 (0.83) 1.40 (0.55) 1.60 (0.63) 1.80 (0.71)
(at the proximal
interphalangeal joint)
12 Hand breadth 7.80 (3.07) 8.70 (3.43) 9.50 (3.74) 6.90 (2.72) 7.60 (2.99) 8.30 (3.27)
(metacarpal)
13 Hand breadth (across 9.70 (3.82) 10.50 (4.13) 11.40 (4.49) 8.40 (3.31) 9.20 (3.62) 9.90 (3.90)
thumb)
15 Hand thickness 2.70 (1.06) 3.30 (1.30) 3.80 (1.50) 2.40 (0.94) 2.80 (1.10) 3.30 (1.30)
(metacarpal)
16 Hand thickness 4.40 (1.73) 5.10 (2.01) 5.80 (2.28) 4.00 (1.57) 4.50 (1.77) 5.00 (1.97)
(including thumb)
17 Maximum grip 4.50 (1.77) 5.20 (2.05) 5.90 (2.32) 4.30 (1.69) 4.80 (1.89) 5.30 (2.09)
diameter (thumb and
middle fingers just
touching)
18 Maximum spread (tip 17.80 (7.01) 20.60 (8.11) 23.40 (9.21) 16.50 (6.50) 19.00 (7.48) 21.50 (8.46)
of thumb to tip of
fifth finger)
19 Maximum functional 12.20 (4.80) 14.20 (5.59) 16.20 (6.38) 10.90 (4.29) 12.70 (5.00) 14.50 (5.71)
spread (tip end
segments of the thumb
and ring fingers)
20 Minimum square 5.60 (2.20) 6.60 (2.60) 7.60 (2.99) 5.00 (1.97) 5.80 (2.28) 6.70 (2.64)
access

Source: Pheasant, S., Bodyspace: Anthropometry, Ergonomics and the Design of Work, Taylor & Francis Ltd.,
London, 1996.
a Reference numbers refer to numbers shown in Figure 4.2 and in Pheastant’s book.

TAF-ER562X_WINGR-09-0401-C004.indd 106 11/9/10 2:23:08 AM


Anthropometry and Biomechanics 107

6 5 4 17
7 10

1
12

2 3 9
13

18

11 15
16 19
8
20

FIGURE 4.2 Graphic illustrations of hand dimension data. The reference numbers shown in these
drawings relate to those listed in the first column of Table 4.3 (From Pheasant, S., Bodyspace:
Anthropometry, ergonomics and the design of work, Taylor & Francis Ltd., London, 1996.)

for use as a preliminary reference during tool and equipment design in which interaction
with the hands is a primary requirement.

4.1.1.2 Foot Data


Individuals use their feet to activate many types of medical devices. Table 4.4 provides
anthropometric foot data for two relevant dimensions (illustrated in Figure 4.3) that were
compiled from several sources for various national populations (Flugel, Greil, and Sommer,
1986; Gordon et al., 1989; Kagimoto, 1990; Pheasant, 1996). These data should be used
as an initial source for designing foot-activated controls, workspace clearances, or other
devices where knowledge of foot size ranges is important.
Examples of medical devices that use the feet are activator switches for electrocauteriza-
tion surgical instruments (that essentially burn human tissue to stop bleeding) and fluor-
oscopes (e.g., to image dye-stained blood as part of an angiogram to determine extent of
coronary artery blockage). One important piece of data from Table 4.4 would be foot length
of a 5th-percentile operator of these activation switches to be sure the pedal is not too long
to be safely operated by someone with a small foot. Table 4.4 shows the 5th-percentile
female dimension to be 21.5 cm.

4.1.1.3 Data Regarding Children


Children also use a wide range of medical devices, such as handheld glucose meters, ambu-
latory enteral feeding pumps, external insulin pumps, and mobility aids, such as wheel-
chairs, crutches, and canes. Thus, designing devices for this population requires knowledge
of their unique physical measurements. Anthropometric data for those younger than age 18
are less detailed than for adult populations. Table 4.5 presents some of the published data for
children (Diffrient, Tilley, and Bardagjy, 1981), with reference measurements graphically

TAF-ER562X_WINGR-09-0401-C004.indd 107 11/9/10 2:23:08 AM


108 Handbook of Human Factors in Medical Device Design

TABLE 4.4
Male and Female Anthropometric Foot Measurements, in cm (in.)
Male Female

Ref. Percentile Percentile


No.a Dimension 5th 50th 95th 5th 50th 95th
1 Foot length (the maximal
length of the right foot,
when standing)
U.S. adults, aged 19–60 24.9 (9.8) 27.0 (10.6) 29.2 (11.5) 22.4 (8.8) 24.4 (9.6) 26.5 (10.4)
British adults, aged 19–35 24.0 (9.5) 26.5 (10.4) 28.5 (11.2) 21.5 (8.5) 23.5 (9.3) 25.5 (10.0)
East German adults, aged 18–59 24.3 (9.6) 26.4 (10.4) 28.5 (11.2) 22.2 (8.7) 24.1 (9.5) 26.0 (10.2)
Japanese adults, aged 18–30 23.4 (9.2) 25.1 (9.9) 26.9 (10.6) 21.7 (8.5) 23.2 (9.1) 24.6 (9.7)
2 Foot breadth (the maximal
breadth of the right foot, at
right angle to the long axis of
the foot, when standing)
U.S. adults, aged 19–60 9.2 (3.6) 10.1 (4.0) 11.0 (4.3) 8.2 (3.2) 9.0 (3.5) 9.8 (3.9)
British adults, aged 19–35 8.5 (3.4) 9.5 (3.7) 11.0 (4.3) 8.0 (3.2) 9.0 (3.5) 10.0 (3.9)
East German adults, aged 18–59 9.1 (3.6) 10.2 (4.0) 11.3 (4.5) 8.3 (3.3) 9.3 (3.7) 10.4 (4.1)
Japanese adults, aged 18–30 9.7 (3.8) 10.4 (4.1) 11.1 (4.4) 8.9 (3.5) 9.6 (3.8) 10.3 (4.1)

Sources: Flugel, F., Greil, H., and Sommer, KI. Anthropologischer Atlas, Tribuene, Berlin, 1986; Gordon, C.C.
et al., 1988 Anthropometric Survey of U.S. Army Personnel: Summary Statistics Interim Report, Natick-TR-
89/027. U.S. Army Natick Research, Development and Engineering Center, Natick, MA, 1989; Kagimoto,
Y., Anthropometry of JASDF Personnel and its Applications for Human Engineering, Aeromedical
Laboratory, Air Development and Test Wing, JASDF, Tokyo, 1990; Pheasant, S., Bodyspace: Anthropometry,
Ergonomics and the Design of Work. Taylor & Francis Ltd., London, 1996. With permission.
a Reference numbers refer to foot dimensions shown in Figure 4.3.

FIGURE 4.3 Graphic illustrations of foot dimension data. The reference numbers shown in these
drawings relate to those listed in the first column of Table 4.4.

TAF-ER562X_WINGR-09-0401-C004.indd 108 11/9/10 2:23:09 AM


TABLE 4.5
Child Anthropometric Data, from Birth to Age 4, in cm (in.), and Weight, in kg (lbs.)

TAF-ER562X_WINGR-09-0401-C004.indd 109
Age (years)
Dimension Birth ½ 1 1½ 2 2½ 3 4
Body length 50.5 (19.9) 66.0 (26.0) 74.9 (29.5) 80.8 (31.8) 86.4 (34.0) 91.2 (35.9) 95.3 (37.5) 103.9 (40.9)
Head length 12.7 (5.0) 15.2 (6.0) 17.5 (6.9) 18.3 (7.2) 19.1 (7.5) 19.3 (7.6) 19.6 (7.7) 19.8 (7.8)
Anthropometry and Biomechanics

Head width 9.7 (3.8) na 13.2 (5.2) na 14.0 (5.5) na 14.2 (5.6) 14.5 (5.7)
Head circumference 35.3 (13.9) 43.9 (17.3) 47.2 (18.6) 49.5 (19.5) 49.8 (19.6) 50.0 (19.7) 50.3 (19.8) 50.5 (19.9)
Neck length na 0.8 (0.3) 1.3 (0.5) 1.8 (0.7) 2.5 (1.0) 2.5 (1.0) 3.0 (1.2) 3.3 (1.3)
Trunk length 21.1 (8.3) 29.5 (11.6) 32.0 (12.6) 33.8 (13.3) 34.5 (13.6) 36.1 (14.2) 36.3 (14.3) 38.1 (15.0)
Shoulder width 15.2 (6.0) 17.8 (7.0) 20.3 (8.0) 21.1 (8.3) 22.4 (8.8) 22.9 (9.0) 23.6 (9.3) 24.6 (9.7)
Chest circumference 33.3 (13.1) 43.7 (17.2) 47.5 (18.7) 49.5 (19.5) 50.8 (20.0) 51.6 (20.3) 52.1 (20.5) 52.8 (20.8)
Abdominal circumference na 41.1 (16.2) 44.5 (17.5) 45.7 (18.0) 46.2 (18.2) 46.7 (18.4) 47.0 (18.5) 51.6 (20.3)
Pelvic width 8.1 (3.2) 11.7 (4.6) 13.0 (5.1) 13.7 (5.4) 14.5 (5.7) 15.2 (6.0) 15.7 (6.2) 17.5 (6.9)
Arm length 19.3 (7.6) 25.4 (10.0) 30.5 (12.0) 34.0 (13.4) 37.1 (14.6) 39.4 (15.5) 41.7 (16.4) 42.4 (16.7)
Upper arm circumference na 13.5 (5.3) 15.2 (6.0) na 16.0 (6.3) na 16.3 (6.4) 16.8 (6.6)
Hand length na na 9.7 (3.8) na 10.7 (4.2) na 11.9 (4.7) 12.4 (4.9)
Leg and thigh length 16.8 (6.6) 20.8 (8.2) 24.4 (9.6) 27.9 (11.0) 31.2 (12.3) 34.3 (13.5) 37.1 (14.6) 43.7 (17.2)
Knee width 3.8 (1.5) na 6.4 (2.5) na 6.6 (2.6) na 6.9 (2.7) 6.9 (2.7)
Weight 3.4 (7.5) 7.6 (16.7) 10.1 (22.2) 11.4 (25.2) 12.6 (27.7) 13.6 (30.0) 14.6 (32.2) 17.2 (38.0)

Source: Diffrient N., Tilley, A.R., and Bardagjy, J.C., Humanscale 1/2/3, MIT Press, Cambridge, MA, 1981. With permission.
na, not available.
109

11/9/10 2:23:09 AM
110 Handbook of Human Factors in Medical Device Design

Shoulder
width
Head
width

Head length
Head circumference

Arm
(chin to trunk)

circumference
Neck length
Trunk length

Chest

Arm length
circumference
Body length

Abdominal
circumference
Pelvic
width
Hand length
Leg and thigh length

Knee width
Knee pivot
to floor

FIGURE 4.4 Reference dimensions for the child measurements presented in Table 4.5. (From
Diffrient N., Tilley, A.R., and Bardagjy, J.C., Humanscale 1/2/3, MIT Press, Cambridge, MA, 1981.
With permission.)

illustrated in Figure 4.4. Table 4.5 lists average anthropometric values for males, from birth
to age four, at half-year increments. Within this age range, female dimensions differ little
from males, making these data valid for both sexes.
4.1.1.4 Additional Data Sources
The resources listed at the end of this chapter provide numerous sources of anthropometric
data, both generally (for specific applications) and pertaining to specific populations. These
should be searched for particular data not present in this chapter. PeopleSize 2000 (Open
Ergonomics Ltd, Leicestershire, United Kingdom) is a device that provides anthropometric
data in an electronic format.
In summary, additional anthropometric data (beyond that contained in this chapter)
should be consulted when designing devices:
• For specific body parts, such the hands or feet
• For use by special groups of individuals whose body measurements may differ
from the adult population, such as infants, children, and adolescents

TAF-ER562X_WINGR-09-0401-C004.indd 110 11/9/10 2:23:09 AM


Anthropometry and Biomechanics 111

4.1.2 MOBILITY AND FUNCTIONAL MEASUREMENTS


The unidimensional data (i.e., for single body parts) shown in Tables 4.2 through 4.5 are
one source for designers of medical equipment and may cover many design applications.
However, the designer must consider functional anthropometric data as well. The term
“functional” involves considering movement of the body in two or more planes because
people move, work, and use devices in three-dimensional space.

4.1.2.1 Mobility
Human mobility (also called flexibility) is dependent on many factors, such as health, phys-
ical fitness, and age. This makes mobility data quite variable. However, one sample of 100
males and 100 females (Houy, 1983; Staff, 1983) provides a consistent basis for determin-
ing the ranges of motion for specific joints and body segments in the population. Table 4.6
presents these data, which are illustrated in Figure 4.5.
An example of how to apply these mobility data to medical device designs involves the foot
acivator switches mentioned previously. They must accommodate the smallest range of human
motion. Table 4.5 shows that 5th-percentile males have the shortest ankle extension range
(21 degrees). Thus, the switch should not require a movement range of more than 21 degrees.

4.1.2.2 Functional Work


Body segment lengths and joint mobility during movement combine to determine func-
tionality, particularly with regard to reaching. Figures 4.6 and 4.7 present reach envelopes
recommended for seated and standing work tasks (Eastman Kodak Company, 1983). These
reach envelopes are for individuals at the 5th percentile, thus representing the maximum
horizontal and vertical distances at which medically related controls, equipment, and other
objects should be placed from individuals to ensure use by a majority of the population.
There are several software packages that use anthropometric data to produce virtual
humans in three-dimensional space. The utility of these types of software is that they allow
users to determine how humans of varying sizes will interact with equipment and worksta-
tions being designed. The more commonly used human modeling packages include Jack’s
Task Analysis Toolkit (Siemens Product Lifecycle Management Software Inc., Plano,
TX), Safework® Pro™ (Safework Inc., Montreal, Canada), ManneQuinPRO™ (NexGen
Ergonomics, Montreal, Canada), and Ramsis and Anthropos (Human Solutions of North
America, Inc., Troy, MI). The appropriate software package to use when designing a medi-
cal device depends on the specific application and working environment.
4.1.2.3 Visual Work
Figure 4.8 integrates illustrations of anthropometric measurements needed for standing and
seated visual work (Eastman Kodak Company, 1983). The medical device designer can use this
information to accommodate the postural and visual needs of both large and small individuals.
Designing medical devices to account for human mobility and function should do the
following:

• Integrate the known degrees of human joint flexibility.


• Consider the user’s health, level of physical fitness, and age, as each can impact
human mobility.
• Understand that functional physical and visual work depends on knowledge of
various body segment lengths operating together.

TAF-ER562X_WINGR-09-0401-C004.indd 111 11/9/10 2:23:10 AM


112 Handbook of Human Factors in Medical Device Design

TABLE 4.6
Male and Female Mobility Measurements, in degrees
Male Female
Percentile Percentile
Movement 5th 50th 95th 5th 50th 95th
Neck Ventral flexion 25.0 43.0 60.0 34.0 51.5 69.0
Dorsal flexion 38.0 56.5 74.0 47.5 70.5 93.5
Right rotation 56.0 74.0 85.0 67.0 81.0 95.0
Left rotation 67.5 77.0 85.0 64.0 77.0 90.0
Shoulder Flexion 161.0 178.0 193.5 169.5 184.5 199.5
Extension 41.5 57.5 76.0 47.0 66.0 85.0
Adduction 36.0 50.5 63.0 37.5 52.5 67.5
Abduction 106.0 123.5 140.0 106.0 122.5 139.0
Medial rotation 68.5 95.0 114.0 94.0 110.5 127.0
Lateral rotation 16.0 31.5 46.0 19.5 37.0 54.5
Elbow-forearm Flexion 122.5 138.0 150.0 135.5 148.0 160.5
Supination 86.0 107.5 135.0 87.0 108.5 130.0
Pronation 42.5 65.0 86.5 63.0 81.0 99.0
Wrist Extension 47.0 62.0 76.0 56.5 72.0 87.5
Flexion 50.5 67.5 85.0 53.5 71.5 89.5
Adduction 14.0 22.0 30.0 16.5 26.5 36.5
Abduction 22.0 30.5 40.0 19.0 28.0 37.0
Hip Flexion 95.0 109.5 130.0 103.0 125.0 147.0
Adduction 15.5 26.0 39.0 27.0 38.5 50.0
Abduction 38.0 59.0 81.0 47.0 66.0 85.0
Medial rotation (prone) 30.5 46.0 62.5 30.5 44.5 58.5
Lateral rotation (prone) 21.5 33.0 46.0 29.0 45.5 62.0
Medial rotation (sitting) 18.0 28.0 43.0 20.5 32.0 43.5
Lateral rotation (sitting) 18.0 26.5 37.0 20.5 33.0 45.5
Knee Flexion (standing) 87.0 103.5 122.0 99.5 113.5 127.5
Flexion (prone) 99.5 117.0 130.0 116.0 130.0 144.0
Medial rotation 14.5 23.0 35.0 18.5 31.5 44.5
Lateral rotation 21.0 33.5 48.0 28.5 43.5 58.5
Ankle Flexion 18.0 29.0 34.0 13.0 23.0 33.0
Extension 21.0 35.5 51.5 30.5 41.0 51.5
Adduction 15.0 25.0 38.0 13.0 23.5 34.0
Abduction 11.0 19.0 30.0 11.5 24.0 36.5

Sources: Houy, D.A., Proceedings of the Human Factors Society 27th Annual Meeting, 377, 1983; Staff, K.R.
Master’s thesis, Texas A&M University, 1983. With permission.

4.1.3 STRENGTH
Designers of medical devices and work environments must also understand that body
strength differs in several ways from one person to another and is dependent on many fac-
tors. When designing functions where user strength is an important issue, one should take
into account those factors that may impact the user’s strength potential. These include (but

TAF-ER562X_WINGR-09-0401-C004.indd 112 11/9/10 2:23:10 AM


Anthropometry and Biomechanics 113

A B

B A
Ankle extension (A),
abduction (B)

Ankle extension (A),


flexion (B)

Knee flexion,
standing

Knee flexion (prone)


A
Knee rotation
Knee flexion, B
medial (A),
kneeling
lateral (B)

Hip flexion

A B
A B

Hip rotation,
sitting,
medial (A),
lateral (B) B A
Hip rotation, prone,
Hip adduction (A), medial (A),
abduction (B) lateral (B)

FIGURE 4.5 Reference illustrations for the mobility data presented in Table 4.6. (From Van
Cott, H.P. and Kinkade, R.G., Human Engineering Guide to Equipment Design, U.S. Government
Printing Office, Washington, DC, 1972. With permission.)
are not limited to) age, gender, health status, body part, body part position, direction of exer-
tion, whether the exertion is applied statically or dynamically, and environmental issues.
Because a multitude of factors determine an individual’s strength capacity, one should
consult references included in this section for a summary of strength guidelines (Woodson,
Tillman, and Tillman, 1992). This source includes information related to (1) ranges of hand
strength for adults and children, (2) recommended upper limits for forces commonly used

TAF-ER562X_WINGR-09-0401-C004.indd 113 11/9/10 2:23:10 AM


114 Handbook of Human Factors in Medical Device Design

cm 20 30 40
Forward reach in. 5 10 15 20
30 75

25

20 50 Height
above
15 the
work
10 25 surface
5
in cm

Centerline, 8–22 cm (3–9 in.) to right


38 cm (15 in.) to right of centerline
52 cm (2` in.) to right of centerline

FIGURE 4.6 Maximum forward reach capability for the right hand of a 5th-percentile female.
Reach envelopes are given as a function of the hand moving away from the body’s centerline. (From
Eastman Kodak Company, Ergonomic design for people at work, Volume I: Workplace, equipment,
and environmental design and information transfer, Lifetime Learning Publications, Belmont, CA,
1983. With permission.)
cm 25 50
Forward reach in. 10 20 30
84
79 200
74
69 175
64
59 150
Height
54 above
49 125 floor
44
39 100
34
29 75
24
in. cm

Centerline, 15–30 cm (6–12 in.) to right


46 cm (18 in.) to right of centerline
61 cm (24 in.) to right of centerline
76 cm (30 in.) to right of centerline

FIGURE 4.7 Maximum one-handed and two-handed forward reach capability of a 5th-percentile
person in a population of both males and females. Reach envelopes are given as a function of the
hand moving away from the body’s centerline. (From Eastman Kodak Company, Ergonomic Design
for People at Work, Volume I: Workplace, Equipment, and Environmental Design and Information
Transfer, Lifetime Learning Publications, Belmont, CA, 1983. With permission.)

TAF-ER562X_WINGR-09-0401-C004.indd 114 11/9/10 2:23:11 AM


Anthropometry and Biomechanics 115

B
Secondary C Maximum height
visual displays 20º of workplace
E 165 cm (65 in.)
Viewing
distance
50 cm (20 in.)
Top of primary displays
114 cm (45 in.)
Primary D
visual A 10º
displays
Top of
keyboard
76 cm
(30 in.)
Work
surface
66 cm
(26 in.)

(a)

B
Secondary
visual displays
C
) Maximum
in. 20º
(20 height of
c mE Viewing secondary
5 0
Distance displays
E 175 cm
50 cm (20 in.) (69 in.)

A Maximum
Primary height of
visual primary
displays displays
157 cm
(62 in.)
D
20º Top of
keyboard keyboard
angle 107 cm
(42 in.)

Work
surface
height
102 cm
(40 in.)
(b)

FIGURE 4.8 Recommended dimensions for seated (a) and standing (b) workstations used for
performing visual tasks. Primary displays refer to those that are most frequently monitored, while
secondary displays are those less critical but still necessary for performing the task. (From Eastman
Kodak Company, Ergonomic Design for People at Work, Volume I: Workplace, Equipment, and
Environmental Design and Information Transfer, Lifetime Learning Publications, Belmont, CA,
1983. With permission.)

TAF-ER562X_WINGR-09-0401-C004.indd 115 11/9/10 2:23:12 AM


116 Handbook of Human Factors in Medical Device Design

on equipment and in control operations, and (3) lifting and carrying strengths. Strength
data related to gender and the capabilities of the extremities are provided below.
4.1.3.1 Strength and Gender Differences
U.S. Army data (U.S. Department of Defense, 1995) are some of the best sources of human
strength information. When both males and females use medical equipment or when only male
strength data are available, designers should consider the following general guidelines:

• When the equipment requires exertions of the upper extremities, note that female
arm and hand strength is slightly more than half (56.5%) that of males.
• For force exertions required by the lower extremities (e.g., legs, feet), male strength
capacity data should be reduced to nearly two-thirds (64.2%) to account for female
strength abilities.
• For devices involving trunk strength, female force requirements should be reduced
to 66.0% of the limits used for males.
4.1.3.2 Strength of the Upper Extremity
The shoulder and elbow joints together allow the upper extremity to move in many direc-
tions. However, strength capability is impacted by the positions of these joints and their
direction of motion. Just as the proper design of a medical device’s physical characteristics
often focuses on the smallest user (e.g., reach capability of a 5th-percentile female), the
lowest strength capability of the user population should provide the basis for setting limits
on the device’s force requirements. Thus, Table 4.7 presents maximum upper extremity
muscle strength data for males at the 5th percentile, using Figure 4.9 as a graphical refer-
ence (adapted from U.S. Department of Defense, 1995). Here, strength data are listed for
both the left and the right side, for pulling and pushing, and for upward, downward, inward,

TABLE 4.7
Fifth-Percentile Male Muscle Strength of the Left and Right Arm, for Various
Elbow Postures and Force Application Directions, in N (lbs.)
1 2 3 4 5 6 7

Degree of Pull Push Up Down In Out


Elbow Flexion Left Right Left Right Left Right Left Right Left Right Left Right
180 degrees 222 231 187 222 40 62 58 76 58 89 36 62
(50) (52) (42) (50) (9) (14) (13) (17) (13) (20) (8) (14)
150 degrees 187 249 133 187 67 80 80 89 67 89 36 67
(42) (56) (30) (42) (15) (18) (18) (20) (15) (20) (8) (15)
120 degrees 151 187 116 160 76 107 93 116 89 98 45 67
(34) (42) (26) (36) (17) (24) (21) (26) (20) (22) (10) (15)
90 degrees 142 165 98 160 76 89 93 116 71 80 45 71
(32) (37) (22) (36) (17) (20) (21) (26) (16) (18) (10) (16)
60 degrees 116 107 98 151 67 89 80 89 76 89 53 76
(26) (24) (22) (34) (15) (20) (18) (20) (17) (20) (12) (17)
Source: Department of Defense, Handbook for Human Engineering Design Guidelines, MIL-HDBK-759C,
1995, Navy Publishing and Printing Office, Philadelphia, PA. With permission.)
Note: Numbers in the first row of this table refer to the diagrams in Figure 4.9.

TAF-ER562X_WINGR-09-0401-C004.indd 116 11/9/10 2:23:14 AM


Anthropometry and Biomechanics 117

2 3
1

180°

150° 4 7

60°
90° 120°

5 6

FIGURE 4.9 Reference strength exertions for the data presented in Table 4.7. (From U.S.
Department of Defense, Handbook for Human Engineering Design Guidelines, MIL-HDBK-759C,
1995, Navy Publishing and Printing Office, Philadelphia, PA. With permission.)

and outward motions. Data are presented for the force application directions at five dif-
ferent elbow flexion angles. Note that comparable female strength data are not available.
Thus, the values in Table 4.7 should be reduced by 50% to 60% (as stated in the previous
section) for equipment designed for use by females. As an example, consider the downward
pull force required to lower a heart monitor attached to a moveable swing arm. Assuming
that the elbow flexion angle of most users would be 150 degrees during this task, Table
4.7 shows that 5th-percentile males have a right-arm force capability of 89 N. Given that
females also are likely to perform this task, a force reduction to 50.3 N (56.5% of 89 N) is
reasonable to account for most possible users.
In summary, medical devices that require force application of the upper extremity should
account for the following:

• The population using the device (e.g., males, females, or a combination)


• Human strength ability and differences due to elbow flexion
• Direction of the force application
• The hand used to apply the force

4.1.3.3 Strength of the Fingers and Hands


Finger and hand strength data are relevant in the design and handling of particular medical
equipment. Table 4.8 presents such data for two groups of males—students and industrial
workers (Kroemer, Kroemer, and Kroemer-Elbert, 1994).

4.1.3.4 Strength of the Feet


As with other body parts, foot strength is a function of age, gender, direction of motion,
and leg position. Table 4.9 shows data related to the required strength for pressing and lift-
ing either a bar or a pedal with the foot (Consumer and Competition Policy Directorate,
Department of Trade and Industry, 2002b). Foot strength data are presented for both males
and females and as a function of age.

TAF-ER562X_WINGR-09-0401-C004.indd 117 11/9/10 2:23:14 AM


118 Handbook of Human Factors in Medical Device Design

TABLE 4.8
Finger and Hand Forces That Two Groups of Males Can Exert, in a Variety of
Hand/Handle Configurations
Mean Forces (Standard Deviations) Exerted by Male
Students (*) or Male Machinists, in N
Digit 1 Digit 2 Digit 3 Digit 4 Digit 5
Hand/Handle Coupling and Description (Thumb) (Index) (Middle) (Ring) (Little) Other
Digit touch One digit touches an object.
– Touch perpendicular to 84 (33)* 43 (14)* 36 (13)* 30 (13)* 25 (10)*
extended digit 131 (42) 70 (17) 76 (20) 57 (17) 55 (16)
– Digits 2–5 all pressed 162 (33)
on one bar
– Tip force, as in 30 (12)* 29 (11)* 23 (9)* 19 (7)*
typing 65 (12) 69 (22) 50 (11) 46 (14)
Palm touch Some part of the palm or 233 (65)
hand touches the object.

Hook grip One finger or several fingers 61 (21)* 49 (17)* 48 (19)* 38 (13)* 34 (10)*
hook(s) onto a ridge or 118 (24) 89 (29) 104 (26) 77 (21) 66 (17)
handle (used where thumb
counterforce is not needed).
– All digits combined 108 (39)*
252 (63)
Tip pinch – The thumb top opposes 50 (14)* 53 (14)* 38 (7)* 28 (7)*
one fingertip. 59 (15) 63 (16) 44 (12) 30 (6)

Pad pinch – Thumb pad opposes (On


the palmer pad of one digits 2
finger (or the pads of and 3) 63 (12)* 61 (16)* 41 (12)* 31 (9)*
several fingers) near the 85 (16)* 34 (7) 70 (15) 54 (15) 34 (7)
tops. 95 (19)
Side pinch – Thumb opposes the 98 (13)*
(radial) side of the 112 (16)
index finger.
Power grasp The total inner hand surface 318 (61)*
grasps the (often 366 (53)
cylindrical) handle, which
runs parallel to the
knuckles and generally
protrudes on one or both
sides from the hand.

Source: Kroemer, K.H.E., Kroemer, H.B., and Kroemer-Elbert, K.E. Ergonomics: How to Design for Ease and
Efficiency, Prentice Hall, Englewood Cliffs, NJ, 1994. With permission.

TAF-ER562X_WINGR-09-0401-C004.indd 118 11/9/10 2:23:15 AM


TABLE 4.9
Foot Strength Capabilities for Pressing and Lifting a Bar and a Pedal, in N (lbs.)
Bar Pedal
Standard Standard
Age (years) Gender Activity Mean Deviation Minimum Maximum Mean Deviation Minimum Maximum

TAF-ER562X_WINGR-09-0401-C004.indd 119
2–5 Male Lift 46.4 (10.4) 15.6 (3.5) 18.2 (4.1) 63.0 (14.2) 42.6 (9.6) 12.3 (2.8) 23.2 (5.2) 58.1 (13.1)
Press 93.5 (21.0) 29.5 (6.6) 44.3 (10.0) 137.0 (30.8) 79.2 (17.8) 35.4 (8.0) 18.5 (4.2) 126.7 (28.5)
Female Lift 48.7 (10.9) 16.9 (3.8) 18.2 (4.1) 63.0 (14.2) 33.2 (7.5) 13.0 (2.9) 23.2 (5.2) 63.0 (14.2)
Press 84.8 (19.1) 33.4 (7.5) 44.3 (10.0) 147.3 (33.1) 80.3 (18.1) 33.6 (7.6) 39.1 (8.8) 131.8 (29.6)
Anthropometry and Biomechanics

6–10 Male Lift 90.0 (20.2) 35.4 (8.0) 58.1 (13.1) 142.8 (32.1) 72.0 (16.2) 24.3 (5.5) 48.1 (10.8) 102.9 (23.1)
Press 169.9 (38.2) 40.5 (9.1) 100.9 (22.7) 203.9 (45.8) 151.4 (34.0) 38.9 (8.7) 106.1 (23.9) 198.7 (44.7)
Female Lift 133.4 (30.0) 41.6 (9.4) 73.0 (16.4) 192.6 (43.3) 98.0 (22.0) 36.3 (8.2) 43.1 (9.7) 147.8 (33.2)
Press 232.9 (52.4) 124.7 (28.0) 95.8 (21.5) 466.5 (104.9) 162.7 (36.6) 64.1 (14.4) 75.2 (16.9) 276.0 (62.0)
11–15 Male Lift 223.9 (50.3) 69.0 (15.5) 117.9 (26.5) 342.1 (76.9) 142.8 (32.1) 45.9 (10.3) 48.1 (10.8) 197.6 (44.4)
Press 466.4 (104.8) 126.0 (28.3) 250.2 (56.2) 636.4 (143.1) 280.4 (63.0) 78.4 (17.6) 147.3 (33.1) 384.1 (86.3)
Female Lift 178.7 (40.2) 50.5 (11.4) 147.8 (33.2) 267.4 (60.1) 118.9 (26.7) 44.2 (9.9) 83.0 (18.7) 182.6 (41.0)
Press 428.4 (96.3) 26.7 (6.0) 394.4 (88.7) 466.5 (104.9) 197.7 (44.4) 107.6 (24.2) 111.2 (25.0) 373.8 (84.0)
16–20 Male Lift 327.1 (73.5) 56.5 (12.7) 242.4 (54.5) 416.8 (93.7) 172.7 (38.8) 52.0 (11.7) 117.9 (26.5) 252.4 (56.7)
Press 657.7 (147.9) 129.0 (29.0) 476.8 (107.2) 898.9 (202.1) 407.6 (91.6) 177.7 (39.9) 167.8 (37.7) 667.2 (150.0)
Female Lift 180.6 (40.6) 80.2 (18.0) 102.9 (23.1) 302.2 (67.9) 98.9 (22.2) 35.1 (7.9) 63.0 (14.2) 152.7 (34.3)
Press 403.6 (90.7) 116.7 (26.2) 296.6 (66.7) 590.0 (132.6) 216.3 (48.6) 128.6 (28.9) 85.5 (19.2) 415.0 (93.3)
21–30 Male Lift 320.0 (71.9) 99.4 (22.3) 147.8 (33.2) 441.8 (99.3) 155.6 (35.0) 33.7 (7.6) 107.9 (24.3) 212.5 (47.8)
Press 632.7 (142.2) 178.4 (40.1) 399.5 (89.8) 873.2 (196.3) 341.4 (76.7) 137.9 (31.0) 147.3 (33.1) 564.3 (126.9)
Female Lift 228.3 (51.3) 49.6 (11.2) 142.8 (32.1) 277.3 (62.3) 136.1 (30.6) 31.5 (7.1) 83.0 (18.7) 177.7 (39.9)
Press 469.0 (105.4) 64.6 (14.5) 389.2 (87.5) 528.2 (118.7) 227.1 (51.1) 66.3 (14.9) 167.8 (37.7) 332.6 (74.8)
31–50 Male Lift 309.4 (69.6) 49.2 (11.1) 247.4 (55.6) 391.9 (88.1) 200.4 (45.0) 39.4 (8.9) 157.7 (35.5) 282.3 (63.5)
Press 673.0 (151.3) 137.9 (31.0) 461.3 (103.7) 873.2 (196.3) 329.2 (74.0) 130.0 (29.2) 173.0 (38.9) 579.7 (130.3)
Female Lift 210.7 (47.4) 59.3 (13.3) 127.8 (28.7) 312.2 (70.2) 129.7 (29.2) 24.4 (5.5) 92.9 (20.9) 167.7 (37.7)
Press 519.2 (116.7) 189.7 (42.6) 209.0 (47.0) 754.8 (169.7) 238.0 (53.5) 106.8 (24.0) 121.5 (27.3) 435.6 (97.9)

continued
119

11/9/10 2:23:16 AM
120

TABLE 4.9 (CONTINUED)


Foot Strength Capabilities for Pressing and Lifting a Bar and a Pedal, in N (lbs.)
Bar Pedal
Standard Standard
Age (years) Gender Activity Mean Deviation Minimum Maximum Mean Deviation Minimum Maximum

TAF-ER562X_WINGR-09-0401-C004.indd 120
51–60 Male Lift 279.8 (62.9) 78.7 (17.7) 197.6 (44.4) 386.9 (87.0) 176.4 (39.7) 31.9 (7.2) 137.8 (31.0) 202.6 (45.5)
Press 592.6 (133.2) 162.5 (36.5) 435.6 (97.9) 811.4 (182.4) 215.5 (48.4) 41.0 (9.2) 162.7 (36.6) 260.5 (58.6)
Female Lift 145.3 (32.7) 42.4 (9.5) 102.9 (23.1) 202.6 (45.5) 109.1 (24.5) 31.6 (7.1) 78.0 (17.5) 152.7 (34.3)
Press 351.9 (79.1) 92.9 (20.9) 245.1 (55.1) 466.5 (104.9) 243.8 (54.8) 45.2 (10.2) 193.6 (43.5) 291.4 (65.5)
61–70 Male Lift 245.4 (55.2) 42.1 (9.5) 207.6 (46.7) 317.2 (71.3) 160.7 (36.1) 21.6 (4.9) 142.8 (32.1) 197.6 (44.4)
Press 559.12 (125.7) 126.1 (28.3) 451.0 (101.4) 770.2 (173.1) 267.7 (60.2) 66.4 (14.9) 173.0 (38.9) 342.9 (77.1)
Female Lift 155.6 (35.0) 39.3 (8.8) 112.9 (25.4) 252.4 (56.7) 120.7 (27.1) 22.0 (4.9) 78.0 (17.5) 147.8 (33.2)
Press 337.4 (75.8) 94.7 (21.3) 106.1 (23.9) 512.8 (115.3) 183.3 (41.2) 57.7 (13.0) 70.0 (15.7) 276.0 (62.0)
71–80 Male Lift 182.6 (41.0) 53.4 (12.0) 73.0 (16.4) 242.4 (54.5) 131.1 (29.5) 38.6 (8.7) 63.0 (14.2) 197.6 (44.4)
Press 417.9 (93.9) 112.8 (25.4) 296.6 (66.7) 605.5 (136.1) 205.0 (46.1) 95.7 (21.5) 85.5 (19.2) 399.5 (89.8)
Female Lift 139.7 (31.4) 38.8 (8.7) 38.1 (8.6) 182.6 (41.0) 128.6 (28.9) 25.6 (5.8) 92.9 (20.9) 162.7 (36.6)
Press 357.2 (80.3) 71.6 (16.1) 214.2 (48.2) 481.9 (108.3) 215.0 (48.3) 67.9 (15.3) 111.2 (25.0) 378.9 (85.2)
81–90 Male Lift 162.7 (36.6) 49.4 (11.1) 127.8 (28.7) 197.6 (44.4) 132.8 (29.9) 70.4 (15.8) 83.0 (18.7) 182.6 (41.0)
Press 296.6 (66.7) 72.8 (16.4) 245.1 (55.1) 348.0 (78.2) 211.7 (47.6) 91.0 (20.5) 147.3 (33.1) 276.0 (62.0)
Female Lift 117.9 (26.5) 33.3 (7.5) 63.0 (14.2) 147.8 (33.2) 107.9 (24.3) 52.1 (11.7) 48.1 (10.8) 187.6 (42.2)
Press 256.4 (57.6) 107.0 (24.1) 157.6 (35.4) 430.4 (96.8) 151.4 (34.0) 25.6 (5.8) 126.7 (28.5) 193.6 (43.5)

Source: Consumer and Competition Policy Directorate, Department of Trade and Industry, Strength Data for Design Safety—Phase 2, London, United Kingdom, 2002.
With permission.
Handbook of Human Factors in Medical Device Design

11/9/10 2:23:16 AM
Anthropometry and Biomechanics 121

4.1.4 SPECIAL CONSIDERATIONS


Individuals with a particular temporary or permanent disability also use many medical
devices. Such populations include children, older adults, wheelchair users, or persons with
specific physical limitations. The reader is urged to consult the references listed in this
chapter as well as Chapter 2, “Basic Human Abilities” for sources of anthropometric infor-
mation for these and other special populations.
Note that not all physical measurements of persons with disabilities are necessarily dif-
ferent from those for populations without disabilities. For instance, anthropometric data
from the United Kingdom (Consumer and Competition Policy Directorate, Department of
Trade and Industry, 2002a) found no significant differences in hand dimensions between
those with disabilities (e.g., individuals whose impairments affected their reaching, dexter-
ity, or manipulation capabilities) and those without disabilities. However, handgrip span
(i.e., the distance between the thumb and little finger) differed significantly between groups.
These measurements are shown in Table 4.10.

TABLE 4.10
Handgrip Spans of Dexterity-Impaired and Unimpaired Males and Females, in cm (in.)
Handgrip Span of Individuals with a Dexterity Impairment
Standard
Mean Deviation Minimum Maximum
Males
Left hand 7.2 (2.8) 2.3 (0.9) 2.3 (0.9) 14.1 (5.6)
Right hand 6.7 (2.6) 2.2 (0.9) 1.6 (0.6) 11.2 (4.4)
Females
Left hand 6.1 (2.4) 2.1 (0.8) 0.5 (0.2) 12.9 (5.1)
Right hand 6.2 (2.4) 2.0 (0.8) 0.8 (0.3) 11.1 (4.4)
Combined
Left hand 6.4 (2.5) 2.2 (0.9) 0.5 (0.2) 14.1 (5.6)
Right hand 6.3 (2.5) 2.1 (0.8) 0.8 (0.3) 11.2 (4.4)
Handgrip Span of Individuals without a Dexterity Impairment
Mean Standard Minimum Maximum
Deviation
Males
Left hand 8.5 (3.3) 2.4 (0.9) 4.4 (1.7) 14.2 (5.6)
Right hand 8.4 (3.3) 2.0 (0.8) 4.6 (1.8) 13.1 (5.2)
Females
Left hand 7.1 (2.8) 1.9 (0.7) 2.8 (1.1) 12.1 (4.7)
Right hand 7.3 (2.9) 1.8 (0.7) 3.2 (1.3) 11.6 (4.6)
Combined
Left hand 7.6 (3.0) 2.2 (0.9) 2.8 (1.1) 14.2 (5.6)
Right hand 7.7 (3.0) 2.0 (0.8) 3.2 (1.3) 13.1 (5.2)

Source: Consumer and Competition Policy Directorate, Department of Trade and Industry, Specific
Anthropometric and Strength Data for People with Dexterity Disability, London, United Kingdom,
2002. With permission

TAF-ER562X_WINGR-09-0401-C004.indd 121 11/9/10 2:23:16 AM


122 Handbook of Human Factors in Medical Device Design

1200 270

Trunk extension
1000 225

Strength (lbs.-ft.)
800
Strength (N) Trunk flexion 180

Handgrip
600 135
Knee extension

Elbow flexion
400 90

31–35 51–55 71–75


Age (years)

FIGURE 4.10 Average strength for three age-groups of males for five different muscle groups.
(From Viitasalo, J.T. et al., Ergonomics, 28, 1566, 1985. With permission.)

One should consider the physical limitations of individuals in special populations, par-
ticularly when designing equipment requiring physical force exertions. Aging affects mus-
cle groups differently. An example of aging’s effects on isometric strength (Viitasalo et al.,
1985) is shown in Figure 4.10. On average, males in their early 50s had about 80% of the
strength of those in their early 30s. By the early 70s, average strength declines to about
60% of what it was four decades earlier. Although data were gathered only for males in this
study, female strength capabilities also are known to decrease with age (see Chapter 18,
“Home Health Care”).
When a medical device is to be used by individuals in a special population, the designer
should do the following:

• Use anthropometric data that are reflective of any unique physical capabilities and
limitations.
• Recognize that users with a disability may not differ from those without a disabil-
ity on every physical aspect.

4.1.5 DESIGN GUIDELINES FOR BODY DIMENSION DATA


If anthropometric information is inappropriately applied during the design of medical
devices, fewer individuals will be able to use the device safely and effectively. The follow-
ing is a list of alternative approaches to assist in correctly applying anthropometric data.
The approach used depends on the device being designed.

4.1.5.1 Designing for Population Extremes


The first approach involves designing medical devices to accommodate users on both ends
of the “size spectrum” with regard to their physical body dimensions. For example, those
persons at the lower end of the spectrum (e.g., the 5th percentile) must be able to reach the
necessary controls in a workstation or wrap their fingers around tool handles. These are

TAF-ER562X_WINGR-09-0401-C004.indd 122 11/9/10 2:23:17 AM


Anthropometry and Biomechanics 123

tasks that can be performed more easily by those having larger body part sizes (i.e., those
above the 5th percentile). Similarly, persons at the higher end of the size spectrum (e.g., the
95th percentile) must be able to fit into a workstation, which can be affected by, for exam-
ple, the legroom at seated workstations or the head clearance when walking underneath
monitoring equipment. By creating medical devices to accommodate persons on each end
of this anthropometry spectrum, the designer ensures that a majority of individuals (i.e.,
those falling between these two ranges of body size extremes) will be able to use these
devices effectively and safely.
For device or workstation dimensions that are critical to its operation, the designer
should determine which of these two groups needs to be accommodated. Consider, for
example, safety cabinet hoods in which pharmacy medications are prepared. Users often
need to access devices at the rear of the cabinet. Thus, that dimension should accommodate
individuals with a short arm reach. By default, those with longer arms also will be able to
reach to this distance. Similarly, the foot cutout at the bottom of the safety cabinet should be
designed to fit those with the largest feet. Those with smaller feet dimensions will naturally
be accommodated.

Guideline 4.1: Physical Accommodation of Most Users


Medical devices should physically accommodate the widest practical range of users. A target
range from the 5th-percentile female to the 95th percentile male is considered the nominally
acceptable range. However, a target range from the 2.5th- or 1st-percentile female to the
97.5th- or 99th-percentile male is preferable, noting that the broader range of accommodation
might require only slight adjustments in medical device proportioning. This is particularly
true for hand-operated devices, such as surgical instruments for which an extra 0.1-inch-wide
grip might increase handling comfort dramatically for many users.

Guideline 4.2: Avoid Use at Ability Extremes


Although an anthropometric analysis might identify maximums for parameters such as reach
and strength, medical devices should not require users to function at the extremes of their
ability. Rather, medical devices should place only moderate physical demands on users to
ensure comfortable interactions. Therefore, a medical device’s weight should be kept well
below the limit of a small female’s lifting capabilities, for example.

Guideline 4.3: Accomodate Use with Protective Clothing


Anthropometric analyses should account for users’ clothing, particularly protective gear (e.g.,
face shields, gloves, boots) (see Chapter 3, “Environment of Use”).

Guideline 4.4: Design for Both Genders


Unless intended for exclusive use by either males or females, medical devices should be
designed for equally effective use by both genders.

Guideline 4.5: Confirmation of Physical Accommodation


While designs may be based on anthropometric analyses, physical accommodation should
be confirmed by checking the design’s fit with users who have average and extreme physical
characteristics (e.g., particularly tall, particularly short, and average height individuals).
This approach recognizes the complexities of human movement in three dimensions that

TAF-ER562X_WINGR-09-0401-C004.indd 123 11/9/10 2:23:17 AM


124 Handbook of Human Factors in Medical Device Design

might not be fully or accurately addressed through anthropometric analyses, even those
employing computer-based human models that permit sophisticated, three-dimensional
analyses.

4.1.5.2 Designing for the Average User


Designers sometimes use an approach to develop a device based on the average (i.e., mean,
or 50th percentile) body size values on one or more body dimensions since anthropometric
data tables commonly contain this information. In some cases this approach is appropriate,
such as for noncritical design elements (e.g., the height of a latch to gain access to an X-ray
machine). However, if particular body size elements are critical to the safe and proper
design of a medical device, then designing for the “average person” is inappropriate. Only
a very small percentage of individuals (male or female) are actually “average” on multiple
anthropometric dimensions. For example, just because someone is “average” in terms of
her standing height does not mean she also will have average arm length or average hand
breadth. Thus, designing medical tools or equipment using “average” (or 50th percentile)
data on several body dimensions likely will produce a device that cannot be used easily by
most people.

Guideline 4.6: Multiple Sizes of Devices


Where practical, medical devices (e.g., certain surgical instruments) should be offered in vary-
ing sizes to accommodate diverse users. This approach can avoid undesirable compromises
associated with producing a “one-size-fits-all” device that ultimately is sized optimally only
for a subset of users. For example, it would be better to produce three different size clamps
with variously spaced and sized finger openings instead of just one that is ill-suited to people
with particularly small or large hands.

4.1.5.3 Designing for Adjustability


As the data tables contained in this chapter show, human body sizes are inherently variable
in a population. This is due to diversity, for example, in age, gender, ethnicity, and health
status. Furthermore, personal preference often affects how individuals use equipment. A
classic example of this is in chair design. Two individuals, even those having the same
body size and shape, may prefer to sit at the same laboratory diagnostic workstation in very
different ways. Thus, even when the designer integrates anthropometric data into a device’s
design, users themselves may handle these devices differently.
Factors related to body size and user preference interact in innumerable ways. This
suggests that, where possible, one should design medical devices and workspaces so that
they are adjustable. In other words, a device designed so that its physical components can
be easily raised/lowered, widened/narrowed, or otherwise manipulated will accommodate
not only the desired range of physical body sizes in the target population but also users’
personal preferences. Many medical devices need to be adjustable to accommodate the full
range of users, whether they are patients, clinicians, or technicians. One example would be
an adjustable arm on a television monitor used to display images as part of an endoscopic
workstation, such as one used during a colonoscopy procedure. The gastroenterologist can
move the arm such that he can easily see the images of the colon on the monitor as he
manipulates the endoscope.

TAF-ER562X_WINGR-09-0401-C004.indd 124 11/9/10 2:23:17 AM


Anthropometry and Biomechanics 125

Thus, the approach for accommodating as large a segment of medical devices users as
possible should follow these guidelines:

Guideline 4.7: Don’t Use Population Averages


Designers may make use of anthropometric data to guide conceptual development but should
not use data on population averages in the detailed design, particularly the design of elements
critical to operational effectiveness, safety, and comfort.

Guideline 4.8: Accomodate User Extremes


Designers should size medical devices to accommodate users at the extremes. For example,
workspaces should be spacious enough to accommodate large individuals, while controls should
be placed sufficiently low on a control panel to be within the reach of small individuals.

Guideline 4.9: Incorporate Adjustability


As necessary, medical devices should include adjustment mechanisms (e.g., a telescoping
headrest) to accommodate users’ needs and preferences.

Guideline 4.10: Controls Placed Within Reach


Controls should be positioned where users can reach them without overextending their arms
(i.e., moving their shoulder joint forward from its normal position) (see Chapter 7, “Controls”).
At a minimum, controls should be placed within the users’ functional reach, which may
involve some stretching and twisting.

Guideline 4.11: Either Handed Use


Only in specific cases should medical devices be designed for one-handed rather than two-
handed use. Device design, for example, should facilitate use by people who might have an
injured hand, joint disorder in one hand, or amputated arm/hand/fingers.

4.1.5.4 Deriving Missing Data


Sometimes needed anthropometric dimensions for a medical device are not available for a
particular user group. Thus, a designer will need to determine the appropriate values. There
are several ways to accomplish this, as detailed below.
Designers commonly derive needed body dimension estimates by conveniently measuring
a few colleagues or individuals thought to be representative of the population of intended
users. On the surface, this technique appears to be simple and cost-effective. However, one
should avoid this approach especially if the measurement in question is critical to the use
and safety of the device. For example, imagine that the engineer who is designing an MRI
system decides to use his own body dimensions to specify the machine’s opening but that
this particular design engineer was a 5th-percentile male. This design would be problematic
since the opening would be too small for larger patients.
An alternative method of determining a needed physical dimension is to measure the
dimension in a representative sample of users, sufficient to account for human body size
variability. This approach often is costly and time consuming, but it should be employed
if the needed measurement is critical to perfecting the operation of the medical device or
workstation. For example, spinal implants, such as rods and braces, are critical devices, and
their designs demand accurate dimensions related to spinal column anatomy and the vari-
ability in size found in the population.

TAF-ER562X_WINGR-09-0401-C004.indd 125 11/9/10 2:23:17 AM


126 Handbook of Human Factors in Medical Device Design

Guideline 4.12: Obtaining Missing Data


Where practical, derive missing data from a representative user population rather than a more
convenient but less representative one. Remember that devices often initially intended for one
user population (e.g., American surgeons) may ultimately be used widely by a broader popula-
tion (e.g., international use by many types of clinicians).

A more common approach to determining the appropriate data involves estimating


a needed dimension based on known measurements. There are several computational
techniques used for this purpose. The assumption with the ratio scaling technique is that
many individual body sizes (such as various segment lengths) are roughly proportional to
one another. For example, determining the sleeve length for surgical gowns to be used in
Europe requires arm length data. If these data are only available for a U.S. population, the
ratio of 5th-, 50th-, and 95th-percentile arm lengths to standing heights for this population
could be used to estimate European arm lengths, given their known standing heights (as
this is a common measurement taken for many groups).
The regression equation technique assumes that there is a linear relationship between
two values. For example, suppose that arm length data (used to design surgical gown sleeve
lengths) exist for the 5th, 50th, and 95th percentiles of a population but that they are needed
for the 1st and 99th percentiles. Applying regression techniques to the known data would
be used to extrapolate to the unknown (i.e., 1st and 99th percentile) values. The designer
should evaluate the assumptions of this approach before using it extensively.
Finally, in the probability statistics approach, a measurement for an entire population
is estimated from data gathered on a small sample. Suppose that, to determine surgical
gown sleeve length for Europeans, 15 such individuals are measured. Clearly, this number
of people will not fully represent all Europeans. However, probability equations (from the
average and variance data in the small sample) can be used to estimate population values
from this sample.
The specifics of these approaches can be found in any number of anthropometric texts,
including Kroemer et al. (1997). A more in-depth review of these techniques will help
the designer choose the appropriate method and develop an accurate estimate of a needed
physical measurement. In order to derive needed missing data most accurately, follow these
guidelines:

Guideline 4.13: Larger Sample Size Preferred


Use anthropometric data based on large sample sizes because such a data set is more likely to
contain data whose ranges are reflective of that population.

Guideline 4.14: Largest Practical Study


If the needed anthropometric data are not available, cannot be reliably derived from exist-
ing sources, and are essential to producing an effective design, conduct the largest practical
anthropometric study to collect the necessary data and ensure its reliability.

Guideline 4.15: Most Pertinent User Population


Designers should use anthropometric data pertaining to the most pertinent user population.
For example, when designing a telemetry workstation, it is better to use data derived from
measurements of nurses than military airline pilots or commercial air traffic controllers.

TAF-ER562X_WINGR-09-0401-C004.indd 126 11/9/10 2:23:17 AM


Anthropometry and Biomechanics 127

4.2 CASE STUDIES IN ANTHROPOMETRIC DESIGN


4.2.1 VIEWING ANGLE DETERMINATION FOR A CARDIAC OUTPUT MONITOR
Designers seeking to improve a cardiac output monitor needed to select an LCD display
that could accommodate an appropriate range of viewing angles. This range was based on
the expected user population height extremes and the expected range of viewing positions
in a typical hospital environment. Designers conducted field visits to hospitals as part of
a contextual inquiry to collect the necessary data. They combined the field observational
data with anthropometric data for 5th- and 95th-percentile users representing population
extremes (for more details, refer to Chapter 9, “Visual Displays”).

4.2.1.1 Analysis
One of the limitations of LCDs is that image quality typically is poor (because of inad-
equate luminance, contrast, or glare) when they are viewed at off-normal viewing angles in
both horizontal and vertical directions.
Horizontal viewing angle is that angle between the monitor and the user’s (left or right)
position from it. Users generally stand directly in front of the display when interacting with
the cardiac output monitor, except in certain situations (e.g., viewing readings while insert-
ing a catheter). The worst-case horizontal angles appear to occur most often when viewing
the cardiac output monitor from a distance. For example, at a patient’s bedside, the monitor
often is set up so that the display faces the foot of the bed. The largest horizontal off-normal
viewing angles (±45 degrees) occur when attempting to view the bedside monitor from the
doorway of the patient’s room.
The cardiac output monitor sometimes is located on a shelf or adjustable rack that places
it at or near the user’s eye height. In this situation, the off-normal (upward or downward)
angle does not pose a significant problem.
Although there are a wide variety of other installation locations for these monitors, those
affecting vertical viewing positions fall into two categories—high-mount position (above
eye height) and low-mount position (below eye height).

4.2.1.2 High-Mount Position


The top of the cardiac output monitor’s viewing area can be mounted above the user’s eye
level. For example, in an operating room, the monitor may be stacked atop other equipment,
such as an anesthesia workstation. The angle at which the user must look up (equivalent to
the downward angle from the display normal line) will be determined by display height,
user eye height, and distance from the user’s eye to the display. Figure 4.11 illustrates this
type of installation.

4.2.1.3 Calculated Viewing Angles for the High-Mount Position


A standard design approach for viewing angle is to accommodate the 5th- to 95th-
percentile male and female U.S. population, or approximately 90% of the overall popula-
tion. Figure 4.11 shows that the worst-case situation will be with the 5th-percentile user.
Table 4.11 presents eye heights of 5th-percentile male and female U.S. adults (Pheasant,
1996) in standing and seated positions, where seating is assumed to be on 61.0 or 76.2 cm
(24–30 inch) stools. “Difference” refers to the distance between the eye height of the spec-
ified population and the height of the display. Viewing angles, shown in the right-most

TAF-ER562X_WINGR-09-0401-C004.indd 127 11/9/10 2:23:17 AM


128 Handbook of Human Factors in Medical Device Design

Viewing
distance
0° Elevation
Q2

Off-normal viewing angle


Q2 height

User’s eye
position
(standing or
seated)
Eye height

Floor

FIGURE 4.11 Viewing angle with heart output monitor in a high vertical position. Q2 is the moni-
tor display being evaluated.
column of Table 4.11, are calculated on the basis of the following assumptions (note that the
height of the display was based on the highest location found in the field observations):

• Display height (top of display): 200.7 cm (79 in.)


• Assumed viewing distance: 45.7 cm (18 in.)
4.2.1.4 Low-Mount Position
In many situations, such as in critical care units, the cardiac output monitor is placed where
the user must look down, such as when the device is placed on a bedside stand. Again,
viewing angle is determined by the height of the display, the user’s eye height, and the
user’s distance from the display. In this situation (Figure 4.12), the largest viewing angle is
required when the display is mounted at its lowest level and the user’s eye position is rela-
tively high (e.g., a taller standing user).
4.2.1.5 Calculated Viewing Angle for the Low-Mount Position
Table 4.12 shows calculated viewing angles based on the following assumptions, given that the
tallest user in the population is represented by the 95th-percentile U.S. male (Pheasant, 1996):

• Display height (bottom of display): 88.3 cm (34.8 in.)


• Assumed viewing distance: 45.7 cm (18.0 in.)

TABLE 4.11
Calculated (Off-Normal) Viewing Angles with High Vertical Position, in cm (in.)
User User Eye Height Difference Viewing Angle (°)
5th-percentile female—standing 142.0 (55.9) 58.7 (23.1) 52.1
5th-percentile female—stool (30 in.) 145.2 (57.2) 55.5 (21.9) 50.5
5th-percentile female—stool (24 in.) 130.0 (51.2) 70.7 (27.8) 57.1
5th-percentile male—standing 159.5 (67.8) 41.2 (16.2) 42.0
5th-percentile male—stool (30 in.) 150.2 (59.1) 50.5 (19.9) 47.8
5th-percentile male—stool (24 in.) 135.0 (53.1) 65.7 (25.9) 55.2

TAF-ER562X_WINGR-09-0401-C004.indd 128 11/9/10 2:23:18 AM


Anthropometry and Biomechanics 129

Viewing
distance
Off-
normal
viewing
angle

Q2 Eye height
0° Elevation
Display
height

Floor

FIGURE 4.12 Viewing angle with heart output monitor in a low-mount position. Q2 is the monitor
display being evaluated.

4.2.1.6 Viewing Angles, Summary, and Implications


Measurements and derivations of horizontal and vertical viewing angles are based on the
following assumptions:

• The sample of observed users is representative of target customer users.


• Typical and worst-case observations are representative of the target viewing
scenarios.
• User populations are represented by 5th- to 95th-percentile U.S. male and female
populations.
• The tasks being performed are not so critical that larger percentages of the popula-
tion (i.e., greater than the central 90%) must be accommodated.

With these assumptions, the viewing envelope for the cardiac output monitor and display
(required maximum viewing angles) were calculated to be as follows:

• Horizontal viewing angle: ±45 degrees


• Viewing angle (viewed from below the display): –5 7degrees
• Viewing angle (viewed from above the display): +64 degrees

4.2.2 KEYBOARD HEIGHT IN A DIAGNOSTIC SYSTEM WORKSTATION


A PC-based workstation must be designed for use with a clinical chemistry analyzer diag-
nostic medical system. It is well known that laboratory technician operators range from
small females to very tall males. The design required a determination of the keyboard

TABLE 4.12
Calculated (Off-Normal) Viewing Angles with Low Vertical Position, in cm (in.)
User User Eye Height Difference Viewing Angle (°)
95th-percentile female—standing 163.0 (64.2) 74.7 (29.4) 58.5
95th-percentile male—standing 182.5 (71.9) 94.2 (37.1) 64.1

TAF-ER562X_WINGR-09-0401-C004.indd 129 11/9/10 2:23:18 AM


130 Handbook of Human Factors in Medical Device Design

FIGURE 4.13 Keyboard of a clinical chemistry analyzer diagnostic medical system.

height adjustment range. Figure 4.13 shows the keyboard workstation and the diagnostic
system. The operators must stand since they need to interact with the system (e.g., place
samples into the appropriate loading areas or load reagents and calibration samples). Thus,
a standing operator occasionally would use the keyboard. The designers wanted to provide
an adjustable keyboard tray to accommodate a reasonable range of users.
Table 4.13 shows data for estimating keyboard height adjustability range. A practical adju-
stability range would be from 5th-percentile Asian female to a 95th-percentile Caucasian
male. These values establish a keyboard height range from 92.7 to 114.0 cm. Note that if
this operation were critical, the range should be extended by use of the 1st- to 99th- percen-
tile values (90.7–117.1 cm).

4.2.3 FINGER CLEARANCE SPACE CALCULATIONS


Clinicians and medical device operators use equipment with their hands placed in restricted
spaces. This case study illustrates several examples using hand dimension data from
Table 4.3 and Figure 4.2. These hand data are used to set design dimensions in critical hand
clearance areas for several medical devices.

TABLE 4.13
Keying Heights Required by 95th-Percentile Males and 5th-Percentile Females,
in cm (in.)
Percentile
Gender 5th 50th 95th Reference
Male 99.6 (39.2) 106.9 (42.1) 114.0 (44.9) White and Churchill (1971)
(computed)
Female (Japanese) 92.7 (36.5) 98.6 (38.8) 104.1 (41.0) National Aeronautics and Space
Administration (1978a,b,c)

TAF-ER562X_WINGR-09-0401-C004.indd 130 11/9/10 2:23:19 AM


Anthropometry and Biomechanics 131

FIGURE 4.14 Hand clearances for sample probe on blood analyzer prototype.

4.2.3.1 Prototype Blood Analyzer—Sample Probe Hand Clearance


Figure 4.14 shows the front surface of a foam core mock-up for a blood analyzer prototype.
Laboratory technicians sometimes need to insert a sample vial during a STAT (immedi-
ate) test of blood chemistry. They insert the vial of blood so that a pickup tube probe can
vacuum-draw an aliquot from the sample. The technician’s hand must clear the spaces
(labeled D1 and D2 in Figure 4.14). Because the sample vial typically is small, data taken
from Table 4.3 can be used to show a first approximation of hand thickness for the 95th-
percentile male (dimension 16). This hand thickness value, including the thumb, is 5.8 cm
and is the minimum design value for D1 in Figure 4.14. Similarly, the minimum clearance
for D2 would be set by dimension 15 (metacarpal hand thickness) for the 95th-percentile
male, which is 3.8 cm. These values represent minimum bounds, and using a larger range
is more likely to improve the design.

4.2.3.2 Portable Infusion Pump Access for Thumb Bolus Activation


Lockboxes commonly enclose home-use, portable infusion pumps to secure patient-con-
trolled administration of pain medication. The patient typically pushes a button at the
end of a pendant to request a bolus (i.e., dose) of medication. A program controls dosage
amount, the allowable lockout interval between doses, and the cumulative dose limit over
a specified time. Sometimes the pendant is not used, requiring that the access opening be
designed so patients can use their thumb or finger to push a bolus button incorporated on
the portable pump itself.
Figure 4.15 shows a pump, the clear plastic lockbox, and the opening for finger activation
of the bolus button. Table 4.3 provides data for thumb breadth at the interphalangeal joint

FIGURE 4.15 Finger access opening in a lockbox for a portable infusion pump. (Photos courtesy
of Abbott Laboratories.)

TAF-ER562X_WINGR-09-0401-C004.indd 131 11/9/10 2:23:20 AM


132 Handbook of Human Factors in Medical Device Design

FIGURE 4.16 Access opening for adding reaction vessels to clinical chemistry analyzer. (Photos
courtesy of Abbott Laboratories.)

(dimension 8) for the largest anticipated thumb size (i.e., the 95th-percentile male). The
minimum opening must be 2.6 cm plus a margin to avoid scraping the thumb against the
plastic and to allow more flexibility in the pushing action. Dimension 15 (hand thickness at
the metacarpal joint) would be appropriate for sizing the opening for the pointing finger if
it were used to activate the bolus button. A person with this finger size (the 95th-percentile
male) would require an even larger opening of 3.8 cm. Thus, this analysis shows the mini-
mum opening to be 3.8 cm.

4.2.3.3 Clinical Chemistry Diagnostic System Access Opening


Designers of a clinical chemistry diagnostic system that measures critical values of human
samples for potassium, calcium, sodium, and so on needed to determine access opening
size for the disposable reaction containers. The opening had to allow technicians to pour
new, small, clear plastic reaction vessels into a sorter mechanism, while also preventing
human access. It had been noted in previous risk analyses for use error that technicians
were tempted to reach into the sorting area to push down on the vessels to create more
space. This pushing action could damage the sorting mechanism, so designers considered
placing a grate at the top of the access opening. The grate (shown in Figure 4.16) needed
to be large enough to allow the free flow of new reaction containers but prevent the most
foreseeably small hand from passing through it.
Table 4.3, dimension 20 (minimum square access), shows 5.0 cm to be the smallest (i.e.,
5th-percentile female) hand of a user of this system. Dimension 12 (hand breadth at meta-
carpal) of 6.9 cm, also for the 5th-percentile female, is another data point to consider in the
design of section dimensions and shapes of grating. In this case study, these analyses led the
designers to abandon the use of a grate that attempted to solve two problems simultaneously
(i.e., easy ingress for the reaction containers while discouraging operator access).

4.3 GENERAL PRINCIPLES OF GOOD BIOMECHANICAL DESIGN


Medical device design must account for human physical capabilities and limitations. Failure
to consider biomechanical design issues can result in excessive muscle strength being
required to use the device, muscle fatigue, reduced endurance and ability to work with the
device for as long as is needed, increased numbers of errors, and longer task performance

TAF-ER562X_WINGR-09-0401-C004.indd 132 11/9/10 2:23:22 AM


Anthropometry and Biomechanics 133

times. Medical devices and workplaces designed using biomechanical principles can
increase productivity, reduce user fatigue, and produce a safer working environment.

4.3.1 CRITICAL DESIGN CONSIDERATIONS


4.3.1.1 Body Posture
Muscle strength in the trunk and extremities is greatest when the human body is in a nearly
fetal-like position, that is, slightly crouched, with the arms and legs bent, the arms near the
body, and the wrists in a nearly neutral posture. The strength capability of the body’s joints
in these positions maximizes the tolerance to loads and forces imposed on them. Obviously,
medical devices rarely are used when the body is in the fetal position. However, the fol-
lowing guidelines present two approaches to minimize the impact of physical activity and
stress on the human body during work tasks:

Guideline 4.16: Minimize Device and Load Weights


Generally, designers should minimize a medical device’s weight and the force required to
manipulate it.

Guideline 4.17: Seek Use in Neutral Postures


Designers should design medical devices so that users can work in as close to neutral postures
(i.e., relaxed joint positions) as possible while performing associated tasks.

These two approaches are ideal, but it is not possible to design every work situation to abide
by these guidelines. However, reducing load weights and force requirements as well as
focusing on correct body joint angles as much as is feasible will reduce stress on the body
and increase efficiency when working with medical devices.

4.3.1.2 Trade-Offs
The human body can be thought of as a system of links as when, for example, the hand/
wrist, elbow, and shoulder joints work together to operate a device. This often makes it dif-
ficult to design devices to minimize the force exerted by a joint and properly position that
joint at the same time. In these situations, follow these guidelines:

Guideline 4.18: Proper Position of Critical Body Part(s)


Designers should determine the body part(s) most relevant for the task, most often used, or
most likely to be stressed and design the given medical devices so that they can keep that body
part or parts in proper (i.e., most comfortable) position.

Guideline 4.19: Stable and Preferred Use Positions


Medical devices should enable users to assume comfortable, stable use positions. Often, it is
advantageous to allow users to interact with a device from either a standing or seated position,
whichever is most comfortable or preferable for them at a given time.

Guideline 4.20: Neutral Hand Position


To avoid joint stress and fatigue, medical devices should enable users to maintain a hand
position as close to neutral as possible, requiring only temporary excursions (if at all) to

TAF-ER562X_WINGR-09-0401-C004.indd 133 11/9/10 2:23:23 AM


134 Handbook of Human Factors in Medical Device Design

100%

Strength
50%

0
0 5 10 min
Endurance

FIGURE 4.17 The relationship between isometric muscle strength and muscle endurance. The
curve illustrates that lower amounts of strength required when performing a physical activity result
in an increased length of time that the muscle can sustain that activity.

accomplish tasks. The hand is neutral when positioned as if one was shaking hands, with a
straight (neither flexed nor extended) wrist and slightly closed hand with fingers curled.

Guideline 4.21: Comfortable Working Posture


Medical devices should allow users to maintain a comfortable working posture. For example,
a standing workstation should enable users to keep their back straight rather than require them
to lean forward, which could cause back strain.

4.3.1.3 Endurance
Muscular endurance is a function of the amount of strength exerted by a muscle or group
of muscles. As shown in Figure 4.17, one can exert a muscle maximally for only a few sec-
onds before fatigue sets in. However, the lower the percentage of one’s maximal strength
that a task requires, the longer that strength level can be sustained before fatigue or physical
discomfort occurs. Muscle endurance also is impacted by the length of time that muscle
is used. Even if the body is in an “ideal” posture during device use and only low levels of
muscle strength are required, the involved muscles still will fatigue if they must be exerted
continually or without adequate rest. Therefore, designers should follow these guidelines:

Guideline 4.22: Minimize Static Postures


To avoid fatigue and strain, medical devices should not require users to maintain a specific
posture for lengthy periods of time.

Guideline 4.23: Minimize Required Forces


Medical devices should not require users to exert forces that could cause strain or rapid fatigue.

Guideline 4.24: Rest Periods between Uses


When a medical device will require prolonged periods of muscle exertion, it should also allow
for brief and intermittent periods of rest. Rest periods can delay the onset of fatigue that may
degrade task performance or lead to the development of cumulative trauma.

TAF-ER562X_WINGR-09-0401-C004.indd 134 11/9/10 2:23:23 AM


Anthropometry and Biomechanics 135

4.3.1.4 Repetitive Motions


Continual use of the same body part to perform a task is known to increase musculo-
skeletal injury risk. Examples of repetitive activities include: (1) continuous data entry on
a computer keyboard, such as used in a computerized physician order entry system; (2) use
of the same finger to activate a tool, such as a grasping device for endoscopic surgery; and
(3) repeated rotation of the shoulder to reach into and out of a surgical field (e.g., hand-
assisted laparoscopic nephrectomy).

Guideline 4.25: Avoid Repetitive Motions


To prevent cumulative trauma disorders (e.g., carpal tunnel syndrome), medical devices should
not require users to perform a substantial number of repeated motions (e.g., twisting a tool)
without appropriately timed rest periods.

Guideline 4.26: Flexible Body Positions


Medical devices should not force users to assume precise body positions (e.g., sitting postures
and hands grips). Rather, they should allow for flexible positioning so that users can vary how
they perform a task to reduce the chances of physical stress.

There are several ways to reduce the physical impact from performing repetitive motions.
Thus, designers should follow prior guidance regarding rest periods and strive to permit
work to be done in as close to neutral body positions as possible. These topics are addressed
in more detail in Chapter 16, “Hand Tools.”

4.3.2 SPECIAL POPULATIONS


Physical capabilities, especially strength and endurance, differ among defi ned sub-
groups within the population. These include children, the aged, or those with certain
physical impairments, such as muscular dystrophy. Although the previously mentioned
biomechanical concepts (e.g., the impact of body posture, development of muscular
fatigue) still are relevant for those groups, baseline differences may exist among these
groups and with respect to the overall population (refer to Chapter 18, “Home Health
Care”). It is not possible to discuss all situations in which physical limitations impact
medical device design. However, as a first step when designing devices for use by a
special population, consult the references at the end of this chapter. Also, consider the
following guidelines:

Guideline 4.27: Analyze and Understand Special Populations


As appropriate, anthropometric analyses should address the requirements of special user
populations, such as children, pregnant women, obese individuals, elders, persons with spe-
cific disabilities (e.g., paralysis, fused joints, arthritis), people with specific medical condi-
tions (e.g., Parkinson's Disease), and people with physical anomalies (e.g., swollen body parts,
growths, deformities).

Guideline 4.28: Accommodate Adaptive Equipment


Medical devices should physically accommodate users’ adaptive equipment, including wheel-
chairs, crutches, splints, eyewear, and hearing aids.

TAF-ER562X_WINGR-09-0401-C004.indd 135 11/9/10 2:23:23 AM


136 Handbook of Human Factors in Medical Device Design

4.3.3 DESIGN GUIDELINES FOR TASKS INVOLVING LIFTING


There are numerous situations in which designers may need to evaluate the risk of injury
to individuals who must perform physically demanding tasks, such as lifting heavy, criti-
cal medical devices onto a mobile stretcher prior to transporting a trauma patient. There
are a variety of ergonomic assessment tools available for use in conducting such evalua-
tions. Each has its strengths and weaknesses, but it is important to use the appropriate tool
to determine if there is injury risk associated with working in an environment or using a
particular medical device.
One category of assessment tools relates to the potential for injury to the low back when
performing a lifting task, such as moving an obese patient from one bed to another. Several
of the most frequently used tools for these types of evaluations are described below.

4.3.3.1 NIOSH Revised Lifting Equation


The National Institute for Occupational Safety and Health (NIOSH) developed guidelines
to determine the recommended (i.e., safe) weight limit for a lifting task (Waters, Putz-
Anderson, Garg, and Fine, 1993). These limits indicate load weights that nearly all healthy
workers could lift without increasing their risk of developing low back pain. The guidance
is based on the input of data for seven factors into a lifting equation:

• Horizontal location of the load, measured as the distance of the hands away from
the midpoint between the ankles
• Vertical location of the load, which is the distance of the hands above the floor
when holding the object
• Vertical distance the load travels during the lift (i.e., the absolute value of the dif-
ference between the vertical heights at the origin and destination of the lift)
• Amount of twisting required during the lift (also called the asymmetry), measured
as the angle between the location of the load when it is lifted and the individual
when standing in a “neutral” body posture
• Lifting frequency of the task, measured as the average number of lifts required per
minute
• Length of time lifting is required, which takes into account both work time and
rest periods
• Quality of the hand-to-object coupling, which assesses the handling ease of the
load

The physical dimensions used as inputs for this equation are illustrated in Figure 4.18.
As an example of its use, consider the task of stocking devices in a hospital supply
room with the following working conditions. Boxes of supplies must be held 15 inches
in front of the body and are initially located 18 inches above the floor. Employees lift
these boxes onto a table 40 inches above the floor, and the lifting can be done directly in
front of the body. The rate of lifting is (on average) once per minute, the process takes
four hours to complete, and there are good handles on the supply boxes. Entering this
information into the NIOSH lifting equation would produce a recommended weight
limit of 9.5 kg (21 lbs.). This indicates that lifting boxes less than 9.5 kg (21 lbs.)
is “safe,” but objects weighing more would increase the risk that individuals could
develop low back pain.

TAF-ER562X_WINGR-09-0401-C004.indd 136 11/9/10 2:23:24 AM


Anthropometry and Biomechanics 137

Asymmetry

Vertical
travel
distance

Vertical
location

Horizontal
location

FIGURE 4.18 Illustration of the physical dimensions used as inputs for the NIOSH revised lifting
equation.

To ease the numerical complexity of these calculations, computer software packages


have been designed to compute these weight limits. NIOSH has developed a list of software
sources that are available on its Web site at http://www.cdc.gov/niosh.

4.3.3.2 ACGIH Lifting Threshold Limit Values


The American Conference of Governmental Industrial Hygienists (ACGIH) has developed
threshold limit values (TLVs) for recommended lifting conditions under which a majority
of workers purportedly can be exposed repeatedly without developing low back and shoul-
der disorders (ACGIH, 2002). Information input into this assessment tool is similar to that
of the NIOSH lifting equation. The TLVs are presented in three tables of weight limits,
determined as a function of the following:
• Task duration, or the total length of time the activity is performed in one day
• Lifting frequency of the task, or the number of lifts performed per hour
• Vertical height zone at the beginning of the lift, or the location of the hands at the
beginning of the lift
• Horizontal location of the load at the beginning of the lift, or the distance of the
hands away from the midpoint between the ankles

Using the previous hospital supply room lifting task example, the TLV under the stated
working conditions is 7.0 kg (15.4 lbs.). This indicates that appropriate control measures
should be implemented if supplies are lifted that exceed 7.0 kg (15.4 lbs.) in weight. Note
that this weight limit is lower than that calculated previously using the NIOSH lifting equa-
tion. This is because different data and methodologies were used to generate the lifting

TAF-ER562X_WINGR-09-0401-C004.indd 137 11/9/10 2:23:24 AM


138 Handbook of Human Factors in Medical Device Design

FIGURE 4.19 The industrial lumbar motion monitor (iLMM). (Photo courtesy of Biodynamic
Solutions, Inc.)

limits. The choice of assessments depends on how each tool’s assumptions compare to the
lifting task being evaluated.

4.3.3.3 Industrial Lumbar Motion Monitor Risk Assessment System


The industrial lumbar motion monitor (iLMM) and its associated risk model determine
the probability that a job will produce a low back injury rate comparable to other lifting
tasks having high numbers of back injuries (Marras et al., 1993). Thus, the physical design
of a medical device may impact how a user interacts with it and, thus, the resulting low
back injury rate. Figure 4.19 shows the iLMM device being worn to record trunk motions
as a lifting task is performed. The iLMM records the instantaneous position, velocity, and
acceleration of the trunk in the sagittal (forward bending), lateral (side bending), and trans-
verse (twisting) planes of motion. The iLMM’s risk model incorporates five factors. Two
factors are taken from the workplace:

• Lift rate (the number of lifts required per hour across all job tasks)
• External load moment, which is a function of the weight of the load and the hori-
zontal distance it is handled from the spine

The other three factors used in the risk model are data derived from the iLMM task
measurements:

• Sagittal flexion (i.e., the maximum amount of forward bending required of the job)
• Twisting velocity (i.e., the average speed of axial rotation)
• Lateral velocity (i.e., the maximum speed of lateral movement)

This analytical approach could be beneficial in the design of some medical devices, such as
patient lifting systems for hydrotherapy, or for the positioning and adjustment of mobile stretch-
ers used in ambulances or evacuation helicopters. The iLMM system of hardware and software
is commercially available, currently through NexGen Ergonomics (Montreal, Canada).

TAF-ER562X_WINGR-09-0401-C004.indd 138 11/9/10 2:23:24 AM


Anthropometry and Biomechanics 139

4.3.3.4 Psychophysical Limits


Psychophysical limits refer to estimates of what individuals believe they can safely lift,
lower, push, pull, and carry. These limits were derived from groups of males and females
who were tested performing these types of tasks, and the subjects reported how much
handling they felt was “safe.” From this, tables were created containing acceptable lifting/
force limits, as a function of (1) distance of the force exertion, (2) frequency of the activity,
(3) vertical height at which the force is applied, and (4) percentage of an industrial popula-
tion believed capable of safely performing the task (Snook and Ciriello, 1991).
These psychophysical limits are some of the only available data regarding safe levels of
pushing, pulling, or carrying. However, the definition of “safe” was derived subjectively
rather than through objective testing. Thus, this information should be used as a start-
ing point when designing medical devices requiring these types of activities. Numerous
ergonomics reference books include these psychophysical tables, and the Web site of the
Liberty Mutual Insurance Company (Boston, MA) offers these tables interactively.

Guideline 4.29: Tools to Assess Physical Limits


Designers should use one or more quantitative assessment tools to ensure that a task does not
exceed safe lifting limits. For example, the NIOSH revised lifting equation or the ACGIH
TLVs should be used for situations in which the lifting will be slow, smooth, and steady (e.g.,
no “jerky” motions). Consider using the Industrial Lumbar Motion Monitor Risk Assessment
System when the lifting performed is highly dynamic and repetitive. Consider using psy-
chophysical limits when the activity requires pushing, pulling, or carrying.

4.3.4 DESIGN GUIDELINES FOR TASKS INVOLVING USE OF THE UPPER EXTREMITY
Many medical devices require activity by the shoulders, hands, and wrists. Thus, it is
important to assess the injury risk to these body parts during device use.

Guideline 4.30: Injury Risk Assessment


Designers should use one or more quantitative assessment tools to ensure that the tasks involv-
ing the shoulders, hands, and wrists do not pose a risk of injury.

Several ergonomic tools are available to perform such assessments. As with lifting, upper
extremity assessment tools also have their strengths and weaknesses. The most appropri-
ate tool used will depend on the device being designed and how it is to be used. The more
common tools to evaluate upper extremity injury risk are described below.

4.3.4.1 Strain Index


The Strain Index (Moore and Garg, 1995) evaluates a job’s risk of producing a musculosk-
eletal disorder to the distal upper extremity (i.e., the hand/wrist joint). Six factors form the
basis of the Strain Index:

• Intensity of exertion, which relates to the job’s force requirements


• Duration of exertion (length of time the force is maintained)
• Efforts required per minute, or the repetitiveness of the job
• Hand/wrist posture, relative to the “neutral” position

TAF-ER562X_WINGR-09-0401-C004.indd 139 11/9/10 2:23:25 AM


140 Handbook of Human Factors in Medical Device Design

• Speed of work (i.e., the job pace)


• Task duration, or the total length of time the job is performed

The Strain Index is available as part of several commercially available computer-based


packages, including ErgoIntelligence (NexGen Ergonomics, Montreal, Canada), ErgoTrack
(ErgoTrack, Efland, NC), Job Evaluator Toolbox (Ergoweb, Inc., Park City, UT), and Job
Hazard Pro (Production Technology, Tampa, FL).

4.3.4.2 Rapid Upper Limb Assessment


The Rapid Upper Limb Assessment (RULA) tool can evaluate work with a medical device
for which there is injury risk to the neck, shoulder, upper and lower arms, and hand/wrist
(McAtamney and Corlett, 1993). It produces a score based on task repetition, posture, and
force that can be compared with a recommended action level. An example of a medical
device that could cause neck strain would be an optometric measurement device in which
the head must be kept steady during peripheral optical field measurements.
RULA-based software is available commercially in several computerized forms, includ-
ing ErgoEASE (Ease, Inc., Mission Viejo, CA), ErgoIntelligence (NexGen Ergonomics,
Montreal, Canada), ErgoSURE Pro (Magnitude, Branchburg, NJ), Job Evaluator Toolbox
(Ergoweb, Inc., Park City, UT), Job Hazard Pro (Production Technology, Tampa, FL), Jack’s
Task Analysis Toolkit (Siemens Product Lifecycle Management Software Inc., Plano, TX),
and Osmond Ergonomic Workplace Solutions (Wimborne, UK).

Guideline 4.31: Unacceptable Postures or Forces


The designer should evaluate if medical device operation exposes users to unacceptable
postures and forces to the upper extremity.

4.3.5 DESIGN GUIDELINES TO DETERMINE STRENGTH REQUIREMENTS


4.3.5.1 Three-Dimensional Static Strength Prediction Program
The 3-Dimensional Static Strength Prediction Program is a software application available
for purchase through the University of Michigan (Ann Arbor, MI). This program deter-
mines static strength requirements for activities such as lifting, pressing, pushing, and pull-
ing (Chaffin and Andersson, 1999). The program provides estimates of the percentages of
males and females in the population who have the strength capability (at the elbow, shoul-
der, torso, hip, knee, and ankle) for a given task. It also gives an estimate of the amount of
spinal compression produced for the activity.

Guideline 4.32: Strength Prediction


Medical designers should evaluate whether acceptable levels of a user population have the
static strength necessary to adequately use a device.

Guideline 4.33: Avoid Sustained Forces


Medical devices should not require users to apply a force for extended periods of time (e.g.,
several minutes) because such exertion can cause fatigue and, in the case of repeated applica-
tions, lead to cumulative trauma.

TAF-ER562X_WINGR-09-0401-C004.indd 140 11/9/10 2:23:25 AM


Anthropometry and Biomechanics 141

Guideline 4.34: Avoid Unnecessarily Precise Forces


Generally, medical devices should not require users to apply unnecessarily precise forces to
exert proper control.

Guideline 4.35: Avoid Pressure Points


Medical devices should not apply undue pressure (i.e., sharp or blunt force) to any body part.
In addition to causing discomfort, excess pressure can reduce blood flow to a body part, injure
soft tissue, and cause other physical ailments.

4.4 CASE STUDIES IN BIOMECHANICAL DESIGN


4.4.1 KNOB TWISTING FORCES ON POLE CLAMPS
Often, critical care medical devices such as infusion pumps, internal feeding pumps, and
continuous blood pressure monitors need to be clamped to a bedside pole. Knobs secure
many pole-clamping devices. The main design consideration is to provide a mechanism
that will tightly secure the device without requiring high torque on the knob so that a
majority of health care workers have the ability to secure and release the clamp. The result
of an unsecured device is that it can slide down the pole, causing personal injury or damage
to the device. A pole clamp that is tightened by an especially strong user may preclude a
weaker user from removing the device without mechanical assistance.
There are three types of grips commonly used for twisting a pole clamp knob. Figure 4.20
shows a finger grip, which results in the least torque possible. Figure 4.21 shows a grip typi-
cal for grasping a jar, for which larger torque forces are possible. Figure 4.22 shows the

FIGURE 4.20 Knob turning with a finger grip. (Photo courtesy of Abbott Laboratories.)

FIGURE 4.21 Knob turning with a jar lid grip. (Photo courtesy of Abbott Laboratories.)

TAF-ER562X_WINGR-09-0401-C004.indd 141 11/9/10 2:23:25 AM


142 Handbook of Human Factors in Medical Device Design

FIGURE 4.22 Knob turning with a handlebar power grip. (Photo courtesy of Abbott
Laboratories.)
most powerful grip, such as that used to grasp a handlebar (Hanson and Israelski, 1977).
The handlebar grip may not be possible for some pole clamps because of restricted space in
the area where the clamp is mounted on the device.
The design issue for this case involves determining the tightening torque force that a
user with limited strength can apply to a pole clamp using a 6.9-cm-diameter round knob,
as shown in these figures.

4.4.1.1 Analysis
In this case study, the exact conditions for which a designer is seeking biomechanical data
do not exist in precisely the desired form. Extrapolation often is needed to use data col-
lected under similar (but not exact) circumstances to estimate the forces needed for design
input. A good source of biomechanical data for people with varying amounts of dexterity
comes from the Department of Trade and Industry (2002). These data were used to derive
Figure 4.23, which shows clockwise tightening torque for 5th-percentile females with no
impairment, as a function of knob diameter.
Torque (in.-lbs.) vs. knob diameter (in.)
5th percentile, combined male and female, non-disabled,
jar lid grip turning clockwise
18.00
Male
16.00
Female
14.00

12.00

10.00

8.00

6.00

4.00

2.00

0.00
0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00 5.50
Knob diameter (in.)

FIGURE 4.23 Torque generated versus knob diameter.

TAF-ER562X_WINGR-09-0401-C004.indd 142 11/9/10 2:23:26 AM


Anthropometry and Biomechanics 143

The texture and shape of a knob will affect the forces generated. For example, knurled
knobs increase friction with the hand and allow one to apply larger forces, compared to
smooth knobs. Triangular knobs allow the generation of even higher torque forces, as do
knobs that have small extensible foldout levers (Kohl, 1983).
To preclude stronger users from tightening something so much that it exceeds the
loosening capabilities of weaker users, the design should employ a torque limiter. An
example of a torque-limiting device would be a slipping clutch in a pole clamp design
similar to automotive gas cap tightening mechanisms. See Chapter 7, “Controls,” for more
information about choosing knob designs. A Department of Trade and Industry (2002)
study showed that males and females at the 5th percentile can apply a force of 5.3 N on a
knurled, 6.9-cm-diameter knob using the lid jar grip (Figure 4.22). Since the population
sampled for this study was evenly split between genders, and female upper body strength
is approximately 70% that of males, a gender-combined force of 5.3 N would be reduced
by 0.82 for females. The resulting 5th-percentile female force capability would be 4.3 N
of torque. If the clamp requires more than 4.3 N of torque to tighten, then less than 95%
of female users will be able to place the device securely on the pole. Larger-diameter
knobs are an option, although knobs with diameters greater than 10.7 cm no longer have an
advantage, as users will begin to have trouble gripping them.
The use of the more powerful handle grip also is more difficult in tighter spaces. Existing
biomechanical data often will not be an exact match, so it may be necessary to conduct a sep-
arate strength measurement study using the parameters of the proposed design. Performing
a special study also would be appropriate if risk levels from the analysis are estimated to be
high enough that it does not seem reasonable to extrapolate from existing data.

4.4.2 BENDING FORCES ON AN AUTOINJECTOR DEVICE


Designers of a penlike device for the autoinjection of a critical drug wanted to know the
reasonably foreseeable bending force that might be exerted by a strong user. As users removed
the device’s protective end caps, the concern was that they might exert excessive torque on
the cylindrical body of the device, by holding each end and bending it (Figure 4.24). As
noted in previous case studies, existing biomechanical data only provides an approximation
of the information needed by the designers.

4.4.2.1 Analysis
Table 4.8 shows values for the power grasp that are a reasonable approximation to the
bending force at issue in this case study. The bottom row describing power grasp is relevant
and shows a maximum mean force of 366 N, with a standard deviation of 53 N. Table 4.14
presents static forces exhibited by the strongest 50th- and 95th-percentile males when

FIGURE 4.24 Autoinjector drug delivery pen, 7 inches long by 1 inch in diameter (longitudinal
bending/breakage with one hand at each end).

TAF-ER562X_WINGR-09-0401-C004.indd 143 11/9/10 2:23:27 AM


144 Handbook of Human Factors in Medical Device Design

TABLE 4.14
Bending Forces Generated by the Strongest Males, in N (lbs.)
One-Hand Pulling Force Mean (50th percentile) Standard Deviation 95th Percentile
Male (machinists) 365.6 (82.2) 52.9 (11.9) 452.8 (101.8)

pulling down a fixed cylindrical handle approximately 2.5 cm in diameter with one hand.
(The original source does not specify the handle diameter.)
The solution to this challenge is to design the device to have a maximum bending force
of 46.2 kg, which can be generated on a cylindrical handle by a 95th-percentile male. There
were no data specific to a two-handed maximum bending force on a cylinder. The approxi-
mation for one-handed operation still is reasonable if one assumes that the second hand
holds the cylinder fixed and makes one end of the cylinder the fulcrum point for calculating
a moment arm of 17.8 cm. If the user were to fix the center of the cylinder on a rigid surface
as the fulcrum, then each hand theoretically could generate the 46.2 kg force over shorter
lengths.

4.4.3 SNAP-ON LID REMOVAL FORCES


This case study examined the forces that could be exerted to remove a lid from a vial. The
vial contained medication for clinician or patient access and did not require a screw-on lid
or a childproof container. The holding force of the lid needed to be sufficient to remain
secure during transportation and storage but at the same time be able to withstand an
attempt to open it by someone without proper tools. Thus, the two design questions were
the following:

• What is the foreseeable smallest force that should be exerted to reliably remove a
snap-on lid from a medical vial?
• What is the foreseeable largest force that can be exerted to remove a snap-on lid
from a medical container that must be securely fastened until it is mechanically
removed?

4.4.3.1 Analysis
Both questions involved the same push forces with the thumb against a snap-on lid
(Figure 4.25). However, quite different strength information was needed. A calculation was
needed of both the 5th-percentile female force and the 95th-percentile male force. Thumb
push against a lid was not found in the published biomechanical literature, but a reasonable
approximation was found in Schoorlemmer and Kanis (1992). As shown in Figure 4.26,
the presumed thumb pushing force positions were calculated for a plunger disk 2.0 cm in
diameter, with disk clearance of 1.5 cm above the tabletop. Test subjects pushed from any
free-posture position desired and in a seated position with elbows at a 90-degree angle.
Values from the free-posture position were larger and seemed to be more appropriate to
how snap-on lids might be handled in real-world conditions. Table 4.15 shows the val-
ues from the Schoorlemmer and Kanis study and the derived values for the 5th-percentile
female force (0.9 kg) and the 95th-percentile male force (16.1 kg).

TAF-ER562X_WINGR-09-0401-C004.indd 144 11/9/10 2:23:27 AM


Anthropometry and Biomechanics 145

FIGURE 4.25 Snap-on lid vial (1.6 inches long by 0.8 inches in diameter).

FIGURE 4.26 Thumb-pushing-force positions.

TABLE 4.15
Calculations of Thumb-Pushing Forces for the 5th and 95th Percentiles
Mean (50th percentile) Standard Deviation (SD) Force (Percentile)
Male 19.5 pounds 9.7 pounds 35.4 pounds (95th)
Female 15.3 pounds 8.1 pounds 2.0 pounds (5th)
Note: Conversions used were 1 N = 0.2247 pound-force; 95th or 5th percentile = mean ±
1.645(SD), assuming normal distributions.

TAF-ER562X_WINGR-09-0401-C004.indd 145 11/9/10 2:23:27 AM


146 Handbook of Human Factors in Medical Device Design

This case study illustrates that, if there is any doubt about the applicability of existing
biomechanical information, one should conduct a specific data gathering study. This is
especially true if the tasks in question are known to be high risk.

4.4.4 PULLING FORCES ON IV TUBING


Designers of intravenous (IV) tubing and accessories wanted to determine a reasonably
foreseeable worst-case pulling force on IV tubing that might stem from patient or user
abuse, whether purposeful or inadvertent. The expected pull forces would determine the
mechanical breaking forces of the tube. The tube is attached to cassette pumping units
that are typically inserted into volumetric infusion pumps. Some infusion pumps do not
use proprietary cassettes for fluid pumping. Instead, they allow the IV tubing to be placed
directly into the infusion pump while the peristaltic action of mechanical “fingers” pumps
the IV fluids through the tubing.

4.4.4.1 Analysis
Inadvertent pulling on tubes can arise from numerous situations. Use scenarios involving
pull forces include the following:

• The user moves while still accidentally holding on to the tubing. This likely could
produce only a very small force (e.g., less than 27 N).
• The user snags the tubing with an article of clothing or a hospital bracelet and
moves with a large pull force before noticing the resistance and stopping. This
force likely would be considerably less than the deliberate forces that have been
published (e.g., less than 623 N standing, as described below).
• The user pulls on tubing while attempting to remove the cassette because of a mis-
understanding of the proper way to eject the cassette. This force could approach or
exceed the measured biomechanical force of 623 N capable from a 95th-percentile
or higher male.
• The user loses his footing and starts to fall while holding on to the tubing. The
user’s full body weight becomes the pulling force. For 95th-percentile males, this
force could be greater than 956 N (White and Churchill, 1971).

A reasonable maximum pulling force for 95th-percentile males (a conservative worst-case


scenario for both patients and nurses) is 623 N. This is derived from the 95th-percentile static
pulling force curve shown in Figure 4.27 (Chaffin, 1972). The maximum force is a function of
the position of the hands (these curves are for standing, two-handed pulls). Note that the max-
imum, deliberate pull force is from a position 50.8 cm above the floor and 50.8 cm in front of
the ankles. The 50th-percentile male force would be 556 N (Chaffin, 1972). The use of more
conservative forces (at higher percentiles such as the 97th or 99th) may be overly restrictive.
However, one would need to extrapolate from these curves to calculate the higher values.

4.4.5 PULLING FORCE TO REMOVE A PLUNGER FROM A VIAL


This case study determines the reasonable force expected from the weakest clinical users
to retract an injector plunger from a medication vial. Figure 4.28 shows a sketch of a medi-
cation vial that is activated by a plunger. The vial typically is threaded onto a medication
container and, in turn, connected to tubing as part of a medication infusion system.

TAF-ER562X_WINGR-09-0401-C004.indd 146 11/9/10 2:23:28 AM


Anthropometry and Biomechanics 147

80

70

60
Vertical hand height above ankles (in.)

50

40

30
75 26
140 100 50
20

10

0
10 20 30 40 50 60 70
Horizontal hand position in front
of ankles (in.)

FIGURE 4.27 Male static pulling forces (95th percentile while standing using both hands in differ-
ent vertical and horizontal hand positions). Two-dimensional surfaces on the x-y plane show forces
in pounds. (From Chaffin, D.B., Some Effects of Physical Exertion, Western Electric/University of
Michigan Report, AT&T, New York, 1972. With permission.)

4.4.5.1 Analysis
Because the goal is to determine the weakest forces likely to occur in the user population,
the focus is on the 5th-percentile female. If the retraction force of the plunger exceeds
the 5th-percentile female value, then more than 5% of female users of these vials will
be unable to retract the plunger. Alternatively, the design could specify a retraction force

Plunger

FIGURE 4.28 Plunger-activated medication vial.

TAF-ER562X_WINGR-09-0401-C004.indd 147 11/9/10 2:23:28 AM


148 Handbook of Human Factors in Medical Device Design

that accommodates those below the 5th percentile (e.g., the 1st percentile). A mechanical
lower bound would be established by the design parameters and the need to avoid inad-
vertent plunger movement during shipping or normal handling. The risk analysis would
assess the consequences of a small number of users being unable to retract the plunger
without a tool.
Information about “hook grip” forces (Table 4.8) is the best match for plunger retraction
force. The force data from use of all digits is a reasonable approximation for the retrac-
tion movement. Depending on the size of the plunger handle, not all fingers could be used.
However, use of at least two fingers is likely, and therefore the “all-digits-combined” condi-
tion is a reasonable choice.
The resulting pulling force for the weaker of the two male populations studied is 108
N, with a standard deviation (sd) of 39 N. Assuming a normal distribution, the calculated
5th-percentile male pulling force is 43.6 N (95th or 5th percentile = mean ± 1.645x[sd],
assuming normal distributions). Since female force data were not published for this hand
configuration, one may apply common conversion factors. Various gender conversion
factors are described in the biomechanical literature, ranging from 0.55 to 0.70 (Kroemer
et al., 1997; Pheasant, 1996). The lowest conversion factor would be the most conservative
for this case study. Thus, the resulting 5th-percentile female plunger retraction force would
be 24.0 N.

ANTHROPOMETRY-RELATED RESOURCES
Brown, R., Rogers, N., Ward, J., Wright, D., and Jeffries, G. (1995). The application of an anthro-
pometric database of elderly and disabled people. Biomedical Sciences Instrumentation, 31,
235–239.
Chung, K. and Weimar, W. (1989). Anthropometric Studies for the Physically Disabled Population—
Vol. II, Spinal Cord Injury (Report No. UVA-REC 102-89). Charlottesville: University of
Virginia, Rehabilitation Engineering Center.
Das, B. and Kozey, J. W. (1994). Structural anthropometry for wheelchair mobile adults. Proceedings
of the 12th Triennial Congress of the International Ergonomics Association, Human Factors
Association of Canada, Toronto, Ontario 3: 63–65.
Diffrient, N., Tilley, A. R., and Bardagjy, J. C. (Henry Dreyfuss Associates). (1981). Humanscale
4/5/6. Cambridge, MA: MIT Press.
Diffrient, N., Tilley, A. R., and Bardagjy, J. C. (Henry Dreyfuss Associates). (1981). Humanscale
7/8/9. Cambridge, MA: MIT Press.
Gordon, C. C., Churchill, T., Clauser, C. E., Bradtmiller, B., McConville, J. T., Tebbetts, I., and
Walker, R. A. (1989). 1988 Anthropometric Survey of U.S. Army Personnel: Summary Statistics
Interim Report (Natick-TR-89/027). Natick, MA: U.S. Army Natick Research Development
and Engineering Center.
Greiner, T. M. (1990). Hand Anthropometry of U.S. Army Personnel (AD-A244-533). Natick, MA:
U.S. Army Natick Research Development and Engineering Center.
Hobson, D. A. and Molenbroek, J. F. M. (1990). Anthropometry and design for the disabled:
Experiences with seating design for the cerebral palsy population. Applied Ergonomics, 21(1),
43–54.
Hobson, D. A., Shaw, C. G., Monahan, L., and Mclaurin, C. (1987). Anthropometric Data for
Design of Specialized Seating and Mobility Devices: A Preliminary Report (pp. 480–482).
RESNA 10th Annual Conference: San Jose, CA: Rehabilitation Engineering Society of North
America.
Kagimoto, Y. (1990). Anthropometry of JASDF Personnel and Its Applications for Human
Engineering. Tokyo: Aeromedical Laboratory, Air Development and Test Wing, JASDF.

TAF-ER562X_WINGR-09-0401-C004.indd 148 11/9/10 2:23:30 AM


Anthropometry and Biomechanics 149

Kouchi, M., Yokoyama, K., Yamashita, J., Yokoi, T., Ogi, H., Yoshioka, M., Atsumi, H., and
Hotta, A. (1994). Human Body Dimensions Data for Ergonomic Design [Anthropometric
data of Japanese adults]. Tsukuba, Japan: National Institute of Bioscience and Human
Technology.
Kozey, J. and Das, B. (1997). The determination of the maximum reach envelope for wheelchair
mobile adults. In B. Das and W. Karwowski (Eds.), Advances in Occupational Ergonomics
and Safety 1997 (pp. 315–318). Burke, VA: IOS Press.
Kroemer, K. H. E. (2005). “Extra-Ordinary” Ergonomics: How to Accommodate Small and Big
Persons, the Disabled and Elderly, Expectant Mothers, and Children. London: Taylor &
Francis.
Nowak, E. (1996). The role of anthropometry in design of work and life environments of the dis-
abled population. International Journal of Industrial Ergonomics, 17, 113–121.
Nowak, E. (1997). Anthropometry for the needs of disabled people. In S. Kumar (Ed.), Perspectives
in Rehabilitation Ergonomics (pp. 302–338). Bristol, PA: Taylor & Francis.
Peebles, L. and Norris, B. (1998). Adultdata: The Handbook of Adult Anthropometric and Strength
Measurements—Data for Design Safety. London: Department of Trade and Industry.
Pryor, H. B. and Thelander, H. E. (1967). Growth deviations in handicapped children: An anthropo-
metric study. Clinical Pediatrics, 6(8), 501–512.
Singh, J., Peng, C. M., Lim, M. K., and Ong, C. N. (1995). An anthropometric study of Singapore
candidate aviators. Ergonomics, 38(4), 651–658.
Skelton, D. A., Greig, C. A., Davies, J. M., and Young, A. (1994). Strength, power and related func-
tional ability of healthy people aged 65–89 years. Age and Aging, 23, 371–377.
Snyder, R. G. (1977). Anthropometry and Biomechanics of Selected Populations (Technical
Report UM-HSRI-77-52). Morgantown, WV: National Institute for Occupational Safety and
Health.
Snyder, R. G. (1977). Anthropometry of Infants, Children, and Youths to Age 18 for Product Safety
Design. Ann Arbor: Highway Safety Research Institute, University of Michigan; Warrendale,
PA: Society for Automotive Engineers.
Steenbekkers, L. P. A. and Molenbroek, J. F. M. (1990). Anthropometric data of children for non-
specialist users. Ergonomics, 33(4), 421–429.
Stoudt, H. W. (1981). The anthropometry of the elderly. Human Factors, 23(1), 29–37.
United States Department of Health and Human Services, National Center for Health Statistics.
(1983). Health and Nutrition Examination Survey II, 1976–1980 (computer file). Ann Arbor,
MI: Inter-University Consortium for Political and Social Research.
White, R. (1982). Comparative Anthropometry of the Foot (TR-83/101). Natick, MA: U.S. Army
Natick Research, Development and Engineering Center.
Woodson, W. E., Tillman, B., and Tillman, P. (1992). Human Factors Design Handbook: Information
for the Design of Systems, Facilities, Equipment, and Products for Human Use. New York:
McGraw-Hill.

BIOMECHANICS-RELATED RESOURCES
Brown, D., Knowlton, R. G., Hamill, J., Schneider, T. L., and Hetzler, R. K. (1990). Physiological
and biomechanical differences between wheelchair-dependent and able-bodied subjects during
wheelchair ergometry. European Journal of Applied Physiology and Occupational Physiology,
60, 179–182.
Chaffin, D. B., Martin, B. J., and Andersson, G. B. (1999). Occupational Biomechanics (3rd ed.).
New York: John Wiley & Sons.
Eastman Kodak Company. (1989). Ergonomic Design for People at Work, Vol. II: The Design of
Jobs, Including Work Patterns, Hours of Work, Manual Materials Handling Tasks, Methods
to Evaluate Job Demands, and the Physiological Basis of Work. New York: Van Nostrand
Reinhold.

TAF-ER562X_WINGR-09-0401-C004.indd 149 11/9/10 2:23:30 AM


150 Handbook of Human Factors in Medical Device Design

Jarosz, E. (1996). Determination of the workspace of wheelchair users. International Journal of


Industrial Ergonomics, 17, 123–133.
Karwowski, W. and Marras, W. S. (1999). The Occupational Ergonomics Handbook. Boca Raton,
FL: CRC Press.
Kenward, M. G. (1971). An approach to the design of wheelchairs for young users. Applied
Ergonomics, 2(4), 221–225.
National Research Council, Institute of Medicine. (2001). Musculoskeletal Disorders and the
Workplace: Low Back and Upper Extremities. Washington, DC: National Academies Press.
Nowak, E. (1989). Workspace for disabled people. Ergonomics, 32(9), 1077–1088.
Salvendy, G. (2006). Handbook of Human Factors and Ergonomics (3rd ed.). New York: John
Wiley & Sons.

REFERENCES
American Conference of Governmental Industrial Hygienists. (2002). 2002 Threshold Limit Values
for Chemical Substances and Physical Agents and Biological Exposure Indices. Cincinnati,
OH: American Conference of Governmental Industrial Hygienists.
Chaffin, D. B. (1972). Some Effects of Physical Exertion (Western Electric/University of Michigan
Report). New York: AT&T.
Chaffin, D. B. and Andersson, G. (1999). Occupational Biomechanics (3rd ed.). New York: John
Wiley & Sons.
Consumer and Competition Policy Directorate, Department of Trade and Industry. (2002a). Specific
Anthropometric and Strength Data for People with Dexterity Disability. London, United
Kingdom: Consumer and Competition Policy Directorate.
Consumer and Competition Policy Directorate, Department of Trade and Industry. (2002b). Strength
Data for Design Safety—Phase 2. London, United Kingdom: Consumer and Competition
Policy Directorate.
Department of Trade and Industry. (2002). Specific Anthropometric and Strength Data for People
With Dexterity Disability. London, United Kingdom: Department of Trade and Industry.
Diffrient, N., Tilley, A. R., and Bardagjy, J. C. (Henry Dreyfuss Associates). (1981). Humanscale
1/2/3. Cambridge, MA: MIT Press.
Eastman Kodak Company. (1983). Ergonomic Design for People at Work, Volume I: Workplace,
Equipment, and Environmental Design and Information Transfer. Belmont, CA: Lifetime
Learning Publications.
Flugel, F., Greil, H., and Sommer, K. I. (1986). Anthropologischer Atlas. Berlin: Tribuene.
Gordon, C. C., Churchill, T., Clauser, C. E., Bradtmiller, B., McConville, J. T., Tebbetts, I., and
Walker, R. A. (1989). 1988 Anthropometric Survey of U.S. Army Personnel: Summary Statistics
Interim Report (Natick-TR-89/027). Natick, MA: U.S. Army Natick Research, Development
and Engineering Center.
Hanson, B. L. and Israelski, E. W. (1977). Human factors engineering in the outside plant: Bringing
out the best. Bell Laboratories Record, February, 30–55.
Houy, D. A. (1983). Range of joint motion in college males. Proceedings of the Human Factors
Society 27th Annual Meeting, Human Factors Society, Santa Monica, CA, 374–378.
Kagimoto, Y., ed. (1990). Anthropometry of JASDF Personnel and its Applications for Human
Engineering. Tokyo: Aeromedical Laboratory, Air Development and Test Wing, JASDF.
Kohl, G. A. (1983). Effects of shape and size of knobs on maximal hand-turning forces applied by
females. Bell System Technical Journal 62(6): 1705–1712.
Kroemer, K. H. E., Kroemer, H. B., and Kroemer-Elbert, K. E. (1994). Ergonomics: How to Design
for Ease and Efficiency. Englewood Cliffs, NJ: Prentice Hall.
Kroemer, K. H. E., Kroemer, H. J., and Kroemer-Elbert, K. E. (1997). Engineering Physiology:
Bases of Human Factors/Ergonomics. New York: Van Nostrand Reinhold.

TAF-ER562X_WINGR-09-0401-C004.indd 150 11/9/10 2:23:30 AM


Anthropometry and Biomechanics 151

Marras, W. S., Lavender, S. A., Leurgans, S. E., Rajulu, S. L., Allread, W. G., Fathallah, F. A., et al.
(1993). The role of dynamic three dimensional trunk motion in occupationally related low
back disorders: The effects of workplace factors, trunk position and trunk motion characteris-
tics on injury. Spine, 18(5), 617–628.
McAtamney, L. and Corlett, E. N. (1993). RULA: A survey method for the investigation of work-
related upper limb disorders. Applied Ergonomics, 24(2), 91–99.
Moore, J. S. and Garg, A. (1995). The strain index: A proposed method to analyze jobs for risk
of distal upper extremity disorders. American Industrial Hygiene Association Journal, 56,
443–458.
National Aeronautics and Space Administration. (1978a). Anthropometric Source Book, Vol. I:
Anthropometry for Designers (NASA Reference Publication 1024). Washington, DC: National
Aeronautics and Space Administration, Scientific and Technical Information Office.
National Aeronautics and Space Administration. (1978b). Anthropometric Source Book, Vol. II: A
Handbook of Anthropometric Data (NASA Reference Publication 1024). Washington, DC:
National Aeronautics and Space Administration, Scientific and Technical Information Office.
National Aeronautics and Space Administration. (1978c). Anthropometric Source Book, Vol. III:
Annotated Bibliography of Anthropometry (NASA Reference Publication 1024). Washington,
DC: National Aeronautics and Space Administration, Scientific and Technical Information
Office.
National Aeronautics and Space Administration. (1989). Man-Systems Integration Standards
(NASA-STD-3000A). Houston: Lyndon B. Johnson Space Center.
Pheasant, S. (1996). Bodyspace: Anthropometry, Ergonomics and the Design of Work. London:
Taylor & Francis.
Robinette, K. M. (2000). CAESAR measures up. Ergonomics in Design, 8(3), 17–23.
Schoorlemmer, W. and Kanis, H. (1992). Operation of controls on everyday products. Proceedings
of the Human Factors Society 36th Annual Meeting (pp. 509–513). Santa Monica, CA: Human
Factors and Ergonomics Society.
Snook, S. H. and Ciriello, V. M. (1991). The design of manual handling tasks: Revised tables of
maximum acceptable weights and forces. Ergonomics, 34(9), 1197–1213.
Staff, K. R. (1983). A Comparison of Range of Joint Mobility in College Females and Males. Master’s
thesis, Industrial Engineering, Texas A&M University.
U.S. Department of Defense. (1995). Handbook for Human Engineering Design Guidelines. MIL-
HDBK-759C. Philadelphia, PA: Navy Publishing and Printing Office.
Van Cott, H. P. and Kinkade, R. G. (1972). Human Engineering Guide to Equipment Design.
Washington, DC: U.S. Government Printing Office.
Viitasalo, J. T., Era, P., Leskinen, A. L., and Heikkinen, E. (1985). Muscular strength profiles and
anthropometry in random samples of men aged 31–35, 51–55 and 71–75 years. Ergonomics,
28(11), 1563–1574.
Waters, T. R., Putz-Anderson, V., Garg, A., and Fine, L. J. (1993). Revised NIOSH equation for the
design and evaluation of manual lifting tasks. Ergonomics, 36(7), 749–776.
White, R. M. and Churchill, E. (1971). The body size of soldiers: U.S. Army anthropometry—1966
(Technical Report 72-51-CE AD 743 465). Natick, MA: U.S. Army Natick Laboratories.
Woodson, W. E., Tillman, B., and Tillman, P. (1992). Human Factors Design Handbook: Information
for the Design of Systems, Facilities, Equipment, and Products for Human Use. New York:
McGraw-Hill.

TAF-ER562X_WINGR-09-0401-C004.indd 151 11/9/10 2:23:30 AM


TAF-ER562X_WINGR-09-0401-C004.indd 152 11/9/10 2:23:30 AM
5 Documentation
John W. Gwynne III, PhD; David A. Kobus, PhD, CPE

CONTENTS
5.1 The Value of Well-Designed Documentation ..........................................................154
5.2 User Attitudes toward Documentation.....................................................................155
5.3 The Regulatory Imperative ......................................................................................156
5.4 Limitations of the Guidance in This Chapter ..........................................................157
5.5 General Principles ...................................................................................................157
5.5.1 Early Involvement Is Key..............................................................................157
5.5.2 Practice User-Centered Design .....................................................................158
5.5.3 Consider Environmental Factors...................................................................159
5.5.4 Conduct a Task Analysis ...............................................................................159
5.5.5 State Documentation Specifications .............................................................159
5.5.6 Produce Draft Documentation ......................................................................160
5.5.7 Iterate Usability Testing and Documentation Development .........................160
5.5.8 Documentation Design Checklist ................................................................. 161
5.6 Special Considerations............................................................................................. 161
5.6.1 Electronic Documentation ............................................................................ 161
5.6.1.1 Information Access .........................................................................163
5.6.1.2 Limitations ......................................................................................164
5.6.2 Documentation for Lay Users .......................................................................164
5.7 Design Guidelines....................................................................................................165
5.7.1 Guidelines for All Documentation ................................................................165
5.7.1.1 Content Guidelines ..........................................................................165
5.7.1.2 Presentation Guidelines ................................................................... 175
5.7.2 Guidelines for Electronic Documentation ....................................................187
5.7.2.1 Navigation .......................................................................................187
5.7.2.2 Hyperlinks .......................................................................................188
5.7.2.3 Language and Readability ...............................................................189
5.7.2.4 Organization and Layout .................................................................189
5.7.2.5 Illustrations ......................................................................................190
5.7.2.6 Web-Specific Graphics ....................................................................190
5.7.2.7 Highlighting .................................................................................... 191
5.7.2.8 Typography ......................................................................................192
5.7.2.9 Physical Characteristics ...................................................................193

153

TAF-ER562X_WINGR-09-0401-C005.indd 153 11/9/10 12:40:50 AM


154 Handbook of Human Factors in Medical Device Design

5.8 Case Studies .............................................................................................................193


5.8.1 Contact Lens Care ........................................................................................193
5.8.2 Blood Glucose Meter Operation ...................................................................195
Resources .........................................................................................................................196
References ........................................................................................................................197

Medical device documentation should give device users clear, accurate, and easy-to-follow
instructions on how to set up, operate, maintain, and troubleshoot their devices. This chap-
ter presents a human factors–based approach to designing documentation to achieve these
goals and thereby promote the safe, reliable, and effective use of medical devices. Human
factors principles for developing documentation are first presented to lay the groundwork
for the documentation design process. Two emerging trends in medical device use—the
increasing prevalence of electronic documentation and use of medical devices by layper-
sons—are considered in terms of their implications for documentation design. An extensive
set of design guidelines is provided that applies to both paper and electronic documenta-
tion, followed by guidelines that take into consideration the unique characteristics of elec-
tronic documentation. Case studies illustrate the value of the guidelines and demonstrate
the benefits of the user-centered approach advocated in this chapter. Finally, additional
resources are provided for readers who desire to explore documentation design topics in
more detail.

5.1 THE VALUE OF WELL-DESIGNED DOCUMENTATION


Ideally, medical devices should be so well designed that no documentation is needed to
operate them safely and reliably. Unfortunately, this design ideal has proved elusive. Device
use can expose users and patients to many potential hazards (e.g., radiation, electric shock)
that can be mitigated or reduced in several ways, including (1) improving the device user
interface (controls, displays, on-device labeling, logic of operation), (2) providing initial
and refresher training, and (3) creating documentation that gives users the information they
need to safely operate and maintain a device (Kaye & Crowley, 2000).
The increasing complexity and power of medical devices have placed greater demands
on the persons who operate, maintain, and repair them. Complete, unambiguous, accurate,
and easily understood documentation is important for all users, whether they are profes-
sionals in a health care setting or laypersons operating and maintaining a device in the
home (see Chapter 18, “Home Health Care”).
Documentation should make users aware of the potential hazards associated with device
use and how to minimize them. Hazards associated with use errors represent a serious
public health concern, as documented by the U.S. Food and Drug Administration’s (FDA’s)
Medical Device Reporting program. The literature (Bogner, 1994; Donchin et al., 1995;
Leape et al., 1991) indicates that the frequency and consequences of hazards arising from
use errors exceed those arising from device failures.
Numerous studies have shown that documentation for medical devices is often inad-
equate. For example, a study conducted by the FDA (Kingsley, 1995) found that nurses and
doctors did not consider medical device information and labeling useful for several reasons,
including (1) inadequate information content and formatting; (2) poor distribution methods;
(3) the availability of other, more effective information sources; and (4) questions about the

TAF-ER562X_WINGR-09-0401-C005.indd 154 11/9/10 12:40:50 AM


Documentation 155

validity of some labeling. Clearer descriptions of device operating procedures and more
specific troubleshooting information were the two improvements most desired by users.
Unfortunately, documentation is often considered only as an afterthought to the device
development process. This lack of integration between the design of the device and the
development of associated documentation results in documentation that fails to provide
users with information they can readily understand and use. This failure has contributed to
poor user attitudes toward documentation, which we now consider in more depth.

5.2 USER ATTITUDES TOWARD DOCUMENTATION


To write effective documentation for medical devices, it is necessary to understand device
users’ attitudes toward documentation and how those attitudes influence their behavior.
Most users typically prefer to avoid reading documentation and instead obtain information
about device operation and troubleshooting from other sources, such as in-service training
programs and training from health care professionals or the technical support staff
(Kingsley, 1999). Professional caregivers generally prefer hands-on training to reading
and following instructions in a user manual (Wiklund, 2004). When users do refer to
documentation, they typically do not read it thoroughly, cover to cover. Instead, they scan
it to locate specific topics of interest.
Given the reluctance of many device users to refer to documentation, special efforts are
needed not only to encourage them to use documentation but also to anticipate their needs
and to create documentation that meets those needs. For instance, documentation should
make it easy to locate information by (1) ordering topics in the same sequence they occur
during device operation, (2) highlighting key words, and (3) organizing topics in a shallow
hierarchy of clearly labeled headings and subheadings.
Thus, the traditional format for medical device use documentation, the user manual,
is not structured to facilitate user’s preferred style of interaction. Most user manuals are
written with the goal of satisfying regulatory requirements for labeling rather than giving
users a clear and usable information source for operating and maintaining a device. User
manuals are consequently looked on with disdain by many users, who consider them to be
a resource of last resort. A well-designed user manual can be a valuable user aid, but its
development requires greater planning and effort than is typically expended.
In addition to a user manual, it is often advisable to furnish a quick reference guide that
describes basic and critical aspects of device operation and troubleshooting. Short, suc-
cinct, easily accessible, and highly relevant documentation will almost always be used more
often than more comprehensive forms of documentation. Given that most medical device
users resort to documentation only when they encounter a problem they cannot solve on
their own or with the help of others, it is imperative to include troubleshooting information.
A quick reference guide, for example, could present both device operating procedures and
troubleshooting help.
It is worth noting that regional differences exist in the use of device documentation.
For instance, Europeans typically refer to user manuals more frequently than Americans
(Wiklund, 2002). User manuals may also provide a preferred way of learning for some
individuals despite a general tendency to the contrary as described above. Excellent device
documentation can also serve as a starting point for in-house training programs. Therefore,
even though documentation may not be widely used, it remains an important aspect of
device design, development, and implementation.

TAF-ER562X_WINGR-09-0401-C005.indd 155 11/9/10 12:40:50 AM


156 Handbook of Human Factors in Medical Device Design

The manufacturer who waits until a medical device is at the final prototype stage before
developing that device’s user documentation is missing out on a real opportunity. The
creation and testing of user documentation often identifies shortcomings in a device’s user
interface. Thus, manufacturers who integrate user documentation with other aspects of
device design (including other types of labeling) will produce a more usable and probably
safer and more effective device.

5.3 THE REGULATORY IMPERATIVE


As alluded to above, governmental regulatory requirements constitute a major motivation
for device manufacturers to create documentation. All medical devices manufactured in
the United States are regulated by the FDA. The FDA requires device manufacturers to
furnish printed documentation with their devices; such documentation is considered an
extension of a device’s labeling, which details the conditions for its safe and effective use
(Wiklund, 2002).
The Federal Food, Drug, and Cosmetic Act (FFDCA) is the law under which the FDA
regulates all medical devices. Section 201(m) of the FFDCA defines labeling as “all labels
and other written, printed or graphic matter (a) upon any article or any of its contain-
ers or wrappers, or (b) accompanying such article” at any time while a device is held for
sale after shipment or delivery for shipment in interstate commerce. The term accompa-
nying is interpreted liberally to include posters, tags, pamphlets, circulars, booklets, bro-
chures, instruction manuals, fillers, quick reference guides, troubleshooting guides, and
so on. Accompanying also includes labeling that is brought together with the device after
shipment or delivery for shipment in interstate commerce (see also Chapter 13, “Signs,
Symbols, and Markings”).
All medical device documentation must comply with the applicable law under the U.S.
Code of Federal Regulations (2003a, 2003b), Title 21, Parts 801 (Labeling) and 809 (In
Vitro Diagnostic Products for Human Use). An important consequence of FDA oversight
is that documentation must meet the criteria set forth by the FDA before a medical device
can receive FDA approval. Likewise, devices produced abroad must meet the regulatory
requirements of those government organizations established to oversee medical device
development and distribution.
All medical devices and their accessories placed on the European Union market since
June 15, 1998, must comply with the requirements of the Medical Devices Directive,
which include labeling requirements. Device manufacturers that market their devices
in Europe must comply with all essential requirements in Annex I to Directive 93/42/
EEC on medical devices as well as with the member states’ individual implementing
legislation. The Medical Devices Directive provides that member states may require the
accompanying information to a medical device to be in their national language or in another
community language. All member states require that safety information be provided in
their official language to ensure understanding by the end user. Device manufacturers
thus face a heavy burden to provide documentation that not only is complete, correct,
and understandable but also has been accurately translated from its language of origin to
the national languages used in the countries where the device is marketed. The penalties
for noncompliance with the requirements of the Medical Devices Directive vary from
the removal of the device from the market to economic penalties and, in some cases, to
criminal proceedings (Pilot, 1999).

TAF-ER562X_WINGR-09-0401-C005.indd 156 11/9/10 12:40:50 AM


Documentation 157

The human factors design guidance presented in this chapter should be used in conjunction
with government regulatory guidance documents to create high-quality documentation
that satisfies regulatory requirements while embodying sound human factors. In this way,
documentation will incorporate the necessary regulatory content while still meeting users’
needs. In the United States, the FDA’s Office of Device Evaluation, Center for Devices
and Radiological Health, has prepared a General Program Memorandum (#G91-1), Device
Labeling Guidance, to assist designers in complying with FDA regulations for medical
device labeling.

5.4 LIMITATIONS OF THE GUIDANCE IN THIS CHAPTER


• The human factors design guidance in this chapter cannot be implemented in a
cookbook fashion. Instead, it should be adaptively applied to each design project.
The effectiveness with which this is done depends on the knowledge and exper-
tise of the individual designer. There is no substitute for the skilled application of
human factors principles and guidelines when developing documentation.
• Understanding the rationale underlying human factors principles is indispensable
for the proper application of the design guidelines in this chapter. No single approach
or format is best for all documentation. The type of documentation being created,
the procedural complexity of the device, the skills and knowledge needed to operate
a device, the target user group, and the anticipated use environment are among the
contextual factors that must be considered when developing documentation.
• Good documentation cannot compensate for bad device design. If a device is
poorly designed in terms of its human factors characteristics, good documentation
will not improve the user interface. For each to be most effective, device design
and documentation design should proceed in parallel, each process benefiting from
the other through successive design–test iterations. Documentation design is best
viewed as part of a larger effort to create devices that are easy, safe, and effective
to use through the application of human factors principles and guidelines.

This chapter furnishes general guidance for developers of medical device documen-
tation. It is not intended to serve as a guide to satisfying governmental regulations for
medical device labeling (although the information presented here is consistent with FDA
publications and the literature). The designer should consult the governmental agencies
that regulate medical devices in their intended market(s) for additional information on their
requirements for device documentation.

5.5 GENERAL PRINCIPLES


This section presents a general approach to creating documentation, based on human factors
principles and guidelines for information presentation, training, documentation design, and
human performance capabilities. This approach consists of a set of largely sequential activities,
most of which apply equally to the development of paper and electronic documentation.

5.5.1 EARLY INVOLVEMENT IS KEY


As discussed previously, the design and preparation of documentation should begin early in
the device design and development process, preferably during the device specification phase.

TAF-ER562X_WINGR-09-0401-C005.indd 157 11/9/10 12:40:50 AM


158 Handbook of Human Factors in Medical Device Design

This will enable the documentation designer to gain a thorough understanding of the device
and its documentation requirements. The designer should critically evaluate (1) the types
of documentation needed for a device, (2) characteristics (educational, physical, sensory,
and cognitive factors) of its intended end users, and (3) the effects of the anticipated use
environment on how a device will be implemented and used.
Not only can documentation designers furnish information useful for device designers,
but a close working relationship between them will help ensure that draft documentation
is available for the usability testing of device prototypes. Unfortunately, documentation
development is often begun only after the device design and development effort is far along,
at which point it is too late for the documentation designer to make meaningful contributions
to the device design and refinement process (Callan, Gwynne, Sawyer, & Tolbert, 1993).

5.5.2 PRACTICE USER-CENTERED DESIGN


Documentation designers must take into account the varied needs, skills, knowledge, and
abilities of increasingly diverse user groups, along with the characteristics of the envi-
ronments in which devices are used. User-centered design emphasizes user requirements,
capabilities, needs, tasks, and goals and incorporates them as early as possible in the design
process, when changes can still be made in a time and cost efficient way (Czaja & Nair,
2006; Meister & Enderwick, 2001; Nemeth, 2004).
From the viewpoint of user-centered design, a key aspect of developing effective docu-
mentation is to characterize the anticipated users of a device in terms of the skills, knowl-
edge, and abilities needed to use the device safely and effectively (see Chapter 2, “Basic
Human Abilities”). Users vary along numerous dimensions relevant to documentation
design, including the following:

• Educational and reading levels, language comprehension, and familiarity with


Web-based navigation
• Physical abilities or disabilities related to visual and auditory perception, color
vision, and motor functioning
• Experience with similar technology or devices, including an evaluation of previous
experience that could produce negative transfer (i.e., any user habits or preconcep-
tions that could interfere with understanding the documentation or learning to use
the device)
• Understanding of operating principles and potential hazards associated with the
technology

The documentation must be designed to accommodate a wide range of skills, knowledge,


and abilities. Horton (1994) identifies four types of medical device users:

• Novice users are beginners. They will have little, if any, experience using medical
devices and have at best only a limited knowledge of related topics.
• Occasional users have previous experience with a device but are not familiar
enough with it to remember details of its operation.
• Transfer users have used similar types of medical devices before but not the one
your documentation covers. These users understand general aspects of using medi-
cal devices but need help to gain proficiency on the new device.

TAF-ER562X_WINGR-09-0401-C005.indd 158 11/9/10 12:40:50 AM


Documentation 159

• Expert users have experience performing the tasks required to operate and maintain
a device. If they refer to documentation, it will most likely be for troubleshooting help
in response to an anomalous situation or to access a rarely used device capability.

For many devices, it is likely that persons from several of these user categories will use the
documentation. Identify the most common user categories for a given device and design the
documentation to be appropriate for those categories.

5.5.3 CONSIDER ENVIRONMENTAL FACTORS


Environmental factors should be considered during documentation design because they
can impose important constraints on how devices are used. Environmental factors include
lighting, noise levels, and the presence of other devices. These factors can act singly or in
combination to affect the ability of device users to understand and follow documentation
for device operation and maintenance. For example, a bedroom in a home may be poorly
lit, making it harder to access and read device-based documentation. Refer to Chapter 3,
“Environment of Use,” for an in-depth treatment of this important topic.

5.5.4 CONDUCT A TASK ANALYSIS


To write effective documentation, it is vital to have a thorough understanding of how a
medical device will be used to achieve its intended purpose. A task analysis is indispens-
able to gaining this understanding. Ideally, a task analysis should be conducted during the
early stages of device design by human factor professionals skilled in task analysis tech-
niques. A task analysis defines in detail all aspects of device operation and maintenance,
including (1) the equipment, supplies, and materials needed; (2) the functions, tasks, and
procedural steps that must be performed to properly set up, calibrate, operate, and maintain
a device; (3) the time needed to perform these functions, tasks, and procedures; (4) the
types of errors that can occur and the likelihood of those errors for the various proce-
dures (Hollnagel, 2006); and (5) the procedures involved in identifying and troubleshooting
device operational problems (Kirwan & Ainsworth, 1992). If operating and maintaining a
device involves tasks that have a large cognitive aspect, the task analysis should identify the
memory, attention, and other mental demands placed on device users.
A task analysis furnishes the basis for writing step-by-step operating instructions, devel-
oping warning and caution statements, and providing complete and accurate device descrip-
tions and other useful information to device users.

5.5.5 STATE DOCUMENTATION SPECIFICATIONS


An important part of creating documentation is enumerating its content and format speci-
fications. These specifications should take into account the characteristics of the intended
users and the environment(s) in which a device will be used. Specifications will vary from
one device to another; however, documentation specifications for most devices will include
the following:
• A list of the types of documentation required, such as user manuals, mainte-
nance manuals, reminder sheets, checklists, and quick reference guides. These

TAF-ER562X_WINGR-09-0401-C005.indd 159 11/9/10 12:40:50 AM


160 Handbook of Human Factors in Medical Device Design

requirements will be based on the type of use that is intended for a device, but a
user manual with complete device information and a quick reference guide that
contains basic information on device operation and troubleshooting are likely to
be needed.
• A specification for readability, including a list of requirements such as table of
contents, indexing, and headings.
• A list of languages to be used and a list of special needs or restrictions on the docu-
mentation, such as limitations on color, font size, and document size.
• An outline for each type of documentation, including a list of expected
illustrations.
• A schedule and plan for developing the documentation that is clearly linked to the
device design schedule.
• A schedule for the test and evaluation of the documentation.
This information facilitates documentation design and development by providing both the
necessary information and timetables for implementing those specifications that coincide
with device design and development.

5.5.6 PRODUCE DRAFT DOCUMENTATION


Once the specifications for documentation have been defined, the next step is to design
draft versions of each type of documentation that will be created for a device. Produce the
user manual in parallel with detailed device design or prototyping to facilitate usability
testing of both the manual and the device. Depending on the nature of the device and its
intended users, the user manual could include procedures for setup, installation, calibra-
tion, normal operation, emergency operation, maintenance, and troubleshooting. A quick
reference guide, a valuable companion to the user manual, can reference specific pages
in the user manual, making it easy for users to obtain additional information as needed.
Development of maintenance and service manuals, if they are necessary, should start at the
same time that prototype devices begin to resemble production models. Development of the
user, maintenance, and service manuals should proceed in parallel with the device design
process to help identify deficiencies in user-interface design. Apparent interface design
shortcomings can be fed back to the device designers to improve the user interface.

5.5.7 ITERATE USABILITY TESTING AND DOCUMENTATION DEVELOPMENT


Usability testing, in which target users refer to the documentation while using a device,
should be done as part of the design process to assess how users perceive and respond to the
documentation (Dumas & Redish, 1993; Nielsen, 1994; Rubin, 1994). The prototype docu-
mentation is tested on a representative sample of the anticipated end users of a device.
Documentation evaluation should focus on content sufficiency, user comprehension, and
documentation usability. Content sufficiency addresses the accuracy and completeness of
the information contained in the documentation. User comprehension concerns the abil-
ity of device users to understand the documentation’s procedural instructions, warnings,
troubleshooting advice, and supporting information.
Testing should involve actually observing users while they attempt to operate the device
using the documentation as a guide. Based on the usability test results, the documentation

TAF-ER562X_WINGR-09-0401-C005.indd 160 11/9/10 12:40:51 AM


Documentation 161

should be redesigned and then retested. Before initial testing, acceptance criteria should be
established (See Chapter 6, “Testing and Evaluation”). This iterative design–test process is
repeated until users can use the documentation effectively. The test and revision of the doc-
umentation should proceed in parallel with the test and development of the actual device.
In this way, a close relationship is established between device documentation, operating
characteristics, and interface features.

5.5.8 DOCUMENTATION DESIGN CHECKLIST


The checklist in Table 5.1 provides a comprehensive summary of the characteristics of
effective documentation and important activities in the documentation design process.
Some items in the checklist may not apply to some types of documentation.

5.6 SPECIAL CONSIDERATIONS


Documentation design efforts must be responsive to relevant trends in health care delivery.
One such trend is the use of electronic means to acquire, store, and transmit health care
information. Electronic documentation will become increasingly common in the future.
The documentation designer must understand the unique properties of electronic documen-
tation to make the most effective use of them. A second trend is the increasing prevalence
of home-based health care. This trend has focused attention on the documentation needs of
lay users, who often have little or no prior background in using medical devices, especially
those that have migrated from professional settings into the home. Providing documenta-
tion that is easy for lay users to understand and follow is crucial to the safe and effective
home use of medical devices.

5.6.1 ELECTRONIC DOCUMENTATION


Computerization in health care delivery and care documentation has become ubiquitous in
recent years. Hospitals are taking an increasingly paperless approach to vital signs record-
ing, laboratory testing, staff communication, and clinical note taking (Wiklund, 2002).
Following this trend, device documentation has also become increasingly computerized,
although regulatory requirements for paper documentation still exist. It is thus important to
be aware of how electronic documentation differs from paper documentation and to provide
design guidelines that take these differences into account.
Electronic documentation is any documentation information (1) stored on videotape,
film, audiotape, or computer storage medium (hard disk, compact disk, floppy disk, DVD);
(2) delivered online via a Web-based or e-mail user interface; or (3) delivered in real time
via visual or audible modalities by the device itself.
The challenge for the documentation designer is to exploit the unique features of these
electronic media to create documentation that enhances the device user’s experience over
and above what can be achieved with paper documentation. For example, electronic docu-
mentation can be developed so that it is adaptive and intelligent. Adaptive documentation
keeps a record of a user’s interaction with it to identify procedures that have been problem-
atic or topics that are of special interest to that user. When the user again seeks information
on these topics, the system can provide the user with additional relevant information based
on earlier interactions without requiring the user to submit detailed requests or queries for
that information.

TAF-ER562X_WINGR-09-0401-C005.indd 161 11/9/10 12:40:51 AM


162 Handbook of Human Factors in Medical Device Design

TABLE 5.1
Documentation Design Checklist
Our documentation: Complies with all labeling regulations
Is written for the type of people who use our device
Tells the user how to get help from the manufacturer
Includes a table of contents
Has general warnings and precautions near the beginning
Briefly describes the purpose of the device
Gives a physical description of the device with a labeled graphic
Explains conditions under which the device should and should not be used
Gives clear setup instructions
Gives clear pre-use checkout procedures
Gives clear and easy-to-follow operating instructions in a step-by-step format
Provides cleaning instructions
Describes maintenance that the user should do
Explains how to store the device
Has a clear, easy-to-use, and easy-to-find troubleshooting section
Has a summary page with all the critical information on it
Has an alphabetized index
Has an easy-to-find date of issue
Includes instructions on any accessories
Is laid out to make sections easy to find
Uses white space and other highlighting techniques to focus user attention on
important information
Is printed in at least 12-point type
Has clear, well-labeled graphics in key places to help users understand the text
Uses proper contrast so that text and graphics can be easily viewed
We have: Done a task analysis for the procedures in our documentation
Selected a suitable format (text, flowchart, list)
Written and formatted procedures correctly
Used appropriate sentence construction and word choice
Tested our documentation to ensure a sixth- to seventh-grade reading level
Properly written and placed specific warnings and cautions
Avoided technical terms and jargon unfamiliar to our users
Our manual: Has a durable distinctive cover
Will stand up to the conditions in which it will be used
Is constructed of nonshiny, durable paper
Is laid out to make things easy to locate and update (e.g., tabs, color codes, binding)
Lays flat on its own
We have: Tested the documentation to make sure that its target users can read, understand, and
follow it
Taken steps to make sure that our documentation is available to our users

Source: Backinger, C. L. and Kingsley, P. A. (1993, August). Write It Right: Recommendations for Developing
User Instruction Manuals for Medical Devices Used in Home Health Care (HHS Publication FDA
93-4258). U.S. Department of Health and Human Services, Food and Drug Administration,
Rockville, MD.

TAF-ER562X_WINGR-09-0401-C005.indd 162 11/9/10 12:40:51 AM


Documentation 163

Although electronic documentation shares many characteristics with paper documenta-


tion, it also possesses several unique features, such as sound, voice, animation, and ease
of updating. These features have important implications for documentation design. While
potentially useful, these features can also distract and confuse users if they are not imple-
mented wisely. The design process for electronic documentation involves the same basic
activities as does the development of paper documentation (e.g., task analysis, require-
ments definition, iterative testing and redesign). However, the unique features of electronic
documentation require that specific questions be addressed in terms of user needs and the
device operating environment. These design questions include the following:

• Are the device procedures (operating, troubleshooting, and so on) complex enough
to justify the costs of an electronic system? Paper documentation may be a better
choice for simple procedures.
• Is electronic documentation practical for the intended use environment (e.g.,
patient’s home, ambulance, emergency room)?
• What form of information delivery would be most practical (e.g., device based,
Web based, CD-ROM, DVD, audiotape)? For example, although device-based doc-
umentation might be more practical in some environments, what recourse would
be available in the event of device failure? If troubleshooting procedures were part
of the device-based documentation, they might not be accessible at the very time
when they are needed most.
• Could user limitations make electronic documentation hard to use? Limitations in
vision, hearing, or manual dexterity may limit the value of electronic documenta-
tion for some user groups.
• Will the intended users have ready access to computers or the other technologies
needed to access electronic documentation? For instance, older persons may not
own a computer or, for that matter, the home electronics needed to play DVDs and
CD-ROMs.

5.6.1.1 Information Access


The ability to access information quickly and efficiently is one of the potential advantages
of electronic documentation. However, from the designer’s viewpoint, information access
in electronic documentation is complicated by the fact that users often do not thoroughly
read most forms of electronic documentation in a linear fashion. Instead, they skip around
from one topic to another. As a consequence, readers rely on visual clues to the organization
of documentation, especially when they are reading to follow procedures or make decisions
(Wright, 1999). The documentation designer has little or no control over what parts of the
documentation a user reads prior to reaching a particular topic or where a user will go next.
Although this is also true for paper documentation, it is more pronounced for electronic
documentation because of its enhanced navigational features, such as hyperlinks.
An important design implication of the increased ease of navigating electronic docu-
mentation is that each topic should be self-contained and “stand alone” as much as possible
from other topics. A given topic should answer a single question or address a single aspect
of device operation. Insofar as possible, do not organize electronic documentation so that
it cumulatively builds on topics presented earlier in the documentation. If a reference is
made to earlier topics, make it easy for the reader to access them, for example, by using
hyperlinks.

TAF-ER562X_WINGR-09-0401-C005.indd 163 11/9/10 12:40:51 AM


164 Handbook of Human Factors in Medical Device Design

Ease of access to important information is obviously critical, but the “how,” “when,”
where,” and “why” require careful planning based on a thorough understanding of user
information requirements and device operating characteristics. Given the linear, procedural
nature typical of medical device operation, sets of procedural steps could be linked directly
to illustrations of related controls and displays. Likewise, long, complicated procedures
could be linked to overview information, place-keeping “roadmaps,” or warnings. But there
are always trade-offs. For example, warnings critical to correct device operation and patient
safety should always appear adjacent to the relevant step. In some cases, space limitations
may dictate that less critical or general warnings be located in a separate section devoted to
warnings. Make a strong effort to ensure that warnings can still be easily accessed, however.
For example, several different steps could be linked to the same general warning. Users could
then preview procedural steps from an earlier section on general warnings or vice versa.

5.6.1.2 Limitations
Electronic documentation may not always be a practical solution. Likely drawbacks of
electronic documentation include higher initial development and implementation costs.
Depending on the form of electronic documentation that is used, other limitations include
ongoing demands for computer storage space, display limitations associated with computer
monitors and device screens, and the need for training for users who are not computer
literate. In this last regard, overcoming initial user resistance to electronic documentation
is a potential stumbling block. Some lay users may not have access to a computer or
simply feel uncomfortable using electronic documentation. Another important drawback
is that electronic documentation may not readily be accessible when it is needed most. For
example, some clinical environments, such as emergency rooms or ambulances, may not
be conducive to the use of electronic materials. In addition, if a device becomes inoperable,
device-based documentation will be unavailable.
Many devices with electronic documentation will also provide paper documentation. The
choice to use paper or electronic documentation should be left to individual users who can
then select the one that meets their particular needs. When both forms of documentation are
to be provided, both should be designed on the basis of established human factors principles
and guidelines. It is best to develop paper and electronic documentation concurrently, relying
on a core set of content that can be expressed in either paper or electronic form, depending
on individual users’ needs and preferences. Following this strategy, discrete units of text
can be developed both as hypertext elements for incorporation into nonlinear, hyperlinked
electronic documentation and as regular text for paper documentation. Importantly, because
actual usage of the two forms of documentation may be different, they generally will not
be identical in structure or format (e.g., a PDF of paper documentation will yield few of the
benefits of full electronic documentation).

5.6.2 DOCUMENTATION FOR LAY USERS


In recent years, changes in health care economics, combined with an aging population,
have resulted in many patients being cared for at home. Traditionally, most medical devices
have been used by health care practitioners in professional settings such as hospitals and
doctors’ offices. However, as patient care has moved into the home, the medical devices
needed for this care have followed, where they often are used by laypersons, including
patients, family members, friends, or other nonprofessional caregivers. Lay users rely on

TAF-ER562X_WINGR-09-0401-C005.indd 164 11/9/10 12:40:51 AM


Documentation 165

whatever firsthand training and follow-up support they receive from health care profession-
als, along with device documentation.
Lay users represent an extremely broad and diverse group in terms of the factors that
influence how well they can operate a medical device. These factors include educational
background, age, technical sophistication, physical capabilities, mental condition, and lack
of prior relevant experience or training. In particular, the aging patient or lay caregiver
presents challenges for designing both safe and usable devices and documentation (Rogers,
1997; Rogers & Fisk, 2001; Rogers, Fisk, & Walker, 1996; Vanderheiden, 2006).
The home environment itself poses additional challenges to safe and effective device
use, including the following (McCarthy et al., 1992):

• Inadequate or inappropriate physical environments for device operation


• Inadequate infection control practices
• Inadequate communication channels
• Lack of quality control standards or procedures

For additional information about the challenges of designing documentation for lay users and
the home environment, refer to Chapter 2, “Basic Human Abilities”; Chapter 3, “Environment
of Use”; and Chapter 18, “Home Health Care.”

5.7 DESIGN GUIDELINES


This section presents human factors design guidelines for creating effective user documen-
tation. These guidelines are organized into two main parts. The first part presents guide-
lines that apply to both paper documentation and electronic documentation. The second
part presents guidelines specifically for electronic documentation.

5.7.1 GUIDELINES FOR ALL DOCUMENTATION


The guidelines in this section are organized into two major categories: content and presenta-
tion. Content refers to the type of information that should be included in device documenta-
tion, such as background information, procedures, risk communication, and supplementary
information. Presentation refers to the manner and style used to convey that information,
which includes language and readability, organization, illustrations, layout, highlighting,
typography, and physical characteristics.
With only a few noted exceptions, these guidelines apply equally to both paper and elec-
tronic documentation. They are appropriate for all forms of documentation, including user
manuals, quick reference guides, online information, and device-based help systems.

5.7.1.1 Content Guidelines


The content guidelines are organized into six categories:

• Background information
• Procedures
• Troubleshooting information
• Warnings and precautions
• Supplementary information
• Translations

TAF-ER562X_WINGR-09-0401-C005.indd 165 11/9/10 12:40:51 AM


166 Handbook of Human Factors in Medical Device Design

Together, these topics address the types of information that device users need to know
to be able to safely and reliably install, set up, calibrate, operate, clean, store, maintain,
troubleshoot, and repair a device. The appropriate content for device documentation can
be determined by several methods, including discussions with device designers, medical
personnel, and end users. Additionally, applying human factors techniques, such as task
analysis, assists in determining the cognitive, perceptual, and physical aspects of device
use. The checklist in Table 5.2 shows the numerous topics that content information can
encompass, depending on the device.

5.7.1.1.1 Background Information


There is often information of which device users should be aware before using a device.
This background information should be presented in the first part of the documentation,
before the procedures on how to operate and maintain a device. Depending on the nature
of the device, some or all of the following guidelines are appropriate for background
information.

Guideline 5.1: Describe the Purpose of the Device


Documentation should describe the purpose of the device, the medical needs of persons
who use it, and indications and contraindications for its use. Explain how any information
that is supplied by the device should be used.

Guideline 5.2: Furnish Overview and Contact Information


A statement at the beginning of the documentation should identify critical information and
instructions to be read before device use. Give an overview of basic device operation and list
manufacturer contact information (toll-free telephone number, e-mail address, Web address,
and so on) to obtain help with device use or problems.

Guideline 5.3: Describe Device Components


Documentation should describe each device component in enough detail so that the user can
understand the role it plays in device operation. Include labeled illustrations that clearly show
each component. Emphasize the need for users to familiarize themselves with all device com-
ponents before using the device. If the device is electronic or mechanical, provide explana-
tions for all major device features and operations, such as the following:
• Operating modes
• Menus and navigation
• Display messages
• Control actuation
• Battery loading and testing
• Cleaning and maintenance
• Calibration
• Device-based help

Guideline 5.4: Provide General Warnings


A warning is a statement that makes a user aware that an adverse health consequence may
result from device use. A warning can also discuss hazards associated with noncompliant

TAF-ER562X_WINGR-09-0401-C005.indd 166 11/9/10 12:40:51 AM


Documentation 167

TABLE 5.2
Checklist of Content Information for Device Documentation
— User Assistance Information (e.g., toll-free phone number)
— Table of Contents
— Glossary
Descriptive Information
— Name, other specific identifiers
— Description of device features and characteristics
— Purpose of device (indications for use)
— When device should not be used (contraindications)
— Potential risks/benefits
— Expectations of device and procedure(s) associated with device
— General and specific warnings and precautions
— Importance of the need to adhere to a care regimen
Operating Information (as applicable)
— Setup instructions
— Calibration instructions
— Patient preparation instructions
— Operating instructions
— Graphics, tables, videos, animation, and pictures to accompany procedural steps
— Importance of the need to monitor the activity of the device
— Cleaning instructions
— Description of maintenance and who should do it
— Storage instructions
— Expected failure time and mode and its effect on the patient
Troubleshooting Information
— Instructions on how to safely dispose of the device
— Instructions on accessories
— Instructions on related, additional devices
Additional Information
— Scientific information/clinical studies
— Disease and self-care information
— Adverse events
— Comparative information (e.g., success rates)
— Warranty
— Travel/international use
— Index
— Date of Printing

Source: Adapted from Food and Drug Administration (2001). Guidance on Medical Device
Labeling: Final Guidance for Industry and FDA Reviewers. Food and Drug
Administration, Center for Devices and Radiological Health, Rockville, MD.

device use or pitfalls in interpreting test results. There are two types of warnings: general and
specific. General warnings consist of crucial information needed before operating a device,
such as the need to stop using a device if certain symptoms arise. Present general warnings
in the background section of the documentation. Avoid technical jargon so that a lay user can
understand the warning. Specific warnings pertain to particular aspects of device operation.

TAF-ER562X_WINGR-09-0401-C005.indd 167 11/9/10 12:40:51 AM


168 Handbook of Human Factors in Medical Device Design

Present specific warnings in the appropriate procedure sections, adjacent to the step to which
they apply.

Guideline 5.5: State Conditions of Device Use


State the environmental conditions required for safe device operation. Identify all conditions
under which a device should not be used. Cite any conditions that can adversely impact device
operation, such as excessive temperature or humidity, and the consequences of operating the
device under those conditions.

Guideline 5.6: Specify Supplies and Materials


Describe and illustrate all the supplies and materials needed to operate and maintain the
device. Specify the quantity, size, and type for all supplies and materials and whether they
accompany the device. Provide information about any special handling required of the sup-
plies, as needed.

Guideline 5.7: Clarify Differences when Multiple Models Exist


When multiple models or versions of the same device exist and their user interface or
operation differs, it is generally better to provide separate and unique documentation
for each model/version. However, if documentation provides information about several
different models or versions of the same device, provide a clear indication of how the
models/versions differ and which aspects of the documentation apply to which model(s)
or version(s).

Guideline 5.8: Describe Patient Preparation


Patient preparation requirements should be clearly stated. For example, some medical devices
require that patients ingest or do not ingest specific substances or foods prior to or during
device use.

Guideline 5.9: Explain Device Storage Conditions


Describe the proper preparation of the device for storage and desired storage conditions.
Emphasize any conditions that could damage the device or cause it to malfunction in sub-
sequent use. State the consequences of improper storage. Indicate how extended storage can
adversely affect the device (Backinger & Kingsley, 1993).

5.7.1.1.2 Procedures
A procedure is a set of instructions that tells a user how to operate, maintain, or troubleshoot
a device. The clear and orderly presentation of procedures is vital to effective documenta-
tion. Understanding and following device operating instructions, which often contain both
text and illustrations, requires a broad range of literacy skills. The following guidelines will
help designers write easy-to-follow and effective operating procedures.

Guideline 5.10: Include All Device Procedures


Include all the procedures that a device user must perform when setting up, operating, main-
taining, or troubleshooting a device. This information should derive from the task analyses
performed during device design and development.

TAF-ER562X_WINGR-09-0401-C005.indd 168 11/9/10 12:40:51 AM


Documentation 169

Guideline 5.11: Provide Clear and Logical Procedure Sequences


Organize procedures into clear and logical steps that can be followed accurately and reliably.

Guideline 5.12: Hierarchical Procedure Descriptions


Procedures should be organized hierarchically so that related steps are grouped into sections.
These sections should be given heading names that reflect their content (Wieringa, Moore, &
Barnes, 1993).

Guideline 5.13: Write Procedures as Step-by-Step Instructions


Procedures should be broken down into a series of short, discrete steps rather than a single
long paragraph. Step-by-step instructions make it easier for users to understand and follow
device operating and maintenance procedures. Each procedure usually includes the following
information, as appropriate (Wieringa et al., 1993):
• Conditions under which the procedure should (or should not) be performed
• Supplies and equipment needed
• Timing requirements for time-critical steps
• Examples to clarify text and illustrations
• Factors that influence device use or output, including special circumstances

Guideline 5.14: Include All Steps and Conditions


For each procedure, specify all the required steps and necessary conditions for safe and effec-
tive device operation and maintenance.

Guideline 5.15: Write Procedures That Match


Users’ Skills and Knowledge
Use a level of detail that that is appropriate for the intended users. The appropriate level of
detail depends on the type of procedure, the frequency with which it is performed, and the
skill and knowledge of the users (Wieringa et al., 1993).

Guideline 5.16: Write Procedural Steps as Positive Statements


Write action instructions as positive statements, telling users in succinct, everyday language
what actions they should perform.

Guideline 5.17: Selectively Use Negative Statements


Concise, nonambiguous negative statements are appropriate when users need to be told what
actions they should be sure not to perform (Wieringa et al., 1993).

Guideline 5.18: Include Rationales for Individual Steps


Providing succinct rationale helps motivate users by explaining why a particular action is
required. Keep rationales brief and include them with the applicable procedural steps.

Guideline 5.19: Avoid Nonessential Content


Do not dilute procedures with lengthy justifications and rationales; they divert the reader’s atten-
tion from the procedural steps. Documentation for lay users should omit extraneous details or
technical content that might draw attention away from important aspects of device operation.

TAF-ER562X_WINGR-09-0401-C005.indd 169 11/9/10 12:40:51 AM


170 Handbook of Human Factors in Medical Device Design

Guideline 5.20: Number the Individual Steps in a Procedure


Procedural steps should be clearly broken out and numbered consecutively in Arabic numer-
als. Do not “bury” procedures in text. Figure 5.1 shows how a sequence of individual actions
is specified to accomplish a procedure step (Callan & Gwynne, 1993).

Guideline 5.21: Use Illustrations to Clarify Text


Clean and unambiguous illustrations will make the procedural steps described by text much
easier to follow. Illustrations also help users who are not proficient in the given language to
better understand how to operate a device. Refer to Section 5.7.2.5 on illustrations in this
chapter for additional guidance.

Guideline 5.22: Link Procedural Text to Illustrations


Explicitly link text and accompanying illustrations. Choose a convention for position-
ing text and illustrations in relation to each other and apply it consistently throughout the
documentation.

Guideline 5.23: Use Explicit Cross-References


Whenever possible, include cross-references explicitly in procedures (Wieringa et al., 1993).
Cross-reference rather than include nonessential information (Horton, 1994).

Guideline 5.24: Minimize Cross-References


Refrain from constant cross-referencing, especially if the information is essential. However,
cross-references may be necessary to ensure accuracy and to simplify complex conditional
information (Wieringa et al., 1993).

Guideline 5.25: Clearly Identify All Cross-References


Specify any additional procedural step(s) or information to which the user should refer.

Guideline 5.26: Verify Cross-References When


Revising Documentation
Whenever documentation is revised, the accuracy of the identification information in its cross-
references must be checked. If inaccurate, cross-references can introduce errors in revised
procedures (Wieringa et al., 1993).

STEP 5 – Place lens on your eye.


(a) Place your lens on the tip of your index finger.
(b) Hold down the lower eyelid with the middle finger
of the same hand and look up.
(c) Place the lens on the lower portion of the white
part of the eye.
(d) Look down and remove your finger from the lens.
The lens should then center itself.

FIGURE 5.1 Example of a procedural step decomposed into specific actionable substeps.

TAF-ER562X_WINGR-09-0401-C005.indd 170 11/9/10 12:40:51 AM


Documentation 171

Guideline 5.27: Reading and Interpreting Test Results


of Home Medical Devices
Lay users of home medical devices need to be told what to do with any test results generated by
the device. Include a description of how test results should be read and interpreted. Guidelines
5.28 to 5.30 describe elements essential to instructions on test results management.

Guideline 5.28: Cautionary Notes on Reading Test Results


Explicitly state any conditions that can influence the reading of test results. Examples of these
conditions include timing of reactions, temperature, and lighting conditions under which
results are read.

Guideline 5.29: Interpreting Test Results


State the range of acceptable results so that invalid results can be readily ascertained. Provide
an interpretation of the meaning of all possible test results. The proper treatment of ambigu-
ous results and any limitations of the test performed by the device are important to consider
when interpreting results.

Guideline 5.30: Acting on Test Results


Describe the action to follow for a given result, such as seeking advice from a health care
professional or retesting to confirm a test result. Make the user aware of the possibility of
obtaining invalid results, the hazards likely to be encountered from acting on them, and ways
to detect and correct them.

5.7.1.1.3 Troubleshooting Information


A major reason that device users refer to documentation is for troubleshooting advice. As
long as a device is operating properly, many users will not consult documentation. Only when
a problem arises that they cannot solve will they seek out device documentation. A complete
and easy-to-use troubleshooting section is therefore extremely valuable for device users.

Guideline 5.31: Provide a Thorough Troubleshooting Analysis


Base troubleshooting information on an analysis that specifies (1) all reasonable types of
device failure, (2) indicators for each type of failure, (3) the probable cause(s) for each
type of failure, and (4) diagnostic procedures for identifying the failure (Inaba, Parsons, &
Smillie, 2004).

Guideline 5.32: Anticipate Likely Problems


Problems can occur during device setup, operation, or maintenance. Provide solutions for
each of these classes of problems (FDA, 2001).

Guideline 5.33: Group Similar Types of Problems


Group similar problems together and highlight each with a group heading. Put the most life-
threatening or critical problems first in each section (FDA, 2001).

Guideline 5.34: Clearly Describe Each Problem


Clearly describe the specific problem, explain its likely cause(s), and provide a clear statement of
the actions the device user should follow to diagnose and correct the problem. Use as few words as

TAF-ER562X_WINGR-09-0401-C005.indd 171 11/9/10 12:40:51 AM


172 Handbook of Human Factors in Medical Device Design

possible to help the user identify the observed problem. Do not confuse the user with the technical
reasons for the problem unless the reasons are important to the corrective action (FDA, 2001).

Guideline 5.35: Write Clear Procedures for User-Fixable Problems


If device users can fix a problem, provide them with clear, step-by-step instructions (refer to
the “Procedures” section in this chapter for guidance on writing these instructions).

Guideline 5.36: Use Everyday Language


Use everyday language descriptions, not technical terminology or error codes, to tell users
what is wrong with a device. Make it easy for users to understand and diagnose a problem.

Guideline 5.37: List and Explain Error Messages


If the device displays error messages, list each one separately, explain what it means, and
describe the actions that are needed to resolve the error condition (FDA, 2001).

Guideline 5.38: Tell Users What to Do if They


Cannot Fix the Problem
If there are problems that users cannot or should not try to fix, provide clear and unambiguous
information to that effect, including any hazards associated with user attempts to correct the
problem. Provide contact information needed to obtain assistance (e.g., toll-free telephone num-
bers, e-mail addresses, Web addresses). Tell users how to report undesirable outcomes (adverse
events), including device malfunctions and injury resulting from device use (FDA, 2001).

5.7.1.1.4 Warnings and Precautions


Warnings and precautions convey information about device operating conditions that pose
potential risks to device users, patients, the environment, or the device itself. The distinc-
tion between warnings and precautions is one of degree of likelihood and seriousness of the
adverse event. Warnings alert users to situations that, if not avoided, could cause death or
serious injury. In contrast, precautions alert users to potentially hazardous situations that, if
not avoided, may result in minor or moderate injury to the patient or device user or damage
to the device but that are unlikely to result in serious injury or death (American National
Standards Institute [ANSI], 2002c).
Refer to Chapter 13, “Signs, Symbols, and Markings,” for an in-depth discussion of
warnings. For more information on warning label design, consult ANSI Z535 (ANSI,
2002a, 2002b, 2002c, 2002d), which contains information on safety color codes, icons, and
safety symbols. Ryan (1991) and Wogalter (2006) provide extensive additional information
on the design and application of warnings.

Guideline 5.39: Label Warnings and Precautions


The signal word “WARNING” is generally used to designate a warning hazard alert. The sig-
nal word “CAUTION” is generally used to designate a precaution hazard alert (FDA, 2001).

Guideline 5.40: Place Warnings and Cautions Immediately


before the Step to Which They Apply
Place warnings and cautions immediately before the step to which they apply to increase the
probability that the user will read them (Wogalter, Godfrey, & Fontenelle, 1987).

TAF-ER562X_WINGR-09-0401-C005.indd 172 11/9/10 12:40:51 AM


Documentation 173

Guideline 5.41: Information Elements for Warnings and Precautions


Warnings and precautions should include four key elements:

• Signal word (WARNING or CAUTION) accompanied by the international standard


safety alert symbol
• Hazard statement: a brief description of the potential hazard
• Consequences: what will likely happen if instructions are not followed
• Description of the actions required to avoid or minimize the hazard (“do’s” and “don’ts”)

The signal word must come first. The other three elements may be presented in any order that
makes sense.

Guideline 5.42: Highlight the Signal Word


The signal word should be highlighted by some combination of bolding, white space, or a
contrasting color to make it stand out from adjacent text and graphics. Red is often used for
warnings, while yellow is the common color for cautions.

Guideline 5.43: Use Standardized Symbols and Icons


Whenever possible, use internationally recognized standards as a source of symbols and icons.
ANSI Z535 contains standardized and widely followed guidance “for the design, application,
and use of signs, colors and symbols intended to identify and warn against specific hazards
and for other accident prevention purposes” (ANSI, 2002a, p. v).

Guideline 5.44: Use the Standard Safety Alert Symbol


The standard safety alert symbol, shown in Figure 5.2, indicates a potential injury hazard. It
is composed of an equilateral triangle surrounding an exclamation mark. This symbol should
not be used to alert persons to property damage (ANSI, 2002b).

Guideline 5.45: Test New Symbols and Icons Prior to Use


If you use nonstandard or custom symbols or icons, test them before including them in your
documentation. Use testing procedures like those described in ISO 9186 (International
Organization for Standardization, 2001) to help ensure that users will interpret your symbols
in the way you intend.

Guideline 5.46: Hazard Statement


Provide a realistic and accurate description of the harmful effects associated with device use.

FIGURE 5.2 Examples of safety alert symbols recommended by ANSI Z535.

TAF-ER562X_WINGR-09-0401-C005.indd 173 11/9/10 12:40:51 AM


174 Handbook of Human Factors in Medical Device Design

Guideline 5.47: Inform of any Unusual Hazards


Be sure to inform users about unusual hazards (i.e., those not generally known by the user
population) as well as any hazards specific to the particular model of the device being used.

Guideline 5.48: Use Warning and Precaution Statements Judiciously


If documentation includes many warnings and precautions, especially those address-
ing risks already known to the user, the warnings and precautions will lose their desired
effect.

Guideline 5.49: Make Content Appropriate for Intended Users


Choose the information content for warnings and precautions carefully and with consider-
ation of the intended users’ backgrounds.

Guideline 5.50: Include Any Relevant Environmental Factors


Describe the environmental factors that can interfere with device use and result in harm-
ful effects. This includes the care needed to avoid damaging a device as a result of use or
misuse.

Guideline 5.51: Locate Warnings and Precautions Appropriately


Warnings and precautions that cannot be taken out of context without losing their meaning
(are closely tied to procedural steps) should be located immediately before the associated step
(FDA, 2001). Warnings and precautions are most effective when integrated into the relevant
task information at an appropriate location. Medical device documentation has traditionally
presented warnings and precautions in a separate section set aside for them. Given the reluc-
tance of device manufacturers to break with tradition, it is recommended that the warnings
and precautions placed in separate sections be those that can be taken out of procedural con-
text without losing their meaning.

5.7.1.1.5 Supplementary Information


Supplementary information consists of information that device users may find helpful over
the course of using a device over time, such as manufacturer contact information, device
technical information, and patient information. Patient information may include a means
for recording test results (especially for home medical devices) and a description of the
treatment or monitoring regimen that a patient is following.

Guideline 5.52: Manufacturer Contact Information


Make manufacturer contact information easy to find, such as by placing it at the beginning
or end of the documentation. Include the device manufacturer’s name and address, customer
assistance and technical support telephone numbers, and e-mail and Web addresses.
Corresponding information can also be furnished for the device distributor.

Guideline 5.53: Device Information


Provide specific and complete information about the device. List the complete brand name,
model number, and date of manufacture of the device. If software is included, list the version
of the software used by the device and its date of release.

TAF-ER562X_WINGR-09-0401-C005.indd 174 11/9/10 12:40:52 AM


Documentation 175

Guideline 5.54: Date of Printing


Provide the date the documentation was issued or revised where it can be found easily (e.g.,
cover or last page). The FDA requires dated labeling for prescription devices and recommends
it for all other devices (FDA, 2001).

Guideline 5.55: Record of Patient Test Results


Some medical devices, especially those used in the home by laypersons, measure some physi-
ological or biochemical process in a series of tests over an extended time period. In these
cases, include a form for recording the date, time, and results of each test, even if the device’s
internal memory maintains a record. It may become corrupted and the data lost.

Guideline 5.56: Frequently Asked Questions (FAQs)


It may be helpful to include a list of frequently asked questions with answers to them. This can
help new or infrequent users of a device to better understand its operation and maintenance.

5.7.1.1.6 Translations
Given the requirement of European Union communities for medical device labeling to be
provided in the national languages of the countries in which a medical device is marketed,
translations take on added importance (see also Chapter 19, “Cross-National and Cross-
Cultural Design of Medical Devices”).

Guideline 5.57: Provide Translations if Users Have


Other Primary Languages
Translate the documentation into other languages for any anticipated user populations.

Guideline 5.58: Take Regional Differences into Account


Make sure translations take into account any regional/idiomatic differences within the same
language (e.g., intra–Latin American variability in Spanish usage).

Guideline 5.59: Be Aware of Conversational Idiosyncrasies


A “straight” translation done without regard to conversational idiosyncrasies of language
usage (e.g., figures of speech) can yield inaccurate and misleading results. The translation
should be done by someone who is thoroughly familiar with the conversational use of the
language(s) involved.

Guideline 5.60: Back Translate to Verify Accuracy


It is advisable to perform a back translation of the translated documentation. Compare
the results with the original text to detect and correct misinterpretations and unintended
nuances of meaning. The back translation should be done by someone other than the original
translator.

5.7.1.2 Presentation Guidelines


In contrast to content guidelines, which focus on what information should be provided to
device users, presentation guidelines specify how that information should be presented.
The way in which information is conveyed contributes to how well people understand,

TAF-ER562X_WINGR-09-0401-C005.indd 175 11/9/10 12:40:52 AM


176 Handbook of Human Factors in Medical Device Design

remember, and comply with it. This section organizes presentation guidelines into six
categories:
• Language and readability
• Organization and layout
• Illustrations
• Highlighting
• Typography and legibility
• Physical characteristics

5.7.1.2.1 Language and Readability


Language is the use of spoken or written words to communicate information. The correct,
succinct, and clear use of words minimizes reading effort and makes documentation under-
standable to as many readers as possible. Readability assesses the extent to which printed
text can be read easily. It depends on writing style (e.g., word choice, sentence length, com-
plexity) rather than information content. Together, language and readability are important
contributors to the usability of documentation. This section presents guidelines for making
documentation easy to read and understand.

Guideline 5.61: Be Concise


Use short (less than two lines), direct, positive sentences or phrases that tell the user what
to do. This is especially important for steps in procedures. However, do not sacrifice reader
comprehension for the sake of brevity.

Poor: Be aware of the fact that dampness may cause rust and adversely affect the operation
of the device.
Better: Dampness can rust the device and it may not operate properly.

Guideline 5.62: Active Voice


Use the active voice rather than the passive voice for telling someone what to do.

Poor: Information is presented by the user manual on how to use the device.
Better: The user manual presents information on device use.

Guideline 5.63: Use an “Inverted Pyramid” Writing Style


Present the most important idea at the beginning of a passage, followed by any necessary
details and supporting information. Assume that the reader could lose interest or become
distracted at any time.

Guideline 5.64: Use Simple, Familiar, Everyday Words


Minimize the use of long words and complicated expressions. Otherwise, users may become
confused and be more error prone.

Guideline 5.65: Describe Device Operating Procedures Concisely


Divide procedures into several short ordered steps instead of fewer longer paragraphs; this
improves comprehension and reading speed.

TAF-ER562X_WINGR-09-0401-C005.indd 176 11/9/10 12:40:52 AM


Documentation 177

Guideline 5.66: Keep Readability at or Below the


Sixth- to Seventh-Grade Level
Documentation written at the sixth- to seventh-grade level will be understood by most device
users. Try to use words with fewer than four syllables and limit sentences to fewer than 25
words (Backinger & Kingsley, 1993). Ask yourself this question: “How should this text
be written so that a bright youngster between 10 and 12 years old would understand it?”
(Wiklund, 1998). Although this guideline may be less important for devices used exclusively
by highly trained physicians, for example, remember that under heavy workload, stress, and
time pressure, cognitive capabilities can be appreciably impaired.

Guideline 5.67: Use Specific and Unambiguous Words or Phrases


Avoid vague, awkward, and ambiguous words or phrases—they may cause users to misunder-
stand the instructions and make errors.

Poor: Respond quickly.


Better: Respond within one minute.
Poor: Device operates unreliably in a cool room.
Better: Device will not operate reliably below 60ºF.

Guideline 5.68: Avoid Medical Jargon


Minimize technical terminology and acronyms (except in cases where they are more recog-
nizable than the spelled-out words). This is especially important for documentation intended
for lay users.

Poor: The device should not be used by ESRD patients.


Better: The device should not be used by patients with end-stage renal disease.

Guideline 5.69: Minimize the Use of Abbreviations


If using abbreviations, define them when they are first used in each section instead of just the
very first time they are used in the documentation. If more than five abbreviations are used
throughout the documentation, provide a table of all abbreviations.

Guideline 5.70: Use Terminology Consistently


Use the same terms to refer to the same device features and actions throughout the documen-
tation, even if synonyms seem appropriate. Terminology changes can confuse users, espe-
cially those unfamiliar with the given language or technology. For example, if you initially
call a component a “dial,” do not call it a “knob” later.

5.7.1.2.2 Organization and Layout


Organization refers to how topics are arranged in relation to each other in documenta-
tion. Organization establishes the logical flow of information necessary for understand-
ing and remembering how to operate a device. Layout refers to how information is
formatted on a page or screen. A well-designed layout makes it easier to locate and use
information.

Guideline 5.71: Provide a Table of Contents


The headings in the table of contents should correspond word for word with the headings and
subheadings in the text. The headings should clearly convey the content of each section of the

TAF-ER562X_WINGR-09-0401-C005.indd 177 11/9/10 12:40:52 AM


178 Handbook of Human Factors in Medical Device Design

documentation and show the flow of actions involved in device use. Page numbers are desig-
nated for each topic to make them easy to locate. Figure 5.3 shows a table of contents from a
contact lens care prototype manual that incorporates these features.

Guideline 5.72: Establish a Logical Sequence of Information


Organize topics according to the sequence of information and actions needed to operate a
device safely and effectively. The exact composition and sequence of information depends on
the type of device, but the following organization is a sound general approach:
A) An Opening section should contain background information on the purpose and func-
tioning of the device. Any significant hazards associated with device use should be cited
here, although they should be reiterated in a later section on warnings. Contact informa-
tion can also be included here.
B) A Procedures section should present step-by-step instructions involved in using a device.
Procedures should be presented in the same order the user would typically follow when
operating the device.
C) A section on Warnings should help users make informed decisions about potential
health hazards associated with device use.
D) A Troubleshooting section should tell users how to deal with problems in device
operation. A tabular format can be helpful for organizing troubleshooting informa-
tion so that it is easy to use. Use one column for signs of trouble and another for user
actions.

Guideline 5.73: Use Clear Titles, Headings, and Subheadings


Titles, headings, and subheadings establish the documentation’s organization. Use labels for
headings and subheadings that clearly describe what information is in a given section. This
will speed location of information, improve user memory for device operation, and reinforce
the order in which procedures should be performed.

Table of Contents
Lens Care Chart ........................................................................................ 2
Introduction .............................................................................................. 3
General Precautions ................................................................................. 4
Putting On Lenses .................................................................................... 9
Taking Off Lenses ..................................................................................... 17
Cleaning and Disinfection ............................................................................ 21
Cold (Chemical) Disinfection ................................................................ 26
Heat (Thermal) Disinfection .................................................................. 28
Emergency Heat Disinfection ................................................................ 30
Enzyme Cleaning ...................................................................................... 32
Lens Case Cleaning and Disinfection ................................................... 33
What To Do When You Have Problems While Wearing Lenses.......... 35

FIGURE 5.3 Example of a table of contents with descriptive headings to convey the organization
and flow of topics.

TAF-ER562X_WINGR-09-0401-C005.indd 178 11/9/10 12:40:52 AM


Documentation 179

Guideline 5.74: Use Indentations to Show Information Relationships


Indentations can more clearly show relationships among the various parts of a section, such
as device operating procedures and the step-by-step instructions needed to perform those
procedures.

Guideline 5.75: Make it Easy to Locate Information


Users should be able to quickly and easily find the information they need. Carefully design
those aspects of documentation that help users locate information, such as page numbers,
page and section headings, section outlines, and indexes.

Guideline 5.76: Include an Index


An index, in alphabetical order at the end of documentation, should include all important
topics covered by the documentation and the page numbers on which they are discussed
(Backinger & Kingsley, 1993).

Figures 5.4 and 5.5 show how poorly laid-out procedures can be improved through the
use of numbered step-by-step instructions and closely linked line drawing illustrations.

5.7.1.2.3 Illustrations
Properly chosen and placed illustrations can greatly facilitate understanding of how to
operate a device. Illustrations are usually remembered better than words (Kobus, Moses, &
Bloom, 1994). Effective illustrations clarify text while reducing the amount of informa-
tion that must be communicated via text. They also reduce readers’ reliance on text, a big
advantage for poor readers or users whose native language is different from the language
used by the documentation.
(a)
Pick up the two electrodes
(with the wires inserted) so
the electrode with the white
wire is in your left hand
while that with the black wire
is in your right hand. Now
turn the electrodes so the
Velcro side is facing down. White end
to the
right side

(b)

1. Attach the electrodes (they will


be under the armpits when
the belt is closed).

2. Check to be sure that the white


wire is on baby’s right. Whitewire
on baby’s
right.

FIGURE 5.4 Comparison of (a) poor mapping and (b) clear mapping of text to illustration.

TAF-ER562X_WINGR-09-0401-C005.indd 179 11/9/10 12:40:52 AM


180 Handbook of Human Factors in Medical Device Design

(a)
Prick your finger with a sterile lancet and massage gently if
necessary to obtain a large drop of blood. Quickly turn your
finger upside-down so that the drop is hanging from the finger.
With practice, you will be able to make the drop quite large
without it actually falling off your finger. Do not proceed to the
next step until your blood sample is ready.

(b)
Obtain a drop of blood as follows:

1. Prick your finger.


To lessen the discomfort,
prick the side of your finger,
not the finger pad.

2. Hold your finger downward and


squeeze it firmly to form a large
hanging drop of blood.

FIGURE 5.5 Original (a) versus improved (b) end-user instructions for medical device use.

Guideline 5.77: Make Illustrations Simple and Clear


Illustrations should be clear, simple, and uncluttered. Represent only simple concepts, such
as required user actions, features of the device, or aspects of its surroundings. Use a separate
illustration for each distinct concept (FDA, 2001).

Guideline 5.78: Confine Action Illustrations to a Single Action


Focus on a single action at a time instead of trying to depict multiple actions in the same
illustration.

Guideline 5.79: Make Comparisons Explicit


If two illustrations are compared, make the difference between them clear. Use a visual
means, such as circles or arrows, to point out the difference (FDA, 2001).

Guideline 5.80: Make Illustrations Large Enough for Easy Viewing


Illustrations should be large enough for users to see the key details and important words easily.

Guideline 5.81: Illustrations Should Clarify Text


Use illustrations to clarify and reinforce the accompanying text. This will help readers to bet-
ter understand the information conveyed by the documentation. Illustrations can also show
aspects of device operation that are hard to express verbally.

Guideline 5.82: Link Illustrations to Text


An illustration should always be accompanied by clear explanatory text. Set off text and its
accompanying illustration by the use of lines, white space, or titles (FDA, 2001). Figure 5.6
shows how call-outs can be used to supplement an illustration of a device interface. The
individual items called out in an illustration can be numbered so that they correspond with a
numbered list in the accompanying text that describes each item.

TAF-ER562X_WINGR-09-0401-C005.indd 180 11/9/10 12:40:52 AM


Documentation 181

(1) Delivery rate display (2) Center (3) Dose


display limit
display

(12) Delivery
rate arrow
selectors Delivery Rate Dose Limit

88 ml/h 88 ml/h
(4) Dose limit
arrow
(11) Volume
selectors
delivered
Volume Delivery Dose Delivery
message (5) Dose
display 88 ml CLEAR 88 ml delivery
display
LIMIT
START RESET SILENCE ON/OFF
CLEAR

(10) Start
button
(9) Reset (8) Clear (7) Silence (6) Limit
button button button on/off/clear
button

FIGURE 5.6 Example of the use of call-outs to illustrate the features of a medical device (in this
case, an infusion pump). Each call-out is keyed to a same-numbered description in the text.

Several types of illustrations can be used in device documentation; each type is best suited
for a particular purpose. The alternatives are discussed in the following guidelines.

Guideline 5.83: Eliminate Unnecessary Details


To maximize comprehension, illustrations should focus on key user interface elements and
omit unnecessary details.

Guideline 5.84: Use Line Drawings for Details


Use line drawings to convey detailed aspects of device operation and device components.
Line drawings emphasize specific details and object dimensions (Bailey, 1989). Line draw-
ings should have dark, sharp lines for good contrast. Procedural steps are often more effec-
tively illustrated with line drawings or cartoons rather than with photographs.

Guideline 5.85: Use Photographs for Realism


Photographs convey the exact appearance of objects and show them in three dimensions.
They are best for realistically depicting actual device operating conditions. However, photo-
graphs may have poor contrast and include unwanted details that can distract the user’s atten-
tion from critical aspects of device operation.

Guideline 5.86: Limit the Use of Exploded Views


Use exploded views only for devices that the user should put together or take apart (FDA,
2001). Exploded views are more difficult to place in context. In addition, some users may
be tempted to dismantle and/or tamper with a device if such a view is provided in the user
manual, for example.

TAF-ER562X_WINGR-09-0401-C005.indd 181 11/9/10 12:40:53 AM


182 Handbook of Human Factors in Medical Device Design

Guideline 5.87: Use Tables for Comparisons


Tables present text or numerical comparisons across more than one attribute or category.
Include instructions on a table’s use and clearly label its sections (FDA, 2001). In general,
quick reference guides do not require tables.

Guideline 5.88: Use Graphs for Trends


Graphs show trends and how one factor varies with one or more other factors. Graphs might,
for example, be used to depict changes in calibrating settings or operating conditions at dif-
ferent environments, temperatures, or altitudes.

Guideline 5.89: Color versus Black-and-White Illustrations


Carefully evaluate the importance of color in understanding device operation when choos-
ing between black-and-white and color illustrations. Color illustrations attract and hold read-
ers’ attention better than black-and-white illustrations because of their lifelike character and
greater conspicuity (Marcus, 1992). In addition, research has found that users have a strong
preference for color illustrations (Callan et al., 1993). This preference results in more atten-
tion being given to instructions containing color illustrations. On the other hand, there is
no evidence that color illustrations increase compliance with device operating procedures,
although color used with other highlighting methods has been shown to improve memory for
task features (Young & Wogalter, 1990).

5.7.1.2.4 Highlighting
Highlighting provides visual relief, stresses important points, and sets off sections and
subsections of text. The guidelines in this section can be used to bring important words and
phrases to the reader’s attention.

Guideline 5.90: Use a Small Number of Highlighting Techniques


In most cases, two or three carefully selected highlighting techniques are better than a larger
number. Readers may become confused or dismissive if presented with many forms of
highlighting.

Guideline 5.91: Use Highlighting Consistently


Use the same technique for the same type of information throughout the documentation.

Guideline 5.92: Use Highlighting Judiciously


The effectiveness of highlighting will be reduced if it is overused. When overused, highlight-
ing can distract users, especially novice, older, or infrequent users.

Guideline 5.93: White Space


White space (i.e., blank space) is a simple yet effective highlighting technique. It can be used
between sections or to set off important information. White space also improves the appear-
ance of documentation and makes it easier to read. Figure 5.7 shows examples of white space
that address the following points:
A) Do not use excessive white space between lines of text. It impairs reading speed, com-
prehension, and legibility. It also lengthens documentation unnecessarily and increases
printing costs.

TAF-ER562X_WINGR-09-0401-C005.indd 182 11/9/10 12:40:53 AM


Documentation 183

Too much white space between lines impairs reading speed,


comprehension, and legibility. Excessive white space also increases
printing costs and makes a set of instructions unnecessarily lengthy—
undesirable consequences for both documentation developer and user.

On the other hand, documentation with insufficient white space often looks
cramped and is hard to read. The lines will blur together for many readers,
especially those with poor vision. This will quickly lead to eye fatigue and a
tendency on the reader’s part to abandon the documentation.

White space can also be used to delimit different sections in a set of


instructions. It can draw the reader’s attention to parts of labeling the
author wishes to emphasize. For example, white space can lead the

reader to specific information that the author wishes to emphasize, like this.

The eye is drawn to information that the author wants the reader to be sure
to notice. Other types of highlighting techniques can be used in conjunction
with white space to enhance its effect.

FIGURE 5.7 Examples of the use of white space.

B) On the other hand, if the amount of space between lines is too small, the lines will blur
together for many readers, especially those with poor vision. Documentation with insuf-
ficient white space often looks cramped and is hard to read.
C) White space also can be used to delimit different sections in a set of instructions. It can
draw the reader’s attention to those parts of the documentation the author wishes to
emphasize.

Guideline 5.94: Bolding


Bolding is another simple yet effective highlighting technique. It can be used in conjunction
with white space to emphasize critical information.

Guideline 5.95: Limit the Number of Colors


The background and major page or screen components should be limited to three to five col-
ors, including shades of gray. Be mindful that medical conventions are associated with some
colors (e.g., red for alarms or arterial blood pressure values) (Wiklund, 1998).

Guideline 5.96: Provide Redundant Coding for Color


Always use color highlighting with another highlighting technique (e.g., bolding). Otherwise,
users with a color vision impairment may not be aware that some text is more important than
other text (Nielsen, 2000).

Given the ease of manipulating documents using word processors, numerous other high-
lighting techniques are available. Carefully evaluate how your documentation will be used
when deciding which, if any, of these other highlighting techniques to use.

TAF-ER562X_WINGR-09-0401-C005.indd 183 11/9/10 12:40:53 AM


184 Handbook of Human Factors in Medical Device Design

5.7.1.2.5 Typography and Legibility


Typography is the arrangement, style, and general appearance of material printed from
type. Typography contributes to a consistent “look and feel” for a document, which helps
users gain familiarity with it and understand how it is organized. Three important factors
for typographical variables are font size, font type, and line characteristics. Also discussed
are color and contrast.

5.7.1.2.5.1 Font Size Figure 5.8 shows several different font sizes and their recommended
applications.

Guideline 5.97: Recommended Font Size for All-Around Use


Use a large enough font size for text to be legible to the intended user audience under a wide
range of use conditions. Twelve-point is an excellent compromise between legibility and the
need to conserve space.

Guideline 5.98: Minimum Acceptable Font Size


Ten-point font is an acceptable minimum size for general audiences. However, it is too small
for older users or others with visual impairments, who will find small fonts hard, if not impos-
sible, to read.

Guideline 5.99: Avoid 9-Point and Smaller Font Sizes


Do not use 9-point and smaller fonts. Text in small fonts is read more slowly than text in larger
fonts. Many readers will tend to skip over it or develop eyestrain.

Guideline 5.100: Larger Fonts have their Place


Many home medical device users are older people who need larger fonts to read the
documentation:

• Fourteen-point font is good for readers with visual impairments and older users.

Do not use 9-point and smaller font. It is too small to be legible to most readers.

10-point font is an acceptable minimum size for general audiences but not for older users
and persons with visual impairments.
12-point font is an excellent compromise between the need to conserve space
and to present legible instructions.
14-point font is good for visually impaired readers and older users.
It is also useful for highlighting important information.
18-point font should be used sparingly, if at all.
Reserve it for critical items you want to be sure the
reader notices.
FIGURE 5.8 Examples of different font sizes and their recommended applications.

TAF-ER562X_WINGR-09-0401-C005.indd 184 11/9/10 12:40:53 AM


Documentation 185

• Use 18-point font sparingly and only for critical items you want to be sure the reader notices.
If used to any great extent, 18-point font would make the documentation very long.

5.7.1.2.5.2 Font Type


Guideline 5.101: Serif versus Sans Serif Fonts
For body text, the choice between a serif font (e.g., Times New Roman) and a sans serif font
(e.g., Arial) is largely a matter of personal taste. Limited differences have been found between
serif and sans serif fonts in terms of legibility or reader performance.

Guideline 5.102: Use Sans Serif Fonts for Headings


Sans serif fonts are preferred for headings due to their “cleaner” appearance.

Guideline 5.103: Do Not Mix Serif and Sans Serif Fonts


When used within the same text passage, a mixture of serif and sans serif fonts can slow read-
ing speed (Boyarski et al., 1998).

Guideline 5.104: Use All Capitals and Italics Judiciously


Use of text in all capital letters or all italics reduces legibility and slows reading speed by as
much as 15% to 20% (Tinker, 1963). Text printed in all capital letters also takes up more space.
All-capital letter phrases can be used sparingly for headings or phrases (e.g., warning signal
words) as a means of emphasis.

Guideline 5.105: Minimize the Use of Multiple Font Types


Frequently switching from one font type to another will distract the reader and reduce reading
speed.

Guideline 5.106: Be Consistent


Pick a small number of commonly used fonts and use them consistently throughout your
documentation. It is common practice to use one font type for body text and another for titles,
headings, and subheadings.

5.7.1.2.5.3 Line Characteristics


Guideline 5.107: Choose a Line Length for Reading Ease
The best line length for documentation printed in 12-point font is 4.0 ± 1.25 inches. Longer
lines may strain the eye as it scans across their entire length, making it easier to jump to the
wrong next line. This is an especially crucial consideration for medical devices, where the
steps of each operating procedure must be performed in their correct sequence.

Guideline 5.108: Avoid Very Short Line Lengths


Do not use a line length less than 2.5 inches. It slows reading speed because of the large num-
ber of back-and-forth eye movements required while reading even a single sentence.

Guideline 5.109: Use Ragged-Right Margins


A ragged-right margin makes it easier to move from one line to the next than does fully justi-
fied, center-justified or right-justified text. Readers can keep track of their place better because
the ragged-right profile helps distinguish one line from another. Center-justified lines produce

TAF-ER562X_WINGR-09-0401-C005.indd 185 11/9/10 12:40:54 AM


186 Handbook of Human Factors in Medical Device Design

a “riverbank” effect that makes it harder to locate the next line. Fully justified text requires
the eye to adjust to variable spacing between words. In addition, fully justified text typically
includes hyphenated words, which further slows reading speed.

Guideline 5.110: Minimize Hyphenation


Minimize the use of hyphenation, especially with short words. Hyphenation requires the
reader to remember the last syllable on the previous line. Persons with limited vision or poor
memory often find this to be difficult.

5.7.1.2.5.4 Color and Contrast


Guideline 5.111: Maintain High Print-to-Background Contrast
Black text on a white background is a universal standard for print contrast. It provides the best
contrast and is easiest to read. However, other color combinations are acceptable as long as
they maintain a high contrast between the printed material and the background. Some accept-
able alternative color combinations are black on yellow, dark blue on white, and dark green on
white (Backinger & Kingsley, 1993; see also Chapter 8, “Visual Displays”).

Guideline 5.112: Be Careful with Color Combinations


Avoid color combinations with poor contrast, such as red text on a black background or light-
colored text on a light background, which may make it hard to read the printed material.

Guideline 5.113: Avoid Highly Saturated Colors


Highly saturated colors are difficult to look at for extended periods without developing eye
fatigue.

Guideline 5.114: Avoid Light Blue and Pastel Colors


Blue is hard for many older people to see. Eight percent of males cannot distinguish pastel
colors (Backinger & Kingsley, 1993).

5.7.1.2.6 Physical Characteristics


The guidelines in this section apply mostly to paper documentation. Physical characteristics
of documentation affect its ease of use and subjective appeal and contribute to how well its
content can be understood, followed, and remembered. Desirable physical characteristics
for documentation stem from two factors: the documentation’s environment of use and its
updating requirements.

Guideline 5.115: Use Heavy Dull-Finish Coated Paper


Paper with a dull finish is better than glossy paper, which can direct distracting reflections
toward the eye. Paper should be heavy enough to prevent show-through and to resist tearing
with repeated use. It should also be coated to make it resistant to spillage and soiling if the
document is likely to be used under conditions in which this is probable.

Guideline 5.116: Choose between Landscape and Portrait Orientation


Choose the orientation that best allows interrelated text, illustrations, and warnings to be
displayed next to each other.

TAF-ER562X_WINGR-09-0401-C005.indd 186 11/9/10 12:40:54 AM


Documentation 187

Guideline 5.117: Choose Appropriate Document Size


User manuals and quick reference guides must be designed for ease of access and use during
device operation, which necessitates a smaller format. Other documents, such as technical
manuals, are more likely to be referred to when a device is not being operated. These docu-
ments are often book length and should be sized accordingly.

Guideline 5.118: Make the Document Easy to Update


Updating a paper document involves adding or deleting pages. Ring binding is ideal for meet-
ing this requirement. Glued or spiral bindings are preferable for documents that will not be
modified (Simpson & Casey, 1988).

Guideline 5.119: Ensure the Document Will Lay Flat


The document should lay flat without assistance when open so that users can have both hands
free to operate the device.

Guideline 5.120: Make the Document Durable


The document should be sufficiently durable to stand up to repeated use over the expected
lifetime of the device.

Guideline 5.121: Make the Document Easy to Clean


The document should be easy to clean if liquids (e.g., blood, saliva, saline) are accidentally
spilled on it. Lamination can preserve the document if it is used frequently or in a wet envi-
ronment (Backinger & Kingsley, 1993).

5.7.2 GUIDELINES FOR ELECTRONIC DOCUMENTATION


In addition to the general design guidelines presented in the preceding section, to create the
most effective devices, document designers should consider features unique to electronic
documentation. The design guidelines in this section address those features.

5.7.2.1 Navigation
Compared with paper documentation, electronic documentation gives users more ways to nav-
igate to information of interest. In paper documentation, organization aids, such as a table of
contents, tabbed pages, or an index, are used to locate information. In addition to these features,
electronic documentation uses other navigation aids, including navigation bars, search func-
tions, and hyperlinks. With ineffective organization and navigation, users can become lost or
disoriented as they move from one part of the documentation to another. The following guide-
lines will help designers create effective navigation schemes for electronic documentation.

Guideline 5.122: Make Documentation Content Explicit


The key to effective navigation is the logical and explicit organization of documentation con-
tent. Users should always know where they are, where they came from, how to return to previ-
ous screens or pages, and where they can go next (Koyani & Nall, 1999).

Guideline 5.123: Use the Same Navigation Aids on All Pages


A consistent navigational “look and feel” will make it easier for users to move around the
documentation (Koyani & Nall, 1999).

TAF-ER562X_WINGR-09-0401-C005.indd 187 11/9/10 12:40:54 AM


188 Handbook of Human Factors in Medical Device Design

Guideline 5.124: Place Meaningful Titles on Every Screen


Titles help users remain oriented and be less likely to get lost (Wiklund, 1998).

Guideline 5.125: Multiple Ways to Access Information


Navigation bars, a key word search function, hyperlinks, and breadcrumb trails are ways to
navigate electronic documentation. A site map or menu index that makes the documenta-
tion’s content explicit also provides ways to access information (Spool, Scanlon, Schroeder,
Snyder, & DeAngelo, 1997).

Guideline 5.126: Use Consistent Placement of


Navigation Aids and Controls
Commands such as “Exit,” “OK,” “Previous,” “Next,” and “Help” should be placed in the
same location on each screen or page.

Guideline 5.127: Allow Direct Access to the Home Page or Screen


Give users an easy way to return to the home page or screen. A “Go to Main Menu” or simi-
larly labeled button or icon that takes users to the home page or screen from anywhere in the
documentation accomplishes this goal.

Guideline 5.128: Provide a Key Word Search Function


Key word search allows users to quickly locate information about a specific topic or term.

5.7.2.2 Hyperlinks
Hyperlinks provide direct access between a given topic and related information. Hyperlinks
can appear in many places in electronic documentation, including tables of contents,
indexes, site maps, and header and body text.

Guideline 5.129: Clearly Indicate Hyperlinks


Always use underlines or some other visual indicator (e.g., a stacked list of items) to indi-
cate text links. Using mouse-overs to designate links can confuse users and slow them
down (Bailey, Koyani, & Nall, 2000).

Guideline 5.130: Reserve Blue-Colored Text for Hyperlinks


Blue-colored text should not be used in hyperlinked or Web documentation unless it is a
clickable hyperlink. On the Web, blue text indicates clickable text. For the same reason, try to
avoid making unclickable text red or purple because these colors often denote visited links.
Use blue underlined text for unvisited links and purple or red underlined text for visited links
(these are standard browser-default link colors) (Nielsen, 1996a, 1996b).

Guideline 5.131: Differentiate Link Labels


Make link labels different from each other to reduce navigation errors (Spool et al., 1997).

Guideline 5.132: Position Links Prominently on the Screen


Place important links (and other important information) high enough on the page that they are
visible without scrolling (Bailey et al., 2000).

TAF-ER562X_WINGR-09-0401-C005.indd 188 11/9/10 12:40:54 AM


Documentation 189

Guideline 5.133: Text Links Preferable to Graphic Links


A text link’s descriptive title increases its predictability and facilitates locating information
better than do graphic links, which are generally more difficult to interpret (Koyani & Nall,
1999). Text links change color after being visited, and graphic links do not, making it harder
for users to keep track of what portions of the documentation they have visited when graphic
links are used (Spool et al., 1997).

Guideline 5.134: Selectively Use Graphic Links


Some international symbols, such as the crossbar through a circle meaning “no,” can be
more immediately recognizable to a global audience than plain text. In such cases, the use of
graphic links may be justified, especially when users are not fluent in the language used by
the documentation.

Guideline 5.135: Avoid Long Link Labels


Use short labels for hyperlinks. A long link label may become wrapped (appear on more than
one line) and appear to be two separate links, resulting in user confusion (Spool et al., 1997).

Guideline 5.136: Use Breadcrumb Trails in Web-Based Documentation


It can be surprisingly easy for users, especially novices, to get lost when navigating electronic
documentation. One way to guard against this tendency is always to show users their location
via a specific type of hyperlink called a breadcrumb trail:
A) Breadcrumb trails show the path the user has followed to get to a given location in
the documentation and allow users to move several levels in a hierarchy at a time. Put
breadcrumb trails near the top of a Web page (Nielsen, 1999).
B) Users often leave and then return to a Web page, viewing other Web pages in the interim.
Breadcrumb trails help users reorient to a Web page when they return to it (Nielsen,
2007).

5.7.2.3 Language and Readability


People typically find it harder to read electronic documentation than paper documenta-
tion. Their eyes tend to fatigue more easily, and reading speed slows by as much as 25%.
Therefore, it is not surprising that most people scan electronic documentation rather than
read it thoroughly (Morkes & Nielsen, 1997).

Guideline 5.137: Facilitate Scanning


Use phrases and short sentences to convey information. Avoid complex sentence construction.

Guideline 5.138: Reduce Word Count


Limit the amount of text to about half the word count (or less) compared to paper documentation.

5.7.2.4 Organization and Layout


Guideline 5.139: Use a Shallow Information Hierarchy
Design electronic documentation so that its information is organized into three or fewer lay-
ers. Put as much important information in or near the top layer as possible. This will give
users faster and easier access to it (Czerwinski, Larson, & Robbins, 1998).

TAF-ER562X_WINGR-09-0401-C005.indd 189 11/9/10 12:40:54 AM


190 Handbook of Human Factors in Medical Device Design

Guideline 5.140: Limit the Number of Topics per Information Layer


To avoid overloading a user’s memory, each layer should have no more than seven topics. If
there are more than seven topics, users will tend to lose focus. Fewer than seven topics likely
will be adequate for all but the most complex devices. For example, documentation that is
three layers deep with seven options per layer contains 343 topics (7 × 7 × 7 = 343), which
should be enough for the large majority of medical devices. Documentation that is three layers
deep with four options per layer contains 64 topics (4 × 4 × 4 = 64), which should be adequate
for many medical devices.

Guideline 5.141: Keep Screen Layout Simple


A simple screen layout will help users find information and perform their tasks efficiently. Use
easily understood buttons, hyperlink labels, and other easy-to-use navigation aids (Lynch &
Horton, 1999).

Guideline 5.142: Avoid Horizontal Scrolling


Format page contents so that horizontal scrolling is not necessary. Horizontal scrolling makes
it harder to understand a page because not all of its content can be viewed at once. Horizontal
scrolling also annoys users.

5.7.2.5 Illustrations
Guideline 5.143: Take Display Size Limitations into Account
The size of the display used to present documentation may restrict the amount of detail that
can be included in illustrations. Simplified line drawings that emphasize the important aspects
of device use will be clearer than photographs in these cases.

5.7.2.6 Web-Specific Graphics


The following guidelines apply to graphics presented via the Web.

Guideline 5.144: A Few Simple Graphics Will Reduce Download Times


Use only graphics that enhance content or lead to a better understanding of the information
being presented. Users will only be willing to wait for graphics to download when the graph-
ics add value for them (Nielsen, 1999; Spool et al., 1997).

Guideline 5.145: Use a Base Resolution of 800 × 600


If graphics look good at 800 × 600 pixel resolution, they will also look good at higher resolu-
tions (Venkatacharya, 2000).

Guideline 5.146: Design Graphics Specifically for the Web


Use the 216 Web-safe colors. Many Web users still use equipment that displays only 256
colors. All Web-safe colors will be visible on all 256-color systems. Using Web-safe colors
eliminates many cross-browser inconsistencies in how colors appear on one browser or plat-
form versus another.

Guideline 5.147: Respect the Four-GIF Limit


Graphics Interchange Format (GIF) is an image compression technique commonly used for
Web graphics. Most Web servers call up four GIFs at a time, so it is a good idea to limit GIFs to
fewer than four per page. By avoiding multiple server calls for GIFs, pages will load much faster.

TAF-ER562X_WINGR-09-0401-C005.indd 190 11/9/10 12:40:54 AM


Documentation 191

Guideline 5.148: For Smaller Images, Do Not Create


Blended, Indistinct Edges or Fades
Keep everything in crisp colors. Provide a strong contrast between the image color and the
background. Do not try aliasing (images with jagged edges) small images or fonts. Using crisp
colors reduces file size and helps images look clean and bold.

Guideline 5.149: Use Aliased Images for Graphics with Text


Aliased images containing text will download faster than antialiased images (images with
smoothed-out edges). Aliased images have a smaller file size. Also, many sans-serif fonts are
easier to read when they are aliased.

5.7.2.7 Highlighting
The potential number of highlighting methods is greater for electronic documentation than
for paper documentation. Most highlighting methods for paper documentation can also be
used for electronic documentation. Highlighting methods unique to electronic documenta-
tion include flash coding, animation, and sound. As with paper documentation, only a small
number of highlighting methods should be used in any given document.

5.7.2.7.1 Flash Coding


Guideline 5.150: Use Flash (Blinking) Coding Sparingly
Use flash coding only to display urgent, high-priority information.

Guideline 5.151: Do Not Flash Text


Flash an icon or border associated with the text or display a “focus” area.

Guideline 5.152: Give Users Control over Flash Coding


Users should be able to acknowledge the event causing the flashing and suppress it if desired
(Fernandes, 1993).

Guideline 5.153: Flash Frequencies and Epilepsy


Try to avoid flash frequencies in the 10- to 50-Hz range (especially 20 Hz) to minimize the
possibility of inducing seizures in persons with photosensitive epilepsy. Sensitivity to this
effect increases with the intensity of the light and the portion of the person’s visual field that
is affected. A flashing screen or larger screen object is worse than a smaller screen object,
such as a cursor. Focusing attention on a flashing object, which is the goal of flash coding,
also increases the risk (http://trace.wisc.edu/docs/consumer_product_guidelines/consumer.
htm; Jeavons & Harding, 1975).

5.7.2.7.2 Animation
Guideline 5.154: Use Animation for Event Sequences
Animation is especially effective in depicting a sequence of actions, such as the steps involved
in device operation (see Chapter 11, “Software User Interfaces”).

Guideline 5.155: Animation Emphasizing Key Task-


Related Information
Animation should support task performance, such as drawing attention to critical aspects of
device use or to newly received information (Spool et al., 1997).

TAF-ER562X_WINGR-09-0401-C005.indd 191 11/9/10 12:40:54 AM


192 Handbook of Human Factors in Medical Device Design

Guideline 5.156: Use Animation to Enhance Explanations


Animation is particularly good for explaining ideas involving changes in time, position, and/
or process. It can convey dynamic aspects of device operation.

Guideline 5.157: Positioning Animation


Do not place animation too close to nonmoving text or it may draw users’ attention away from
the text.

Guideline 5.158: Animating Text


Control the timing of animated text so that it remains readable.

5.7.2.7.3 Sound
Guideline 5.159: Effective Use of Sound
Sound can be effective if it is carefully applied; otherwise, it will be only a distraction. Sound
should be used to do the following:
A) Provide users with information, such as the actual voice messages or tones generated by
a medical device
B) Reinforce the presentation of written or graphical information
See also Chapter 10, “Alarm Design,” for more information.

Guideline 5.160: Integrate Sound with Text and Illustrations


Use sound to help users understand how to operate a device, not just to get their attention. For
example, providing the same feedback tones that a device makes during its operation is an
effective use of sound, whereas playing background music is not.

Guideline 5.161: Provide Redundant Coding for Sound


Always supplement sound with a visual representation of the information it conveys. Some
users may have hearing impairments, use their devices in a noisy environment, or use them
in a setting that requires sound to be set at a low volume. See Chapter 3, “Environment of
Use.”

5.7.2.8 Typography
Guideline 5.162: Text on Computer Monitors
For text that appears on computer monitors, keep the following points in mind (Rubenstein, 1988):
• Text appearance will vary with resolution and monitor size. For example, 10-point font
will look smaller on a 1,024 × 768 resolution monitor than it will on an 800 × 600
resolution monitor. Users can adjust text appearance by self-selecting resolutions and
monitors.
• Point size is not comparable across font styles (see Section 5.7.1.2.5.1). Different styles
may have very different size letters at the same point size. Test view the actual fonts in
their intended point sizes to determine if they are legible.

Guideline 5.163: Use Web-Specific Fonts for Online Documentation


If documentation is presented via the Web, use fonts that have been developed specifically for
use on Web pages, such as Georgia (serif) and Verdana (sans serif).

TAF-ER562X_WINGR-09-0401-C005.indd 192 11/9/10 12:40:54 AM


Documentation 193

Guideline 5.164: Avoid Background Images, Textures, and Patterns


Nonneutral backgrounds can make it harder to read the overlying text. If a background image
must be used, blur it and make it a light shade or largely transparent so that the overlying text
stands out (Spool et al., 1997).

5.7.2.9 Physical Characteristics


Most of the physical characteristics of electronic documentation relate to the visual and
auditory features that have already been addressed. However, users may, on occasion, desire
to print out all or a portion of an electronic document.
Guideline 5.165: Create Two Versions of Online Documentation
Optimize one version for online viewing using a screen-oriented style sheet. Chunk it into
several separate files joined by hyperlinks. The second version should contain the entire docu-
ment in a single file in a format suitable for printing, such as PDF. This printer-friendly version
should preserve the pagination, formatting, and typography found in the electronic version
to facilitate easy cross-referencing from one version to the other. It should be printable on
standard letter-sized paper (8.5 × 11.0 inches) and not use specialized fonts or symbols that
may not be available on all printers. If the document is intended for international distribution,
make it compatible with A4 paper size (8.25 × 11.7 inches) (Nielsen, 2000).

5.8 CASE STUDIES


The following case studies demonstrate how to apply the human factors principles and
guidelines in this chapter to create effective device documentation.

5.8.1 CONTACT LENS CARE


A human factors evaluation of manufacturers’ documentation for cleaning soft contact
lenses was performed because of an unexpectedly high incidence of eye health problems
among soft contact lens users. This evaluation identified numerous deficiencies in both
informational content and presentation format that contributed to poor user compliance.
This noncompliance was responsible, at least in part, for microbial infection, allergic
reactions, and corneal trauma. Furthermore, a human factors usability study found that
experienced lens wearers made substantial numbers of errors when using manufacturers’
documentation to guide their lens care.
A task analysis was conducted to obtain a detailed description of the actions necessary
for compliant lens care. The tasks evaluated included procedures for hand washing, lens
removal, lens cleaning, lens disinfection, and reinsertion of the lens or placement in a lens
case. The task analysis also identified the most likely errors associated with each of these
actions. This information was then combined with human factors documentation design
guidelines to improve lens care documentation with the following features:
• Lengthy procedures were decomposed into a series of simple individual steps.
Each step features simple, direct procedural text, an accompanying illustration,
and rationales explaining the reason for performing various steps. Figure 5.9 shows
a representative page from the prototype documentation.
• Text was written at the grade 5 level and printed in a clear, legible font.
• Page layout and highlighting techniques presented the information in a clear and
simple way.

TAF-ER562X_WINGR-09-0401-C005.indd 193 11/9/10 12:40:54 AM


194 Handbook of Human Factors in Medical Device Design

Putting on lenses

Step 1 – Wash, rinse, and dry hands.

A. Wash your hands with soap that does not


have any oils or perfumes.
This will keep residues from getting on
your lenses.

B. Rinse your hands thoroughly after


washing. Always wash your hands before
handling your lenses. This will
C. Dry your hands with a lint-free towel. remove dirt and oils that could get
This will help keep lint and dust from on the lenses. Hand washing also
getting on your lenses and irritating helps prevent eye infections.
your eyes.

FIGURE 5.9 Sample lens care booklet page showing the step-by-step instructions for a contact
lens care procedure.

• Warnings and precautions were linked to the applicable procedural steps.


• Detailed line drawings illustrated the most critical steps.
• A quick reference guide was also provided that summarized lens care procedures
in an easy-to-follow format (see Figure 5.10).
Using this new documentation, both novice and experienced contact lens wearers made
significantly fewer lens care errors and reported a greater understanding of why they should
perform various actions as part of their lens care regimen.

Daily lens care Weekly lens care

1. Clean 2. Disinfect lenses Be sure to perform both of the


lenses As recommended for your lenses, following procedures once a week.
use either Cold or Heat disinfection.
a. Wash hands.
b. Place lens in Cold (chemical) Heat (thermal) Clean/disinfect Enzyme clean
palm. disinfection disinfection lens case lenses
c. Rub lens – a. Fill lens case a. Fill lens case a. Boil a pan of a. Follow
each side for with disinfecting with saline tap water. enzyme cleaner
20 seconds. solution. solution. instructions.
d. Rinse lens b. Boil empty
with saline b. Soak lenses b. Place case in case & caps for b. Clean and
solution. for 4 hours. disinfection unit. at least 10 disinfect lenses
minutes. as for daily lens
e. Place lens in c. Rinse lens c. Start care.
lens case. before wearing. disinfection. c. Air dry case &
f. Go to Step caps.
d. Rinse lens
(2). before wearing.

FIGURE 5.10 Quick reference guide summarizing the major procedures and steps in soft contact
lens care.

TAF-ER562X_WINGR-09-0401-C005.indd 194 11/9/10 12:40:54 AM


Documentation 195

5.8.2 BLOOD GLUCOSE METER OPERATION


Portable blood glucose meters have become a major tool for self-care by persons with
diabetes. Used properly, these devices provide accurate blood glucose measurements. The
results are used as a basis for various interventions, such as injecting insulin when a high
glucose reading is obtained. Accurate readings are therefore crucial to the well-being of
meter users. Providing users with clear and easy-to-follow documentation for meter use and
maintenance is critical to obtaining accurate readings.
The order in which topics are presented in documentation is a major determinant of
its usability. The table of contents for a prototype blood glucose user manual, shown in
Figure 5.11, shows that its topics follow a logical flow from general considerations about
device use to the specific procedures for operating and maintaining the device. A section on
warnings described hazards involved in device use, along with advice on how to minimize
them. A troubleshooting section was provided to help users identify and remedy problems
encountered during device use. Closing sections contained supporting information and an
index for locating specific topics.
The text that described the operating and maintenance procedures for the device consists
of a numbered sequence of steps, each of which was labeled with a short title, an excerpt
of which is shown in Figure 5.12. Each procedural step is described by one or more short
sentences or phrases written in a positive, active voice. A line drawing that depicts the
actions described by the text accompanies each step. The illustrations enable poor readers
or nonnative English speakers an alternative way to understand device operation. Warnings,
when applicable, are placed next to the relevant step but set off from the text and illustrations
to increase their salience.

TABLE OF CONTENTS

1. Introduction.............................................................................................. 1
Advantages of Your Blood Glucose Meter........................................... 1
Preparing to Use Your New Meter........................................................ 2
The Meter Kit............................................................................................. 3
Features of Your Blood Glucose Meter................................................ 6
Display Messages....................................................................................... 7
2. Using Your Meter.................................................................................. 10
Installing the Batteries........................................................................... 10
Calibrating the Meter............................................................................. 12
Doing a Blood Glucose Test................................................................. 15
3. Maintaining Your Meter....................................................................... 19
Replacing the Buffer Solution Reservoir............................................. 19
Replacing the Test Cap.......................................................................... 22
4. Warnings................................................................................................ 25
5. Troubleshooting Guide......................................................................... 26
Troubleshooting Display Messages..................................................... 27
Troubleshooting High or Low Test Results....................................... 30
6. Specific Performance Characteristics.................................................. 33
7. Technical Information.......................................................................... 34
8. Device Warranty.................................................................................... 35
9. Index....................................................................................................... 36

FIGURE 5.11 Table of contents showing the logical organization of information in a user manual
for a portable blood glucose meter.

TAF-ER562X_WINGR-09-0401-C005.indd 195 11/9/10 12:40:55 AM


196 Handbook of Human Factors in Medical Device Design

2. Slide calibration 3. Apply calibration 4. Wait 15 seconds,


bar to the right solution to test cap. then read display.
Slide the calibration Place one drop of After 15 seconds, the
bar to the right until it calibration solution unadjusted reading
locks into place. The inside the white ring on from the 200 mg/dL
meter is now ready to the test cap. calibration solution is
be calibrated. displayed for 3
“CALIBRT” is seconds.
displayed for 15
seconds. “CAL OK” is displayed
when the automatic
calibration is
completed.

FIGURE 5.12 Sample page from a prototype blood glucose meter user manual describing the
steps involved in the device’s calibration procedure.

RESOURCES
The guidelines and principles presented in this chapter are consistent with two FDA medi-
cal device guidance documents—Backinger & Kingsley (August 1993) and Food and Drug
Administration (April 2001)—both cited below. Other useful sources include:
American National Standards Institute/Association for the Advancement of Medical Instrumentation
(ANSI/AAMI). (2010). Human Factors Engineering Guidelines and Preferred Practices for
the Design of Medical Devices. ANSI/AAMI HE-75-2010. Arlington, VA: Association for the
Advancement of Medical Instrumentation.
American National Standards Institute/Association for the Advancement of Medical Instrumentation
(ANSI/AAMI). (2001). Human Factors Design Process for Medical Devices. ANSI/AAMI
HE-74-2001. Arlington, VA: Association for the Advancement of Medical Instrumentation.
Callan, J. R., Gwynne, J. W., Sawyer, D., and Tolbert, M. T. (1993, December). Principles of Medical
Device Labeling (NTIS PB 94-126851). San Diego, CA: Pacific Science & Engineering
Group. Retrieved June 29, 2009, from http://www.fda.gov/downloads/MedicalDevices/
DeviceRegulationandGuidance/Guidance Documents/UCM095300.pdf.
Inaba, K., Parsons, S. O., and Smillie, R. (2004). Guidelines for Developing Instructions. Boca
Raton, FL: CRC Press.
For additional information on the role of human factors in medical device design and devel-
opment, see the following:
Sawyer, D. (1996). Do It by Design: An Introduction to Human Factors in Medical Devices.
Washington, DC: U.S. Department of Health and Human Services, Food and Drug
Administration.

TAF-ER562X_WINGR-09-0401-C005.indd 196 11/9/10 12:40:55 AM


Documentation 197

REFERENCES
American National Standards Institute. (2002a). American National Standard for Safety Color Code.
ANSI Z-535.1-2002. New York: American National Standards Institute.
American National Standards Institute. (2002b). American National Standard for Environmental and
Facility Safety Signs. ANSI Z-535.2-2002. New York: American National Standards Institute.
American National Standards Institute. (2002c). American National Standard for Safety Symbols.
ANSI Z-535.3-2002. New York: American National Standards Institute.
American National Standards Institute. (2002d). American National Standard for Product Safety
Signs and Labels. ANSI Z-535.4-2002. New York: American National Standards Institute.
Backinger, C. L. and Kingsley, P. A. (1993, August). Write It Right: Recommendations for Developing
User Instruction Manuals for Medical Devices Used in Home Health Care (HHS Publication
FDA 93-4258). Rockville, MD: U.S. Department of Health and Human Services, Food and
Drug Administration. Retrieved June 29, 2009, from http://www.fda.gov/downloads/Medical
Devices/DeviceRegulationandGuidance/GuidanceDocuments/ucm070771.pdf.
Bailey, R. W. (1989). Human Performance Engineering (2nd ed.). Englewood Cliffs, NJ: Prentice
Hall.
Bailey, R. W., Koyani, S., and Nall, J. (2000, September 7–8). Usability testing of several health
information Web sites. In National Cancer Institute Technical Report. Bethesda, MD: National
Cancer Institute.
Bogner, M. S. (Ed.). (1994). Human Error in Medicine. Hillsdale, NJ: Lawrence Erlbaum
Associates.
Boyarski, D., Neuwirth, C., Forlizzi, J., and Regli, S. H. (1998). A study of fonts designed for screen
display. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems
(pp. 87–94). New York: ACM Press/Addison-Wesley.
Callan, J. R. and Gwynne, J. W. (1993, September). Human Factors Evaluation of Contact Lens
Labeling for Wearers of Soft Contact Lenses: Task VI final report. San Diego, CA: Pacific
Science & Engineering Group.
Callan, J. R., Gwynne, J. W., Sawyer, D., and Tolbert, M. T. (1993, December). Principles of Medical
Device Labeling (NTIS PB 94-126851). San Diego, CA: Pacific Science & Engineering
Group. Retrieved June 29, 2009, from http://www.fda.gov/downloads/MedicalDevices/
DeviceRegulationandGuidance/GuidanceDocuments/UCM095300.pdf.
Czaja, S. J. and Nair, S. N. (2006). Human factors engineering and systems design. In G. Salvendy
(Ed.), Handbook of Human Factors and Ergonomics (3rd ed., pp. 32–49). Hoboken, NJ: Wiley.
Czerwinski, M., Larson, K., and Robbins, D. (1998). Designing for navigating personal Web infor-
mation: Retrieval cues. In Proceedings of the Human Factors and Ergonomics Society 42nd
Annual Meeting (pp. 458–462). Santa Monica, CA: Human Factors and Ergonomics Society.
Donchin, Y., Gopher, D., Olin, M., Badihi, Y., Biesky, M., Sprung, C., Pizov, R., and Cotev, S.
(1995). A look into the nature and causes of human errors in the intensive care unit. Critical
Care Medicine, 23(2), 294–300.
Dumas, J. S. and Redish, J. C. (1993). A Practical Guide to Usability Testing. Norwood, NJ: Ablex.
Fernandes, K. (1993). User Interface Specifications for the Joint Maritime Command Information
System (JMCIS), Version 1.3. San Diego, CA: Naval Command, Control, and Ocean Surveillance
Center, RDT&E Division, Code 4221.
Food and Drug Administration. (2001, April). Guidance on Medical Device Patient Labeling: Final
Guidance for Industry and FDA Reviewers. Rockville, MD: Food and Drug Administration,
Center for Devices and Radiological Health. Retrieved June 29, 2009, from http://www.fda.
gov/MedicalDevices/DeviceRegulationandGuidance/GuidanceDocuments/ucm070782.htm.
Hollnagel, E. (2006). Task analysis: Why, what, and how. In G. Salvendy (Ed.), Handbook of Human
Factors and Ergonomics (3rd ed., pp. 373–383). Hoboken, NJ: Wiley.
Horton, W. (1994). Designing and Writing Online Documentation: Hypermedia for Self-Supporting
Products (2nd ed.). New York: Wiley.

TAF-ER562X_WINGR-09-0401-C005.indd 197 11/9/10 12:40:55 AM


198 Handbook of Human Factors in Medical Device Design

Inaba, K., Parsons, S. O., and Smillie, R. (2004). Guidelines for Developing Instructions. Boca
Raton, FL: CRC Press.
International Organization for Standardization. (2001). ISO 9186:2001. Graphical Symbols—
Test Methods for Judged Comprehensibility and for Comprehension. Geneva: International
Organization for Standardization.
Jeavons, P. M. and Harding, G. F. A. (1975). Photosensitive Epilepsy. London: Heinemann.
Kaye, R. and Crowley, J. (2000, July). Medical Device Use–Safety: Incorporating Human Factors
Engineering into Risk Management. Rockville, MD: Food and Drug Administration, Center
for Devices and Radiological Health.
Kingsley, P. A. (1995, November). Draft Initial Report on Medical Device Labeling: Health
Care Practitioners’ Medical Device Information and Labeling Needs, Results of Qualitative
Research. Rockville, MD: U.S. Department of Health and Human Services, Food and Drug
Administration, Center for Devices and Radiological Health.
Kingsley, P. A. (1999, August). Draft Report on Medical Device Labeling: Patients’ and Lay
Caregivers’ Medical Device Information and Labeling Needs, Results of Qualitative Research.
Rockville, MD: U.S. Department of Health and Human Services, Food and Drug Administration,
Center for Devices and Radiological Health.
Kirwan, B. and Ainsworth, L. K. (Eds.). (1992). A Guide to Task Analysis. London: Taylor & Francis.
Kobus, D. A., Moses, J. D., and Bloom, F. A. (1994). Effect of multimodal stimulus presentation of
recall. Perceptual and Motor Skills, 78, 320–322.
Koyani, S. J. and Nall, J. (1999, November). Web site design and usability guidelines. In National
Cancer Institute, Communication Technologies Branch Technical Report. Bethesda, MD:
National Cancer Institute.
Leape, L. L., Brennan, T. A., Laird, N., Lawthers, A. G., Localio, A. R., Barnes, B. A., Hebert, L.,
Newhouse, J. P., Weiler, P. C., and Hiatt, H. (1991). The nature of adverse events in hospi-
talized patients: Results of the Harvard Medical Practice Study II. New England Journal of
Medicine, 324(6), 377–384.
Lynch, P. J. and Horton, S. (1999). Web Style Guide: Basic Design Principles for Creating Web Sites.
New Haven, CT: Yale University Press. Retrieved June 29, 2009, from http://www.webstyleguide.
com/wsg3.index.html.
Marcus, A. (1992). Graphic Design for Electronic Documents and User Interfaces. New York: ACM
Press.
McCarthy, E. J., Backinger, C., Kaczmarek, R., Withiam-Wilson, M., Alexander, G., Coppola, A.,
Greishaber, M. F., Kingsley, P., Porter, B., Riley, K., et al. (1992, June). Medical Devices Used
in Home Health Care (HHS Publication FDA 92-4252). Rockville, MD: U.S. Department of
Health and Human Services, Food and Drug Administration.
Meister, D. and Enderwick, T. P. (2001). Human Factors in System Design, Development, and
Testing. Mahwah, NJ: Lawrence Erlbaum Associates.
Morkes, J. and Nielsen, J. (1997). Concise, SCANNABLE, and Objective: How to Write for the Web.
Retrieved April 13, 2010, from http://www.useit.com/papers/webwriting/writing.html.
Nemeth, C. P. (2004). Human Factors Methods for Design: Making Systems Human-Centered. Boca
Raton, FL: CRC Press.
Nielsen, J. (1994). Usability Engineering. San Francisco: Morgan Kaufmann.
Nielsen, J. (1996a). Marginalia of Web Design. Retrieved April 13, 2010, from http://www.useit.
com/alertbox/9611.html.
Nielsen, J. (1996b). Top Ten Mistakes in Web Design. Retrieved April 13, 2010, from http://www.
useit.com/alertbox/9605.html.
Nielsen, J. (1999). When Bad Design Elements Become the Standard. Retrieved April 13, 2010,
from http://www.useit.com/alertbox/991114.html.
Nielsen, J. (2000). Designing Web Usability. Indianapolis: New Riders.
Nielsen, J. (2007). Breadcrumb Navigation Increasingly Useful. Retrieved April 13, 2010, from
http://www.useit.com/alertbox/breadcrumbs.html.

TAF-ER562X_WINGR-09-0401-C005.indd 198 11/9/10 12:40:55 AM


Documentation 199

Pilot, L. R. (1999, May). Medical device labeling in the European Union. Medical Device and
Diagnostic Industry. Retrieved June 29, 2009, from http://www.devicelink.com/mddi/
archive/99/05/012.html.
Rogers, W. A. (1997). Designing for an Aging Population. Santa Monica, CA: Human Factors and
Ergonomics Society.
Rogers, W. A. and Fisk, A. D. (Eds.). (2001). Human Factors Interventions for the Health Care of
Older Adults. Mahwah, NJ: Lawrence Erlbaum Associates.
Rogers, W. A., Fisk, A. D., and Walker, N. (1996). Aging and Skilled Performance. Mahwah, NJ:
Lawrence Erlbaum Associates.
Rubenstein, R. (1988). Digital Typography: An Introduction to Type and Composition for Computer
System Design. Reading, MA: Addison-Wesley.
Rubin, J. (1994). Handbook of Usability Uesting: How to Plan, Design, and Conduct Effective Tests.
New York: Wiley.
Ryan, J. P. (1991). Design of Warning Labels and Instructions. New York: Van Nostrand Reinhold.
Simpson, H. and Casey, S. M. (1988). Developing Effective User Documentation: A Human Factors
Approach. New York: McGraw-Hill.
Spool, J. M., Scanlon, T., Schroeder, W., Snyder, C., and DeAngelo, T. (1997). Web Site Usability:
A Designer’s Guide. North Andover, MA: User Interface Engineering.
Tinker, M. A. (1963). Legibility of Print. Ames: Iowa State University Press.
U.S. Code of Federal Regulations. (2003a, April 1). Title 21 Food and Drug Administration, DHHS,
Part 801—Labeling.
U.S. Code of Federal Regulations. (2003b, April 1). Title 21 Food and Drug Administration, DHHS,
Part 809—In vitro diagnostic products for human use, Subpart B—Labeling.
Vanderheiden, G. C. (2006). Design for people with functional limitations. In G. Salvendy (Ed.),
Handbook of Human Factors and Ergonomics (3rd ed., pp. 1387–1417). Hoboken, NJ:
Wiley.
Venkatacharya, P. S. (2000, October). Techniques for Creating User-Friendly Enterprise Portals:
Developing for the Oracle Internet Platform. Retrieved April 13, 2010, from http://otn.oracle.
com/products/iportal/pdf/tech_create_user_friend_eportals.pdf.
Wieringa, D., Moore, C., and Barnes, V. (1993). Procedure Writing: Principles and Practices.
Columbus, OH: Battelle Press.
Wiklund, M. E. (1998, May). Making medical device interfaces more user-friendly. Medical Device
and Diagnostic Industry. Retrieved June 29, 2009, from http://www.devicelink.com/mddi/
archive/98/05/032.html.
Wiklund, M. E. (2002, January). Medical device user manuals: Shifting toward computeriza-
tion. Medical Device and Diagnostic Industry. Retrieved June 29, 2009, from http://www.
devicelink.com/mddi/archive/02/01/003.html.
Wiklund, M. E. (2004, February). Intuitive design: Removing obstacles also increases appeal.
Medical Device and Diagnostic Industry. Retrieved June 29, 2009, from http://www.devicelink.
com/mddi/archive/04/02/005.html.
Wogalter, M. S. (Ed.). (2006). Handbook of Warnings. Mahwah, NJ: Lawrence Erlbaum
Associates.
Wogalter, M. S., Godfrey, S. A., and Fontenelle, G. A. (1987). Effectiveness of warnings. Human
Factors, 29, 599–612.
Wright, P. (1999). Writing and information design of healthcare materials. In C. Candlin & K.
Hyland (Eds.), Writing: Texts, Processes and Practices. London: Addison-Wesley Longman.
Young, S. L. and Wogalter, M. S. (1990). Comprehension and memory of instruction manual warn-
ings: Conspicuous print and pictorial icons. Human Factors, 32, 637–649.

TAF-ER562X_WINGR-09-0401-C005.indd 199 11/9/10 12:40:55 AM


TAF-ER562X_WINGR-09-0401-C005.indd 200 11/9/10 12:40:55 AM
6 Testing and Evaluation
Edmond W. Israelski, PhD

CONTENTS
6.1 Purpose ....................................................................................................................202
6.1.1 Constraints and Limitations of This Guidance...........................................203
6.2 Types of Usability Tests ...........................................................................................203
6.2.1 Formative Usability Testing........................................................................203
6.2.2 Summative Usability Testing ......................................................................205
6.3 Principles of Good Usability Test Design ............................................................... 206
6.3.1 Critical Usability Test Design Considerations ........................................... 206
6.4 Usability Testing: Overview ....................................................................................207
6.4.1 Usability Test Plan Content ........................................................................207
6.4.1.1 Purpose ....................................................................................... 209
6.4.1.2 Setting ........................................................................................ 209
6.4.1.3 Participants ..................................................................................212
6.4.1.4 Prototypes or Simulations ...........................................................212
6.4.1.5 Methodology ...............................................................................213
6.4.1.6 Tasks ............................................................................................ 214
6.4.1.7 Usability Objectives .................................................................... 216
6.4.1.8 Data Collection ............................................................................ 217
6.4.1.9 Data Analysis .............................................................................. 217
6.4.1.10 Reporting ..................................................................................... 218
6.5 Logistics................................................................................................................... 219
6.5.1 Testing Locations ........................................................................................ 219
6.5.2 Number of Participants ............................................................................... 219
6.5.3 Recruiting Activities ...................................................................................220
6.5.4 Testing Staff Size ........................................................................................220
6.5.5 Session Length ............................................................................................220
6.5.6 Video Recording .........................................................................................221
6.5.7 Note Taker ..................................................................................................221
6.5.8 Language Translators..................................................................................221
6.5.9 Data Logging Software...............................................................................222
6.5.10 Screen Capture ...........................................................................................222
6.5.11 Eye Scan Capture........................................................................................222
6.6 Protocol-Related Activities ......................................................................................224
6.6.1 Participant Orientation................................................................................224
6.6.2 Consent Forms ............................................................................................224

201

TAF-ER562X_WINGR-09-0401-C006.indd 201 11/9/10 12:49:41 AM


202 Handbook of Human Factors in Medical Device Design

6.6.3 Nondisclosure Forms ..................................................................................224


6.6.4 Pretest Questionnaire ..................................................................................225
6.6.5 Participant Training prior to Testing ..........................................................225
6.6.6 Directed Tasks ............................................................................................225
6.6.6.1 Think-Aloud Protocol..................................................................225
6.6.6.2 Codiscovery .................................................................................226
6.6.6.3 Self-Exploration ...........................................................................226
6.6.7 Interviews ...................................................................................................227
6.6.8 Posttest Questionnaire ................................................................................227
6.6.8.1 Examples of Questionnaire Items................................................227
6.6.9 Debriefing ...................................................................................................228
6.6.10 Record Keeping ..........................................................................................228
6.6.11 Testing Team Debriefing.............................................................................229
6.7 Sources of Test Bias .................................................................................................229
6.7.1 Common Testing Mistakes .........................................................................229
6.8 Supplemental Usability Evaluation Methods ...........................................................229
6.8.1 Cognitive Walk-Through ...........................................................................230
6.8.2 Expert Review ............................................................................................230
6.8.3 Heuristic Review.........................................................................................230
Appendix 6.A: Examples of Usability Test Plans ............................................................231
Appendix 6.B: Usability Test Report Checklist ...............................................................237
Appendix 6.C: Statistical Justification for Sample Sizes in Usability Tests ................... 240
Appendix 6.D: Frequently Asked Questions ...................................................................245
Resources .........................................................................................................................248
References ........................................................................................................................248

Testing and evaluation covers usability testing and all the considerations needed to plan,
execute, analyze, and report on these studies. This chapter gives guidance on principles for
developing usability tests as well as advice on how variables are chosen, controlled, mea-
sured, and analyzed. Usability testing is a cornerstone of the best practices for the design
of medical devices.

6.1 PURPOSE
This chapter provides key methodological details on how to plan and conduct usability
evaluations to obtain valid and reliable data to judge the safety and effectiveness of the
medical device user interfaces. As described in Human Factors Design Process for Medical
Devices (ANSI/AAMI HE-74-2001), usability testing is only one part of good human fac-
tors design methodology, and it must be used in conjunction with other important methods,
such as task analysis, user profiling, use environment analysis, use error risk analysis, and
iterative user-centered design. Usability testing alone will not ensure the development of
safe, effective, and easy-to-use medical devices. This chapter covers a range of best prac-
tices for conducting usability tests. A great deal can be learned with simple usability tests
that do not require a major financial investment or elaborate equipment. The usability
testing program should be executed systematically and controlled for sources of bias and
unreliability.

TAF-ER562X_WINGR-09-0401-C006.indd 202 11/9/10 12:49:41 AM


Testing and Evaluation 203

6.1.1 CONSTRAINTS AND LIMITATIONS OF THIS GUIDANCE


Usability testing should never be considered a substitute for product validation. Usability
testing can verify that design outputs match design inputs; that is, the actual design has
been executed to match the design specifications as described in the product development
plan. Further, usability testing can validate that device usability meets usability objectives.
If the usability objectives have been validated with users through rigorous user research,
then the usability test can contribute to the validation of the user interface. Medical device
validation should include laboratory testing and possibly clinical testing to evaluate device
efficacy, reliability, safety, and performance.
Another important limitation of usability testing is the false belief that user-interface
quality can be tested into a device. Iterative or multiple rounds of design and redesign
lead to usability improvements, but if up-front user needs analysis (also called contextual
inquiry) is not conducted to understand the users, their tasks and goals, and the use environ-
ment, then usability testing alone may never allow valid usability goals or objectives to be
met. This may be true no matter how many rounds of iterative design, redesign and testing
are conducted.
Usability testing can be employed during both user-interface design verification and
validation. User-interface verification can ensure that design outputs meet design inputs for
measures of usability. User-interface validation can ensure that device use is consistent with
user requirements. This may be accomplished by having acceptance criteria for the final
summative usability test based on the following:

• Valid usability objectives that truly reflect user requirements


• Rigorous use of risk analysis and risk management in the selection and prioritiza-
tion of usability objectives as acceptance criteria
• User-interface verification and validation being only one part of overall device
verification and validation, which should include all product effectiveness and
safety measures

Table 6.1 provides definitions for the terms used in this chapter.

6.2 TYPES OF USABILITY TESTS


Usability testing includes two major categories.

6.2.1 FORMATIVE USABILITY TESTING


Formative usability testing is performed early with simulations and early working proto-
types and explores whether usability objectives are attainable. Formative testing typically
does not have strict acceptance criteria.

• Exploratory testing—Tests users performing high-level tasks or walking through


the tasks using low-fidelity simulations (e.g., paper or foam models). Concepts are
being tested at this stage of development (e.g., a usability test of a computer simu-
lation of a touch-screen user interface to a heart monitor or paper sketches of the
navigational buttons and menus for such a device).

TAF-ER562X_WINGR-09-0401-C006.indd 203 11/9/10 12:49:41 AM


204 Handbook of Human Factors in Medical Device Design

TABLE 6.1
Definitions
Term Definition
Contextual inquiry* The process of observing and working with users in their normal environment to
better understand the tasks they do and their workflow.
Effectiveness** The accuracy and completeness with which users achieve specified goals.
Efficiency** The resources expended in relation to the accuracy and completeness with which
users achieve goals. Efficiency in the context of usability is related to
“productivity” rather than to its meaning in the context of software efficiency.
Formative usability Usability testing that is performed early in the design process with simulations
testing or the earliest working prototypes and explores whether usability objectives are
attainable but does not have strict acceptance criteria.
Prototyping and iterative The design process that involves the rapid turnaround of user-interface
design prototypes or simulations that are usability tested and improved in an iterative
cycle until usability objectives are attainable.
Summative usability Usability testing that is performed in the late stages of design. These tests
testing contribute to design verification and validation. It is a recommended best
practice to have formal acceptance criteria (e.g., usability objectives for human
performance and satisfaction ratings).
Task analysis* Task analysis is a family of systematic methods that produce detailed
descriptions of the sequential and simultaneous manual and intellectual
activities of personnel who are operating, maintaining, or controlling devices or
systems.
Usability inspection Inspection methods involve analytical reviews and systematic walk-throughs of
methods user interactions with simulated or working user-interface designs to identify
usability problems.
Usability objectives* Usability objectives (or goals) are a desired quality of a user-device interaction
that may be expressed in written form, stipulating a particular usability attribute
(e.g., task speed) and performance criteria (e.g., number of seconds).
Usability testing* Procedure for determining whether the usability goals have been achieved.
Usability tests can be performed in a laboratory setting, in a simulated
environment, or in the actual environment of intended use.
User Any person who interacts with the device.
User interface The hardware and software aspects of a device that can be seen (felt, heard, or
otherwise perceived) by the user and the commands and mechanisms the user
employs to control device operation and input data. The interface includes
labeling, instructions for use, and training materials.
User group** Subset of users who are differentiated from other users by factors such as age,
culture, or expertise that are likely to influence usability (also described as a
distinct user group).
Use environment* The actual conditions and settings in which users interact with the device or
system.
Use error Use error is characterized by a repetitive pattern of failure that indicates that a
failure mode is likely to occur with use and thus has a reasonable possibility of
predictability of occurrence. Use error can be proactively identified through the
use of techniques such as usability testing and hazard analysis. Use error can be
addressed and minimized by the device designer.

TAF-ER562X_WINGR-09-0401-C006.indd 204 11/9/10 12:49:41 AM


Testing and Evaluation 205

TABLE 6.1 (CONTINUED)


Definitions
Term Definition
Use error risk analysis Risk analysis method focused on the use error component of fault and hazard
analysis for medical devices.
User error User error is characterized by an isolated pattern of failure that indicates a failure
mode that is due to fundamental errors by humans and has no reasonable
possibility of being predicted. User error is not readily preventable and cannot
feasibly be addressed by the device designer.
User profiles* Summary of the mental, physical, and demographic traits of the end-user
population as well as any special characteristics, such as occupational skills and
job requirements, that may have a bearing on design decisions.

*From Human Factors Design Process for Medical Devices (ANSI/AAMI HE-74-2001).
**From Common Industry Format for Usability Test Reports (ANSI/NCITS 354-2001).

• Assessment testing—Tests that give the users realistic tasks to perform on working
prototypes or more fully developed simulations, usually without patients attached
to the device (e.g., usability testing of the feel and control of a working prototype
of a handheld glucose meter).
• Comparison (contrast) testing—Tests comparing two or more design alternatives
(e.g., a test to measure the effectiveness and alerting properties of two sets of com-
peting auditory alarms for an infusion pump).
• Comparison (competitive) testing—Tests that gather usability performance of com-
petitors’ products. These tests could be part of the design exploration to under-
stand the best features of existing devices or could be used to support marketing
claims for a device (e.g., usability test comparing task success rate and time to run
a blood chemistry test using a variety of on-market handheld point-of-care blood
analyzers).

6.2.2 SUMMATIVE USABILITY TESTING


Summative usability testing is performed late in design as part of formal device verification
and validation. It is a recommended best practice to have formal acceptance criteria (e.g.,
usability objectives for human performance and user satisfaction ratings).

Guideline 6.1: Summative Usability Testing Sample Size


Summative usability testing should use a sufficient sample size to allow performance of sta-
tistical tests on the reliability of the results.

If iterative rounds of formative usability testing are performed as recommended in this


chapter and in the usability literature, then typically there will be few usability surprises
uncovered during late-stage summative testing. Recommendations for statistically justifi-
able sample sizes are given in Appendix 6.A. The remaining sections of this chapter cover
best practices for both formative and summative usability testing.

TAF-ER562X_WINGR-09-0401-C006.indd 205 11/9/10 12:49:41 AM


206 Handbook of Human Factors in Medical Device Design

6.3 PRINCIPLES OF GOOD USABILITY TEST DESIGN


In usability tests, representatives of the intended user population interact with one or more
device models, prototypes, or production units to assess ease of learning, efficiency of use,
ease of use, ease of remembering, safety, and/or user appeal (Nielsen, 1994), among the
many possible attributes of interest. Usability tests can be performed in a laboratory setting,
in a simulated environment, or in the actual use environment. Usability testing, especially
when conducted in the actual use environment, is a particularly effective way to detect use
errors. However, because the subject populations typically are small, use errors having
an inherently low probability of occurrence across the entire user population may not be
detected. Statistically, it can be shown that small sample size usability tests are not power-
ful enough to detect low-frequency use errors. For this reason, the use of additional comple-
mentary techniques, such as risk analysis and usability inspection methods, is essential.
Testing some types of devices to the limit of their functional capabilities may place patients
at risk, which is unacceptable.

6.3.1 CRITICAL USABILITY TEST DESIGN CONSIDERATIONS


Usability testing should be done carefully and systematically, or else the resulting data may
not be valid or reliable. Poor data may lead to poor design decisions and ultimately an error-
prone and unsafe medical device.

Guideline 6.2: Usability Testing Protocol Development


Usability testing protocols should be developed in collaboration with professional-level human
factors specialists. Nonspecialists under the direction and training of professionals can handle
the actual execution and reporting of the test.

There are many factors and characteristics that must be considered in designing a usabil-
ity testing protocol. Table 6.2 summarizes these characteristics.
A usability test has some very high-level attributes that are outlined below and explained
in more detail in subsequent sections of this chapter.

TABLE 6.2
Characteristics of a Typical Usability Test
Test Plan Content Logistics Protocol Activities Data
• Purpose • Testing locations • Orientation • Task times
• Setting • Number of • Consent • Task completion
• Participants participants • Nondisclosure • Efficiency
• Prototypes • Recruiting activities • Prequestionnaire • Significant errors
• Simulations • Testing staff size • Type of training • Ratings
• Methods • Session length • Directed tasks • Rankings
• Tasks • Video recording • Self-exploration • Verbal comments
• Usability • Note taker • Interview • Questionnaires
objectives • Language translator • Postquestionnaire • Videos
• Data collection • Data logging software • Debriefing • Photo images
• Data analysis • Screen capture
• Reporting methods • Eye scan capture

TAF-ER562X_WINGR-09-0401-C006.indd 206 11/9/10 12:49:42 AM


Testing and Evaluation 207

Guideline 6.3: Usability Testing Plan


A usability testing plan should specify the test subjects, sample size (and rationale), tasks to
be tested, test environment, device to be tested, performance measures, and specific testing
procedure.

The sample size should be small but sufficient (explained in more detail in Appendix
6.C). For many usability tests, the size may be as small as five to eight for early formative
testing (one round) and 15 to 20 for later summative testing against usability objectives.
The test should focus on realistic tasks based on user scenarios derived from previous task
analysis and risk analysis. The later stage tests should occur in a realistic simulation of the
intended use environment (e.g., lighting, noise, other equipment, and so on) and use the
actual device, a prototype, or a sufficiently interactive simulation. Testing should focus on
actual user performance (task completion, time, errors, and so on), although participant opin-
ions can provide valuable information. The test is typically recorded (audio and/or video).

6.4 USABILITY TESTING: OVERVIEW


This section gives specific guidance on the details of designing and executing a usability
test. Table 6.3 presents a high-level description of the most common types of usability
tests.

Guideline 6.4: Frequency of Usability Testing


Usability testing should be applied iteratively at various stages in the user-centered design
process.

Figure 6.1 suggests where in the device development process usability testing and its vari-
ous forms should be applied. In particular, formative usability testing is an important part
of the product design cycle as part of the process of specifying user-interface design out-
put. In an iterative process, the final user-interface specification can be achieved through
successive usability tests, each attempting to measure whether the usability requirements
(objectives) can be attained. Summative usability testing is an important part of verifica-
tion (design outputs meet design inputs) and validation (user requirements in the form of
usability objectives are met).

6.4.1 USABILITY TEST PLAN CONTENT


As noted earlier, usability testing is a scalable process. For simple, low-risk products, sim-
ple and inexpensive testing may be sufficient. For newer technology or more complex user
interfaces, a more comprehensive usability testing plan may be required. The higher the
risk indexes estimated from risk analysis, the greater the need for more comprehensive
usability testing. For example, a simple drug injection device might require only a single
formative test and one final summative usability test involving a few simple tasks. In con-
trast, a complex new infusion pump might require five or more iterative formative tests
using computer simulations of the interface screens as well as a series of investigational
tests of display readability and alarm effectiveness. The number of tasks included in the
usability tests would be sufficient to address the most critical and high-risk use scenarios.

TAF-ER562X_WINGR-09-0401-C006.indd 207 11/9/10 12:49:42 AM


208 Handbook of Human Factors in Medical Device Design

TABLE 6.3
Types of Usability Testing and Considerations
Minimum
When in Sample Size
Type Description Design Cycle (per group) Considerations

Formative Usability Testing


Exploratory High-level test of users Conceptual 5–8 Based on simulations of early
performing tasks. design or higher concepts; could be low-fidelity
paper prototypes or foam core
models. Useful to use
think-aloud protocols.
Assessment Gives representative Iterative 5–8 Early designs or computer
users real tasks to throughout the simulations. Used to gauge
perform (1:1 or two design cycle whether usability objectives
working together). are attainable.
Comparative Compares two or more During design 5–8 Used to decide if one user
and contrast design alternatives. interface concept or prototype
is better than others.
Comparative Tests against During design 5–8 Used to learn about
and competitors’ user competitors’ devices, best
competitive interfaces. features. May be done early
during conceptual design or
near the end of design cycle.

Summative Usability Testing


Validation Tests real users and real End of design At least 15–20 Used to verify and validate
tasks with market cycle (often larger) interface design; usability
ready device. objectives have acceptance
criteria. Should include training,
documentation, and labeling.

The summative usability test(s) would need to be comprehensive in terms of tasks and the
number of different representative user groups included.

Guideline 6.5: Usability Test Plan Design


Whether the usability test is simple or comprehensive, the usability test plan should describe
the goals for the usability test and how acceptance criteria will be determined. It is very
important that the usability test plan describes the following:

Purpose Tasks Methodology


Setting Usability objectives Prototypes/simulations
Participants Data collection Data analysis
Reporting

Appendix 6.B contains a Usability Test Report Checklist that describes the important
elements of a usability test plan. In the following, additional information is provided on

TAF-ER562X_WINGR-09-0401-C006.indd 208 11/9/10 12:49:42 AM


Testing and Evaluation 209

Concept Design Design


Verification Validation
phase input output
Design
control Perform Test output
activities Design Design Test against
studies & against
requirements specifications user needs
analysis input

Contextual Task analysis Prototyping/ Expert Production


inquiry simulations reviews units (or
Human User profiles equivalent)
Literature Iterative design
factors reviews Use environment
Cognitive
activities Formative walkthroughs Summative
Complaints Heuristic review usability usability
analysis testing Summative testing
Risk analysis usability
Market Risk analysis testing
Usability Field studies
research Cognitive
objectives
walkthroughs Risk analysis

FIGURE 6.1 Usability evaluation in the device design control process.

the contents of each section of the usability test plan, which is sometimes described as a
usability testing protocol.

6.4.1.1 Purpose
It is important to explain the high-level objectives or purpose for conducting the usability
test. The objectives should at least address the following questions:

• Is it exploratory or a design alternatives comparison test?


• Is it an early-stage formative test or a late-stage summative test?

6.4.1.2 Setting
The test setting can place constraints on how a test can be conducted. Testing in the actual
clinical environment might mean that some control will be lost over the variables manipu-
lated in the test. If a professional usability testing facility or laboratory is used, then more
experimental control can be achieved, for example, by systematically varying levels of
ambient light in studying a nurse’s ability to read a patient monitor. However, a laboratory
can never re-create the full contextual richness of the actual use environment. In a clinical
setting, there might be only one or two levels of ambient light. A trade-off needs to be made
among test setting choices.
High-fidelity simulations, such as mock operating rooms and intensive care units, can
provide advantages for usability testing. High fidelity in this context means that the testing
environment closely resembles the actual environment of use. These settings often include a
computer-controlled patient simulator; a full suite of medical devices, accessories, and sup-
plies; and even confederate clinicians who act like they are delivering actual patient care.
Such high-fidelity test environments may be particularly valuable for summative usability
testing performed late in the design cycle as described in Table 6.3. Figure 6.2 shows a
realistic patient simulation complete with computer-controlled patient mannequins. Figure
6.3 shows a typical usability lab as seen from the control room, which is the room for con-
trolling the videotaping equipment and systems being tested. Figure 6.4 shows a typical

TAF-ER562X_WINGR-09-0401-C006.indd 209 11/9/10 12:49:42 AM


210 Handbook of Human Factors in Medical Device Design

FIGURE 6.2 Patient simulator lab configured for usability testing.

FIGURE 6.3 Typical usability lab as seen from an observation and control room.

TAF-ER562X_WINGR-09-0401-C006.indd 210 11/9/10 12:49:42 AM


Testing and Evaluation 211

VCR &
camera
controls

Additional Participant
Observation Observer Side
Side
Table
Sliding
glass
door

FIGURE 6.4 Floor plan of typical usability testing lab.

floor plan of a usability testing lab. Observers and technicians can view the test participants
through one-way glass. The participants only see a mirror from their side of the one-way
glass. Finally, Figure 6.5 shows a person operating the control console of a portable usabil-
ity testing and videotaping system that is easily carried into actual clinical settings, which
in this case is a diagnostic laboratory.
Recruiting of clinical test participants would be easier in a clinical setting, but then clini-
cians could be more easily called away to attend to a critical, clinical situation unless they
are recruited to participate in the usability test while off shift. In an off-site professional
location, unplanned interruptions are less of a problem.
One consideration in choosing the proper setting is whether the test moderator should
be in the same room as the test participant. When together, the chances of the moderator
influencing the results are greater because there are more opportunities for the observer’s
body language and gestures to lead the participant, for example, inadvertently giving them
biasing hints on how to complete a task. However, some tasks require a fair amount of
physical interaction between moderator and participant, which necessitates physical co-
location. See Wiklund (1994) for a detailed discussion regarding usability testing facility
layouts and floor plans.

FIGURE 6.5 Moderator operating a portable usability testing control console.

TAF-ER562X_WINGR-09-0401-C006.indd 211 11/9/10 12:49:43 AM


212 Handbook of Human Factors in Medical Device Design

6.4.1.3 Participants
Guideline 6.6: Usability Test Participants
Usability test participants should be representative of the most likely users of the device.
Participants should be selected on the basis of the personal characteristics described in the
user profiles that are typically gathered during the initial contextual inquiry phase of design.

User profiles include information about training, skills, experience levels, demographics,
job type, and motivation, among other defining characteristics. Sometimes, predesign anal-
ysis will also describe unique user groups for which the device is targeted. For example, an
infusion pump might have users who are nurses, doctors, technicians, and patients, each of
which is a unique user group. The number of test participants selected from each unique
user group depends on whether the test is formative or summative, as described more com-
pletely in the “Number of Participants” section.

Guideline 6.7: Device Development Team Members


Product development team members and others with special product knowledge should not be
usability test participants because they will have biases and are distinctly different from real
device users. Section 6.5.3 describes considerations in recruiting test participants.

6.4.1.4 Prototypes or Simulations


Early prototypes or simulations are best derived from the user-interface requirements
obtained in the user analysis conducted in the early development stages. Contextual inquiry
is typically used in these early stages. Employ usability testing and usability inspection
methods to assess how closely the early prototype designs match user needs and usability
objectives. Gather usability feedback to improve the design for further testing in an iterative
fashion. Some recommended types of prototypes and simulations include:

• Paper-based, low-fidelity simulations (e.g., rough hand-sketched paper screen shots


of a handheld glucose meter that are placed in front of a user and sequenced as a
working model might behave). See Synder (2003) for more detail on the methods
of paper prototyping.
• Wizard of Oz simulations whereby a human simulates a not-yet-functional sys-
tem and mimics its operation during a usability test. For example, to simulate
a voice-activated surgical robot, a human pretends to be a speech recognition
system by listening to the words spoken by a test subject and typing them into
a computer that in turn displays these words to the test subject.
• Computer simulations (typically screen based, e.g., Flash™, Macromedia
Director™, Visual Basic™, Java, or HTML). Such computer prototypes can realis-
tically simulate intended user interactions while sacrificing full fidelity (e.g., touch
screen to mimic actual hardware buttons on an infusion pump) yet at a lower cost.
• Horizontal and vertical simulations. Horizontal simulations are broad and show the
range of navigation options and high-level features (e.g., the high-level navigation
bar for a computer workstation that controls a blood gas analyzer). Vertical simula-
tions are narrow and deep and show how one particular feature works in depth (e.g.,
a full set of functions for all the steps in calibrating a computer-controlled immuno-
assay diagnostic system). Hybrid simulations combine horizontal simulations with

TAF-ER562X_WINGR-09-0401-C006.indd 212 11/9/10 12:49:43 AM


Testing and Evaluation 213

a limited set of vertical simulations of critical device features (e.g., a touch-screen


infusion pump simulating the global navigation and two features, such as delayed
therapy and bolus delivery).
• Working early prototypes of the actual device or parts of it. The caveat is that early
software can be unstable and recovery steps need to be created if the software
crashes during the usability tests.

6.4.1.5 Methodology
The method description section of a usability protocol and subsequent report is much like
the “methodology” section of any scientific report.

Guideline 6.8: Description of Test Methods


The usability test plan should describe the usability study methodology so that another
researcher or developer has sufficient information to be able to replicate the test. It should
cover all of the items discussed in the “Test Plan Contents” (6.4.1) and “Logistics” (6.5) sec-
tions of this chapter.

A usability test is similar in many ways to a psychological experiment. Most of the rules
and considerations for good experimental design are applicable. For example, if several
independent variables are being manipulated in the usability test (e.g., some subjects get
training and others do not, and the subjects are from two levels of relevant medical device
experience), then a balanced experimental design is required to avoid “confounding” of the
independent variables. Confounding occurs when two or more independent variables are
mixed in such a way that their individual impact on the outcome measures cannot be deter-
mined. For example, if a study were designed to compare reading accuracy of a cathode ray
tube (CRT) versus an LCD display but used larger character sizes on the CRT screen, then
the variables of display type and character size would be confounded. In this case, it would
not be possible to determine whether better reading scores for the CRT were due to better
image quality or to having larger display characters compared to the LCD screen.
The usability study design should take into account the goals of the study (i.e., the
key questions it is intended to answer) and the methods to be employed. An early issue
is how independent variables in the design will be controlled. The independent variables
can be varied systematically by choosing multiple levels (those of most importance for
design), held constant by using only one level, or randomized and possibly presented in
counterbalanced orders to control for learning order effects (see Section 6.4.1.6.2, “Task
Presentation Order”).

6.4.1.5.1 Experimental Design


A key study design attribute is how the participants will be allocated across the indepen-
dent variables. In “within-subjects” desi