&EPA
EPA 600/R-08/062 | May 2008 | www.epa.gov/ord
United States
Environmental Protection
Agency
Scientific and Ethical
Approaches for
Observational Exposure Studies
Office of Research and Development
National Exposure Research Laboratory
-------
-------
EPA 600/R-08/062
May 2008
Scientific and Ethical Approaches for
Observational Exposure Studies
National Exposure Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
-------
Notice
This document has been subjected to the Agency's peer and administrative review and has
been approved for publication as an EPA document. Mention of trade names or commercial
products does not constitute endorsement or recommendation for use.
-------
Abstract
Researchers conduct observational human exposure studies to understand how and the extent
to which people come into contact with chemicals and environmental stressors in their everyday
lives, through the air they breathe, the food and liquids they consume, and the things they touch.
The U.S. Environmental Protection Agency's (EPA's) National Exposure Research Laboratory
(NERL) has conducted observational human exposure studies for several decades and uses the
information and data from these studies to improve the Agency's understanding of human
exposures to chemicals and other stressors and ultimately to support efforts to improve public
health. Because these studies involve people as research participants, they are complex and raise
numerous scientific and ethical issues that have to be addressed prior to and during their design
and implementation. To ensure mat EPA's research continues to be based on the most up-to-date
science and the highest ethical standards, the Agency has developed this document mat contains
state-of-the-science approaches for conducting observational human exposure studies. This
document is not meant to represent an official Agency "guidance document" but, rather, serves as
a resource tool and source of information for NERL and other researchers on which to rely as
they develop and conduct observational human exposure studies.
-------
Foreword
This document is intended as a resource and reference for the U.S. Environmental Protection
Agency's (EPA's) National Exposure Research Laboratory (NERL) scientists as they develop
and implement observational human exposure studies. The authors recognize that this document
also may prove to be useful to others involved in exposure science research, but the document is
not meant to represent an official Agency "guidance document" and should not be used for that
purpose.
Observational human exposure studies involve the collection of information about individuals
and the environment around them. NERL scientists and managers take the protection of human
subjects who participate in these studies very seriously. The steps needed to ensure protection of
the human subjects are often complex, and the specific actions will vary depending on the
objectives of the study and details about the participants.
This document does not provide solutions to all scientific and ethical issues that may arise as
such studies are undertaken. That is, it is not possible to identify or address all potential issues in
advance or to develop a comprehensive checklist for all such studies. Rather, this document
attempts to present and discuss the types of issues that will need to be considered and addressed
as NERL researchers plan and implement observational human exposure studies. The researchers
will need to work with others—the study team, institutional review board members, EPA's
Human Subjects Research Review Official, the participants and their community, and other
stakeholders—to identify and address all of the relevant issues for their particular study in order
to ensure mat the specific elements of the study will respect, safeguard, and protect the human
research subjects.
As EPA employees, NERL scientists face both regulator}' and moral obligations to ensure the
protection of the human subjects participating in observational research. The regulatory
requirements are set forth in EPA's human subjects regulations (40 CFR 26). NERL scientists are
resolved to meet both the "letter" of the law as set forth in the regulations and also the "spirit"
that derives from the most up-to-date thinking and consensus on these sensitive issues. This
document provides information on regulatory requirements and the state of the science for a
number of issues associated with observational human exposure studies to help NERL scientists
meet their goal of conducting these studies based on the most up-to-date and sound science and
the highest ethical standards.
To gather information for the scientific and ethical approaches for observational human
exposure studies, NERL convened an expert panel workshop on November 28 and 29, 2006, to
discuss the state of the science. The 11-member panel discussed their ideas for the content of this
document and the state of the science for various elements of observational human exposure
studies. The panel agreed that the document planned by EPA should include the following six
major topic areas:
(1) identifying elements to be considered in study conceptualization,
(2) ensuring protection of vulnerable groups,
(3) addressing privacy and other concerns related to observational human exposure studies,
(4) creating an appropriate relationship between the participant and investigator,
IV
-------
(5) building and maintaining appropriate community and stakeholder relationships, and
(6) designing and implementing strategies for effective communication.
The structure and content of the current document follow the recommendations of the Expert
Panel. These recommendations include pragmatic steps that NERL scientists can undertake
during the development and implementation of observational human exposure studies.
-------
Acknowledgments
This document was developed through the joint efforts of many U.S. Environmental
Protection Agency (EPA) staff. Roy Fortmann, Larry Cupitt, Kent W. Thomas, and Peter Egeghy,
researchers in the National Exposure Research Laboratory, were primary authors of the
document. Linda Sheldon provided input to the development of the document and contributed
through discussions with the authors on issues associated with observational human exposure
studies. Davyda Hammond assisted in the revision of the external review draft document. Larry
Cupitt had primary responsibility for editing the document and coordinating the peer reviews.
Internal EPA reviewers included Pamela Williams, Warren Lux, Michael Firestone, Gary Bangs,
Julian Preston, Hugh Tilson, Ronald Williams, Cathy Fehrenbacher, and Deirdre Murphy. The
EPA Human Studies Review Board, a Federal advisory committee, reviewed the external review
draft document and provided advice and recommendations that were addressed in the final
revision of the document. John Barton with Computer Sciences Corporation performed the copy
edits of the external review and final drafts. Public comments solicited on the external draft were
addressed in the final revision of the document.
VI
-------
Contents
Notice ii
Abstract iii
Foreword iv
Acknowledgments vi
Executive Summary 1
1. Introduction, Purpose, and Scope 7
1.1. Observational Human Exposure Studies 7
1.2. Ethical Issues in Observational Human Exposure Studies 8
1.3. Purpose of This Document 11
1.4. Process for Developing the Document 13
1.5. Organization of the Document 13
2. Elements to Be Considered in Study Conceptualization and Planning 17
2.1. Scientific Value of a Proposed Observational Human Exposure Study 18
2.1.1. Defining the Study Problem 18
2.1.2. Justifying the Study 19
2.1.3. Identifying a Research Team To Plan and Implement the Study 20
2.2. Ensuring Scientific Validity of the Research Study 20
2.2.1. Study Design 20
2.2.2. Feasibility 21
2.2.3. Sample Size Determination 22
2.2.4. Representativeness of the Sample 22
2.2.5. Information Collection Rule 23
2.2.6. Quality Assurance Project Plan 24
2.2.7. The Study Design Document 24
2.3. Ethical Issues in Ensuring Fair Subject Selection 24
2.4. Ensuring a Favorable Risk-Benefit Ratio 25
2.4.1. Designing in Benefits for the Participants 25
2.4.2. Assessing Benefits and Risks of Study Participation 25
2.5. Independent Scientific and Ethical Review 26
2.5.1. Scientific Peer Review 27
2.5.2. Conflicts of Interest 27
2.5.3. Develop the Human Subjects Protocol for Institutional Review Board Review 28
2.5.4. Ethical Review 29
2.5.5. Internal U.S. EPA Review of Scientific and Ethical Issues 29
2.6. Informed Consent 29
2.7. Ensuring That Participant Behaviors Are Not Changed Adversely Because of
Being in the Study 30
2.8. Criteria and Standards for Monitoring Scientific and Ethical Issues 30
3. Ensuring Protection of Vulnerable Groups 35
3.1. Identification of Vulnerable Groups 35
VII
-------
3.2. Justification for Involving Vulnerable Persons in Observational Research 36
3.3. Minimal Risk and Vulnerable Groups 37
3.4. Research Involving Children 37
3.5. Women as Research Subjects 38
3.6. Other Potentially Vulnerable Groups 39
4. Privacy, Confidentiality, and Other Concerns Related to Observational
Human Exposure Studies 41
4.1. Privacy Issues 41
4.2. Confidentiality 42
4.2.1. Confidentiality of Information 43
4.2.2. Confidentiality of Participation 44
4.3. Collateral Observations 44
4.3.1. Potential Nonstudy Hazards in the Residence 44
4.3.2. Collateral Observations with Mandated Reporting Requirements 45
4.3.3. Hazard Communication 45
4.3.4. Planning and Staff Training 46
4.4. Third-Party Issues 46
4.4.1. Determining Whether a Third Party Is a Human Subject 47
4.4.2. Informing Third Parties of Research Activities 47
4.4.3. Research Results and Third Parties 47
4.5. Data and Safety Monitoring and Oversight 48
5. Creating an Appropriate Relationship Between the Participant and Researcher 51
5.1. Informed Consent 52
5.1.1. Information 52
5.1.2. Comprehension 54
5.1.3. Voluntary Participation 55
5.2. Payments to Research Participants 56
5.2.1. Types and Amounts of Payments Offered in Research Studies 56
5.2.2. Regulations and Guidance Regarding Payment to Research Participants 58
5.2.3. Payments When Children or Other Vulnerable Populations Are Involved 59
5.2.4. Payments in Observational Human Exposure Studies 61
5.3. Research Rights and Grievance Procedures 61
5.3.1. Ombudsman 61
5.3.2. Community Advisory Board 62
5.4. Creating a Supportive Environment for Research and Interaction 62
5.5. Recruitment Strategies 62
5.6. Retention Strategies 63
5.7. Ensuring Recruitment or Retention Methods Will Not Lead to Unacceptable Risk 64
6. Building and Maintaining Appropriate Community and Stakeholder Relationships 67
6.1. Approaches to Community Involvement 68
6.2. Issues in Community Involvement 71
6.2.1. Defining "Community" 71
6.2.2. Identifying Who Represents the Community 71
6.2.3. Building Relationships and Trust 72
6.2.4. Importance of Language and Communications with the Community 73
6.2.5. Recognizing and Addressing Cultural Differences 73
6.2.6. Honesty, Power Relationships, and Partnerships 74
6.2.7. Building aLasting Infrastructure 74
6.3. Community Involvement in Observational Human Exposure Studies 75
6.4. Identifying and Interacting with Other Stakeholders 75
VIII
-------
7. Designing and Implementing Strategies for Effective Communication 81
7.1 Communication Strategy and Implementation Plan 81
7.2 Individuals and Groups Involved in the Communications 82
7.3 Communications Timetables—When To Communicate 83
7.4 Communicating at Different Levels 85
7.5 Communications Materials 85
7.6 Informing the Study Participants and Communities 87
7.7 Reporting Study Results to the Participant and Community 87
7.8 Reporting Unanticipated Results or Observations 90
7.9 Anticipating and Responding to Criticism 91
7.10 Responding to the Media, Public Inquiries, and Other Stakeholders 92
Appendix A: Additional Discussion of Observational and Exposure Terminology and
Examples of Previous NERL Observational Human Exposure Studies 95
Appendix B: The Process for Development of This Document: Description of the
Expert Panel Workshop, the External Peer Review by the
Human Studies Review Board, and Public Comment 107
Appendix C: Recommended Content of aHuman Subjects Protocol Ill
Appendix D: Recommendations for Enhancing Public Trust 113
Appendix E: List of Acronyms and Abbreviations 115
Appendix F: Glossary 117
IX
-------
-------
Executive Summary
Scientists at the U.S. Environmental Protection
Agency's (EPA's) National Exposure Research
Laboratory (NERL) have conducted observational
human exposure studies for several decades to
understand how people come into contact with
chemicals and other stressors in their everyday lives,
through the air they breathe, the food and liquids they
consume, and the things they touch. These studies are
performed to determine what chemicals people are
exposed to, the concentrations of the chemicals, the most
important sources contributing to people's exposures,
the routes and pathways of exposure, and the factors mat
have the biggest impact on exposure. The studies help
explain when, where, why, how, and how often people
are exposed to chemicals and other stressors (e.g.,
allergens, viruses, mold, radiation, noise) in their
everyday environments as they go about their daily
activities. Information from these studies helps EPA
improve the understanding of people's exposures to
chemicals and other stressors and, ultimately, supports
EPA's efforts to protect public health.
NERL scientists and managers take the protection of
human subjects very seriously. Because observational
human exposure studies involve people as research
participants, NERL researchers must act to ensure the
protection of the human subjects throughout the study.
Such studies are often complex, and the specific actions
will vary depending on the objectives of the study, the
details of the study design and human subjects research
protocol, and the details about the participants and the
communities in which they live. To ensure that the
actions of NERL researchers will properly respect,
safeguard, and protect the rights and welfare of the
participants in their research, NERL scientists need to be
knowledgeable about the scientific and ethical issues
that may arise as they plan and conduct their research,
and they also need to be diligent in the application of the
most up-to-date and sound scientific approaches and of
the highest ethical standards to their research.
This document, therefore, was prepared by NERL
scientists as a resource and reference for EPA's NERL
scientists as they develop and implement observational
human exposure studies. The authors recognize that this
document also may prove to be useful to others involved
in exposure science research, but that this document
does not meet the definition of an official Agency
"guidance document" (it does not set forth "a policy on a
statutory, regulatory or technical issue or an
interpretation of a statutory or regulatory issue") and
should not be used for that purpose.
As EPA employees, NERL scientists face both
regulatory and moral and ethical obligations to ensure
the protection of the human subjects participating in
their observational research. The regulatory
requirements are set forth in EPA's human subjects
regulations (40 CFR 26). The moral obligations derive
from the ethical principles of biomedical ethics. NERL
scientists and managers are resolved to meet both the
"letter" of the law as set forth in the regulations and also
the "spirit" that derives from the most up-to-date
thinking and consensus on these sensitive issues.
This document provides information on regulatory
requirements and ethical issues to consider when
performing human subjects research. Knowledge about
these requirements and issues will help NERL scientists
meet their goal of conducting observational human
exposure studies based on the most up-to-date and sound
science and the highest ethical scientific standards.
The ethical and moral issues associated with human
subjects research has long been the subject of a great
deal of thought and discussion, both in the United States
and abroad. Issues in biomedical ethics continue to be
discussed and debated in today's headlines. Spurred by
the atrocities of World War II concentration camps and
by the disclosure of unethical treatment of
undereducated African-American men and other
vulnerable groups by medical staff in the United States,
the U.S. and world communities were prompted to
establish ethical principles for medical and scientific
experiments that involve people as participants. In the
United States, the Belmont Report (U.S. DREW, 1979)
is the foundational document in the development of the
-------
ethics of human subjects research. This report lays out
the fundamental ethical principles behind research that
involves humans as research subjects. These three basic
principles, (1) respect for persons, (2) beneficence, and
(3) justice, have become the cornerstones for regulations
involving human subjects. Ethicists have expanded on
those principles since 1979, translating them into ethical
requirements that any human subjects research must be
both ethically acceptable and scientifically sound.1
EPA's Science Advisory Board has affirmed, "Bad
science is always unethical" (U.S. EPA, 2000).
In an effort to ensure mat NERL's observational
human exposure studies are founded on the ethical
principles of respect for persons, beneficence and
nonmaleficence, and justice and adhering to the principle
that bad science is always unethical, scientists and
managers from NERL have assembled this document as
a resource and reference for NERL exposure scientists.
These same scientists and managers have sought expert
advice, including input from an expert panel workshop;
review and comment on the external review draft of the
document by the Human Studies Review Board (HSRB),
a Federal advisory committee; public comment on the
external review draft of the document; and public input
about the state of the science for scientific and ethical
approaches for design and implementation of
observational human exposure studies.
A number of references, both from the bioethics
literature and from U.S. regulations, have proven useful
to the authors as they have developed this document.
Those references are listed in Table 1-4, which is
replicated below.
The authors also have relied extensively on the
advice of an expert panel that convened in November
2006 to provide advice and guidance about the staicture
and content of this document. The Expert Panel
consisted of 11 nationally recognized authorities from
diverse fields — exposure science, environmental health,
bioethics, epidemiology, community-based research,
law, research in minority communities, public health,
toxicology, pediatrics, and children's environmental
health.
The Expert Panel Workshop resulted in suggestions
for both the structure and the content of this document
(ERG, 2007). Following the advice of the Expert Panel,
this document is organized in seven sections.
Section 1. Introduction: The background for
observational human exposure studies, the scope of
this document, and the important scientific and ethical
issues that are critical to human subjects and
observational research.
Section 2. Elements to be considered in study
conceptualization: Incorporating ethical concerns into
the scientific effort from the onset and integrating
them throughout all phases of study planning and
implementation. As shown in the text box below, the
planning process involves the initial identification of
the research question and justification of the research
effort during the problem conceptualization phase. If
human subjects research is justified for the study, the
scientific and ethical approaches are described in the
study design and the human subjects research
protocol. The basic elements that should be included
in the study design and in the human subjects research
protocol are described, and researchers are advised to
consider alternative and innovative study designs mat
maximize the benefits to the study participants and
their community. Information is provided on both
scientific peer review and ethical review, and the
integration of the two. Mandated procedures for
review by external peers, by Institutional Review
Boards (IRBs), and by the EPA Human Subjects
Research Review Official (HSRRO) are detailed.
See, for example, the writings of Beauchamp and Childress in Principles of
Biomedical Ethics (Beauchamp and Childress, 2001) and the discussion of
"What Makes Clinical Research Ethical?" by Emanuel, Wendler, and Grady
(Emanueletal.,2000).
Define & Justify
Planning
&
Scoping
Develop Study
Design
Develop Human
Subjects Protocol
Independent
Review
Agency
HSRRO
Review
HSRRO \ Disapprove
Review / NO study
Approve**'Conduct Study
Implement ^\c"te"a Exceeded
& Monitor ^~ Terminate'study
Stages in the Development of a NERL
Observational Human Exposure Study
-------
Table 1-4. Important References in Developing This Document:
Some Recent Developments in Defining the Ethics of Conducting Research Involving Human Participants
Year
Event/Report
Description
1979
The Belmont Report: Ethical
Principles and Guidelines for
the Protection of Human
Subjects of Research (U.S.
DHEW, 1979)
The Belmont Report attempts to summarize the basic ethical principles identified by the legislatively
created National Commission for the Protection of Human Subjects of Biomedical and Behavioral
Research. It is a statement of basic ethical principles and guidelines that should assist in resolving
the ethical problems that surround the conduct of research with human subjects. The three basic
ethical principles are (1) respect for persons, (2) beneficence, and (3) justice.
1991
The Common Rule
40CFR26, SubpartA
The Common Rule is a short name for "The Federal Policy for the Protection of Human Subjects"
and was adopted by more than a dozen Federal departments or agencies in 1991. Each agency
incorporated the policy into its own Code of Federal Regulations (CFR), with EPA adapting it in Title
40CFR Part 26, SubpartA.
1993
The Institutional Review
Board Guidebook (U.S. HHS,
1993)
The document is intended as a resource and a reference document for IRB members, researchers,
and institutional administrators. It is not designed to tell IRBs whether or not specific protocols should
be approved; rather, the Guidebook points out issues to which IRBs should pay attention and
presents, wherever possible, areas where ethicists have arrived at a consensus on the ethical
acceptability of a particular activity or method.
2000
What Makes Clinical
Research Ethical? (Emanuel
etal.,2000)
This journal article lays out seven areas of concern that need to be addressed if clinical research is
deemed to be ethically acceptable: (1) social or scientific value, (2) scientific validity, (3) fair subject
selection, (4) favorable risk-benefit ratio, (5) independent review, (6) informed consent, and
(7) respect for potential and enrolled subjects.
2001
Principles of Biomedical
Ethics:(Fifth Edition
(Beauchamp and Childress,
2001)
A classic text in biomedical ethics. Core chapters discuss respect for autonomy, nonmaleficence,
beneficence, and justice. The chapter on professional-patient relationships discusses issues
important to privacy, confidentiality, and protection of subjects. The fifth edition is an update that
reflects developments in philosophical analysis, as well as developments in science and medicine.
2002
International Ethical
Guidelines for Biomedical
Research Involving Human
Subjects (CIOMS, 2002)
Developed by the Council for International Organizations of Medical Sciences particularly for use in
developing countries, the guidelines relate mainly to ethical justification and scientific validity of
research; ethical review; informed consent; vulnerability of individuals, groups, communities, and
populations; women as research subjects; equity regarding burdens and benefits; choice of control in
clinical trials; confidentiality; compensation for injury; strengthening of national or local capacity for
ethical review; and obligations of sponsors to provide health care services.
2003
Protecting Participants and
Facilitating Social and
Behavioral Sciences
Research (NRC, 2003)
This NRC publication targets policymakers, research administrators, research sponsors, IRB
members, and investigators. It examines three key ethical issues: (1) obtaining informed, voluntary
consent from prospective participants; (2) guaranteeing the confidentiality of information collected
from participants, which is a particularly challenging problem in social sciences research; and
(3) using appropriate review procedures for minimal-risk research.
2005
Ethical Considerations for
Research on Housing-
Related Health Hazards
Involving Children. (NRC &
IOM, 2005)
This National Research Council and Institute of Medicine report reviews the challenges and ethical
issues in conducting housing-related health hazards research in the wake of the Maryland Court of
Appeals ruling in the case of Grimes v. Kennedy Krieger Institute that has led to substantial
controversy and confusion. The ruling highlighted a range of potential ethical concerns, such as
issues involving adequacy of informed consent, parents' perception of risk, duties of researchers to
child subjects and their parents, the role of IRBs, and the authority of parents to provide permission
for their children to participate in research. This report offers much needed recommendations and
practical guidance for the ethical conduct of this type of research.
2006
EPA adds Additional Human
Subjects Protections at
40 CFR 26
EPA added additional human subjects protections in the Code of Federal Regulations to govern its
actions. Subparts B through D apply to research conducted or supported by EPA and are directly
applicable to NERL and this document. Subpart B prohibits research involving intentional exposure
of children, pregnant women (and their fetuses), or nursing women. Subparts C and D provide
additional protections for observational research involving pregnant women and their fetuses
(Subpart C) and for children (Subpart D). Subparts K through M and O through Q apply to EPA's use
of third-party human research data.
2008
International Ethical
Guidelines for
Epidemiological Studies
(CIOMS, 2008)
This document builds on the CIOMS (2002) document (see above) and extends the discussion to
address the special features of epidemiological studies.
-------
Sections. Ensuring protection of vulnerable groups:
Protections afforded by EPA's human subjects rules
and the ethical concerns of involving such groups in
observational research. Special requirements for the
protection of potentially vulnerable groups, including
children, prisoners, pregnant women, handicapped
persons, mentally disabled persons, and economically
or educationally disadvantaged persons, throughout
the planning and implementation process are
described.
Section 4. Addressing privacy and other concerns
related to observational human exposure studies:
Ethical issues and regulatory requirements concerning
privacy, including third-party involvement and
observations of nonstudy hazards. Unlike clinical
research that is conducted in an institutional setting,
observational human exposure studies take place in
the participants' "personal" environments as they go
about their everyday lives, presenting an even greater
challenge in meeting the ethical obligation to respect
the privacy of the participants.
Section 5. Creating an appropriate relationship between
participant and investigator: Issues surrounding
recruitment, informed consent, payment, and the
researcher's need to support the welfare of the
participants. An appropriate relationship built on
openness and trust requires strong and effective
bidirectional communication. Informed consent
ensures that the participant understands the range of
risks associated with participation and the voluntary
nature of participation, and provides essential
protections to the participant. Recent observations by
various national and international review committees
on the appropriate level of payment to research
participants are presented. This is a complex ethical
issue, balancing the issue of fairness against the
possibility of undue influence and the loss of free
consent. Other topics include participant recruitment,
retention strategies, research rights, and grievance
procedures.
Section 6. Building and maintaining appropriate
community and stakeholder relationships: Involving
the community in the research effort to improve the
research both scientifically and ethically. Various
approaches are discussed related to issues such as
defining the community, identifying who represents
the community, recognizing and addressing cultural
differences, and the importance of language, power
relationships, and partnerships.
Section 7. Designing and implementing strategies for
effective communication: Ongoing, interactive
dialogue among researchers, participants, the
community, stakeholders, and the public to establish
effective communications and to foster a relationship
of trust. To facilitate information dissemination to
participants, communities, and stakeholders, this
section of the document describes communication
strategies, implementation plans, communication
groups, timetables, communication materials, and
other tools available to researchers.
This document does not and, indeed, could not
provide solutions to all scientific and ethical issues that
may arise as observational human exposure studies are
undertaken. No document could identify and address all
potential issues in advance, nor is it possible to develop
a comprehensive checklist for all such studies. Rather,
this document attempts to present and discuss the types
of ethical and scientific issues that will need to be
considered and addressed as NERL researchers plan and
implement observational human exposure studies. The
researchers will need to work with others — the study
team, IRB members, EPA's HSRRO, the participants
and their community, and other stakeholders — to
identify and address all of the relevant issues for any
particular study. The authors are confident that this
document will be helpful to NERL scientists in their
endeavors to assure that all of NERL's observational
human exposure studies will respect, safeguard, and
protect the participants in that research.
References
Beauchamp TL, Childress, JF (2001). Principles of
Biomedical Ethics: Fifth Edition. New York, NY: Oxford
University Press.
CFR (Code of Federal Regulations) (2006). 40 CFR Chapter I
Environmental Protection Agency Part 26 Protection of
Human Subjects. U.S. Code of Federal Regulations.
Available:
http://www.access.gpo.gov/nara/cfr/waisidx 06/40cfr26 06.
html [accessed 12 June 2007].
CIOMS (The Council for International Organizations of
Medical Sciences) (2008). International Ethical Guidelines
for Epidemiological Studies. World Health Organization.
Geneva, Switzerland. Available:
http://www.cioms.ch/080221feb_2008.pdf [accessed 25
April 2008].
CIOMS (The Council for International Organizations of
Medical Sciences) (2002). International Ethical Guidelines
for Biomedical Research Involving Human Subjects. World
Health Organization. Geneva, Switzerland. Available:
-------
http://www.cioms.ch/frame guidelines nov 2002.htm
[accessed 12 June 2007].
Emanuel EJ, WendlerD, Grady C (2000). What Makes
Clinical Research Ethical? JAMA 20:2701-2711.
ERG (Eastern Research Group) (2007). Report on the
Workshop to Discuss State-of-the-Science Approaches for
Observational Exposure Measurement Studies. Held in
Durham, North Carolina on November 28-29, 2006. Final
Report: January 25, 2007. (PB2007-108905).
NRC (National Research Council) (2003) Protecting
Participants and Facilitating Social and Behavioral
Sciences Research. Washington, DC: The National
Academies Press. Available:
http://books.nap.edu/catalog.php7record id=10638
[accessed 12 June 2007].
NRC & IOM (National Research Council and Institute of
Medicine) (2005). Ethical Considerations for Research on
Housing-Related Health Hazards Involving Children.
Washington, DC: The National Academies Press. Available:
http://books.nap.edu/catalog.php7record id=l 1450
[accessed 12 June 2007].
U.S. DHEW (U.S. Department of Health, Education, and
Welfare) (1979) The Belmont Report: Ethical Principles
and Guidelines for the Protection of Human Subjects of
Research. National Commission for the Protection of
Human Subjects of Biomedical and Behavioral Research.
Available: http://ohsr.od.nih.gov/guidelines/belmont.html
[accessed 12 June 2007].
U.S. EPA (U.S. Environmental Protection Agency) (2000)
Comments on the Use of Data from the Testing of Human
Subjects: A Report by the Science Advisory Board and the
FIFRA Scientific Advisory Panel. United States
Environmental Protection Agency, Science Advisory Board,
EPA-SAB-EC-00-017. Available:
http://vosemite.epa.gov/sab/SABPRODUCT.NSF/E3AEOF3
OD7430436852571A7005325E8/$File/ec0017.pdf [accessed
12 June 2007].
U.S. HHS (U.S. Department of Health and Human Sendees)
(1993). Protecting Human Research Subjects: Institutional
Review Board Guidebook. Office for Human Research
Protections. Available:
http://www.hhs. gov/ohrp/irb/irb_guidebook.htm [accessed
12 June 2007].
-------
-------
SECTION 1
Introduction, Purpose, and Scope
Observational human exposure studies are an
important research tool for understanding people's
contact with pollutants and other stressors in the
environment, that is, their exposure.2 Such studies allow
researchers to collect information about people's
exposures to chemicals and other stressors under real-
world conditions during their normal day-to-day
activities. Exposures occur through the air we breathe,
the food we eat, the water and beverages we drink, and
the surfaces that we touch as we go about our daily
routines. To understand and characterize people's
exposures to chemicals, two things have to be known:
(1) the concentrations of the chemicals in the
environment that people inhale, ingest, or touch; and (2)
the human activities that bring people into contact with
the media containing the chemicals.3 This document
addresses issues associated with observational human
exposure studies that are conducted by the National
Exposure Research Laboratory (NERL) of the U.S.
Environmental Protection Agency (EPA) in an effort to
understand and characterize the exposures that people
encounter as they go about their daily lives.
Because observational human exposure studies
involve human participants, they are complex in their
design and implementation. As in all research involving
human participants, observational human exposure
studies carry both regulatory obligations for the
protection of human subjects (40 CFR 26) and ethical
obligations to the study participants: namely, to respect
2 Exposure, as it is used throughout this document, is a technical term that is
defined as the "contact of a chemical, physical, or biological agent with the
outer boundary of an organism [e.g., a person]. Exposure is quantified as the
concentration of the agent in the medium in contact integrated over the time
duration of that contact." (The definition is taken from Guidelines for
Exposure Assessment [EPA/600/Z-92/001, May 1992]). Seethe Glossary for
more information and the definition of additional terms.
3 The term "chemical" is used in this document as a surrogate term for all
stressors, including chemical, physical, or biological agents.
their autonomy, to not inflict harm (nonmaleficence), to
avoid harm and to maximize their benefits (beneficence),
and to treat all participants fairly (justice) (See, for
example, Principles of Biomedical Ethics, Beauchamp
and Childress, 2001). Ethical obligations have to be
carefully considered as they relate to the scientific
elements of these studies. Therefore, it is important that
researchers recognize and understand these obligations
and use the most up-to-date scientific and ethical
approaches in the design and implementation of
observational human exposure studies.
1.1 Observational Human Exposure Studies
As we are using the term in this document,
"observational human exposure studies" are studies that
involve the collection of environmental samples, data,
and information from study participants in their
everyday environments as they go about their normal
activities. These are studies where the NERL researchers
do not intentionally try to control the study variables or
outcomes, but instead merely observe both the variables
and the outcomes. They involve neither the deliberate
exposure of participants nor the control of environmental
conditions in a way that impacts the participants'
naturally occurring exposures. This scientifically based
definition needs to be distinguished at the outset from
the broader regulatory term "observational" so that the
scope and limits of this document are clear. EPA
Regulation 40 CFR 26 (Protection of Human Subjects)
at Subpart B, first, defines research involving intentional
exposure of a human subject as "a study of a substance
in which the exposure to the substance experienced by a
human subject participating in the study would not have
occurred but for the human subject's participation in the
study" [40 CFR 26.202(a)]. It then goes on at Subpart C
to define observational research as "any human research
that does not meet the definition of research involving
intentional exposure of a human subject" (40 CFR
-------
26.302). By defining "observational" so broadly, a
regulatory category is created that encompasses a range
of study types whose only common feature is mat they
do not entail "intentional exposure" as defined in the
regulations. Moreover, some of these "observational"
study types do not involve the study of exposures at all
but rather "observe" other sorts of phenomena. The
observational human exposure studies with which we are
concerned here, on the other hand, are exposure studies,
but they represent a narrower group of studies as defined
scientifically at the beginning of this paragraph.
Although they would generally be expected to fall within
the regulatory category of "observational research", they
are not synonymous with it. (For a more thorough
discussion of the scientific and regulatory meanings of
observational and exposure, with examples, please see
Appendix A.)
Observational human exposure studies are
performed for many different purposes. They have been
used extensively in the fields of social behavioral,
economic, biological, medical, epidemiological, and
exposure research to collect information that relates one
or more variables (e.g., exposure to a chemical) to its
result (e.g., the concentration of an exposure biomarker
in blood). There are many examples of observational
human exposure studies that have been conducted over
the past decade, and the reader is referred to the Journal
of Exposure Science and Environmental Epidemiology
and to Environmental Health Perspectives for examples
of the objectives, designs, and results of these studies.
NERL researchers have conducted and relied on
observational human exposure studies for more than
three decades. Examples of studies conducted or funded
by NERL are listed in Appendix A, Table A-l. Common
goals in those studies included those that follow.
• Identify which chemicals or other stressors mat people
are exposed to during their normal activities in the
environments that they occupy.
• Measure the concentrations of the chemicals to which
people are exposed.
• Identify the most important routes and pathways of
exposure.
• Identify the factors that impact people's exposures
(i.e., determine the when, why, how, and how much
that people are exposed to chemicals in the
environment).
These studies involve many different types of data
collection efforts and typically include observations,
measurements, and information on the following items.
• Chemical concentrations in environmental media (air,
water, soil, floor dust, and dust on surfaces)
• Chemical concentrations in the diet (food and
beverages).
• Biomonitoring (measurements of biomarkers of
exposure in urine, blood, and saliva)
• Time, location, and activity information
• Information on personal activities, product use, diet,
occupation, and other factors that may impact
exposure
• Information on the characteristics of the environments
mat study participants occupy (homes, schools,
offices, public access buildings, etc.)
The information obtained in observational human
exposure studies is used to better understand people's
contact with chemicals in the environment and to
improve exposure assessments and risk assessments.
This information is also essential for developing risk
mitigation strategies and for developing educational
materials and programs for reducing exposures and risks
to chemicals or other stressors in the environment (see
Table 1-1).
1.2 Ethical Issues in Observational Human
Exposure Studies
By definition, observational human exposure studies
involve human subjects. Whenever their research
involves human subjects, EPA researchers are required
to ensure the protection of the study participants by
complying with the Agency's human subjects rules as
set forth in 40 CFR 26.
The Common Rule (Subpart A of the rules)
represents basic regulatory actions (common to more
than a dozen Federal departments or agencies) that are
intended to ensure the protection of all human subjects.
The central requirements of the Common Rule are
twofold:
(l)that people who participate as subjects in covered
research are selected equitably and give their fully
informed, fully voluntary written consent; and
(2) that proposed research be reviewed by an
independent oversight group referred to as an
institutional review board (IRB) and approved only if
risks to subjects have been minimized, and risks are
reasonable in relation to anticipated benefits, if any, to
the subjects and to the importance of the knowledge that
may realistically be expected to result.
8
-------
Table 1-1. Examples of the Impact of Observational Human Exposure Studies on Pollution Levels and Regulatory Actions
Pollutant
Participate Matter (PM)
Volatile Organic Compounds
(VOCs)
Formaldehyde
Observational Study Result
Observational panel studies demonstrated
the appropriateness of ambient measurement
of fine particles as a surrogate for a
population's longitudinal exposure to fine PM.
EPA's Total Exposure Assessment
Methodology (TEAM) studies found levels of
about a dozen common organic pollutants to
be 2- to 5-times higher inside homes than
outside. Use of products containing organic
chemicals may result in very high and
persistent pollutant levels.
Studies found elevated formaldehyde levels
indoors and helped identify indoor sources.
Impact/Action/Result
Resolved questions in MAS review of PM science and provided a
"generally consistent finding that ambient particle concentrations are
a key determinant of the longitudinal variation in personal exposure."
(NRC, 2004). These results have been instrumental in support of the
National Ambient Air Quality Standard for PM (U.S. EPA, 1 999).
EPA, States, and the Consumer Product Safety Commission worked
together to influence manufacturers to voluntarily reduce emissions of
toxic chemicals from consumer products, building materials, and
furnishings, and to develop mitigation strategies and educational
materials to teach people how to reduce their contact with chemicals
indoors. As a result, contact with toxic chemicals indoors has been
reduced (see www.cpsc.gov/CPSCPUB/PUBS/450.html).
EPA worked with HUD, CPSC, and other agencies to limit
formaldehyde in building or consumer products and to educate the
public on how to reduce exposures (see www.
epa.gov/iaq/formalde.html).
EPA has adopted additional protections for children
and pregnant or nursing mothers in Subparts B through
D. These sections apply to all research either conducted
or funded by EPA and are, therefore, directly applicable
to NERL's observational human exposure studies.4
Subpart B prohibits EPA from conducting or supporting
research that involves intentional exposure of "a
pregnant woman (and, thereby, her fetus), a nursing
woman, or a child." NERL researchers conducting (or
funding) observational human exposure studies must
comply with all of these regulatory requirements,
including seeking review and approval by an IRB and by
the Agency's Human Subjects Research Review Official
(HSRRO) before beginning any human subjects
research. EPA's human subjects rules also define a
variety of fundamental terms—from "human subject" to
"research" to "intentional exposure" to "observational
research." Understanding these regulatory definitions is
vital for NERL researchers to comply with the
regulatory requirements.5
To more effectively ensure the protection of human
subjects, NERL scientists and managers need to
understand the ethical principles and issues that
prompted the development of the regulatory
requirements in the first place and to be knowledgeable
4 Subparts K, L, M, O, P, and Q of 40 CFR 26 set basic ethical requirements
that have to be met if human subjects data from a person or group external to
EPA and not funded by EPA (a third party) are to be used by EPA in specified
rulemaking actions. These subparts do not apply to NERL researchers and
will not be discussed further in this document.
5 The Glossary (Appendix F) lists definitions for a number of important terms;
definitions that come from the regulatory language are identified with their
specific CFR citation.
about the most recent thinking and guidance on
protection of human subjects.
The Belmont Report (U.S. DREW, 1979) is a
foundational document in the development of the ethics
of human subjects research in the United States. Because
of the adverse publicity and political embarrassment
arising from the unethical treatment of African-
American men in the Tuskegee Syphilis Study, Congress
passed the National Research Act of 1974, which called
on the Department of Health, Education, and Welfare
(DREW) to codify its rules on human subjects research
and established the National Commission for the
Protection of Human Subjects of Biomedical and
Behavioral Research. The commission was charged with
identifying the basic ethical principles that should
underlie human subjects research. The commission
published the Belmont Report in 1979. This report
established three basic principles: (1) respect for
persons; (2) beneficence; and (3) justice, which have
become the cornerstones for regulations involving
human subjects (see Table 1-2).
In 1981, the Department of Health and Human
Services (HHS) issued regulations based on the Belmont
Report. Ten years later, the core HHS regulations
(Subpart A) were adopted by almost all of the Federal
departments and agencies that conducted or sponsored
human subjects research as the "Common Rule."
Since 1991, ethical thought and regulatory processes
for the protection of human subjects have continued to
evolve and grow. For example, many ethicists expand
the elements contained in the principle of beneficence
from the Belmont Report into two principles:
(1) beneficence, meaning to prevent or remove harm and
-------
to maximize the possible benefits; and
(2) nonmaleficence, meaning not to inflict harm
(Beauchamp and Childress, 2001).
Table 1-2. The Belmont Report-
Principles and Recommendations
Ethical Principle
Respect for Persons
• Individuals should be
treated as autonomous
agents.
• Persons with diminished
autonomy are entitled to
protection.
Beneficence
• Human subjects should
not be harmed.
• Research should
maximize possible
benefits and minimize
possible harms.
Justice
• The benefits and risks of
research must be
distributed fairly.
Regulatory Manifestation
Informed Consent
• Subjects must be given the
opportunity to choose what will or
will not happen to them
• The consent process must include
(1) information,
(2) comprehension, and
(3) voluntariness
Assessment of Risks and
Benefits
• The nature and scope of risks and
benefits must be assessed in a
systematic manner.
Selection of Subjects
• There must be fair procedures and
outcomes in the selection of
research subjects.
Iii 2000, Emanuel, Wendler, and Grady considered
the ethical principles involved in clinical research and
proposed seven ethical requirements to be addressed in
research with humans (Emanuel et al., 2000). Their
published article specifically addressed clinical research,
but the issues are similar for observational human
exposure studies. Their ethical requirements are
summarized and briefly explained in Table 1-3. The
requirements are a logical extension of the ethical
principles enunciated in the Belmont Report and
manifest themselves in additional requirements for
social or scientific value; for processes to ensure the
scientific integrity of the research; and for independent
review of the design, the subject population, and the
risk-benefit ratio. The principle of respect for subjects
also includes additional emphasis on the welfare of the
subjects.
More recently, there has been increased scrutiny and
discussions of the ethics of research involving human
participants,6 and a number of respected institutions have
6 The term "human participants" often is used in this document. It denotes the
importance of the study participant being actively engaged in a partnership
with the researchers to address the objectives and goals of the study. The term
should be considered to be synonymous with the term "human subject" as
used in the Common Rule and in documents used to describe regulatory
requirements for studies involving human subjects.
addressed many important scientific and ethical issues
on this topic, including the National Research Council
(NRC) in its report, Protecting Participants and
Facilitating Social and Behavioral Sciences Research
(NRC, 2003), a joint NRC and Institute of Medicine
(NRC & IOM, 2005) committee in the report on Ethical
Considerations for Research on Housing-Related Health
Hazards Involving Children; the Council for
International Organizations of Medical Sciences
(CIOMS) under the World Health Organization in its
International Ethical Guidelines for Biomedical
Research Involving Human Subjects (CIOMS, 2002) and
in the International Ethical Guidelines for
Epidemiological Studies (CIOMS, 2008); and the
National Ethics Advisory Committee of New Zealand in
the Ethical Guidelines for Observational Studies:
Observational Research, Audits and Related Activities
(NEAC, 2006).
Table 1-3. Seven Ethical Requirements for
Clinical Research
From Emanuel, Wendler, and Grady (2000)
Requirement
Social or
scientific value
Scientific
validity
Fair subject
selection
Favorable risk-
benefit ratio
Independent
review
Informed
consent
Respect for
potential and
enrolled
subjects
Explanation
Evaluation of a treatment, intervention, or theory
that will improve health and well-being or
increase knowledge
Use of accepted scientific principles and
methods, including statistical techniques, to
produce reliable and valid data
Selection of subjects so that stigmatized and
vulnerable individuals are not targeted for risky
research, and the rich and socially powerful are
not favored for potentially beneficial research
Minimization of risks; enhancement of potential
benefits and risks to the subject are
proportionate to the benefits to the subject and to
society
Review of the design of the research trial, its
proposed subject population, and risk-benefit
ratio by individuals unaffiliated with the research
Provision of information to subjects about the
purpose of the research and its procedures,
potential risks, benefits, and alternatives, so that
the individual understands this information and
can make a voluntary decision whether to enroll
and continue to participate
Respect for subjects by
• permitting withdrawal from the research,
• protecting privacy through confidentiality,
• informing subjects of newly discovered risks
or benefits,
• informing subjects of results of the research,
and
• maintaining welfare of subjects.
10
-------
Collectively, these documents have reaffirmed the
basic ethical principles asserted in the Belmont Report
and have attempted, in some cases, to expand scientific
and ethical reasoning and understanding to define
approaches for dealing with additional elements of
human subjects research. These additional elements,
which often have been identified because of specific
incidents or case studies, include issues such as those
described below.
• Payment to participants—How much is adequate and
fair, without being an undue inducement?
• Non-study hazards—What is the researcher's
responsibility to identify hazards in the home that are
not part of the study?
• Third-party issues—Are there people other than the
participant who may be impacted during the study and
by the study results? If so, what needs to be done to
protect their interests and to respect them as persons?
• Community involvement—How should the
community be involved in the design and
implementation of studies?
These documents, together with EPA's regulatory
requirements for the protection of human subjects, serve
as important references for the subsequent sections of
this document (see Table 1-4).
1.3 Purpose of This Document
This document is meant to serve as a resource of
current scientific and ethical information for NERL
researchers as they develop and conduct observational
human exposure studies. The increased scrutiny of
research studies involving human participants makes it
imperative that researchers ensure that their research
protocols for protection of human subjects in
observational human exposure studies incorporate the
most up-to-date ethical approaches. Protocols for
protecting study participants in research studies have
been developed by experts in both academia and various
Federal agencies and adopted by the research
community because they ensure that observational
research meets the highest ethical and scientific
standards. However, because ethical and scientific
approaches for human subjects research continue to be
refined and evolve over time, there is a continuing need
to evaluate the latest approaches and ensure that
researchers are using state-of-the-science approaches in
their design and implementation of such studies.
The purpose of this document is to provide
information that researchers in EPA's Office of
Research and Development's NERL can use in the
design and implementation of observational human
exposure studies to ensure the protection of the human
study participants. It is intended to be a resource tool for
NERL's exposure science researchers, but it is not
intended to serve as a "guidelines" document or a "how-
to" checklist. This document does not meet the Office of
Management and Budget (OMB) definition for an
official Agency "guidance document" ("Guidance
document - an agency statement of general applicability
and future effect, other than a regulatory action ... that
sets forth a policy on a statutory, regulatory or technical
issue or an interpretation of a statutory or regulatory
issue" [U.S. OMB, 2007]). The authors, as researchers,
perceived the need to provide this document for
themselves, their co-workers, and their collaborators, but
do not presume to speak for the whole of the Agency.
The authors have tried to (1) identify major areas and
elements of observational human exposure studies for
which ethical issues need to be considered, (2) provide
information on the state of the science for selected
approaches for applying ethical principles to the conduct
of these studies, and (3) provide sources of information
that researchers can use in the design and
implementation of these studies.
The emphasis of this document is to identify and
discuss ethical issues and approaches in observational
human exposure studies. As the document title implies
and as discussed in Section 2, scientific and ethical
issues are intrinsically bound together in research
involving human subjects. Therefore, it is essential that
scientific and ethical issues be considered together, not
separately. In this document, scientific issues and
approaches are discussed as they relate to and impact
ethical issues. However, it is beyond the scope of this
document to present a comprehensive discussion of the
scientific approaches for observational human exposure
studies. To include a comprehensive discussion of
scientific approaches in this document would reduce the
utility of the document by increasing the complexity of
the presentation and the length of the document.
This document does not provide solutions to all
scientific and ethical issues that may arise as such
studies are undertaken. That is, it is not possible to
identify or address all potential issues in advance or
develop a comprehensive checklist for all such studies.
Rather, this document attempts to present and discuss
the types of issues that will need to be considered and
addressed as NERL researchers plan and implement
observational human exposure studies. The researchers
will need to work with others—the study team, IRB
members, EPA HSRRO, the participants and their
community, and other stakeholders—to identify and
11
-------
Table 1-4. Important References in Developing This Document:
Some Recent Developments in Defining the Ethics of Conducting Research Involving Human Participants
Year
Event/Report
Description
1979
The Belmont Report: Ethical
Principles and Guidelines for
the Protection of Human
Subjects of Research (U.S.
DHEW, 1979)
The Belmont Report attempts to summarize the basic ethical principles identified by the legislatively
created National Commission for the Protection of Human Subjects of Biomedical and Behavioral
Research. It is a statement of basic ethical principles and guidelines that should assist in resolving
the ethical problems that surround the conduct of research with human subjects. The three basic
ethical principles are (1) respect for persons, (2) beneficence, and (3) justice.
1991
The Common Rule
40CFR26, SubpartA
The Common Rule is a short name for "The Federal Policy for the Protection of Human Subjects"
and was adopted by more than a dozen Federal departments or agencies in 1991. Each agency
incorporated the policy into its own Code of Federal Regulations (CFR), with EPA adapting it in Title
40CFR Part 26, SubpartA.
1993
The Institutional Review
Board Guidebook (U.S. HHS,
1993)
The document is intended as a resource and a reference document for IRB members, researchers,
and institutional administrators. It is not designed to tell IRBs whether or not specific protocols should
be approved; rather, the Guidebook points out issues to which IRBs should pay attention and
presents, wherever possible, areas where ethicists have arrived at a consensus on the ethical
acceptability of a particular activity or method.
2000
What Makes Clinical
Research Ethical? (Emanuel
etal.,2000)
This journal article lays out seven areas of concern that need to be addressed if clinical research is
deemed to be ethically acceptable: (1) social or scientific value, (2) scientific validity, (3) fair subject
selection, (4) favorable risk-benefit ratio, (5) independent review, (6) informed consent, and
(7) respect for potential and enrolled subjects.
2001
Principles of Biomedical
Ethics:(Fifth Edition
(Beauchamp and Childress,
2001)
A classic text in biomedical ethics. Core chapters discuss respect for autonomy, nonmaleficence,
beneficence, and justice. The chapter on professional-patient relationships discusses issues
important to privacy, confidentiality, and protection of subjects. The fifth edition is an update that
reflects developments in philosophical analysis, as well as developments in science and medicine.
2002
International Ethical
Guidelines for Biomedical
Research Involving Human
Subjects (CIOMS, 2002)
Developed by the Council for International Organizations of Medical Sciences particularly for use in
developing countries, the guidelines relate mainly to ethical justification and scientific validity of
research; ethical review; informed consent; vulnerability of individuals, groups, communities, and
populations; women as research subjects; equity regarding burdens and benefits; choice of control in
clinical trials; confidentiality; compensation for injury; strengthening of national or local capacity for
ethical review; and obligations of sponsors to provide health care services.
2003
Protecting Participants and
Facilitating Social and
Behavioral Sciences
Research (NRC, 2003)
This NRC publication targets policymakers, research administrators, research sponsors, IRB
members, and investigators. It examines three key ethical issues: (1) obtaining informed, voluntary
consent from prospective participants; (2) guaranteeing the confidentiality of information collected
from participants, which is a particularly challenging problem in social sciences research; and
(3) using appropriate review procedures for minimal-risk research.
2005
Ethical Considerations for
Research on Housing-
Related Health Hazards
Involving Children. (NRC &
IOM, 2005)
This National Research Council and Institute of Medicine report reviews the challenges and ethical
issues in conducting housing-related health hazards research in the wake of the Maryland Court of
Appeals ruling in the case of Grimes v. Kennedy Krieger Institute that has led to substantial
controversy and confusion. The ruling highlighted a range of potential ethical concerns, such as
issues involving adequacy of informed consent, parents' perception of risk, duties of researchers to
child subjects and their parents, the role of IRBs, and the authority of parents to provide permission
for their children to participate in research. This report offers much needed recommendations and
practical guidance for the ethical conduct of this type of research.
2006
EPA adds Additional Human
Subjects Protections at
40 CFR 26
EPA added additional human subjects protections in the Code of Federal Regulations to govern its
actions. Subparts B through D apply to research conducted or supported by EPA and are directly
applicable to NERL and this document. Subpart B prohibits research involving intentional exposure
of children, pregnant women (and their fetuses), or nursing women. Subparts C and D provide
additional protections for observational research involving pregnant women and their fetuses
(Subpart C) and for children (Subpart D). Subparts K through M and O through Q apply to EPA's use
of third-party human research data.
2008
International Ethical
Guidelines for
Epidemiological Studies
(CIOMS, 2008)
This document builds on the CIOMS (2002) document (see above) and extends the discussion to
address the special features of epidemiological studies.
12
-------
address all of the relevant issues for their particular
study to ensure that the specific elements of the study
will safeguard and protect the human research subjects.
In addition to being an information resource for
NERL researchers, this document provides useful
information for contractors and grantees funded by
NERL to consider during the design and implementation
of their exposure science research. Although not its
intended audience, this document also may prove to be
useful to other researchers, within and outside of EPA,
who are involved in observational human exposure
studies.
1.4 Process for Developing the Document
This document was written by exposure science
researchers in EPA's NERL, with substantial input from
experts within and outside of the Agency. Information
relevant to the process and the document has been
routinely posted on the EPA Web site at
http://www.epa.gov/nerl/sots.
NERL staff began this work by hosting a series of
stakeholder meetings in the summer of 2006 to seek
input on the content and format of the document. In
November 2006, NERL convened an expert panel to
provide its advice and guidance about the scope and
content of this document. The Expert Panel consisted of
11 nationally recognized authorities in diverse fields:
exposure science, environmental health, bioethics,
epidemiology law, community-based research, research
in minority communities, public health, toxicology,
pediatrics, children's environmental health, etc. Details
about the Expert Panel and the workshop can be found in
Appendix B. The summary report from the Expert Panel
may be accessed online at
http://www.epa.gov/nerl/sots/workshop-report.pdf
The structure and content of the current report
follow the recommendations of the Expert Panel.
Specifically, the Expert Panel recommended that this
document should include the following six major topic
areas:
(1) elements to be considered in study
conceptualization,
(2) ensuring protection of vulnerable groups,
(3) addressing privacy and other concerns related to
observational human exposure studies,
(4) creating an appropriate relationship between the
participant and investigator,
(5) building and maintaining appropriate community
and stakeholder relationships, and
(6) designing and implementing strategies for effective
communication.
These recommendations include pragmatic steps that
NERL scientists can undertake during the development
and implementation of observational human exposure
studies. Note that each step may require consideration
and application of multiple ethical and scientific
principles, and the same ethical principle may be
fundamental to several of the topic areas. As a result, the
same ethical principle may be discussed in several
sections throughout this document.
Using the advice of the Expert Panel, an internal
review draft of the document was written. Based on the
comments from internal EPA reviewers, the internal
review draft was revised, and an external review draft
was prepared. The external review draft was submitted
for peer review by EPA's Human Subjects Review
Board (HSRB), a Federal advisory committee consisting
of a panel of experts chartered to review and advise the
Agency on the scientific and ethical underpinnings of
human subjects research efforts. The external review
draft also was announced in the Federal Register and
made available for public review and comment. The
document subsequently was revised in response to
comments from the HSRB and from the public in
preparation for publication and release as an EPA report.
EPA's response to comments was posted to the draft
report's docket (EPA-HQ-ORD-2007-0972) at
www.regulations.gov.
1.5 Organization of the Document
The document is organized along the lines that the
Expert Panel recommended. It has seven sections, an
introduction followed by a section addressing each of the
major topic areas. The content of each section also is
based on recommendations from the Expert Panel
Workshop. Because the authors concluded that the
discussion for each topic area needed to be complete in
and of itself (i.e., capable of standing independently
without having to reference other sections), there may be
some issues or topics that are discussed in several
sections. Appendixes include additional descriptions of
NERL observational human exposure studies, details
about the process for developing this document, a list of
acronyms and abbreviations, a glossary, and other
supplemental information. The main body of the
document includes the following sections.
• Introduction, Purpose, and Scope (Section 1) lays out
the background for observational human exposure
studies, the scope of the document, and some of the
important scientific and ethical issues that are critical
to human subjects and observational research.
13
-------
• Study Conceptualization and Planning (Section 2)
establishes that ethical concerns are to be incorporated
in the scientific effort from the very beginning and
includes ethical issues such as justifying the study
because of its social and scientific merit and ensuring
scientific validity and independent review.
• Protection of Vulnerable Groups (Section 3) discusses
some of the special protections afforded to vulnerable
groups by EPA's human subjects rules and the ethical
issues of involving such groups in observational
research.
• Ensuring Privacy and Confidentiality (Section 4) lays
out the ethical issues and the regulatory requirements,
including observations of nonstudy hazards and the
recently discussed issues of third-party involvement or
concerns.
• The Relationship Between the Participant and the
Researcher (Section 5) builds on the ethical principles
of respect for persons and beneficence to discuss the
issues around recruitment, informed consent,
payment, and the researcher's need to support the
welfare of the participants.
• Community and Stakeholder Relationships (Section 6)
begins with the principles of fairness, justice, and
equity and of respect for persons to develop
approaches to demonstrate respect for culture and to
empower the participants' community to endure,
including the need to build trust in the community and
with stakeholders through open and honest
communications and legitimate power sharing.
• Strategies for Effective Communication (Section 7)
builds on the presumption of an ongoing, interactive
dialogue and exchange of ideas between researchers
and the participants, community, and stakeholders and
focuses on steps that the researcher needs to take for
effective communications. The section discusses
communication strategies, implementation plans,
communication tools, reporting of results, and
approaches for effective communications, two-way
communications between the researchers, participants,
community, and other stakeholders.
References
Beauchamp TL, Childress, JF (2001). Principles of
Biomedical Ethics: Fifth Edition. New York, NY: Oxford
University Press.
CFR (Code of Federal Regulations) (2006). 40 CFR Chapter I
Environmental Protection Agency Part 26 Protection of
Human Subjects. U.S. Code of Federal Regulations.
Available:
http://www.access.gpo.gov/nara/cfr/waisidx 06/40cfr26 06
.html [accessed 12 June 2007].
CIOMS (The Council for International Organizations of
Medical Sciences) (2008). International Ethical Guidelines
for Epidemiological Studies. Geneva, Switzerland: CIOMS.
Available: http://www.cioms.ch/080221feb 2008.pdf
[accessed 25 April 2008].
CIOMS (The Council for International Organizations of
Medical Sciences) (2002). International Ethical Guidelines
for Biomedical Research Involving Human Subjects.
Geneva, Switzerland: CIOMS. Available:
http://www.doms.cti/frame guidelines nov 2002.htm
[accessed 12 June 2007].
Emanuel EJ, WendlerD, Grady C (2000). What Makes
Clinical Research Ethical? JAMA 283(20): 2701-2711.
NEAC (National Ethics Advisory Committee) (2006). Ethical
Guidelines for Observational Studies: Observational
research audits and related activities. Wellington, New
Zealand: Ministry of Health. Available:
http://www.neac.healtli.govt.nz/moh.nsf/indexcm/neac-
resources-publications-ethicalguidelines.
NRC (National Research Council) (2004) Research Priorities
for Airborne P'articulate Matter: IV. Continuing Research
Progress. Washington, DC: The National Academies Press.
Available:
http://books.nap.edu/openbook.php7record_id= 10957
[accessed 16 August 2007].
NRC (National Research Council) (2003) Protecting
Participants and Facilitating Social and Behavioral
Sciences Research. Washington, DC: The National
Academies Press. Available:
http://books.nap.edu/catalog.plip7record id=10638
[accessed 12 June 2007].
NRC & IOM (National Research Council and Institute of
Medicine) (2005). Ethical Considerations for Research on
Housing-Related Health Hazards Involving Children.
Washington, DC: The National Academies Press. Available:
http://books.nap.edu/catalog.php?record_id= 11450
[accessed 12 June 2007].
U.S. DHEW (U.S. Department of Health, Education, and
Welfare) (1979) The Belmont Report: Ethical Principles
and Guidelines for the Protection of Human Subjects of
Research. National Commission for the Protection of
Human Subjects of Biomedical and Behavioral Research.
Available: http://ohsr.od.nih.gov/guidelines/belmont.html
[accessed 12 June 2007].
U.S. EPA (U.S. Environmental Protection Agency) (1999).
Air Quality Criteria for Particulate Matter. Environmental
Protection Agency, National Center for Environmental
14
-------
Assessment, Office of Research and Development.
EPA/600/P-99/002.
U.S. EPA (U.S. Environmental Protection Agency) (1992).
Guidelines for Exposure Assessment. Environmental
Protection Agency, Risk Assessment Forum. EPA/600/Z-
92/001. Available:
http://oaspub.epa.gov/eims/eimsco mm.getfile?p_download_
id=429103 [accessed 15 April 2008].
U.S. HHS (U.S. Department of Health and Human Services)
(1993). Protecting Human Research Subjects: Institutional
Review Board Guidebook. Office for Human Research
Protections. Available:
http://www.lihs.gov/ohrp/irb/irb guidebook.htm [accessed
12 June 2007].
U.S. OMB (U.S. Office of Management and Budget) (2007).
Final Bulletin for Agency Good Guidance Practices,
Section 1(3). Federal Register 72:16 (25 January 2007) p.
3432. Available:
http://www.whitehouse. gov/omb/fedreg/2007/012507_good
guidance.pdf [accessed 6 December 2007].
15
-------
-------
SECTION 2
Elements to Be Considered in
Study Conceptualization and Planning
Consideration of the scientific and ethical
approaches for observational human exposure studies
begins at the very beginning of the study and continues
throughout it. Because such studies involve human
participants, researchers will have to consider the ethical
issues associated with the regulatory requirements for
human subjects review and approval. Beyond that,
however, ethical principles and issues should be an
integral part of all elements of the study and should be
included as soon as a study is proposed.
In laying out seven requirements for determining
whether a research trial is ethical, Emanuel, Wendler,
and Grady (Emanuel et al., 2000) listed the requirements
in rough chronological order, from conception of the
research to formulation of the plan and protocol to
implementation of the research study. This section
follows their outline as this document highlights areas
that NERL exposure science researchers should consider
as they develop plans for an observational human
exposure study. Table 2-1 shows the ethical
requirements and the topics in this section relevant to
each requirement.
The first stage in the research process is to
understand the state of exposure science and EPA's
programmatic needs for exposure data. NERL scientists
and managers must decide whether an observational
human exposure study is necessary and justified to meet
the Agency's need. If so, then NERL staff will begin to
plan for a study. A variety of important issues will need
to be considered (identifying and enlisting stakeholders
and community representatives, forming a research
team, maximizing benefits for participants, precluding
conflicts of interest, etc.).
The planning phase will culminate in the
development of a science-based study design in an
ethically sound human subjects research protocol.
Adhering to the principal that "Bad science is always
unethical" (U.S. EPA, 2000), the research must first be
judged to be scientifically sound to meet ethical
standards. Peer review by independent and
knowledgeable experts will be used by NERL to assess
the scientific validity of its proposed research. But
passing scientific peer review is just the first hurdle:
sound science is not necessarily ethical. NERL's
observational human exposure studies also must meet
both the ethical requirements set forth in human subjects
regulations and the ethical standards demanded of
responsible researchers by their peers and society. Peer
review will focus on the study design and the science but
also necessarily will incorporate relevant ethical
considerations. IRB review will focus on ethics and the
protection of the human research participants but also
necessarily will incorporate evaluation of the adequacy
of the study design and other relevant aspects of the
science. The principle that unsound science is unethical
science demands that exposure of human subjects to any
research risk whatsoever, even minimal risk, cannot be
justified if the research will not answer the scientific
questions that motivated the research in the first place.
After independent reviews evaluate both the
scientific and ethical aspects of the proposed research,
EPA policy requires that the proposed study undergo
internal EPA review, and mat the human subjects
research protocol undergo evaluation by the Agency's
Human Subjects Research Review Official (HSRRO).
The HSRRO, located in the EPA Office of the Science
Advisor, ensures that Agency studies comply with the
Common Rule and all directives of 40 CFR 26 providing
guidance and principles to govern Federal agency
sponsored human subject research.. Only after HSRRO
approval can any research actually begin. As the NERL
study is implemented, project data and concerns of the
participants will be monitored on a continuing basis and
compared with previously established standards and
17
-------
Table 2.1. Seven Requirements for Determining Whether a Research Trial is Ethical
From Emanuel, Wendler, and Grady (2000)
Requirement
Explanation
Topics in this Section
Social or
scientific value
Evaluation of a treatment, intervention, or theory that
will improve health and well-being or increase
knowledge
Defining the study problem
Justifying the study
Identifying a research team
Scientific validity
Use of accepted scientific principles and methods,
including statistical techniques, to produce reliable
and valid data
Study design
Feasibility
Sample size determination
Representativeness of the sample
Information collection rule
Quality assurance
Study design document
Fair subject
selection
Selection of subjects so that stigmatized and
vulnerable individuals are not targeted for risky
research, and the rich and socially powerful are not
favored for potentially beneficial research
Ethical issues in fair subject selection
Favorable risk-
benefit ratio
Minimization of risks; enhancement of potential
benefits and risks to the subject are proportionate to
the benefits to the subject and to society
Designing in benefits for participants
Assessing benefits and risks of study participation
Independent
review
Review of the design of the research trial, its
proposed subject population, and risk-benefit ratio by
individuals unaffiliated with the research
Scientific peer review
Conflicts of interest
Developing the protocol for ethical review
Ethical review
EPA review of scientific and ethical issues
Informed
consent
Provision of information to subjects about the
purpose of the research, its procedures, potential
risks, benefits, and alternatives, so that the individual
understands this information and can make a
voluntary decision whether to enroll and continue to
participate
Informed consent
Ensuring that participant behaviors are not adversely changed
Respect for
potential and
enrolled subjects
Respect for subjects by
• permitting withdrawal from the research,
• protecting privacy through confidentiality,
• informing subjects of newly discovered risks or
benefits,
• informing subjects of results of the research, and
• maintaining welfare of subjects
Establishing criteria and standards for monitoring scientific and
ethical issues during a study
criteria to evaluate whether the study is on target for
meeting its objectives, or if some unforeseen
circumstances indicate that the study should be stopped
immediately on either scientific or ethical grounds.
2.1 Scientific Value of a Proposed
Observational Human Exposure Study
NERL's planning for an observational human
exposure study begins with an assessment of the state of
exposure science and of EPA's programmatic needs for
exposure data. NERL scientists and managers must
decide whether an observational human exposure study
is necessary scientifically and whether it is justified to
meet the Agency's need.
2.1.1 Defining the Study Problem
Observational studies historically have been
performed for many different purposes and in many
different fields of research—social behavioral,
economic, biological, medical, epidemiological, and
exposure science. NERL has used observational human
exposure studies to understand how people come into
contact with pollutants in their everyday lives, with the
ultimate goal of protecting public health. NERL's
exposure research program addresses critical science
needs directly related to Agency goals for protection of
human health. The research program is driven by key
exposure science questions that may be generated from a
number of different sources, including legislative
mandates (e.g., the Food Quality Protection Act, the
18
-------
Clean Air Act, the Safe Drinking Water Act), program
offices or research planning groups in the Agency,
scientific peers and researchers, or collaborators.
Communities also may identify concerns about
exposures in their locales. NERL's observational human
exposure studies collect data to improve exposure and
risk assessments, to develop risk management strategies,
and to substantiate informational and educational
materials for use by EPA program offices (e.g., Office of
Pollution Prevention and Toxic Substances, Office of
Air and Radiation, Office of Children's Health
Protection).
Emanuel et al. (2000) contend that an ethical
research study must provide a worthwhile social or
scientific value. Ideally, observational human exposure
studies can provide both a scientific value and a social
value to the participants and their community when
feasible. Whenever possible, researchers should work
with communities to develop studies that can help
address community problems and maximize the benefit
to the participants and the community, both of which
also assume a burden for participation in a research
study.
The study problems addressed in past observational
human exposure studies conducted or supported by
NERL have varied substantially, as described in
Appendix A of this document. As shown in Table A-l,
NERL's studies have addressed exposures to particulate
matter (PM), air toxics, persistent organic chemicals, and
nonpersistent chemicals including pesticides. The studies
have examined single routes of exposure (air) and
multimedia exposures, including dietary exposure. The
studies range from small-scale pilot studies to large
probability-based samples. They have included cross-
sectional, longitudinal, and convenience samples. Table
A-l demonstrates that many of the studies are small in
scale and were intended to test a methodology to see if it
may prove useful for a subsequent large-scale
probabilistic human exposure research effort. Both
small-scale and large-scale human exposure studies do
involve human subjects. This means that the studies
must meet both ethical standards and regulatory
requirements. Regardless of the study's scale, the
scientific study design must be technically sound and
appropriate to meet the objectives of the study. But the
nature and objectives of the scientific inquiry will be
different when the study is designed to test a
methodology versus when the study is intended to
measure a representative distribution of people's
exposures. Defining the study problem is a critical and
fundamental first step in the scientific process, because it
will establish the objectives that the research will be
designed to achieve and the uses to which the research
data will be put. The details of the scientific design of
the study will naturally be strongly influenced by the
objectives that the research is intended to meet.
2.1.2 Justifying the Study
Justification of any human study includes both a
scientific and an ethical justification. In the list of seven
ethical requirements that must be met for human subjects
research to be considered ethically acceptable, four of
those requirements, (1) respect for subjects, (2) informed
consent, (3) favorable risk-benefit ratio, and (4) fair
subject selection, are founded on the traditional ethical
principles enunciated in the Belmont Report and
codified in the Common Rule (Emanuel et al., 2000).
But three requirements, (1) social or scientific value,
(2) scientific validity, and (3) independent review, apply
directly to the scientific aspects of the study. Similarly,
Guideline 1 from the CIOMS (2002) document reiterates
the foundational principle that "scientifically invalid
research is unethical." Beyond the traditional ethical
expectations of respect for, protection of, and fairness to
the research subjects, CIOMS requires investigators and
sponsors to ensure that the research be "scientifically
sound," that it "conform to generally accepted, scientific
principles," and that all researchers be "qualified" and
"competent."7 Text Box 2-1 identifies a number of
elements that should be considered in justifying an
observational human exposure study. As discussed
below, a critical element to support justification of both
the scientific and ethical elements of a study is the use of
independent scientific and ethical peer review.
Guideline 1 states "research can be ethically justifiable only if it is carried
out in ways that respect and protect, and are fair to, the subjects of that
research and are morally acceptable within the communities in which the
research is carried out. Moreover, because scientifically invalid research is
unethical in that it exposes research subjects to risks without possible benefit,
investigators and sponsors must ensure that proposed studies involving human
subjects conform to generally accepted scientific principles and are based on
adequate knowledge of the pertinent scientific literature." The commentary on
the guideline goes on to say, "Among the essential features of ethically
justified research involving human subjects, including research with
identifiable human tissue or data, are that the research offers a means of
developing information not otherwise obtainable, that the design of the
research is scientifically sound, and that the investigators and other research
personnel are competent. The methods to be used should be appropriate to the
objectives of the research and the field of study. Investigators and sponsors
must also ensure that all who participate in the conduct of the research are
qualified by virtue of their education and experience to perform competently
in their roles. These considerations should be adequately reflected in the
research protocol submitted for review and clearance to scientific and ethical
review committees."
19
-------
Text Box 2-1. Elements to be Considered in
Justifying a Study
The research problem and questions to be addressed in the
study
The objectives of the study or the hypotheses to be tested
A discussion of why human participants are required for the
study, including a discussion of alternative designs that were
considered
Available information on the need for the study (i.e., it is not
redundant and the research question has not been already
answered)
Available information from the scientific literature
demonstrating the relevance of the proposed study
A discussion of the general technical approach and scientific
soundness of the approach
An assessment of the needed competencies and
qualifications of all personnel involved in conducting the
research
The likelihood of success in meeting the study goals and
objectives (including an evaluation of the accuracy, precision,
and quality assurance of the data needed to attain the study
goals and objectives)
Justification for the investment of time and money
2.1.3 Identifying a Research Team
To Plan and Implement the Study
Once the study problem has been defined and
justified, the next step in developing the research study
is to form the research team. The team should be diverse,
including the technical experts (e.g., exposure
researchers, statisticians, chemists), stakeholders, and
representatives and members of the community in which
the study likely will be performed. Information on
identifying potential conflicts of interest among
researchers early in the planning stage is discussed in
Section 2.5.2. For scientific, ethical, and practical
reasons, the community should be appropriately
involved throughout the study, including throughout the
planning phase. Information on identifying and engaging
community members in the process is described in
Section 6 of this document. The joint NRC-IOM
committee reviewing ethical issues for research
conducted in the homes and communities of the
participants (like much of NERL's human exposure
research) concluded that "When researchers discuss a
planned study with community representatives,
understand their concerns and needs, and respond to
them, protocols can be strengthened both scientifically
and ethically" (p. xii, NRC & IOM, 2005).
2.2 Ensuring Scientific Validity of the
Research Study
To facilitate scientific and ethical review, the
research team members should develop a comprehensive
and detailed study design that describes the technical
approach for the observational study. Although the
format and scope may vary depending on the specific
study, there are a number of basic elements generally
included in the study design.
Translating the information developed in defining
the problem and justifying the study into a real,
workable, feasible study design and human subjects
protocol is an iterative process involving input from all
of the members of the research team. Scientific and
technical expertise is required to assure the scientific
integrity of the research, including developing the
conceptual model8 for the effort and devising a reliable
sampling and analysis plan. Stakeholder input is critical
to assuring that the generalizable research information
from the study actually will be applicable for addressing
the study problem. Community input is particularly
important during the planning stage because the
community representatives can provide valuable
information about the community members (the future
study cohort), the cultures of the community, community
values, community concerns, feasibility of working in
the community, information needed to develop the
technical approach, and information on important factors
like pollutant sources and other stressors in the
community. (Additional considerations for
communicating and working with both the participants
and the community in which they live are the topics of
Sections 5 through 7 of this document.)
In developing the study design and the human
subjects protocol, the research team often will have to
deal with a variety of complex issues, including how to
maximize benefits for participants, the community, and
the stakeholders, and how to ensure the integrity,
generalizability, and representativeness of the study.
2.2.7 Study Design
In epidemiology, the concept of study design has
been structured to include (1) experimental studies, like
drug trials, where the variables are isolated and
controlled (See discussion on experimental studies in
Appendix A.) and (2) observational studies where the
variables are not controlled intentionally, but are simply
A conceptual framework or model is often an effective approach to describe
the relationship between the predicted exposures of the population and the
population stressors, laying out the predicted pathways and routes of exposure
(e.g., see Cohen Hubal et al. 2000). A conceptual model often is illustrated by
a block diagram that represents the major scientific processes and interactions.
The model is often very useful in developing an analysis plan that describes
the hypotheses or objectives of the study, identifies the data needed to address
the objectives, and specifies the analyses that will be done to test the
hypotheses or address the objectives.
20
-------
observed along with the outcome or response measures.
Clearly, the NERL human exposure observational
studies (see Table A-l) fall into the latter category. Such
studies may be cross-sectional studies (where a subset of
the population is observed at the same time); cohort
studies (which measure the same variables and response
measures in a group of people repeatedly over time);
case control studies (where the observed population is
separated into two groups based on the outcome or
response measure [the case group and the control group],
and exposure variables between the two groups are
analyzed for differences); or studies that combine
approaches. The research may collect data from the
participants once (e.g., in a large-scale, randomized
design to collect a representative sample of the
population, like many of the National Human Exposure
Assessment Survey [NHEXAS] studies); over a few
days (e.g., NERL's Particulate Matter [PM] panel
studies); from a cohort in different seasons or years (e.g.,
NERL's Detroit Exposure Aerosol Research Study
[DEARS]; the NHEXAS study in Maryland); or a series
of cross-sectional studies repeated one after another with
a different group of participants each time. (This latter
case often is necessitated to minimize time and expenses
associated with being in the field and because of
limitations of sampling equipment or field technicians to
assist in the collection of the data.)
Although much of the literature on study design
involves experimental studies (like drug trials), there is
much that can be learned about the design of
observational studies in epidemiology, including
understanding their strengths and weaknesses. Web sites,
like "Epidemiology for the Uninitiated" (Coggin et al.,
1997, available at www.bmj .com/epidem/epid.html),
provide useful basic information on important topics for
observational studies and their design.
The specific details of the study design and the
sampling approach to be used will depend very much on
the objectives to be achieved in the study. Based on the
objectives, a good study design must first clearly state
the research question (or hypothesis to be tested) and
must also define in advance what outcome or response
measure will be used to test the question or hypothesis.
Study designs that are vague or which propose to test
some outcome measures against a variety of potential
factors do not represent sound science. Statistically, a
correlation is likely to be found between an outcome and
some factor if enough factors are tested—whether there
is an actual biological or physical relationship or not.
This highlights the importance of statistics as an integral
part of the study design process. Including individuals
with appropriate statistical knowledge and experience on
the research team from its very beginning is critical to a
scientifically sound study design. (See, for example,
Dallal, 2001, "Some Aspects of Study Design" in The
Little Handbook of Statistical Practice at
www.StatisticalPractice.com. and at www.tufts.edu/~gda
llal/LHSP.HTM).
The study design must carefully consider each item
of data or information mat is to be collected during the
study and how the data will be used in the analysis. All
of the data needed to complete the analysis and test the
hypothesis must be collected, otherwise the research
objectives cannot be met, and the study should not go
forward. Alternatively, data that are unrelated to the
study analysis should not be collected. Collection of
such data imposes a burden on both the participants and
research staff with no known potential for benefit.
Collecting, validating, and archiving such data wastes
time and money and inappropriately burdens the
participants. If there is no solid analysis plan for a
particular data item, such data should not be collected. If
the researchers actually believe that some factor may
represent a potentially overlooked variable, then they
should collect the information to test that hypothesis
correctly. But, just collecting the data "while we can,"
with no plan for how to analyze and use the data, wastes
resources, imposes unnecessary burden, and often proves
tempting as a statistical correlation hunt. Finally, the
study design must be described in sufficient detail to be
properly evaluated. Preparation of the study design
document is described in a later section, but the
overriding goal is to provide enough information to
allow proper review of the science behind the study
design by the research team, the scientific peer
community, and the IRB panel members.
2.2.2 Feasibility
The authors consider the evaluation of the feasibility
of accomplishing the study to be one of the most critical
components of the development of the study design.
Evaluating the feasibility of the research project involves
considering not only whether there are sufficient
resources to accomplish the study, but whether or not the
study is feasible from a scientific and ethical perspective.
If the research team concludes that the study is not
feasible, there will be no further effort to develop the
study. There may be practical limitations that preclude
conduct of the study as initially conceived. Because
"scientifically invalid research is unethical"
(Guideline 1, CIOMS, 2002), it is essential that scientific
and ethical considerations be considered together. Text
21
-------
Box 2-2 includes some examples of the types of
questions that may be asked when evaluating the
feasibility of a studv.
Text Box 2-2. Is the Study Feasible?
What are the sample size requirements? Can enough
participants be enrolled into the study from the proposed
community, considering the eligibility criteria and anticipated
response rate? What is the predicted retention rate if this is a
repeated measurements study? Is that acceptable?
Is the research question well defined, and can we measure the
variables and the outcome metrics?
Are the measurement methods specific and accurate enough
to accomplish the research?
Does the analysis plan provide results to test the research
questions and to meet the study objectives?
Will the community be receptive to this study?
Are there cultural mores, societal values, or other factors
associated with the community that would make it difficult to
conduct the study in the community? Is the study morally
acceptable to the community?
Is there a community structure that will allow the research
team to engage the community in the study?
What is the burden on the community?
What is the burden on the participants? Is it acceptable?
What is the risk-benefit balance? Is it acceptable, considering
the burden on the participants?
Are there sufficient resources available to conduct a study of
this size?
Are resources available for community outreach and sustained
interactions with the community? Are resources available to
support community members involved in the study?
2.2.3 Sample Size Determination
One critical issue in assuring that an observational
human exposure study is scientifically valid (and,
thereby, not invalid and unethical) is the issue of sample
size. EPA's Science Advisory Board has stated, "Bad
science is always unethical; research protocols that are
fundamentally flawed, such as those with sample sizes
inadequate to support reasonable inferences about the
matter in question, are unjustifiable" (p. 2, U.S. EPA,
2000).
A study has to have an adequate size to meet the
study objectives. If the sample size is too small, the
results may not be statistically significant, and the results
may not be either valid or generalizable. Such a result
would be a waste of resources or cause undue burden on
study participants without generating the intended
generalizable knowledge mat will benefit society. On the
other hand, if the study sample size is larger than
necessary to meet a study objective, this also may result
in a waste of resources or the imposition of needless
burden on participants.
Sample size determination is an important step in
planning a study, but it can be a difficult task (Lentil,
2001). Sample size determination also may be
confounded by other issues that will reduce the number
of measurements that may be used in the analysis. For
example, participants in a longitudinal study may "drop
out" over time (some may move, others may tire of
participation, etc.), so that the number of participants at
the end of the study will be less than the number that
started the study. The study design must account for the
attrition of participants over time and plan ahead to
recruit enough participants to ensure that a statistically
useful number of participants complete the study. Often
NERL scientists have used historical retention rates to
estimate the sample size needed for a study. (See
discussion of retention rates in NHEXAS publications.)
In addition, some participants or their environments
may not have concentrations above the detection limit.
Depending on the nature of the research question, data
below the analytical detection limit may, or may not, be
useful in addressing the research question: sample size
may need to be increased to account for missing values
and values below detection. Pilot studies and
environmental studies (with no participants as human
subjects) may prove useful in understanding the range of
concentrations to be expected and may provide insight
into how to adjust the sample size appropriately. In
surveys, like NERL's National Human Activity Pattern
Survey (NHAPS) for example, some randomly selected
participants simply may refuse to participate, decreasing
the number of samples and perhaps biasing the data.
It is critical mat sample size be determined at the
time of study conceptualization and planning and not
after the study already has been conducted. Researchers
must include experts with the appropriate statistical
expertise on the research team at its very inception. For
additional information and insight, readers should refer
to biostatistics books. Web sites, and other references
(Lenth, 2001; Castelloe, 2000; Kraemer and Thiemann,
1987; Van Belle and Fisher, 2004; Wackerly et al., 2001;
Dallal, 2001) about this very important topic.
2.2.4 Representativeness of the Sample
In addition to concerns about the sample size being
sufficient to provide statistically significant results,
researchers also must be concerned about the individuals
who participate in a study and of what group or
population they may be representative. Research, as
defined in the Common Rule, is "a systematic
investigation ... designed to develop or contribute to
generalizable knowledge." How "generalizable" the
22
-------
results of an observational study may be depends on the
representativeness of the sample (the participants). A
review of former NERL studies (Table A-l) shows a
wide variety of approaches to selecting the participants.
both probability sampling and nonprobability sampling.
Some of the studies have involved randomized sampling
(e.g., NHAPS, NHEXAS), selecting a cohort of interest
and then using a randomized sample (e.g., Children's
Total Exposure to Persistent Pesticides and Other
Persistent Organic Pollutants [CTEPP], DEARS) or a
convenience sample (e.g., some panel studies),
purposeful selection (judgment sampling) by the
research team (for methods testing studies), recruiting
the whole population in a locale or of a particular cohort
(e.g., as proposed for the National Children's Study
[NCS]), or simply a convenience sample (in small pilot
studies).
The approach used to select the participants has
depended on the objectives of the study. Research to
understand and describe the distribution of exposures in
the general population (like NHEXAS) or of the
population or cohort in a particular state or region (like
CTEPP or Agricultural Health Study [AHS]), has
required a large number of randomly-selected
participants. Randomized selection is recognized as a
valid statistical method to get a sample that is
representative of the larger population from which the
participants were selected. Pilot studies performed to
evaluate a method or to estimate the likely range of
exposure concentrations often have employed either
purposeful sampling, where the researchers use
information on the relevant characteristics of the
population to select those participants who will exhibit a
wide range of activity levels or potential exposure
concentrations, or convenience sampling, where the
researchers select the most accessible members of a
population
NERL's observational human exposure studies also
have routinely collected information about the
participants" activities by using questionnaires and
surveys. Survey design is both a science and an art
because the design of surveys is based on statistics and
science, but designing a good and effective questionnaire
is often an art that requires understanding the individuals
being surveyed. Text Box 2-3 identifies some of the
areas of art involved in designing an effective
questionnaire.
Understanding the process for selecting participants
and the statistical-scientific requirements of
questionnaire design are both components of survey
sampling and design. A variety of references can provide
the researcher with information about the issues in
survey sampling and design. (See
http://home.ubalt.edu/ntsbarsh/Business-stat/stat-
data/Surveys.htm and www.statpac.com/surveys/, for
example.) But the research team must include or have
access to the appropriate survey statistics expertise as it
plans the research study.
Text Box 2-3. Questionnaire Considerations
Questions should be simple and in a language the individual
can understand.
Questions should be unambiguous.
Questions should be relevant to the study.
Questions should not be too personal.
The questionnaire should be as short as possible.
The questions should not be leading (e.g., internally
suggestive of the answer).
Questions should follow a logical order.
2.2.5 Information Collection Rule
As Federal employees. NERL researchers also must
be aware of Information Collection Rule requirements.
The Paperwork Reduction Act stipulates mat every
Federal agency must obtain approval from OMB before
collecting the same or similar information from 10 or
more members of the public. An Information Collection
Request (ICR) is required if the same or similar
information is being collected from 10 or more non-
Federal respondents within a 12-month period, even if
the information collection is voluntary1. Generally, any
survey, questionnaire, monitoring, reporting, or
recordkeeping requirement imposed on non-Federal
respondents by EPA will require an ICR. Information
collections associated with all cooperative agreements
funded by the EPA require an ICR. When an ICR is
required, it must be approved by OMB before the
collection begins, regardless of whether the collection of
information is mandatory, voluntary, or required to
receive a benefit. The principal investigator must prepare
an ICR and submit it to the appropriate Office of
Environmental Information Desk Officer.
An ICR
• describes the information to be collected,
• provides justification for why the information is
needed, and
• estimates the time and cost for the public to answer
the request.
Information about ICRs and their requirements is
available to NERL and other EPA staff members at
http://intranet.epa.gov/icrintra/index.html.
23
-------
2.2.6 Quality Assurance Project Plan
Data of unknown or uncertain quality can undermine
the scientific integrity of a study and render an otherwise
sound study invalid. NERL scientists must be diligent in
the implementation of the procedures and processes
specified in a well-developed quality assurance project
plan (QAPP). A discussion of quality assurance
programs and QAPPs is outside of the scope of this
document. There are many good references on the topic,
including the EPA Web site, www.epa.gov/qualitv/.
2.2.7 The Study Design Document
The study design document should contain sufficient
detail to allow independent review and assessment of the
scientific soundness of the study and the approaches that
will be followed to ensure that the study meets the
highest scientific and ethical standards. The research
team can meet regularly to specifically evaluate the plan.
It should be noted that a study design is not the same as
an implementation plan. The latter includes an even
greater level of detail describing how the study will be
performed and includes protocols and operating
procedures. Text Box 2-4 lists a number of elements that
may be appropriate to include in a study design.
2.3 Ethical Issues in Ensuring
Fair Subject Selection
One of the ethical principles of human subjects
research is that selection of participants should be fair. In
Section 2.2.4, the discussion centers on understanding
what population or cohort the participants in a research
study represents, and how such a selection may be
influenced by the research questions and study
objectives. That consideration actually represents the
first component of ensuring mat subject selection is fair.
That is, the scientific goals of the study should be the
primary basis for determining what groups or individuals
should be recruited and enrolled in a study. Participants
should not be recruited either because of privilege or
because of vulnerability or their inability to look out for
their own interests properly. Similarly, groups or
individuals should not be excluded peremptorily without
consideration of the risks and benefits to them as
individuals. Section 3 of this document discusses
protection of vulnerable groups and concludes that
NERL researchers should include vulnerable groups in
observational human exposure studies only if their
participation is critical to the success and applicability of
the research. Even then, EPA and NERL researchers will
have to meet stringent standards for protecting the rights
and safety of the vulnerable participants. "The essence of
Text Box 2-4. Elements That May Be Included in a
Study Design
Introduction and background, including the purpose and scope
of the study
The desired outputs and outcomes of the study, including the
objectives and the hypotheses to be tested
A brief description or overview of the study
The technical approach and conceptual model that accounts for
- sources of the chemicals being studied;
- potential routes and pathways of exposure;
- factors that may impact exposure and other relevant
stressors;
- selection and characteristics of the study participants;
eligibility criteria; and recruitment, retention, and payment
approaches;
- justification for sample size, the methodology for selecting
participants, and the sampling methods;
- characteristics of the community in which the study will be
performed;
- environmental conditions, factors, or end points to be
measured, including sampling and analysis approaches and
methods (with description of expected performance);
- survey design and questionnaires and other survey
instruments, as applicable (with description of prior use and
validation in similar studies);
- pilot studies that may be undertaken;
- quality assurance project plan and quality control;
- timeframe for the study;
- exposure scenarios to be considered;
- burden of the study on the participants;
- resources available; and
- feasibility
Discussion of alternative study designs and approaches
considered and reasons for rejecting other approaches and
selecting the one proposed
An analysis plan that considers
- Information and data needs, including data storage, security,
access, and release;
- nature of the measurement data (e.g., variability, quality
assurance);
- how the collected data will be used, and how the proposed
analyses will address objectives of the study; and
- hypotheses to be tested and statistical power and sample
size required to test the hypotheses
Resources required or available
Project organization and management, including team members
and roles and responsibilities
Schedule
fairness in human subjects research is that scientific
goals, considered in dynamic interaction with the
potential for and distribution of risks and benefits,
should guide the selection of subjects." (Emmanuel et al,
2000). On the other hand, as discussed in Section 3,
many study problems specifically address exposures of
vulnerable groups to chemicals and other stressors.
Researchers should not avoid research studies that will.
24
-------
of necessity, include vulnerable groups to address the
study hypotheses or objectives simply to avoid the more
stringent requirements for working with these groups.
2.4 Ensuring a Favorable Risk-Benefit Ratio
2.4.1 Designing in Benefits for the Participants
Study designs vary depending on the objectives of
the study, existing knowledge on the research question,
and the hazard being studied (NRC & IOM, 2005).
Recent ethical discussions about study designs in human
subjects research (cf., Recommendation 7.1, p. 143,
NRC & IOM [2005] and Emanuel et al. [2000]) support
the development of innovative study designs to
maximize the benefit9 to the study participants, as well
as to the community and the greater society beyond.
Observational human exposure studies generally collect
data that contribute to generalizable knowledge that will
benefit the community and society as a whole, but they
often do not provide obvious direct benefit to study-
participants. Therefore, it is important to include
elements in the study design that can offer benefits to the
participants wherever possible. This is not always
straightforward, but one way that participants, as well as
communities, can benefit from these studies is by
incorporating strong educational components into the
conduct of the research. For example, brochures, videos,
and other materials that educate study participants on
safety around the home or on how to reduce their
exposure to chemicals can be distributed during the
study. EPA's program offices, including the Office of
Children's Health Protection, the Office of Pollution
Prevention and Toxics, the Office of Pesticide Programs,
the Office of Drinking Water, and others have Web sites
with substantial amounts of informational and
educational materials available that could be distributed
to study participants. Other organizations, such as the
American Lung Association, the American Cancer
Society, the American Academy of Pediatrics (AAP),
and various environmental groups, have materials of
which study participants may not be aware that could be
used as educational materials when relevant.
In addition, approaches that provide direct benefits
to study participants will need to be tailored to the
particular study population and community. Feedback
from potential participants in focus groups and input
from community representatives may be useful in
identifying these approaches.
Payment to participants is never considered a benefit of a study.
2.4.2 Assessing Benefits and Risks of Study
Participation
For all research involving human participants, the
Common Rule requires researchers to ensure that
potential risks "are reasonable in relation to the
anticipated benefits," and mat risks are minimized (40
CFR 26.111). It is most useful if the assessment of
benefits and risks is begun early in the scoping and
planning phase of a study.
Unlike some biomedical research that involves the
study of interventions or procedures that hold out the
prospect of direct diagnostic, therapeutic, or preventative
benefit for the study participants, observational human
exposure studies often do not have a similar prospect of
direct benefit to the participant. Therefore, the risk-
benefit balance is based on the balance between the risks
to the participants and the expected benefits to society
(generalizable knowledge). The risks to participants
must be reasonable [40 CFR 26.111(a)(2)] in relation to
the importance of the knowledge gained. This
assessment of the risk-benefit balance, therefore, needs
to be performed in the initial planning of the study to be
included in the justification for the study (Section 2.2).
If there is no prospect of direct participant benefit,
and the study participants are children, moreover, EPA is
permitted to conduct or support only those observational
human exposure studies that meet both the regulatory
definition of "observational" and the regulatory
definition of "minimal risk." The latter is defined in the
Common Rule at 40 CFR 26.102(i) and reiterated in
Subpart D of the EPA Rule at 40 CFR 26.402(g):
^Minimal risk means that the probability and magnitude
of harm or discomfort anticipated in the research are not
greater in and of themselves than those ordinarily
encountered in daily life or during the performance of
routine physical or psychological examinations or tests."
In applying this definition, EPA adheres to the
consensus standard that the reference population for this
definition is normal children living in safe, healthy
environments. In its discussion of the perception of risks
and benefits, the NRC & IOM (2005) report on housing
health hazards in children notes mat the children
participating in these studies may be at risk for physical
harms or adverse health outcomes because they live in
housing (or otherwise occupy environments) with health
hazards. However, such risks are not introduced by the
research but, rather, would be present whether or not the
children were involved in a research study. As a
consequence, the study still would meet the regulatory
criteria for minimal risk as long as the research itself
introduced no risks over and above those minimal risks
25
-------
experienced by normal children living in safe healthy
environments.
However, the existence of greater than minimal
background risks that are not introduced by the research,
nonetheless, raises additional ethical considerations. The
joint NRC & IOM Committee on Research on Housing-
Related Health Hazards Involving Children discussed
the ethical arguments that arise when scientists conduct
research that observes children in poor-quality housing.
They point out that a researcher's first duty of
beneficence under the Common Rule requires that the
risks of the research actions be proportionate to
["reasonable in relation to"; 40 CFR 26.111(a)(2)] the
benefits of the research, and that the risks be minimized.
They acknowledge, however, mat some have argued that
the "best interests of the child" also obligates researchers
to "rescue" children from harm and to provide better
living conditions. They conclude that, properly applied,
the ethical principle of beneficence does indeed direct
researchers who observe serious harms to child subjects
to take steps to try to prevent the harms. However, they
also argue that the researcher's duty does not extend to
"personally and directly prevent harm by removing the
child from the harmful environment" (p. 60, NRC &
IOM, 2005). They conclude instead mat "it is unrealistic
and unfair to hold individual research investigators
responsible for ameliorating the social circumstances
that they study" and that "a nuanced balancing of the
benefits and risks of research" is an ethically sound
approach that is firmly established in Federal regulations
(p. 60, NRC & IOM, 2005). Balancing the ethical
obligation to mitigate risks or harms observed during
research with the reasonable limits on an investigator's
moral responsibility for the social circumstances
surrounding the research will be the subject of later
sections of this document, particularly Section 4.3.1.
Assessing the risks and benefits of the research
study can be very difficult for the researchers, especially
because the researchers and the community or
participants may perceive the risks and benefits quite
differently. (See the discussion in NRC & IOM [2005],
for example.) To understand the community's
perspective better, the researcher may find it helpful to
discuss the assessment of risks and benefits with
members of the research team, community
representatives, and relevant stakeholders. The research
team should consider the use of a community advisory
board (CAB) to provide input to the assessment of the
risks and benefits of the study. The group could include
individuals who are representative of the population to
be studied, community representatives, exposure
scientists, and bioethicists. The group should include
experts familiar with the human subjects research
regulations, preferably including someone who has
served on IRBs. Obtaining input from the group can be
accomplished by submitting the study concept and
general study design to the group for review and
feedback, even before a full study design has been
developed. (See the discussions of CABs in Sections 5
and 6.) Ultimately, it will be the review by the members
of the IRB that will determine whether the balance is
appropriate and justifiable.
2.5 Independent Scientific and
Ethical Review
Because issues of science and ethics are intrinsically
bound together in human subjects research (Emanuel et
al., 2000; CIOMS, 2002), it is important that scientific
and ethical reviews be considered together, not
separately. Scientific reviews are performed to ensure
the scientific soundness of the study, whereas ethical
reviews are performed to ensure proper action and the
protection of the human subjects in a research study. A
study that is not scientifically sound could expose study
participants to unnecessary risk or inconvenience and
burden, with no additional societal benefits (i.e., no
increase in generalizable knowledge). EPA's Science
Advisory Board has stated that "bad science is always
unethical" (U.S. EPA, 2000), and CIOMS declares that
"scientifically invalid research is unethical" (CIOMS,
2002).10 It is clear, therefore, that the ethical review has
to consider the scientific aspects of the study also.
There may be multiple levels of review during
development of the study design and human subjects
research protocol for an observational human exposure
study. The research team is responsible for the design of
the study and for ensuring that adequate peer review is
CIOMS (2002) Guideline 2 asserts "Ethical review committees—All
proposals to conduct research involving human subjects must be submitted for
review of their scientific merit and ethical acceptability to one or more
scientific review and ethical review committees. The review committees must
be independent of the research team, and any direct financial or other material
benefit they may derive from the research should not be contingent on the
outcome of their review. The investigator must obtain their approval or
clearance before undertaking the research. The ethical review committee
should conduct further reviews as necessary in the course of the research,
including monitoring of the progress of the study." The CIOMS document
continues, "According to the Declaration of Helsinki (Paragraph 11), medical
research involving humans must conform to generally accepted scientific
principles, and be based on a thorough knowledge of the scientific literature,
other relevant sources of information, and adequate laboratory and, where
indicated, animal experimentation. Scientific review must consider, inter alia,
the study design, including the provisions for avoiding or minimizing risk and
for monitoring safety. Committees competent to review and approve scientific
aspects of research proposals must be multidisciplinary."
26
-------
performed to evaluate both the scientific and ethical
approaches for the study. Following completion of a
draft study design, researchers should engage a diverse
group of experts to review the study design and human
subjects aspects. The scope of the study should dictate
the level of the review (i.e., internal or external
independent peer review). A small pilot study to
evaluate measurement methods or to collect screening
level data in preparation for a large study may not
require as extensive review as a larger study.
When the scientific soundness of the study has been
evaluated and found to be feasible, and the final study-
design is completed, the human subjects research
protocol should be developed and submitted to the 1KB
for review and approval. For studies conducted or
supported by EPA, additional review and certification of
the human subjects research protocol is required by EPA
Order 1000.17 Al. (available online at
www.epa.gov/oamrtpnc/forms/1000 17a.pdf). Review
and approval of the protocol and associated documents
must be obtained from EPA's HSRRO, located in the
EPA Office of the Science Advisor, before any work
begins. Any changes made to the protocol after IRB
approval must be submitted to and approved by the IRB.
2.5.1 Scientific Peer Review
For all studies, regardless of the scope, the research
team should solicit review and comment on the scientific
approach by experts external to the research team. A
peer review panel consisting of individuals who were not
involved in the design of the study can be formed to
review the scientific soundness of the study. It is
important for the panel to consist of individuals with
experience and background appropriate to the study and
to include members with knowledge of the ethical
principles for protection of human subjects in these types
of studies. The panel also would benefit from including
someone with sufficient background and expertise in
statistics to evaluate whether the study design, sample
size, and proposed data analyses are appropriate and
adequate to address the study objectives or test the
hypotheses. For small studies, the peer review panel may
consist of individuals within the organization conducting
the study if they have not been involved in developing
the study design. For larger and complex studies, it is
recommended that an external peer review panel be
convened to review both the scientific and ethical
soundness of the study design.
For research conducted or sponsored by NERL,
human subjects research efforts will undergo both a
scientific review and an ethical review. The director of
the division conducting or funding the observational
research is the manager with the primary responsibility
for ensuring mat the scientific and the ethical reviews are
conducted, and that the review comments are properly
addressed. The study design will be reviewed for
scientific quality by independent and knowledgeable
reviewers. Depending on the scope of the study, the
appropriate NERL associate director or the NERL
laboratory director will make the final determination
about (1) the process for selecting scientific peer
reviewers (including the range of disciplines to be
included), (2) the nature and scope of the review process
(e.g., charge to the reviewers and scope of the review;
letter reviews, convening a peer panel, or both; the size
and nature of the panel review; etc.), and (3) the
adequacy of the responses to the scientific review.
2.5.2 Conflicts of Interest
It is recommended that potential conflicts of interest
among researchers or study participants be identified at
all stages of study planning and implementation, but
particularly early in the study during the planning stage.
There can be many sources of potential conflicts of
interest, including sources of project funding, pressures
to publish, consulting arrangements of the investigators,
employment of investigators' family members with
affected parties, participation in affected advocacy
groups, collaborations or relationships with experts on
the IRB or other independent review committees,
institutional conflicts for any contractors who may be
involved, or a wide range of other situations.
Any situations that constitute actual conflicts of
interest and all potential or apparent conflicts of interest
must be reported to the IRB for their review and
resolution. The CIOMS (2002) guidelines for research
protocols involving human subjects specif}' that all
sponsors of the research be identified, and that the
protocol include actions to disclose and address potential
conflicts of interest. Concerns about conflicts of interest
also need to be identified and discussed with the
researchers, community, and other stakeholders to make
a determination of the existence of conflicts, and how
they should be avoided or handled.
Even if actual conflicts of interest do not exist,
researchers should recognize that mere can be perceived
conflicts of interest that can be just as damaging as real
conflicts of interest. Perceptions by participants,
community members and representatives, stakeholder
groups, and the public may be substantially different
from the reality of the situation. This is especially likely
to occur when external sources, such as industry, are
27
-------
involved in funding research. Even though researchers
ma}' develop agreements with funding organizations that
ensure researcher autonomy, a perception may exist that
the funding organization will bias the study (Resnik and
Wing, 2007). Concerns about perceived conflicts of
interest should be discussed with the IRB and other
relevant review committees, in addition to the
researchers, the community, and other stakeholders. The
IRB must be made aware of any circumstances that may
give rise to actual conflicts of interest or to the
appearance or perception of a conflict of interest. The
IRE's recommendations about how to resolve any such
conflicts must be strictly adhered to by NERL
researchers.
2.5.3 Develop the Human Subjects
Protocol for IRB Review
IRBs may have specific format requirements for
their human subject research protocols. Traditionally, the
human subjects research protocols for research
conducted or funded by NERL have included
descriptions of the project, including title and description
of the research; the duration of the project; the type of
data to be collected; the objectives of the study; the
number of samples; a description of the participants and
participant recruitment procedures; the informed consent
procedures and forms; estimates of participant risk and
burden, an assessment of benefits and the risk-benefit
ratio; and actions to protect the participants. CIOMS has
developed a comprehensive list of items that they
recommend for inclusion in a human subjects research
protocol (Appendix 1, CIOMS, 2002). Many of the
items that they identify are also useful for observational
human exposure studies. (The CIOMS items can be
found in Appendix C of this document.) The authors
recommend that anyone developing a human subjects
protocol for observational human exposure studies
review and utilize the CIOMS list of topics, as
appropriate. Text Box 2-5 identifies a number of topics
that should be considered in development of the human
subjects research protocol.
In addition, the authors" experience leads them to
suggest that three additional topics beyond those from
the CIOMS (2002) document also may need to be
considered in a human subjects protocol: (1) approaches
to minimize changes in participant behavior because of
participation in the study (see Section 2.3.4);
(2) approaches to minimize therapeutic misconception
(see Section 5.4.1); and (3) actions to involve the
community in a community-based participator}' research
Text Box 2-5. Potential Topics in a
Human Subjects Research Protocol
1. Title
2. Summary in lay language
3. Justification for the study
4. Ethical issues and proposed resolution
5. Summary of previous research
6. Affirmation of Belmont Principles and 40 CFR 26 compliance
7. Previous history or use of the protocol
8. Information on the location and demographics of research
9. Information on funding organization, researcher partners,
and collaborators
10. Names, qualifications, and experience of investigators
11. Objectives, hypotheses, assumptions, and variables
12. Study design
13. Sample size and power and statistical analysis plan
14. Criteria and justification for subject selection
15. Justification for use of vulnerable groups, if any
16. Process of recruitment
17. Actions to involve the community in a community-based
participatory research program
18. Description and explanation of any and all interventions
19. Measurements or data to be collected
20. Any clinical or other tests
21. Rules or criteria for removing subjects or terminating the
study
22. Adverse events and unanticipated problems—reporting and
responses
23. Potential benefits to subjects or others
24. Expected benefits of the research to the population
25. Informed consent process and responsibilities
26. Protections for the consent/assent of vulnerable participants
27. Efforts to minimize "therapeutic misconception"
28. Approaches to minimize changes in participant behavior
29. Payments
30. Plans for informing subjects about items that could affect
subjects' willingness to continue in the study
31. Plans to inform subjects about the results of the study
32. Privacy and confidentiality
33. Security of personal information and when, how, and by
whom private information can be revealed
34. All foreseen uses of personal data or biological materials
35. Procedures for data and safety monitoring and oversight of
the study and the criteria for identifying, reporting, and
responding to adverse events, including ethical breaches,
environmental measures in excess of reporting standards,
and collateral observations, and criteria for prematurely
terminating the study if necessary
36. A list of the references cited in the protocol
37. The source and amount of funding
38. Protocols for dealing with financial or other conflicts of
interest
39. Schedule
40. Arrangements with sponsors regarding publication rights and
procedures
41. Reasons for not publishing the study findings
42. Procedures for preventing falsification of data
28
-------
(CBPR) effort, as appropriate (see Section 6, especially
Section 6.10).
2.5.4 Ethical Review
In the United States, ethical reviews of studies
involving human subjects are performed by IRBs. The
Common Rule specifies requirements (40 CFR 26.107 -
115) for IRB membership, IRB functions and operations,
IRB review of research, and other details related to IRB
review and approval of research. Emanuel states that
"the independent ethical review of [human subjects
research] should involve individuals with training in
science, statistics, ethics, and law, as well as reflective
citizens who understand social values, priorities, and the
vulnerability and concerns of potential subjects"
(Emanuel et al., 2000). It is beyond the scope of this
document to include detailed discussions on IRB
membership, operations, processes, etc. The reader is
referred to the Common Rule, as well as a number of
other available references (e.g., CHRP, 2007; CFR,
2006; U.S. HHS, 1993; NRC, 2003).
It is essential that research with human subjects be
carried out or strictly supervised by suitably trained,
qualified, and experienced investigators. For all research
subject to the Common Rule, these qualified researchers
are expected to prepare a human subjects research
protocol (as in Section 2.5) and to submit the protocol to
be ethically and scientifically appraised by one or more
suitably constituted IRBs, independent of the
investigators.
There are a number of other issues associated with
IRBs that may impact researchers conducting
observational human exposure studies. As an example,
there has been concern about the transparency of IRBs.
Questions have been raised about what information the
IRB should make available to the public regarding
membership on the IRB for review of individual
projects, the discussions held with the researchers, the
IRB's concerns about the research protocol, the
researchers' response, etc. Should this information be
documented in files mat the researchers can make
available to the participants, community, stakeholders,
and the public? At the present time, there is no clear
approach as to how to address these issues. Because
these issues are associated with the IRB, not the
researcher, it is outside the scope of this document to
recommend approaches for IRBs to address these
concerns. IRB processes and procedures will continue to
evolve as recommended by various committees and
workgroups (e.g., as reported in NRC & IOM, 2005;
NRC, 2003; U.S. HHS, 1993).
All human subjects research conducted or sponsored
by NERL is subject to both the 40 CFR 26 requirements
and procedures set forth in EPA Order 1000.17 Change
Al (www.epa.gov/oamrtpnc/forms/1 OOP 17a.pdf). The
EPA order establishes as policy that all research will
comply with the Common Rule and with the order. All
human research studies must be reviewed and approved
by the EPA HSRRO before the work can begin.
In NERL, the director of the division conducting or
funding the research is the manager with the primary
responsibility for developing the human subjects
research protocol and for having that protocol reviewed
by an independent IRB acceptable to the EPA HSRRO.
The protocol also will be reviewed by the NERL
HSRRO and by the appropriate NERL associate director
before it is submitted to the IRB. Under 40 CFR 26.109,
the IRB can demand changes to the research protocol
and is the final authority for approving or disapproving
the research activity.
2.5.5 Internal EPA Review of Scientific and
Ethical Issues
After IRB approval is obtained, the division director
will be the primary manager responsible for preparing a
request for review and approval or exemption of the
human subjects research by the EPA HSRRO. The
division director will ensure mat the request is consistent
with EPA Order 1000.17 Al and all other policies or
procedures that the EPA HSRRO may have established.
The EPA HSRRO shall be the final authority for
approving or disapproving the research effort. The EPA
HSRRO may request additional reviews or establish
additional policies and procedures for seeking review
and approval. No human subjects research will
begin—not even recruiting of potential
participants—until the EPA HSRRO has approved or
exempted the research.
2.6 Informed Consent
Informed consent is discussed extensively in
Section 5 of this document. The major focus is that
informed consent is a process, not a form, that "should
be an on-going, interactive dialogue between research
staff and research participants involving the disclosure
and exchange of relevant information, discussion of that
information, and assessment of the individual's
understanding of the discussion" (Recommendation 4.1,
IOM, 2002). These comments emphasize how important
true two-way communication is to comprehension, the
second pillar in the informed consent process. Informed
consent is built on three "pillars:" (1) information; (2)
29
-------
comprehension; and (3) voluntary participation, or
"voluntariness" (U.S. DREW, 1979). Informed consent
requires "provision of information to subjects about the
purpose of the research, its procedures, potential risks.
benefits, and alternatives, so that the individual
understands this information and can make a voluntary
decision whether to enroll and continue to participate"
(Emanuel et al., 2000).
2.7 Ensuring That Participant Behaviors Are
Not Changed Adversely Because of Being
in the Study
The goal of observational human exposure studies is
to collect information on people's exposures to
chemicals in their real-world environment as they carry
on their normal daily activities. Researchers who
conduct these studies, however, recognize that
participation in a study may affect behavior. This cannot
always be avoided, as simply agreeing to participate in a
study may impact the participant's activities and
schedules. For example, this occurs when technicians
visit homes to collect samples or when participants are
asked to collect samples (e.g., food, urine), or to
complete surveys, activity logs, or questionnaires. These
types of changes in behavior may or may not affect the
outcome of the study.
Some changes in behavior can affect the study
outcome. The Hawthorne Effect is a well-recognized
phenomenon. It is an effect on an outcome variable
caused by the fact that the participants of the study know
they are participating in the study. The Hawthorne Effect
originally referred to the increase in worker productivity-
observed when a worker is singled out and made to feel
important; the increased productivity was not related to
the environmental factors mat were being studied. The
effect was described based on a series of industrial
productivity studies from 1927 to 1932. Similarly, some
changes in participant behaviors may change the
observations, measurements, and conclusions from
observational human exposure studies. For example.
participants may do more cleaning in their home because
they do not want the researchers to think they are poor
housekeepers; this could affect the measurement of
environmental concentrations in the home. In a study of
chemicals from consumer products, participants may
think mat because the researchers are studying the
products, the products must be harmful. Therefore, study
participants may elect not to use the products during the
study in the same manner as they would normally.
Alternatively, prospective participants may choose to
use more of the household product to qualify for the
study. As a result, the participant's exposure to the
chemicals could be either more or less than "normal."
Any change in a participant's behavior that is related
to the research question being addressed in the study
may impact the study results. Researchers should try to
anticipate how a study may impact participant behaviors
and ensure that the study design and implementation
protocols do not cause changes in behavior that may
cause harm to a participant during a study. A number of
study elements with the potential to influence
participants' behavior are listed in Text Box 2-6.
Text Box 2-6. Study Elements That
Could Affect People's Behavior
Eligibility criteria
Recruiting approach and materials
Enrollment approach
Payments
Retention strategy
Types of measurements made and data collected
Protocols for data collection
Protocols for visits to homes
Interactions with participants
Communications
It is very difficult to predict in advance how these
elements may be interpreted and acted on by the
participants. Researchers may learn from the experiences
of others, including the "'lessons learned" from experts
and their publications. They may wish to engage the
community representatives (see Section 6) in a thorough
discussion of the issue. Community-based focus groups
or pilot studies also may demonstrate how the various
elements of the study may have an unintended impact.
Additionally, researchers can be very careful in the
informed consent process (see Section 5.1), to ensure
mat participants not only know, but that they understand
the facts of the study (Gilbert, 2006), and that they
comprehend that the goal is to observe and measure the
participant's exposures during their normal, everyday
activities.
2.8 Criteria and Standards for Monitoring
Scientific and Ethical Issues
Ethical requirements do not end when the
participants sign the consent form and agree to
participate. Indeed, as was stated above, informed
consent is a process, not a form, and the process is an
on-going dialogue between participants and researchers
mat continues throughout the study and beyond. In
discussing respect for both potential and enrolled
30
-------
subjects, Emmanuel et al. (2000) identify five actions
that demonstrate respect for subjects, including
(1) permitting withdrawal from the research,
(2) protecting privacy through confidentiality,
(3) informing subjects of newly discovered risks or
benefits,
(4) informing subjects of results of other relevant
research, and
(5) maintaining welfare of subjects.
The authors conclude that these requirements can be
met by diligent exercise of data and safety monitoring
and oversight of the research effort. (Data and safety
monitoring and oversight also are discussed throughout
Section 4, but especially in Section 4.5 entitled "Data
and Safety Monitoring and Oversight" and in
Section 7.8, "Reporting Unanticipated Results and
Observations." The reader also is referred to those
sections.)
Creation of data and safety monitoring and oversight
organization and procedures, combined with careful
consideration of actions to ensure the ethical protection
of participants, is perhaps the most important aspect of
planning and implementing an observational human
subjects study. The monitoring must include both the
technical aspects of the study—like planning for actions
if unsafe environmental conditions are observed—and
also the ethical aspects of the study, such as those items
listed above. The team of people involved in data and
safety monitoring and oversight have to establish and
implement mechanisms to get feedback on a continuing
basis from the participants, in addition to monitoring and
reviewing the scientific data.
CIOMS recommends that all human subjects
research protocols contain "A description of the plans
for statistical analysis of the study, including plans for
interim analyses, if any, and criteria for prematurely
terminating the study as a whole if necessary"
(Appendix A, CIOMS, 2002). To be consistent with this
recommendation, the research team will need to develop
and implement an approach for monitoring the scientific
and ethical issues during the study, so that changes can
be made to the study, or the study can be stopped if
necessary. Criteria and standards need to be established
against which study activities and results can be
evaluated, and these criteria and standards need to be
incorporated into the study design, the human subjects
research protocol, and the QAPP.
In developing an approach to monitor scientific and
ethical issues during the study, the research team may
choose to
• identify the individual, team, advisory committee, or
data safety monitoring board (DSMB) responsible for
monitoring the progress and results of the study;
• develop roles and responsibilities;
• develop a schedule and timeline for the activities to be
conducted;
• develop goals for interim data analysis and prepare an
analysis plan;
• identify what data will be analyzed, how it will be
processed and validated, and who will perform the
analyses;
• develop a plan for reporting interim results to the
research team;
• develop standards for reporting scientific and ethical
issues to the research team; or
• develop criteria for evaluating scientific and ethical
issues that arise during the study.
In a well-designed observational study for which the
research team has adequately prepared, it is unlikely that
there will be scientific issues requiring mat the study be
stopped. Nonetheless, it is important for criteria to be
established for when the study needs to be changed or
terminated. An example might be the participant
retention rate. In a study with repeated measurements, a
certain sample size is required to obtain statistically
significant results. If the retention rate is poor, and too
many participants drop out of the study, it may not be
possible to meet the study objectives, and early
termination of the study may be warranted (see Text
Box 2-7 for issues that warrant early termination).
However, it is anticipated that the study design would
include contingency planning (for example, related to
replacement).
Developing criteria for study elements that may have
associated ethical concerns as a study progresses will no
doubt be more difficult than reviewing the measurement
data. Nonetheless, assuring the ethical safety of the
participants is critically important. There are no standard
formulas for dealing with ethical concerns. For example,
if the privacy of a number of study participants is
compromised by a technician conducting the
measurements in their homes, what criteria should be
used to evaluate the severity of the issues? How many
landlord-participant problems are too many before the
study needs to be changed to exclude participants who
rent their dwellings? Despite the challenges, the team of
people involved in the data and safety monitoring and
oversight effort should work diligently, with input from
the community, to establish open and continuing
channels of communication with participants, the
community, and stakeholders - with the goal of ensuring
that their involvement in the research effort is, and
31
-------
remains to be, based on the three pillars of infonned
consent: (1) that the parties are full}' infonned of all of
the relevant and useful information, (2) that the
information is understood by the parties, and (3) that all
parties continue to participate voluntarily. With diligent
effort to continue an open dialogue with the involved
parties - combined with thoughtful review and oversight
of the technical study results and procedures - the
monitoring and oversight team can ensure that
participants are free to withdraw from the research at any
time; are fully informed about the technical study
results; understand any new information about relevant
risks and benefits; and that the privacy and
confidentiality of the subjects is properly protected.
Through these efforts, the monitoring and oversight team
will ensure that the welfare of the research subjects
remains a focus of the study effort.
Text Box 2-7. Examples of Issues That May Cause a
Study To Be Stopped Early
> Participant recruiting and enrollment—low response rates,
disproportionate enrollment of select groups, problems
associated with advertising, inadequate selection criteria
> Informed consent—difficulties with the process and materials,
poor comprehension
> Participation—poor response to questionnaires, poor
compliance with researcher requests in data collection
activities
• Burden—higher than predicted
• Changes in participant behaviors—potential changes because
of participation in the study
> Grievances—participant issues
• Retention—high dropout rates
> Community issues—poor interactions, lack of support
> Third-party issues—problems with landlords, spouses, or
others
• Collateral observations—identification of nonstudy hazards,
difficulty reporting
> Unanticipated results—high contaminant concentrations
measured, unexpected results
> New data indicating that participation in the study (or
observations measured in the study) represent a risk to
participants or others
References
Castelloe, JM (2000). Sample size computations and power
analysis with the SAS system. Paper 265-25 in Proceedings
of the 25th Annual SAS Users Group International
Conference. Gary, NC: SAS Institute, Inc. Available:
http://support.sas.com/rnd7app/papers/powersamplesize.pdf
[accessed 16 April 2008].
CFR (Code of Federal Regulations) (2006). 40 CFR Chapter I
Environmental Protection Agency Part 26 Protection of
Human Subjects. U.S. Code of Federal Regulations.
Available: http://www.access.gpo.gov/nara/cfr/waisidx 067
40cfr26_06.html [accessed 12 June 2007].
CIOMS (The Council for International Organizations of
Medical Sciences) (2002). International Ethical Guidelines
for Biornedical Research Involving Human Subjects.
Geneva, Switzerland: CIOMS. Available at
http://www.cioms.ch/frame_guidelines_nov_2002.htm
[accessed 12 June 2007].
Coggon D, Rose G, Barker DJP (1997). Epidemiology for the
Uninitiated, Fourth Edition. London: BMJ Publishing
Group Ltd. Available at
http://www.bmi.com/epidein/epid.htinl [accessed 29
January 2008].
Cohen Hubal EA, Sheldon LS, Zufall, MJ, Burke JM, Thomas
KW (2000). The challenge of assessing children's
residential exposure to pesticides. Journal of Exposure
Analysis and Environmental Epidemiology 10(6 part 2):
638-649.
Dallal, GE (2001). Some aspects of study design. In The Little
Handbook of Statistical Practice at
www.StatisticalPractice.com, and at
http://www.tufts.edu/~gdallal/STUDY.HTM [accessed 29
January 2008]
Euianuel EJ, WendlerD, Grady C (2000). What Makes
Clinical Research Ethical? .Z4M4 283(20): 2701-2711.
Gilbert, SG (2006). Supplementing the traditional institutional
review board with an environmental health and community
review board. Environmental Health Perspectives 114(10):
1626-1629.
Kraemer HC, Thiemann S (1987). How Many Subjects?
Statistical Power Analysis in Research. Newbury Park, CA:
Sage Publications.
Lenth RV (2001). Some practical guidelines for effective
sample size determination. American Statistician 55(3):
187-193. Available:
http://www.stat.uiowa.edu/teclirep/tr303.pdf [accessed 13
June 2007].
NRC (National Research Council) (2003). Protecting
Participants and Facilitating Social and Behavioral
Sciences Research. Washington, DC: The National
Academies Press. Available:
http://books.nap.edu/catalog.php7record id=10638
[accessed 12 June 2007].
NRC & IOM (National Research Council and Institute of
Medicine) (2005). Ethical Considerations for Research on
Housing-Related Health Hazards Involving Children.
Washington, DC: The National Academies Press. Available:
32
-------
http://books.nap.edu/catalog.php7record id=l 1450
[accessed 12 June 2007].
OHRP (Office for Human Research Protections) (2007). U.S.
Department of Health and Human Services [Online].
Available: http://www.Mis.gov/ohrp/ [accessed 13 June
2007].
Resnik DB, Wing S (2007). Lessons learned from the
Children's Environmental Exposure Research Study. Am J
Public Health 97(3): 414-8.
U.S. DHEW (U.S. Department of Health, Education, and
Welfare) (1979). The Belmont Report: Ethical Principles
and Guidelines for the Protection of Human Subjects of
Research. Washington, D.C.: National Commission for the
Protection of Human Subjects of Biomedical and
Behavioral Research. Available:
http://ohsr.od.nih.gov/guidelines/belmont.html [accessed
June 2007].
U.S. EPA (U.S. Environmental Protection Agency) (2000).
Comments on the use of data from the testing of human
subjects: A report by the Science Advisory Board and the
FIFRA Scientific Advisory Panel. Environmental Protection
Agency, Science Advisory Board. EPA-SAB-EC-00-017.
Available:
http://www.epa.gOv/scipoly/sap/meetings/1999/november/e
c0017.pdf [accessed 16 April 2008].
U.S. HHS (U.S. Department of Health and Human Services)
(1993). Protecting Human Research Subjects: Institutional
Review Board Guidebook. Office for Human Research
Protections. Available:
http://www.lihs.gov/ohrp/irb/irb guidebook.htm [accessed
12 June 2007].
Van Belle G, Fisher L (2004). Biostatistics: A Methodology
for the Health Sciences (2nd Edition) Hoboken, NJ: John
Wiley & Sons.
Wackerly DD, Mendenhall W, Scheaffer RL (2001).
Mathematical Statistics with Applications (6th Edition).
Pacific Grove, CA: Duxbury Press.
33
-------
-------
SECTION 3
Ensuring Protection of Vulnerable Groups
Concern for the protection of vulnerable groups is
fundamental to modern ethical thought and guidelines.
The Belmont Report was "meant to provide broad
principles that could be used to generate specific rules
and regulations in response to [U.S.] research scandals
such as Tuskegee and Willowbrook.11 It focuses on
informed consent, favorable risk-benefit ratio, and the
need to ensure that vulnerable populations are not
targeted for risky research" [emphasis added] (Emanuel
et al, 2000).
The Common Rule requires IRBs to assure that
"additional safeguards have been included in the study to
protect the rights and welfare of these [vulnerable]
subjects" [at 40 CFR 26.111(b) in CFR, 2006a]. If an
observational human exposure study includes vulnerable
research participants, it is essential that the investigators
be cognizant of the special issues and requirements of
research involving vulnerable populations. Researchers
have to justify the involvement of vulnerable
populations in the research study and include the
appropriate safeguards for protection of their safety and
welfare. The Common Rule protections are discussed
further in the IRB guidebook (U.S. HHS, 1993). EPA
regulations include not only the general protections for
vulnerable populations found in the Common Rule
(Subpart A) but also define additional protections for
children and for pregnant or nursing women (and their
fetus or nursing child) in Subparts B, C, and D (CFR,
2006a).
The section begins by identifying or defining
vulnerable groups and then discusses ethical issues that
For more information about these and other research scandals, see Ethical
and Policy Issues in Research Involving Human Participants, Vol. I, Report
and Recommendations of the National Bioethics Advisory Commission,
Bethesda, MD, August, 2001. See p. 153 for information about the
Willowbrook State School experiments. The report is available at
www.bioethics.gov/reports/past commissions/nbac human part.pdf
(Accessed September 3, 2007).
may be important in conducting observational human
exposure studies involving those groups, especially
children and pregnant women. The discussions about the
ethical issues are based largely on EPA's human subjects
regulations and on the recommendations from the
Council for International Organizations of Medical
Sciences document, International Ethical Guidelines for
Biomedical Research Involving Human Subjects
(CIOMS, 2002).
3.1 Identification of Vulnerable Groups
In the United States, human subjects regulations
(45 CFR 46 and 40 CFR 26) do not formally define
vulnerable populations. Instead, the Common Rule gives
examples of potentially vulnerable groups (see Text Box
3-1). In addition, HHS extends added human subjects
protections to pregnant women, human fetuses,
neonates, prisoners, and children as vulnerable groups
(45 CFR 46, Subparts B, C, and D, see CFR, 2006b).
Analogous but somewhat more stringent protections for
children, pregnant or nursing women, and fetuses are
specified in Subparts B, C, and D of the EPA Rule (40
CFR 26). The regulations do not preclude other groups
from being considered vulnerable, however, and the
National Institutes of Health (NIH), in its Human
Participant Protections Education for Research Teams
online tutorial (NIH, 2002), lists students or employees
and terminally ill or comatose patients as potentially
vulnerable groups.
CIOMS defines vulnerable persons as those who are
relatively (or absolutely) incapable of protecting their
own interests. Vulnerability here refers to a substantial
incapacity to protect one's own interests owing to such
impediments as lack of capability to give informed
consent, lack of alternative means of obtaining medical
care or other expensive necessities, or being a junior or
subordinate member of a hierarchical group. Vulnerable
persons may have insufficient power, intelligence,
resources, strength, or needed attributes to protect their
35
-------
own interests (CIOMS, 2002) (see Text Box 3-2).
Because of their incapacity to protect their own interests,
ethically perceptive researchers will plan and implement
special provisions for the protection of the rights and
welfare of the vulnerable persons.
Text Box 3-1. Potentially Vulnerable Groups
Identified in U.S. Regulations
Common Rule:
Examples of vulnerable
groups
(40 CFR 26)
EPA extends stringent
protections to these
groups
(40 CFR 26)
HHS extends additional
protections to these
groups
(45 CFR 46)
Additional vulnerable
groups in NIH training
materials
• Children
• Pregnant women (and their
fetuses)
• Nursing women (and their
neonates)
• Prisoners
• Handicapped persons
• Mentally disabled persons
• Economically disadvantaged
persons
• Educationally disadvantaged
persons
• Children
• Pregnant women (and their
fetuses)
• Nursing women (and their
neonates
• Children
• Pregnant women and fetuses
• Nursing women and neonates
• Prisoners
• The terminally ill
• Students and employees
• Comatose patients
Although Federal regulations define vulnerability in
terms of the person's ability to protect their own
interests or particular sensitivity to risks because of
physical condition, the lay public may perceive a broader
definition of vulnerability as it relates to education,
economics, social status, and other factors. As shown in
Text Boxes 3-1 and 3-2 and described above, the
concept of vulnerability is broader than that presented in
the Common Rule. It is not adequate to simply check the
list in Text Box 3-1 to identify if a potentially vulnerable
group is included in an observational study. The
researchers should assess the potential vulnerability of a
study population within the study by evaluating the
characteristics (e.g., socioeconomic status) of the study
population within the context of the study by considering
the various design elements of the study, as discussed in
Section 2.
Text Box 3-2. Potentially Vulnerable Groups
Identified in International Guidance
(Council for International Organizations of
Medical Sciences, 2002)
Junior or subordinate members of a hierarchical group;
examples include employees, students, members of the armed
forces, police, and others who work for, or closely with re-
searchers; they may have expectations of preferential
treatment if they agree to participate or fear of disapproval or
retaliation if they refuse to participate in a study.
Elderly persons, who may acquire attributes that define them
as vulnerable with advancing age.
Residents of nursing homes.
People receiving welfare benefits or social assistance.
People with low or no incomes (poor and unemployed).
Homeless persons.
Nomads.
Refugees or displaced persons.
Some ethnic and racial minority groups.
People with incurable diseases (in clinical studies).
The politically powerless.
Members of communities unfamiliar with modern medical
concepts (applies to clinical studies)
3.2 Justification for Involving Vulnerable
Persons in Observational Research
The Common Rule requires IRBs to ensure that the
selection of subjects is equitable [40 CFR 26.11 l(a)(3)]
and instructs the IRB to consider the "purposes of the
research and the setting in which the research will be
conducted." CIOMS goes further and recommends that
"Special justification is required for inviting vulnerable
individuals to serve as research subjects" (CIOMS,
2002).12
In the commentary on Guideline 13 in CIOMS (2002). the committee states
that the central problem presented by plans to involve vulnerable persons as
research subjects is that such plans may entail an inequitable distribution of
the burdens and benefits of research participation. Classes of individuals
conventionally considered vulnerable are those with limited capacity or
freedom to consent or lo decline to consent. They are the subject of specific
guidelines in the CIOMS document (Guidelines 14 and 15) and include
children, and persons who, because of mental or behavioral disorders, are
incapable of giving informed consent. Ethical justification of their
involvement usually requires that
• the research could not be carried out equally well with less vulnerable
subjects;
• the research is intended to obtain knowledge that will lead to improved
diagnosis, prevention, or treatment of diseases or other health problems
characteristic of or unique to, the vulnerable class—either the actual
subjects or other similarly situated members of the vulnerable class;
• research subjects and other members of the vulnerable class from which
subjects are recruited will ordinarily be assured reasonable access to any
diagnostic, preventive, or therapeutic products that will become available
as a consequence of the research;
• the risks attached to interventions or procedures that do not hold out the
prospect of direct health-related benefit will not exceed those associated
36
-------
CIOMS recommendations, although written to
address biomedical research, also generally are
applicable to observational human exposure studies. The
authors of this document consider the CIOMS
requirement that the research could not be carried out
equally well with less vulnerable subjects to be
particularly important. EPA and NERL researchers
should include vulnerable groups in observational
human exposure studies only if their participation is
critical to the success and applicability of the research.
Even then, EPA and NERL researchers will have to meet
stringent standards for protecting the rights and safety of
the vulnerable participants. For example, EPA
regulations governing observational research with
children are even more stringent than the CIOMS
guideline. If such research does not hold out the prospect
of direct benefit to the child, no increase whatsoever
over minimal risk is permitted.
However, many observational human exposure
studies are developed specifically to study the exposures
of selected vulnerable groups to chemicals and other
environmental stressors in everyday environments. So,
researchers should be prepared to address the issues
associated with vulnerable groups in observational
research. Furthermore, as discussed in Section 3.4, there
has been increased concern in recent years that exclusion
of vulnerable groups from research studies is not ethical.
Failure to conduct research with vulnerable groups may
deprive them of the benefits of research. NIH, for
example, has a policy (NIH, 1998) with a goal of
increasing participation of children in research.
3.3 Minimal Risk and Vulnerable Groups
EPA has codified protections for children, pregnant
or nursing women, and fetuses in Subparts B, C, and D
of the EPA human subjects rule (40 CFR 26). Subpart B
strictly prohibits research involving intentional exposure
of children or pregnant or nursing women (and,
therefore, exposure of her fetus).
EPA's regulations do allow for observational
research involving fetuses and pregnant women (40 CFR
26 Subpart C) or children (40 CFR 26 Subpart D) but
with additional protections in place and with strict
limitations on research that presents more than minimal
risk (CFR, 2006a).13 When considering vulnerable
groups, The Institutional Review Board Guidebook (U.S.
HHS, 1993) states that "IRBs should therefore determine
whether the proposed subject population would be more
sensitive or vulnerable to the risks posed by the research
as a result of their general condition or disabilities. If so,
the procedures would constitute more than minimal risk
for those subjects:'
When conducting observational human exposure
studies, it is recommended that researchers consult these
regulations and guidebooks. NERL researchers also will
need to ensure that all of the requirements in Subparts B,
C, and D of the EPA Human Subjects Rule are met.
3.4 Research Involving Children
Children long have been recognized as a vulnerable
group in research studies. EPA and HHS both extend
special protections to children (CFR, 2006a,b). There are
many books, reports, and research manuscripts that
specifically address issues associated with research
involving children (e.g., NRC & IOM. 2005; IOM,
2004; Kodish, 2005; NRC, 2003; AAP, 2003).
CIOMS has drafted guidelines for including children
in biomedical research (Guideline 14, CIOMS, 2002).
The guidelines require an investigator to provide the
assurances shown in Text Box 3-3 before undertaking
research involving children.
Text Box 3-3. Assurances Required by CIOMS Before
Research Involving Children May Begin
• the research might not equally well be carried out with adults;
• the purpose of the research is to obtain knowledge relevant to
the health needs of children;
• a parent or legal representative of each child has given
permission;
• the agreement (assent) of each child has been obtained to the
extent of the child's capabilities; and
• a child's refusal to participate or continue in the research will be
respected.
The participation of children in some observational
human exposure studies is critical to characterizing
children's exposures to chemicals in the environment. It
is well recognized that children are not "little adults,"
and that their exposures to chemicals differ (and, in
with routine medical or psychological examination of such persons unless
an ethical review committee authorizes a slight increase over this level of
risk (Guideline 9); and,
when the prospective subjects are either incompetent or otherwise
substantially unable to give informed consent, their agreement will be
supplemented by the permission of their legal guardians or other
appropriate representatives.
Minimal risk is defined at 40 CFR 26.102(i) and again at 40 CFR
26.402(g). It "means that the probability and magnitude of harm or discomfort
anticipated in the research are not greater in and of themselves than those
ordinarily encountered in daily life or during the performance of routine
physical or psychological examinations or tests."
37
-------
some cases, are higher) from those of adults. Children
are behaviorally and physiologically different from
adults. Their interaction with their environment, through
activities such as playing on floors, mouthing of hands
and objects, and handling of food, may increase contact
with contaminated surfaces. Children have
proportionately higher breathing rates, relative surface
area, and food intake requirements that also may
increase exposure. Differences in absorption,
metabolism, storage, and excretion may result in higher
biologically effective doses to target tissues. Immature
organ systems may be more susceptible to toxicological
challenges. Windows of vulnerability, when specific
toxicants may permanently alter the function of an organ
system, are thought to exist at various stages of
development. Because the factors influencing children's
exposures to chemicals are not characterized well
(Cohen Hubal et al., 2000), it is sometimes important
that observational human exposure studies involve
children.
Because children are so vulnerable, there long has
been concern about including them in research studies,
and biomedical research often excluded children.
However, in recent years, there has been concern that
excluding children from research is not ethical. NIH's
Policy and Guidelines on the Inclusion of Children as
Participants in Research Involving Human Subjects
(NIH, 1998) has a goal of increasing participation of
children in research. The policy of NIH is that children
must be included in all human subjects research
conducted or supported by NIH, unless there are
scientific and ethical reasons not to include them.
Proposals or applications to NIH for research have to
present an acceptable justification if children will be
excluded from a research study. Of course, as discussed
above, if the research topic is irrelevant to children, the
CIOMS guidelines would recommend that they be
excluded from the research.
Observational human exposure studies conducted by
NERL are not expected to involve greater than minimal
risk. It will be the responsibility of the NERL
researchers to present adequate information for the IRB
to demonstrate that the research does not involve greater
than minimal risk. Researchers designing observational
research studies should carefully evaluate the risks and
benefits specific to their study and the participants
involved. In developing the study design and human
subjects protocols, researchers need to ensure that the
protocols ensure the protection of the rights and welfare
of the participant children, and mat risks and harm are
minimized. The perception of risks and benefits, both by
the individual and by the family or community, may
influence the risk-benefit determination. It may prove
useful for the research team to consult with other
experienced researchers who have conducted similar
studies and with members of the IRB to ensure that the
information included in the human subjects research
protocol is adequate for the IRB's review.
It is recommended that researchers consider all of
the potential issues associated with involvement of
children in their studies in developing the study design
and research protocols, including the role of the family.
EPA's human subjects rule for observational research
not involving greater than minimal risk to children (40
CFR 26.404) (i.e., the kinds of observational human
exposure studies that NERL exposure research is likely
to entail) focuses on obtaining assent of the children and
permission of their parents or guardians. But the role of
the family goes far beyond their involvement in the
informed consent process. In observational human
exposure studies, even when children are the
participants, the parents or guardian play a key role in
the collection of data and information during the study.
For studies with very young children, family members
supply all of the information relevant to the child. NERL
researchers need to ensure that both the child and the
parents or guardians and other caregivers are informed
fully and are willing participants. Without their willing
participation, the research cannot be successful.
3.5 Women as Research Subjects
Women are routinely included as research
participants in observational human exposure studies.
However, pregnant women and their fetuses are
vulnerable groups and require special protections. EPA's
human subjects rule prohibits intentional dosing studies
and provides additional controls for observational
research (40 CFR 26, Subparts B and C).
CIOMS (2002) includes two guidelines for
biomedical research involving women as research
subjects. The first of these, number 16, states that
women should not be excluded from biomedical
research because of the potential for becoming pregnant
during a study. The document continues, "A general
policy of excluding from such clinical trials women
biologically capable of becoming pregnant is unjust in
mat it deprives women as a class of persons of the
benefits of new knowledge derived from the trials." The
second relevant CIOMS guideline, number 17, asserts
mat, if involved in a research study, pregnant women
should be fully informed, and included only if the
38
-------
research benefits pregnant women and is thoroughly
supported by reliable evidence in animal studies.
Although the CIOMS guideline specifically
addresses biomedical research, the ethical concepts
behind the guidelines generally may be applicable to
observational human exposure studies. EPA's human
subjects rule is completely consistent with the HHS rule
in adding additional protections for pregnant women and
fetuses involved in observational research (40 CFR
26.304 and 45 CFR 46.204). These additional
protections (specified in 45 CFR 46.204 subparagraphs a
through j) reflect the CIOMS recommendations by-
requiring: availability of data from previously conducted
studies to assess the risk to pregnant women and fetuses;
scientific necessity for inclusion of pregnant women and
fetuses (i.e., providing benefit to the woman or fetus, or
producing important, but otherwise unobtainable,
biomedical knowledge); that risk is reduced to the least
possible level for achieving the objectives of the
research; and other protections.
3.6 Other Potentially Vulnerable Groups
FIHS specifies additional protections for prisoners as
a potentially vulnerable group in Subpart C of 45 CFR
26. Additional requirements for other vulnerable groups
in research studies are not specifically defined in either
EPA's or FIHS' human subjects rules. Nonetheless, other
groups (as discussed in Section 3.1) may be considered
to be vulnerable and, as such, may warrant additional
consideration and protection as required in the Common
Rule. For these other potentially vulnerable groups, such
as employees, students, handicapped persons, mentally
disabled persons, and economically or educationally
disadvantaged persons, nursing home residents or
otherwise incapacitated elderly, etc., the Common Rule
requires researchers and IRBs to fully evaluate the
protocols to ensure that the safety and welfare of the
groups will be protected. As discussed in Section 3.1, It
also should be noted that, although Federal regulations
define vulnerability in terms of the ability to protect
one's own interests, the lay public may perceive a
broader definition of vulnerability as it relates to
education, economics, social status, and other factors.
The researcher should evaluate vulnerability in this
broader context to ensure that adequate safeguards are
included for potentially vulnerable populations that do
not meet the definition of the Federal regulations.
References
AAP (American Academy of Pediatrics) (2003). Pediatric
Environmental Health. (2nd Edition). Elk Grove Village, IL:
American Academy of Pediatrics.
CFR (Code of Federal Regulations) (2006a). 40 CFR Chapter
I Environmental Protection Agency Part 26 Protection of
Human Subjects. U.S. Code of Federal Regulations.
Available:
http://www.access.gpo.gov/nara/cfr/waisidx 06/40cfr26 06.
html [accessed 12 June 2007].
CFR (Code of Federal Regulations) (2006b). 45 CFR Subtitle
A Department of Health and Human Services Part 46
Protection of Human Subjects. U.S. Code of Federal
Regulations. Available:
http://www.access.gpo.gov/nara/cfr/waisidx_06/45cfr46_06.ht
ml [accessed 13 June 2007].
CIOMS (The Council for International Organizations of
Medical Sciences) (2002). International Ethical Guidelines
for Biomedical Research Involving Human Subjects.
Geneva, Switzerland: CIOMS. Available at
http://www.cioms.ch/frame guidelines nov 2002.htm
[accessed 12 June 2007].
Cohen Hubal EA, Sheldon LS, Burke JM, McCurdy TR, Berry
MR, Rigas ML, Zartarian VG, Freeman NCG (2000).
Children's exposure assessment: a review of factors
influencing children's exposure, and the data available to
characterize and assess that exposure. Environ Health
Perspect 108(6): 475-486.
Emanuel EJ, WendlerD, Grady C (2000). What makes
clinical research ethical? JAMA 283(20): 2701-2711.
IOM (Institute of Medicine) (2004). Ethical Conduct of
Clinical Research Involving Children. Washington, DC:
The National Academies Press. Available:
http://www.nap.edu/catalog.php?record_id=10958 [accessed
16 April 2008].
Kodish E (Ed.) (2005). Ethics and Research with Children: A
Case-Based Approach. New York: Oxford University Press.
NBAC (National Bioethics Advisory Commission) (2001).
Ethical and Policy Issues in Research Involving Human
Participants, Volume I: Report and Recommendations of the
National Bioethics Advisory Commission. Bethesda, MD:
National Bioethics Advisory Commission. Available:
http://www.bioethics.gov/reports/past commissions/nbach
uman_part.pdf [accessed 3 September 2007].
39
-------
NIH (National Institutes of Health) (2002) Human Participant
Protections Education for Research Teams [online tutorial].
NIH (National Institutes of Health) (1998). NIH Policy and
Guidelines on The Inclusion of Children as Participants in
Research Involving Human Subjects. Available:
http://grants.nih.gov/grants/guide/notice-files/not98-
024.html [accessed 13 June 2007].
NRC (National Research Council) (2003). Protecting
Participants and Facilitating Social and Behavioral
Sciences Research. Washington, DC: The National
Academies Press. Available:
http://books.nap.edu/catalog.php7record id=10638
[accessed 12 June 2007].
NRC & IOM (National Research Council and Institute of
Medicine) (2005). Ethical Considerations for Research on
Housing-Related Health Hazards Involving Children.
Washington, DC: The National Academies Press. Available:
http://books.nap.edu/catalog.php7record id=l 1450
[accessed 12 June 2007].
U.S. HHS (U.S. Department of Health and Human Services)
(1993). Protecting Human Research Subjects: Institutional
Review Board Guidebook. Office for Human Research
Protections. Available:
http://www.lihs.gov/ohrp/irb/irb guidebook.htm [accessed
12 June 2007].
40
-------
SECTION 4
Privacy, Confidentiality, and Other Concerns Related
to Observational Human Exposure Studies
Observational human exposure studies are designed
to describe people's contact with pollutants as they go
about their everyday lives. Of necessity then, these
studies take place in the locations that participants often
consider to be personal and private. Clinical research
studies generally are conducted in a research facility, a
clinic, a hospital, or some other institutional or medical
setting. Survey research may be conducted by mail, over
the phone, or in another '"neutral"' setting. But,
observational human exposure studies are conducted in
the participants' "personal" environment—their home,
daycare center, school, vehicle, workplace, or other
environments that people occupy during their routine
daily activities. This difference in the research setting
means that researchers involved in observational human
exposure studies have an even greater challenge in
meeting the ethical obligation to respect the privacy of
the participants.
When exposure science researchers like those at
NERL enter a home to carry out their studies, the
"expectations and constraints may be strikingly different
man when research is carried out in a medical setting"
(p. 64, NRC & IOM, 2005). The legal precept of
freedom from unreasonable search and seizure and the
historic and deeply rooted principle mat "a man's home
is his castle" contribute to a belief in the "sanctity of the
home" (see the discussion on pp. 62-66, NRC & IOM.
2005).
The joint NRC and IOM report Ethical
Considerations for Research on Housing-Related Health
Hazards Involving Children discusses the ethical issues
associated with entering a participant's home to conduct
research and explores the researchers' responsibilities
that derive from conducting research in people's homes
(NRC & IOM, 2005). These housing-related discussions
are particularly relevant to observational human
exposure studies, which often include environmental and
biological measurements in people's homes or personal
locations. Many of the topics identified in that report are
discussed in this section (Text Box 4-1).
Text Box 4-1. Topics in Section 4
Privacy Issues
Confidentiality
Confidentiality of Information
Confidentiality of Participation
Collateral Observations
Potential Nonstudy Hazards in the Residence
Collateral Observations with Mandated Reporting
Requirements
Hazard Communication
Planning and Staff Training
Third-Party Issues
Determining Whether a Third Party is a Human Subject
Informing Third Parties of Research Activities
Research Results and Third Parties
Data and Safety Monitoring and Oversight
4.1 Privacy Issues
Privacy refers to an expectation that a person is free
from intrusion into personal matters and is free from the
presence or view of others. The Institutional Review
Board Guidebook defines privacy as "control over the
extent, timing, and circumstances of sharing oneself
(physically, behaviorally, or intellectually) with others"
(U.S. HHS, 1993). Beauchamp and Childress find that
the right to privacy is based on the principle of respect
for autonomy. "We often respect persons by respecting
their autonomous wishes not to be observed, touched, or
intruded upon.... A loss of privacy occurs if others use
any of several forms of access, including intervening in
zones of secrecy, anonymity, seclusion, or solitude" (pp.
295-296, Beauchamp and Childress, 2001).
41
-------
Although research participants may agree to allow
researchers to enter their home or other zone of personal
space to conduct their research measurements, they have
not abrogated their right to privacy. "When individuals
voluntarily grant others some form of access to
themselves, their act is an exercise of the right to
privacy, not a waiver of that right" (p. 297, Beauchamp
and Childress, 2001). Researchers should remember that
they are guests in the homes for a specific purpose.
"When people visit a home, there are social expectations
about what is acceptable behavior. People who are
invited into a home are expected to be sensitive to and
respectful of the host's customs and values"' (p. 65, NRC
& IOM, 2005).
By their very nature, observational human exposure
studies encroach on the privacy of a research participant.
Entry in a participant's home (or other personal zones)
does represent a loss of privacy, but researchers should
be careful to ensure that their presence does not become
a violation of the individual's right to privacy. The
relationship between the researcher and the participant
may be complicated, and there may be conflicts between
the researcher's role and their ethical obligations (NRC
& IOM, 2005). In entering a participant's personal
space, it may be difficult, or impossible, to avoid making
observations unrelated to the research question, thereby
further intruding on the participant's personal privacy.
Indeed, there may be ethical and legal obligations for the
researchers to respond to those observations. Beauchamp
and Childress suggest that "policies carefully specify the
conditions of access that will and will not count as a loss
of privacy or a violation of the right to privacy. The
policy should accurately define the zones mat are
considered private and not to be invaded, and should
also identify interests that legitimately may be balanced
against privacy interests" (Beauchamp and Childress,
2001).
Observational human exposure studies also may
infringe on the privacy of other individuals, for example,
other members of the participant's family or household.
Researchers should strive to minimize the intrusion and
loss of privacy and to show respect for the privacy of
study participants and third parties at all times (see Text
Box 4-2 for a list of relevant privacy issues). It is
incumbent on the researcher to recognize privacy issues
in the design and implementation of the research study.
The NRC & IOM report suggests that researchers
anticipate the ethical issues that arise from conducting
research in a person's home, and that they take steps to
correct them (1) by thinking through the issues as part of
the study design; (2) by discussing the issues during the
informed consent process; and (3) by ensuring that the
frontline staff that enter a participant's home
"understand their role as members of the research team,
how that role differs from the role of neighbor or friend,
and how they should respond when they make
observations that are not part of the protocol" (p. 66,
NRC & IOM, 2005).
Text Box 4-2. Privacy Issues
Researchers should develop an anticipatory plan for how to
deal with privacy issues during the study. The plan should
include a list of potential observations that could be of concern
and a plan for how they will be handled.
The plan needs to address both the legal and ethical
obligations of the researcher in response to situations where
privacy is compromised.
Privacy issues will vary depending on the culture of the
population being studied. What one individual or group may
find as an invasion of privacy, another group may not have a
concern about.
Privacy issues involve individual participants and may extend
to third parties, including the community.
Researchers may find a meeting with community
representatives to learn about the community residents and
potential privacy issues to be helpful. Community
representatives can help the researcher identify potential
privacy issues and offer advice on how to address them.
Researchers may wish to respect the privacy of occupants
sharing the study participant's household or other study
locations by providing advance notification of study visits and
by giving them the opportunity not to be present during those
visits.
Field staff should be trained on how to minimize breaches of
privacy and how to handle privacy issues.
The informed consent process and form should address how
the researcher will handle privacy issues such as collateral
observations of household hazards
4.2 Confidentiality
Confidentiality and privacy are not the same thing.
Confidentiality refers to limits on the dissemination of
information disclosed by a person in a special
professional relationship, such as the doctor-patient
relationship or the participant-researcher relationship
(Beauchamp and Childress, 2001). The Institutional
Review Board Guidebook defines confidentiality as
"pertains to the treatment of information that an
individual has disclosed in a relationship of trust and
with the expectation that it will not be divulged to others
without permission in ways that are inconsistent with the
understanding of the original disclosure" (U.S. HHS,
1993). Emanuel et al. (2000) state that one way to
respect the privacy of the participants is "by managing
the information in accordance with confidentiality
42
-------
rules/' Confidentiality in research also may extend to
limiting dissemination of the knowledge that an
individual is participating in a research study.
As part of the research planning process, researchers
are responsible for developing procedures to protect
confidentiality and to define limits on the researcher's
ability to provide or protect confidentiality. Explaining
plans or procedures for protecting confidentiality and
their limits14 to prospective research participants is an
integral part of the informed consent process.
4.2.7 Confidentiality of Information
Many types of information may be collected in
observational human exposure studies. Information may
be collected through questionnaires, staff observations of
residential or other environments, diaries, personal
sample collection, environmental or residential sample
collection, and collection of biological specimens.
Measurement data from the collected samples become
part of the information for a participant. The specific
information to be obtained to address the research
questions should be determined in the development of
the study design and research protocol.
Disclosure of information that can be linked to an
individual may cause harm or distress to that individual.
Researchers are responsible for developing safeguards to
protect the confidentiality of information and physical
samples collected from research participants (see, for
example, Guideline 18, CIOMS, 2002). 15
Researchers also should be aware that certain
combinations of information from a study may
sometimes lead to the indirect identification of the
individual. Certain combinations of demographic
information, for example, may make it relatively simple
Beauchamp and Childress (2001) discuss when one may be ethically
justified in infringing on an individual's privacy and confidentiality—for
example, because of risks to others evidenced by biomarkers of infectious
disease. They also discuss similar ethical issues that may arise in regard to
genetic data. CIOMS (2002) Guideline 18 provides suggestions for
safeguarding or disclosing genetic information. If exposure scientists
collaborate with medical researchers or epidemiologists and obtain such
information, they need to be cognizant of the relevant ethical issues and of the
CIOMS guidelines.
Guideline 18 states. "The investigator must establish secure safeguards of
the confidentiality of subjects" research data. Subjects should be told the
limits, legal or other, to the investigators' ability to safeguard confidentiality
and the possible consequences of breaches of confidentiality." Additional
CIOMS commentary on the confidentiality guideline states. "Prospective
subjects should be informed of limits to the ability of researchers to ensure
strict confidentiality and of the foreseeable adverse social consequences of
breaches of confidentiality. Some jurisdictions require the reporting to
appropriate agencies of, for instance, certain communicable diseases or
evidence of child abuse or neglect. . . . These and similar limits to the ability
to maintain confidentiality should be anticipated and disclosed to prospective
subjects."
to identify an individual. Precise geographic location
information may be sufficient to pinpoint a residence.
Researchers may use several strategies to reduce the
likelihood of indirect identification when study results
are reported (see, also, Text Box 4-3).
• Redact from publications, reports, or public data sets
information mat might be used to indirectly identify a
research participant.
• Generalize exact information; for example, replace
birth date with age or year of birth or classify age as
part of a range.
• Aggregate information across individuals; for
example, only report data in cells of sufficient size to
make individual linkages unlikely.
• Reduce the specificity of geographic coordinate
information to a level that a specific residence or other
location can not be identified.
Text Box 4-3. Approaches for Protecting
Personally Identifiable Information
Developing procedures for safeguarding information prior to
collecting the information
Ensuring that data or samples are anonymous by not
collecting or by destroying identifying information or linkages
Restricting access to identifying information to only those
requiring access
Assigning codes to participants, data, and samples rather than
using identifiers
Physically separating identifying information and linkage files
from other study information
Securing identifying information in locked files with limited
access
Restricting identifying information from computers that are
networked with other computers or electronic systems
Restricting identifying information from computers that are not
kept in secure locations with limited access
Training research staff members on human subject protection
and on information security procedures
Another step that can help protect confidentiality is
to obtain a Certificate of Confidentiality. Certificates of
Confidentiality are issued by NIH (2002) to protect
identifiable research information from forced disclosure.
They allow the investigator and others who have access
to research records to refuse to disclose identifying
information on research participants in any civil,
criminal, administrative, legislative, or other proceeding,
whether at the Federal, State, or local level. Certificates
of Confidentiality may be granted for studies collecting
information that, if disclosed, could have adverse
consequences for subjects or damage their financial
standing, employability, insurability, or reputation. By
protecting researchers and institutions from being
43
-------
compelled to disclose information that would identify
research subjects, Certificates of Confidentiality help
achieve the research objectives and promote
participation in studies by assuring confidentiality and
privacy to participants. Any research project that collects
sensitive, personally identifiable information and mat
has been approved by an 1KB is eligible for a certificate.
Identifying information is broadly defined as any item or
combination of items in the research data that could lead
directly or indirectly to the identification of a research
subject.
Federal funding is not a prerequisite for a certificate.
A Certificate of Confidentiality does not diminish,
however, the investigator's need to protect the
personally identifiable information as described above.
4.2.2 Confidentiality of Participation
In some types of research, the knowledge that a
person is participating in a particular research study
could, potentially, put the participant at risk for harm or
distress. This topic is discussed in The Institutional
Review Board Guidebook, with special emphasis on
behavioral and social research mat deals with sensitive
topics (U.S. HHS, 1993). The guidebook describes the
need for additional safeguards to protect and prevent
disclosure of the identity of participants, including the
use of Certificates of Confidentiality for sensitive
matters.
Observational human exposure studies often pose
particular challenges with regard to limiting
dissemination of the knowledge of an individual's
participation in the study. Visiting the research
participant's residence to collect samples or to make
observations will necessitate informing other family
members or occupants about the visit and study
procedures. Research participants may be asked to wear
personal monitors over time periods ranging from a day
to a week or more. Wearing these devices in public
places, schools, or workplaces may identify them as a
study participant or generate questions regarding the
activity. Field staff visits to the participant's home or
setting up outdoor sample collection devices around the
home also might disclose their participation. And, in
some cases, third parties outside of the home have to be
asked for permission or be informed that monitoring
activities are taking place.
Researchers and IRBs should consider whether
knowledge of an individual's participation by others
might create potential for harm or distress in an
observational human exposure study. Such risks might
be limited to possible discomfort in attracting unwanted
attention; this may be particularly taie for adolescents.
However, in some cases, the potential risks could be
greater, for example, in cases where participation could
provoke an adverse reaction from a landlord or
employer. Oftentimes study protocols can be structured
to minimize these potential risks. Through the informed
consent process, prospective participants should be made
aware of the limits of the researcher's ability to protect
knowledge of their participation in the study and of the
possible risks of disclosure.
4.3Collateral Observations
In the course of conducting an observational human
exposure study, research staff may observe potentially
unsafe conditions or situations that are unrelated to the
research study. Such "collateral observations" may
involve physical hazards in the study participant's
residential environment or evidence of situations, such
as child abuse, mat have to be reported to proper
authorities. In preparing for the research study, it is
recommended that researchers carefully plan for possible
collateral observations, including their identification,
staff training, and hazard communication and reporting.
This may be a major element in the data and safety
monitoring and oversight for the study. The informed
consent process should reflect procedures used to
manage collateral observations. Potential participants
should be informed of situations in which confidentiality
might be breached, such as statutory requirements for
reporting abuse or imminent harm to self or others.
4.3.1 Potential Nonstudy Hazards in the
Residence
Research staff conducting observational human
exposure studies often will spend time in and around
study participant residences. In the course of visiting a
residence or conducting study-related observations,
research staff may observe potential hazards unrelated to
the research being performed (see Text Box 4-4). Some
hazards may be associated with the potential for physical
injury, whereas others may be related to exposure to
chemical or biological agents. Some situations may be
potential hazards only for young children, whereas other
conditions may present potential hazards for all residents
or occupants.
The NRC & IOM recommend that researchers
should consider such foreseeable observations and
potential hazards in advance, develop responses to the
risks, and submit the proposed plans to the IRB for
review to ensure that they are appropriate "in the context
of the research and the affected community." The NRC
44
-------
& IOM also advise that field staff should be trained in
how to assess and respond to such risks
(Recommendations 7.3 and 7.4, NRC & IOM, 2005).
For other behaviors and risks that have not been
specifically identified in advance, procedures should be
included in the data and safety monitoring and oversight
provisions of the study design and research protocol to
address these issues. The fundamental ethical principle
of beneficence would motivate researchers who observe
serious harms to take steps to try to prevent those harms,
even for observations that are not directly related to the
study. The steps that they may take can range from
immediate action to prevent an imminent and serious
danger to statutory reporting of observations (see
Section 4.3.2) to reporting the observation to the data
and safety monitoring and oversight authority for advice
on how to respond (see Section 4.5). (The reader is also
referred to pages 59-61 and 134-144 of the NRC & IOM
[2005] report for a more thorough discussion of
researchers" responsibilities in such cases.)
Text Box 4-4. Potential Hazards that Might Be
Encountered in a Residential Environment
• Unsecured firearm
• Uncovered electrical outlets
• Unprotected stairways
• Missing child-protective cabinet latches
• Lack of window guards
• Missing or inoperable smoke alarm
• Housing code violations
• Chipping or flaking paint—potential for lead exposure in older
homes
• Malfunctioning or unvented combustion appliances—potential
for carbon monoxide exposure
• Unsecured poisons or other dangerous products
• Excessive mold growth
4.3.2 Collateral Observations with
Mandated Reporting Requirements
Some collateral observations may have statutory
requirements for reporting to designated authorities.
Examples of such observations include
• observed child or elder abuse or evidence of such
abuse or neglect,
• statements or actions of intent to harm self or others,
and
• certain communicable diseases.
Because different reporting statutes pertain in
different states, it is necessary for researchers to leani
and understand the applicable reporting requirements for
the study location. In the case of abuse, it is also
important to understand what actions or situations are
considered abusive in a particular state. Although direct
physical harm or violence might be obvious to a research
staff member, there are other conditions of neglect that
might be more difficult to recognize or to know when to
report.
Study participants should be made aware of statutory
requirements for reporting collateral observations as part
of the informed consent process. Researchers should
include the reporting requirements in the informed
consent form (discussed in Section 5) and should ensure
that the study participant fully comprehends this
information and the impact on their privacy and
confidentiality.
4.3.3 Hazard Communication
It is difficult for researchers to determine when and
how to communicate with study participants or third
parties about collateral observations related to potential
hazards. A hazard might present such an imminent threat
to health or safety that staff would need to communicate
immediately with the participant or take action to
mitigate the threat. In some cases, such as instances of
abuse with attendant statutory reporting requirements, it
may be necessary to breach confidentiality. More often,
however, a potential hazard identified as a result of
collateral observation may not be an imminent threat or
pose a potential risk that is situation dependent or is
related to third parties. A number of considerations in
hazard communication come into play regarding
confidentiality, privacy, the ability of the researcher to
provide accurate and effective information regarding the
hazard and hazard mitigation, and the ability of the study
participant or others to effectively mitigate the hazard
without unintended adverse consequences. The National
Academy of Sciences Committee on Ethical Issues in
Housing-Related Health Hazard Research Involving
Children, Youth, and Families discussed many of these
issues in depth (NRC & IOM, 2005).
Different communities, cultures, or demographic
groups can have different risk perceptions, which may
affect how collateral observations are assessed and
reported from one study location to the next. The AAP
Committee on Environmental Health has prepared
information regarding perception, identification, and
communication of environmental health risks (AAP,
2003). Researchers likely will benefit from including
community members on the research team in developing
the study design and research protocol or from
consultation with community boards regarding
identification of hazards and hazard communication.
45
-------
It is important that any advice that the researcher
might provide to study participants regarding hazard
mitigation should be carefully considered.
Considerations in recommending an action may include
whether the mitigation approach has been shown to be
effective, whether the study participant can understand
and effectively implement the action, and whether
unintended adverse consequences might result from
taking an action. In some cases, it may be reasonable to
refer the participant to another organization that can
provide expert advice or assistance.
4.3.4 Planning and Staff Training
As part of the study planning process and protocol
development, it is important that researchers be
cognizant of the kinds of collateral observations mat
might occur in the implementation of the study protocol
and to develop plans as to how such observations would
be handled. Researchers may choose to include a
systematic approach in hazard identification, such as
using a home-hazard checklist that becomes an ancillary
part of the study protocol. Alternatively, collateral
observations could be handled on a case-by-case basis.
Staff experience and training is a critical
consideration for managing collateral observations. Staff
members that visit study participant residences may not
have expertise or experience in identifying many of the
potential hazards without adequate training. All staff
involved in a study, particularly those responsible for
field data collection, should be trained on identification
and reporting of collateral observations. This training
should be study specific and likely will include, but not
be limited to
(1) identification of actions (e.g., child abuse) that have
(state-specific) statutory reporting requirements,
(2) conditions of neglect that may adversely impact
study participants or third parties,
(3) environmental hazards and situations that may be
associated with imminent harm (e.g., combustible
materials near an open flame, unsecured firearms
accessible to very young children),
(4) policies and procedures for reporting or intervention
by members of the research team, and
(5) local and state reporting requirements.
It is advisable that this training be developed in
consultation with community representatives who can
provide input on potential hazards and situations that
may be encountered in the study community, local
norms and attitudes about potential interventions and
reporting, and local agencies available to assist on these
types of issues. It is also especially important to consider
staff experience and training in hazard communication.
Consistency in communication is very important, and
researchers may decide to use materials prepared by
other organizations that have expertise regarding a
particular hazard.
Another important element of planning for field data
collection and training of research staff is on hazards and
situations that the field staff may encounter during their
field work. Although the previous discussion highlights
the need to be prepared to report potentially negligent or
illegal behaviors, these same behaviors may place the
research staff in imminent danger during the conduct of
their work in residences and communities. The research
team should develop a plan for identifying potential
situations, hazards, and dangers that may place the
research team at risk of imminent harm. This should
generally involve working with community
representatives to identify the hazards and situations mat
may be of concern. For example, a researcher's
observation of illegal drugs in a study residence may
place the researcher in potential danger. Similarly, if
studies are conducted in areas with high crime rates,
researchers need to be aware of the potential dangers and
have a plan for addressing them. In some cases,
situations may arise because of to inadvertent actions.
For example, if an area has a high rate of daytime break-
ins, the presence of researchers in the area may trigger
calls to local authorities when the researchers, who are
strangers, are observed in the neighborhood. Feedback
from the community representatives should be solicited
when developing a plan for responding to situations such
as these. Comprehensive training should be developed
for research field staff to assist them in identifying
potentially dangerous situations and in responding to
such situations.
4.4 Third-Party Issues
Third-party issues can arise in observational human
exposure studies in two ways. First, the study may
collect limited information about or related to
individuals other than the study participants. Second,
study activities may affect or involve people or
organizations other than the study participants.
Examples of activities that may involve or affect
third parties in observational human exposure studies
could include, but are not limited to the following types.
• Asking the participant about demographic,
occupational, smoking, or product use information for
other household members
• Collecting residential environmental samples in
multiperson households
46
-------
• Collecting environmental samples in common areas of
multifamily housing units
• Collecting personal or environmental samples in a day
care, school, health care, or occupational setting
• Measuring chemical occurrences or concentrations
that may be of interest or import to other household
members or to the community
• Collecting activity or dietary information about a
community
It is important for researchers and research staff to
understand whether and to what extent the research
involves or affects third parties, and how third-party
involvement might affect the study participants. Several
examples of possible third-parties are listed in Text Box
4-5. Study planning; 1KB review; and communication
before, during, and after the study should take third-
party issues into account.
Text Box 4-5. Possible Third-Parties in
Exposure Studies
Household members not enrolled in the study
Relatives
Care givers for children or elders
School staff
Employers
Other members of the community
Building managers or facility operators
Landlords
4.4.1 Determining Whether a Third Party
Is a Human Subject
It is up to the IRB to determine whether a third
party is a human subject afforded human subject
protections under the Common Rule. A third party
would meet the Common Rule definition of a human
subject [40 CFR 26.102(f)] if individually identifiable
private information about them is collected (CFR, 2006).
When this occurs, the informed consent of the third party
must be obtained, or, if certain criteria are met, the IRB
may determine that informed consent may be waived. It
can be difficult to determine whether information about
a third party is both individually identifiable and private.
Discussions of this issue and recommendations for
determining whether third-party information is
identifiable and private have been submitted to the FIHS
Office of Human Research Protections by NTH (2001)
and by the National Human Research Protections
Advisory Committee (NHRPAC, 2002).
Whether or not a third party is determined to be a
human subject, the researcher should treat research
information about a third party as confidential.
4.4.2 Informing Third Parties of
Research Activities
Obtaining permission from or informing third parties
of certain types of activities may be needed in some
observational human exposure studies. For example,
household members living with a study participant need
to be informed about home study visits and residential
sample collection activities. Study activities that occur
outside of the participant's home or yard may require
informing or gaining permission from third parties. A
study may include collection of environmental samples
(i.e., ambient air, dust, soil) from outdoor common areas
of multifamily housing where the study participant lives.
Issues regarding privacy, permission, and payments for
third parties in housing-related studies have been
discussed in the NRC & IOM (2005) report.
Observational human exposure studies also may
include cases when study participants are asked to
collect personal samples (i.e.. wearing a personal air
monitor) over a time period that includes time they
spend in a school, day care, or workplace. Such
monitoring might require informing or gaining
permission from an organization's staffer an employer.
In each case, the researcher and IRB have to consider
whether obtaining permission from or informing a third
party is appropriate and, if so, to define the procedures
for doing so. The researcher and IRB have to also
consider the potential impact of third-party knowledge
of research activities on confidentiality and risk for the
study participant and have to ensure that it is clearly and
fully explained in the informed consent process.
4.4.3 Research Results and Third Parties
Prior to initiating a research study, researchers
should consider whether research results may be
provided to third parties. In some studies, there may be
reasons to inform household members living with a
study participant about specific residential measurement
results. In community research studies, aggregated or
summary research results may provide a benefit to the
community. In this case, it would be beneficial to seek
out the advice of community representatives regarding
results reporting prior to the study. Researchers also
should determine whether there are State or local
reporting requirements for some types of measurement
results above specified action levels (i.e., blood-lead
levels, heavy metal concentration in soil). It is important
that the researcher and IRB ensure that confidentiality
and privacy of study participants are carefully
47
-------
considered in any case where reporting study results to
third parties is contemplated or may be required. Ideally,
the informed consent process would make clear whether,
under what conditions, and how research results might
be provided to third parties.
4.5 Data and Safety Monitoring and
Oversight
The Common Rule requires for IRB approval that,
"When appropriate, the research plan makes adequate
provision for monitoring the data collected to ensure the
safety of subjects." [40 CFR 26.11 l(a)(6)].
Data and safety monitoring plans (DSMPs) are
developed and applied in all clinical trial research
studies. Clinical trials are prospective studies designed to
answer specific questions about the effects or impact of
particular biomedical or behavioral interventions. The
DSMPs are used to insure the safety of participants, the
validity of data, and appropriate termination of studies
for which significant risks or benefits have been
uncovered or when it appears that the trial cannot be
concluded successfully (NIH, 1998; NCI, 2001; U.S.
FDA, 2001). Depending on the study scope and potential
risks and benefits, a data safety monitoring board
(DSMB) may be created to assess procedures for data
and safety monitoring and to independently assess safety
and outcomes on an ongoing basis during the study.
Formal independent monitoring boards or
committees, like DSMBs, have not seen widespread use
in observational human exposure studies, although much
of the information included in DSMPs often has been
captured in the research protocols. Researchers and IRBs
may, however, consider using monitoring and oversight
boards to help assure participant safety and research
integrity in observational human exposure studies,
particularly in complex longitudinal studies and in
studies that include vulnerable subjects.
At least two NIH institutes have developed
guidelines for monitoring and oversight in the
observational research that they sponsor.
(1) The National Heart, Lung, and Blood Institute
(NHLBI) has developed an interim policy on the
creation and role of observational study monitoring
boards (OSMBs) for observational research
sponsored by mat institute (NHLBI, 2007). OSMBs
may be established for large or complex studies on a
case-by-case basis. The role of the OSMB is "to help
assure the integrity of the study by closely
monitoring data acquisition for comprehensiveness,
accuracy, and timeliness; and monitoring other
concerns such as participant confidentiality."
(2) The National Eye Institute (NEI) has developed
guidelines for data monitoring and oversight
committees (DMOCs) for observational research
(NEI, 2001). The role of the DMOC is to "assist the
NEI and the study investigators in protecting the
interests of study participants and in preserving the
integrity and credibility of the study."
When appropriate, formal procedures for routine
monitoring of scientific and ethical issues will need to be
incorporated into observational research and approved
by the IRB to ensure participant safety and the integrity
of the research. Even though most observational human
exposure studies are considered low-risk, there is often a
need to determine whether appropriate threshold values
for biological or environmental levels of chemicals exist
or can be determined that, if the threshold value is
exceeded, it would trigger reporting or other actions. The
safety of measurement procedures and equipment also
has to be considered. Unanticipated adverse events also
may be encountered in observational research.
Participant consent and understanding of the research
effort, participant recruitment, participant retention, and
data accuracy and quality should all be monitored to
ensure the scientific integrity of research results.
The authors already have discussed (Section 2.8) the
needs (1) to establish, in advance, criteria and standards
for monitoring the research program in regard to both
scientific and ethical issues; (2) to establish who will
monitor and oversee the research progress (the
monitoring and oversight authority, be it an individual,
team, or review committee); and (3) to establish the
roles, responsibilities, and authorities of the researchers
and of the monitoring and oversight authority. The
planning also should include steps to meet the IOM
recommendations mat researchers should "anticipate
risks and behaviors that may be observed in the home...
[and] develop anticipatory plans that specify how to
assess and respond to risks when they are identified, and
educate their staffs about the plan" (Recommendation
7.3, p. 144, NRC & IOM, 2005).
Once the procedures and organization for monitoring
and oversight of the observational study are approved by
the IRB, it is the responsibility of the researchers and of
the monitoring and oversight authority to ensure that the
planned actions are implemented. Implementation of the
monitoring and oversight function may include the
following items.
• Ensuring that procedures for identifying, reporting,
and responding to anticipated or unanticipated adverse
events and safety issues are in place and are being
followed
48
-------
• Assessing and responding to risks when they are
identified
• Evaluating the performance and knowledge of the
staff regarding identification of potential risks and the
actions they should take
• Implementing procedures for monitoring the informed
consent process, participant behaviors, participant
recruitment, participant retention, procedures to
protect privacy and confidentiality, and other human
requirements for adherence to the research protocol
and compliance with ethical standards and with EPA's
human subjects rules
• Ensuring that measurements and samples are collected
as planned, and that data are reported on a timely basis
• Evaluating whether the observed measurements
exceed the pre-established threshold values and, if so,
ensuring that reporting procedures and plans to
respond to the potential risks are completed on a
timely basis
• Ensuring mat quality assurance plans that define
procedures for assessing and ensuring study protocol
compliance are being met
• Ensuring data quality targets are met through
independent internal or external auditing requirements
• Taking all warranted oversight actions to ensure the
safety of the participants and the integrity of the study,
including terminating the research study if appropriate
References
AAP (American Academy of Pediatrics Committee on
Environmental Health) (2003). Pediatric Environmental
Health. (2nd Edition). Elk Grove Village, IL: American
Academy of Pediatrics.
Beauchamp TL, Childress JF (2001). Principles of Biomedical
Ethics. (5th Edition). New York, NY: Oxford University
Press.
CFR (Code of Federal Regulations) (2006). 40 CFR Chapter
I—Environmental Protection Agency Part 26 Protection of
Human Subjects. U.S. Code of Federal Regulations.
Available:
http://www.access.gpo.gov/nara/cfr/waisidx 06/40cfr26 06.
html [accessed August 2007].
CIOMS (Council for International Organizations of Medical
Sciences) (2002). International Ethical Guidelines for
Biomedical Research Involving Human Subjects. World
Health Organization. Geneva, Switzerland. Available:
http://www.cioms.ch/frame_guidelines_nov_2002.htm
[accessed June 2007].
Emanuel, EJ, WendlerD, Grady C (2000). What makes
clinical research ethical? JAMA 283(20): 2701-2711.
NCI (National Cancer Institute) (2001). Essential Elements of
a Data and Safety Monitoring Plan for Clinical Trials
Funded by the National Cancer Institute. Available:
http://www3.cancer.gov/rrp/dsm ess.html [accessed April
2007].
NEI (National Eye Institute) (2001). National Eye Institute
Guidelines for Data Monitoring and Oversight of
Obsen'ational Studies. Available:
http://www.nei.nih.gov/funding/datastudies.asp [accessed
April 2007].
NHLBI (National Heart Lung and Blood Institute) (2007).
Monitoring Boards for Data and Safety (Interim Policy).
Policy effective date: 14 May 2007. Available:
http://public.nhlbi.nih.gov/ocr/home/GetPolicv.aspx?id=8
[accessed September 2007].
NHRPAC (National Human Research Protections Advisory
Committee) (2002). Clarification of the status of third
parties when referenced by human subjects in research.
Available:
http://www.lihs.gov/olirp/nlirpac/documents/tliird.pdf
[accessed April 2007].
NIH (National Institutes of Health) (2002). Extramural
Projects -Application Instructions (03/15/2002) - Detailed
Application Instructions for Certificate of Confidentiality
Involving Extramural Research Projects. Certificates of
Confidentiality. Available:
littp://grants.nili.gov/grants/policy/coc/index.litm. [accessed
April 2007]
Nffl (National Institutes of Health) (2001). Protection of third
party information in research: Recommendations of the
National Institutes of Health to the Office for Human
Research Protections. Available:
http://bioetliics.od.nili.gov/nili tiiirdjartv rec.html
[accessed March 2007].
Nffl (National Institutes of Health) (1998). NIH Guide: NIH
Policy for Data and Safety Monitoring. Available:
http://grants.nih.gov/graiits/guide/notice-files/iiot98-
084.html [accessed April 2007].
NRC & IOM (National Research Council and Institute of
Medicine) (2005). Ethical Considerations for Research on
Housing-Related Health Hazards Involving Children.
Washington, DC: The National Academies Press. Available:
littp://books.nap.edu/catalog.php?record id=11450
[accessed June 2007].
49
-------
U.S. FDA (U.S. Food and Drag Administration) (2001). Draft U.S. HHS (U.S. Department of Health and Human Services)
Guidance for Clinical Trial Sponsors on the Establishment (1993). Protecting Human Research Subjects: Institutional
of Clinical Trial Data Monitoring Committees. U.S. Review Board Guidebook. Office for Human Research
Department of Health and Human Services, Food and Drag Protections. Available:
Administration, Rockville, MD. Available: http://www.lihs.gov/ohrp/irb/irb guidebook.htm [accessed
http://www.fda.gov/cber/gdrns/crindatmon.pdf. [accessed June 2007].
April 2007].
50
-------
SECTION 5
Creating an Appropriate Relationship Between
the Participant and Researcher
In observational human exposure studies, the
researcher and the participant routinely interact with
each other, often in the participant's home or other
private setting and often repeatedly over a number of
days. The nature and setting of the interactions mean that
exposure researchers should give special consideration
to the many scientific and ethical issues that shape the
relationship between participants and the researchers. In
these studies, it is recommended that a strong
relationship, built on openness and trust, should be
developed between the researcher and participant. The
nature of that relationship and the ethical principles
underpinning an appropriate relationship are the focuses
of this section of the document.
This relationship should be established on the ethical
values of respect for the participant's autonomy and
respect for their welfare. Emanuel and his co-authors
find that these two ethical values translate into specific
responsibilities for an ethical researcher in regard to
informed consent and respect for potential and enrolled
subjects (Emanuel et al, 2000). They describe the ethical
principles for these responsible actions thusly, "Respect
for potential and enrolled subjects is justified by multiple
principles including beneficence, nonmaleficence, and
respect for persons. Permitting subjects to withdraw and
providing them additional information learned from the
research are key aspects of respecting subject autonomy.
Protecting confidentiality and monitoring well-being are
motivated by respect for persons, beneficence, and
nonmaleficence." Section 4 already has described some
of the particular concerns regarding privacy,
confidentiality, and other issues related to observational
human exposure studies. This section further describes
elements of the relationship between researchers and
participants that are important to consider and address
during design and implementation of a study.
Of course, the relationship between the researchers
and the individual participants does not exist in isolation.
The researcher-participant relationship may influence
and be influenced by the relationship with the
community in which the participant lives. Good, two-
way communications are critical for the development
and nourishment of an appropriate researcher-participant
relationship. Although those two topics are the subject of
the next sections of this document, elements from those
topics unavoidably will color the discussions in this
section as well.
Researcher training is a key component for
conducting research that incorporates human subject
protections and fosters appropriate researcher-participant
relationships. Most organizations require basic human
subjects training on the essential elements for processes
and procedures for research with human subjects. More
in-depth training will improve researcher understanding
in areas of the informed consent process, observational
techniques, community-based research, and other topic
areas. Such training will benefit principal investigators
throughout the research study from the design stage
through communication of research results. Training is
also important for staff mat will work directly or
indirectly with research participants or their samples and
data. This document can be used as both a training tool
for researchers as well as a resource for designing
training courses. There are a number of sources of
training on human subjects protection. The
Collaborative Institutional Training Initiative (CITI,
http://www.citiprogram.org/) is a subscription service
providing research ethics education to many institutions.
Other training, such as that provided by the National
Cancer Institute (http://cme.cancer.gov/clinicaltrials/lear
ning/humanparticipant-protections.asp) and the U.S.
Department of Health and Human Services
(http://www.hrsa.gov/humansubjects/default.htm). are
available on-line.
51
-------
5.1 Informed Consent
In observational human exposure studies, informed
consent ensures that the participant accurately
understands the range of risks and benefits (if any)
associated with participation; emphasizes the voluntary-
nature of their participation; and provides essential
protections to the participant. The three '"pillars" of the
informed consent process are (1) information;
(2) comprehension; and (3) voluntary participation, or
"voluntariness" (U.S. DREW, 1979/ Informed consent
requires "provision of information to subjects about the
purpose of the research, its procedures, potential risks,
benefits, and alternatives, so that the individual
understands this information and can make a voluntary
decision whether to enroll and continue to participate"
(Emanuel et al., 2000).
The NRC & IOM document Ethical Considerations
for Research on Housing-Related Health Hazards
Involving Children (NRC & IOM, 2005) contains a
comprehensive and very useful discussion of informed
consent procedures and requirements in Chapter 6. The
IOM report, Responsible Research: A Systems Approach
to Protecting Research Participants, also includes a
thoughtful discussion of participant-investigator
interactions and the informed consent process (IOM,
2002). CIOMS also includes recommendations for both
the process and content of informed consent (CIOMS,
2002). Some of the important points from those
documents are summarized in Text Box 5-1, but the
reader should refer to those documents for additional
information about this topic.
Federal regulations governing research that is either
Federally conducted or Federally funded (i.e., all human
subjects research at NERL) are codified in the Common
Rule. The regulations set forth requirements for both the
content of an informed consent and the process for
obtaining and documenting an individual's informed
consent (see Text Box 5-2).16 General regulatory
requirements for the elements of informed consent are
codified in the Common Rule at 40 CFR 26.116(a)(l)-
(8) (CFR, 2006). The regulations also prescribe the use
of a written consent form and describe how informed
consent is to be documented (at 40 CFR 26.117). The
regulatory requirements for informed consent highlight a
number of issues that a NERL researcher needs to
consider in developing and administering the informed
consent process and the consent form document. The
discussion of these issues, arising from regulatory
requirements or identified in recent writings on ethical
considerations in human subjects research, is grouped
below, under the three pillars of informed consent:
(1) information, (2) comprehension, and (3) voluntary
participation.
Text Box 5-1. Recommendations for Informed Consent
In the United States and Internationally
(1) Revitalize Informed Consent (IOM, 2002)
Informed consent is a process, not a form. It is an on-going,
interactive dialogue between research staff and research
participants with disclosure and exchange of relevant
information, including assessment of understanding.
Ethics Review Boards should ensure that the focus of both the
informed consent process and the consent forms is on
informing and protecting participants, not the research
institution.
(2) Strengthen Process of Parental Permission and
Children's Assent if Children Are Involved in Research
(NRC& IOM, 2005)
The process begins with a community-based discussion and
concludes with an assurance that individual parents
understand the essential elements of the research.
Educate parents on issues critical to informed decision making
and assess their degree of understanding.
Use informational materials in the form that is most
appropriate to convey information to potential participants.
Consult with community representatives to ensure that
information is complete, clear, and understandable, and that
any payment will not be exploitive.
Expand the perspective about what information about risks
and benefits is needed to make an informed choice.
Ethically, it is critical to assure that participants and parents
understand the crucial features of the research effort.
(3) Informed Consent Is a Decision To Participate by an
Informed, Competent Individual Without Undue Coercion,
Influence, or Intimidation (CIOMS, 2002)
Informed consent is a process that takes time and resources.
It is not a ritual recitation of text from a form but informative
communication in language that suits the individual's level of
understanding.
Consent is ethically obtained, considering language and
cultural issues, and is documented.
Material changes in conditions or procedures require that
informed consent be renewed.
Consent should explain how samples will be used in current
research, and, if applicable, how samples may be used in
future research.
5.1.11nformation
Some items that researchers should keep in mind as
they provide information to the study participants are
summarized below. These items may be based on
regulatory requirements or currently may be
recommendations as ethical "best practices."
An IRB may waive informed consent under some very limited conditions.
See 40 CFR 26.116(c) and (d).
52
-------
Text Box 5-2. Common Rule Requirements:
Elements of Informed Consent
(1) An explanation of the purposes of the research
(2) The expected duration of the subject's participation
(3) A description of the procedures to be followed and
identification of any experimental procedures
(4) A description of any reasonably foreseeable risks or
discomforts to the subject
(5) A description of any reasonably expected benefits to the
subject or others
(6) A disclosure of appropriate alternative procedures that might
be advantageous to the subject
(7) A description of the extent that confidentiality will be
maintained
(8) For research involving more than minimal risk, an
explanation about whether compensation or medical
treatments are available if injury occurs
(9) An explanation of whom to contact with questions about the
research or to report a research-related injury
(10) A statement that participation is voluntary, refusal to
participate will involve no penalty, and the subject may
discontinue participation at any time without penalty or loss
of benefits to which the subject is otherwise entitled
(11*) A statement that the particular treatment or procedure may
involve risks to the subject (or to the embryo or fetus, if the
subject is or may become pregnant) that are currently
unforeseeable
(12*) Anticipated circumstances under which the subject's
participation may be terminated by the investigator without
regard to the subject's consent
(13*) Any additional costs to the subject that may result from
participation in the research
(14*) The consequences of a subject's decision to withdraw from
the research and procedures for orderly termination of
participation by the subject
(15*) A statement that significant new findings developed during
the course of the research that may relate to the subject's
willingness to continue participation will be provided to the
subject.
(16*) The approximate number of subjects in the study
* Included if appropriate [40 CFR 26.116(b)]
The information "shall be in language understandable
to the subject" (40 CFR 26.116). This may require
forms to be written and administered in different
languages during a study. For example, the National
Children's Study (NCS) plans to produce all consent
materials in English and Spanish, with other
translations made available as needed (NCS, 2007).
Ethically, "language understandable to the subject"
goes beyond simply using the appropriate language;
the researcher is compelled also to consider
readability and vocabulary (e.g., avoiding jargon or
terminology mat may be clear to the researcher but
which may confuse or intimidate the potential subject)
if the information is to be understandable.
Information may be presented orally in addition to an
appropriately written document (40 CFR 26.117).
Participants often find discussions with research staff
more useful man written consent forms (p. 103, NRC
& IOM, 2005). NCS plans to pilot test an interactive,
computer-based audio/video consent tool and to
compare it with traditional written informed consent
approaches (NCS, 2007).
The information being discussed, including the
explanation of the purpose of the research and the
description of the study procedures, should be written
or discussed at a level that the participant can
understand.17 The National Institutes of Health
recommends writing consent forms as "plain language
documents that explain the research in an honest,
straightforward way" and suggest that doing so will
help enhance public trust (Recommendation 11, NIH,
2005).
The consent form should contain sufficient
information to describe the study procedures, but not
so much information that it causes confusion and
results in the participant not understanding the study.
There is not agreement on what the appropriate level
of information is. IRBs do not agree on the level of
information; some require lengthy descriptions of the
study, whereas others prefer concise information.
Ultimately, the IRB dictates the language of the
informed consent document, and the researcher will
need to comply. It will benefit the researcher to
discuss the consent process with their IRB when they
develop the consent form document and process (p.
108, NRC & IOM, 2005).
Iii observational human exposure studies, information
about the risks of the hazards being studied needs to
be conveyed to the participants during the consent
process. Information should be provided to the study
participant on what hazards pertinent to the topic of
the study may be present in the participant's
environment, particularly those microenvironments
being studied, what hazards will continue to exist in
those microenvironments after the research is
completed, and how those hazards may adversely
affect the participant's health (NRC & IOM, 2005).
The informed consent process should describe
whether any study results will be provided to
A survey of IRBs found that their readability standards ranged from 5th- to
lOth-grade level (Paasche-Orlow et al., 2003). Interestingly, the same report
found that, 92% of the time, the sample consent forms provided by the IRBs
did not meet their own readability standards. The NRC & IOM report (p. 107,
NRC & IOM, 2005) discusses an NCI effort to simplify informed consent
forms by using text targeted at 8th-grade reading level. More information
about the NCI template may be found at
www.nci.nili.gov/clinicaltrials/understandina'simplification-of-infonned-
consent-docs;;page2 (accessed September 12. 2007).
53
-------
participants and, if so, how and when (p. 101, NRC &
IOM, 2005).
• For studies involving children as participants, it
generally is regarded as desirable that the informed
consent process should involve both parents
(assuming mat there are two competent parents
available), and that the consent of both parents be
obtained, if possible. The IRB may decide that the
permission of one parent is sufficient, but only when
certain risk-benefit conditions are met.18
• For studies involving children as participants, it is
desirable for those children with sufficient capacity to
be involved in the consent process. Moreover, it is
generally accepted mat the child's assent be obtained
whenever this is developmentally possible and
otherwise appropriate. Under both the EPA Rule and
the FiHS Rule, the IRB is responsible for determining
that adequate provisions have been made for soliciting
the assent of the children when, in the judgment of the
IRB, the children are capable of providing assent.
Assent, however, may be waived in those restricted
circumstances in which consent may be waived under
the Common Rule. Those circumstances and the
required IRB documentation are described in the
Common Rule at 40 CFR 26.116(d).
• The consent form should clearly state that
participation is voluntary and that study participants
may "discontinue participation at any time without
any penalty or loss of benefits to which the subject is
otherwise entitled" [40 CFR 26.116(a)(8)]. If a subject
chooses to withdraw from a study, the consequences
of their decision and the process for orderly
withdrawal should be clearly explained [40 CFR
26.116(b)(4)].
• The consent form should address any foreseeable
potential future use of samples and data (CIOMS,
2002). For example, effects of environmental
exposures on gene expression are potentially very
important. Therefore, biologic specimens for DNA
analysis may be obtained from participants in future
studies. But, it is recognized mat human genomic data
are private, intimate, and sensitive, and they create
See the EPA Rule at 40 CFR 26.406(b) for observational research with
children and the HHS Rule at 45 CFR 46.408(b) for all research with children.
Under the EPA Rule, greater than minimal risk observational research with
children that does not hold out the prospect of direct benefit to the child is not
permitted under any circumstances. Under the HHS Rule, greater than
minimal risk research involving children without the prospect of direct benefit
is permitted in very limited circumstances, but the consent of both parents is
required in those cases (unless one parent is deceased, unknown, incompetent,
or not reasonably available, or when only one parent has legal responsibility
for the care and custody of the child).
special concerns about the potential for
discrimination, stigmatization, and impact on future
employment or insurance. The informed consent
process needs to explain what the plans may be for
such specimens and recognize the rights of the
subjects to decide about any such future use, including
having the material destroyed. The informed consent
process needs to explicitly discuss obtaining
permission from participants on behalf of themselves
and their child to obtain specimens for genetic
analysis.
5.1.2 Comprehension
Research participants frequently fail to understand
the research protocols in which they agree to participate
(NRC & IOM, 2005). In considering the ethical issues
raised by the Grimes v. Kennedy Krieger case, the NRC
& IOM committee "realized that the crucial issue
regarding consent was not what information was
contained in the consent forms, but rather what the
parents understood about the study and the hazards
present in the home before and after the study" (p. 19,
NRC & IOM, 2005). The committee laments that "IRBs
place their attention on consent forms rather than on the
process of providing and discussing information" (p.
103, NRC & IOM, 2005). The IOM recommends that
"the informed consent process should be an on-going,
interactive dialogue between research staff and research
participants involving the disclosure and exchange of
relevant information, discussion of that information, and
assessment of the individual's understanding of the
discussion" (Recommendation 4.1, IOM, 2002). These
comments emphasize how important true two-way
communication is to comprehension, the second pillar in
the informed consent process.
The following items are a variety of issues
concerning comprehension that NERL scientists should
keep in mind as they develop an informed consent
process in collaboration with the research team, the IRB
and other peer reviewers, and EPA's HSRRO. These
items may be based on regulatory requirements or may
simply be recommendations as ethical "best practices."
• Researchers need to assume responsibility for
developing an interactive dialogue with participants
for the exchange and discussion of relevant
information as a part of the informed consent process,
not just for conveying information. The dialogue
should be ongoing, continuing throughout the research
project (IOM, 2002).
• The consent form and its content are only one part of
the overall consent process. An equally important part
54
-------
is how information is conveyed to the participant
outside of the written form itself. Participant
comprehension is contingent on all elements of a
comprehensive consent process that involves ongoing
information exchange between researchers and
participants, as well as a written informed consent
document (NRC & IOM, 2005).
The most effective way to improve comprehension is
by talking one-on-one with study participants.
"Having a study team member or a neutral educator
spend more time talking one-on-one to study
participants appears to be the most effective way of
improving research participants' understanding"
(Flory and Emanuel, 2004).
The information being exchanged (e.g., explanation of
the purpose of the research, description of the study
procedures) should be written at a level that the
participant can understand (NRC & IOM, 2005).
The researcher should describe the benefits of
participation in the study [40 CFR 26.116(a)(3)], but
should not promise any outputs or outcomes that he or
she cannot deliver. Participants often misunderstand
the purpose of the research. The researchers also
should attempt to reduce the likelihood of "therapeutic
misconception"19 or related misunderstandings in
which the participant anticipates a benefit that does
not really exist, such as reduction of the hazard in an
observational study (NRC & IOM, 2005).
The consent procedure should include some test of the
participants to demonstrate that they truly understand
the information that is being conveyed (IOM, 2002).
Tools to assess comprehension have been developed,
but, as described in NRC & IOM (2005), there are no
standard mechanisms for assessing comprehension.
Tests for appropriate grade-level language can be
performed, but additional comprehension testing
should be considered as well (Flory and Emanuel,
2004).
Researchers need to develop innovative approaches to
improve comprehension. Multimedia, such as video or
graphics, may be used but have had limited success in
the past (NRC & IOM, 2005; Flory and Emanuel,
2004). NCS currently is developing a highly-
sophisticated video consent tool that may be able to
Therapeutic misconception is a term that refers to an inaccurate
understanding on the part of a research participant that a direct therapeutic
benefit will be provided by virtue of participation in a clinical trial.
Researchers performing observational human exposure studies should be
aware of the potential for misunderstandings to arise that are analogous to the
misunderstanding represented by the therapeutic misconception. Ensuring
comprehension of the study and its expected results is important to this issue.
serve as a model going forward. The video
presentation will include embedded questions to
assess the participant's understanding of the key
elements of NCS and what their participation will
involve (NCS, 2007).
• Development of written materials with appropriate
languages and comprehension levels is only part of
the communication challenge. In communities where
languages other than English are spoken, it will be
important that the research team be able to
communicate orally with participants, often in ad hoc
situations. Research protocols should address how
translations will be accomplished. The use of
untrained persons such as co-workers usually will not
meet requirements for full understanding of human
subject protections. Researchers should make plans
for trained staff or trained community members,
associated or affiliated with the research study, to be
available for translations.
5.1.3 Voluntary Participation
The third pillar of informed consent is voluntary
participation. The Belmont Report emphasizes mat
participants "should understand clearly the range of risk
and the voluntary nature of participation" [emphasis
added]. The ethical principles of respect for persons and
their autonomous decisions morally obligate the
researcher to ensure mat an individual's decision to
participate in a human research study is truly voluntary
and uncoerced (Emanuel et al., 2000). A number of
study characteristics may affect whether the participant's
actions are truly voluntary.
• Access to study-dependent benefits or care that would
otherwise not normally be received may impair
voluntariness.
• Voluntary participation also may be compromised
when there is an existing relationship between the
researcher and participants, such as employer and
employee or teacher and student.
• Restricted voluntariness may be an intrinsic part of
belonging to certain vulnerable groups, including
children, prisoners, handicapped persons, mentally
disabled persons, and economically or educationally
disadvantaged persons, or members of the military, for
example. When research participants come from such
groups, additional protections to insure voluntariness
in the context of the research may be required (see
also 40 CFR 26, Subparts B, C, and D).
• Payments as incentives may have undue influence and
are discussed below.
55
-------
• Whether payments will lead to a coerced decision to
participate often is difficult to determine without input
from people from similar socioeconomic backgrounds
as the participants (p. Ill, NRC & IOM, 2005).
Researchers should work with community
representatives to develop a consent process that will
be maximally effective in providing information,
ensuring and documenting comprehension, and
ensuring that participation is voluntary (also see
Section 6).
Researchers should remember that obtaining
informed consent should be "an on-going, interactive
dialogue . . . involving the disclosure and exchange of
relevant information" (IOM, 2002): it is not simply
having a consent form signed. The process is most
effective when the researcher spends time with potential
participants to discuss the study and to answer questions.
5.2 Payments to Research Participants
The decision whether to pay research participants,
including the appropriate level of payment,20 is a
complex ethical issue. Monetary or nonmonetary
payments are not ethical if they constitute an undue
inducement for participants to assume research risks mat
they would not otherwise accept. On the other hand, it
may be ethically appropriate to offer reasonable
payments in some research studies. Indeed, payment of
human subjects for their participation in scientific
research is a common practice in the United States, with
a history of well over 100 years (Grady, 2005). The
difficulty for researchers and IRBs is that there is often
little clear and uniform guidance for determining what
constitutes "undue inducement" or "reasonable"
payment for any particular research study, population,
and level of risk. Additional considerations regarding
payment to participants arise when working with
vulnerable populations, including children.
There are many terms that may be found in the literature to describe both
monetary and nonmonetary payments, like payment, remuneration,
compensation, recompense, incentive, inducement, reimbursement, and
reward. Each of these words has its own definition, and each word also may
carry specific connotations. In general, these terms refer to money or other
items that "are often given to acknowledge the time and inconvenience of
participating in research or to reimburse participants for any costs they incur.
The term compensation is often used in the context of compensation for
research-related injuries" (p. 112, NRC & IOM, 2005). The authors have
chosen to use "payment" as the general term for monetary and nonmonetary
items provided to research subjects for their participation in the research. They
may occasionally refer to compensation for research-related injuries, but the
text should make it clear when they do.
5.2.1 Types and Amounts of Payments Offered
in Research Studies
Payments have been offered in a wide variety of
research studies, ranging from clinical trials to
behavioral and social research to observational human
exposure studies. Payments can take various forms,
including monetary payments (e.g., cash, gift
certificates, reimbursement for expenses), nonmonetary
payments (e.g., gifts, valuable information), or nothing
at all (e.g., for a sense of altruism). Grady (2005) has
described payment for the participation of research
subjects as serving from one to four purposes: (1) an
incentive, (2) compensation, (3) reimbursement, or
(4) reward. Text Box 5-3 describes the roles that
payments may serve, together with potential advantages
and disadvantages or ethical concerns. The text comes
mostly from Grady (2005) but builds also on the NRC &
IOM document and other writings. (A number of authors
have addressed issues associated with payments,
including Ackerman, 1989; Dickert et al., 2002;
Emanuel 2004; Erlen et al., 1999; Fry et al., 2005;
Grady, 2005; Grady et al., 2005; Iltis et al., 2006; IOM,
2004; NRC & IOM, 2005; Russell et al., 2000;
VanderWalde, 2005; Weise et al., 2002; Wendler et al.,
2002). A number of specific issues or concerns
regarding participant payments have been identified.
Many researchers and ethicists argue that it is often
appropriate to provide reasonable payment and have
done so for many years. In the view of an NRC expert
panel, the principles of "justice, fairness, and gratitude
support payment to those who bear the burdens of
research on behalf of society" (NRC, 2004).
Direct reimbursement may be made to participants
for out-of-pocket expenses for costs directly associated
with participation in a study. These might include
transportation costs, parking fees, or child care costs.
When payment for time and burden is provided, it is
often in the form of monetary payments. Different
approaches may be considered for determining
reasonable amounts for payment, including a set
payment for each visit, a small daily payment, payment
at the prevailing minimum hourly wage, or payment at
some other hourly rate appropriate for the
community—perhaps a prevailing rate for unskilled
labor (Emanuel, 2004). Incentives to encourage
enrollment are sometimes used when participants will
receive little or no direct benefit from the research and
can take the form of monetary or non-monetary
payments. Incentives are kept modest so as not to impart
undue influence. Researchers need to consider the
possible effects of incentive payments on the potential
56
-------
Text Box 5-3. Payments for Participation of Research Subjects
(based on Grady, 2005)
Payment Serves As
Incentive
Compensation
Reimbursement
Reward
Amount Determined By
Supply and demand; market
rates
Standardized "wage" for time
and effort, suggested to be
commensurate with wages for
unskilled, but essential jobs;
additional payment for extra
burdens such as endurance of
uncomfortable procedures
Provide "wage" for time and
effort; base level of payment
on some small fraction of
participant's or the
community's income, or on an
appropriate hourly rate for the
location, or on community-input
about the appropriate level of
payment
Actual out-of-pocket cost to
participant related to
participation, such as parking,
transportation, child care, cost
of food samples, etc.
Token of appreciation at end of
study
Potential Advantages
(a) More rapid recruitment
(b) Completion bonuses encourage
subject retention and high
completion rate
(c) Possibility of profit for participants
(d) Little or no financial sacrifice by
subject
(a) Recognizes contributions of
participants
(b) Uniform payment across studies
(c) Equal pay for equal work
(d) Less risk of undue inducement
(a) Recognizes contributions of
participants
(b) Decreases potential that payment
would be undue inducement in
some communities or locations
and insufficient in others
(c) Provides equivalent payment for
participation across communities
and locations (markets)
(d) Less risk of undue inducement
across study
(a) Lowers barriers to participation
(b) Reduces burden and impact of
research on participants
(a) Expresses gratitude for
contribution made
(b) Not market dependent
(c) Avoids undue inducement
Potential Disadvantages
(a) Undue inducement possibly resulting
in incomplete assessment of risks
and benefits by subject; subject
concealing information to ensure
enrollment and retention
(b) Competition between studies; better
funded studies more likely to meet
recruitment goals
(c) Different levels of payment at
different locations for multicenter
research because of supply and
demand
(a) May have little impact on recruitment
(b) Might undercompensate some
subjects in relation to regular wage
and preferentially attract others
(a) Different levels of payment at
different locations in multicenter
research or in different communities
(b) Unequal pay for equal work may
violate one's sense of justice
(a) Few disadvantages
(b) Possible differences in costs
experienced by different participants
(a) Probably no impact on recruitment
(b) No basis for consistency
for differential recruitment that could result in bias in the
study sample.
Determining appropriate level of payments or
incentives for participants in a research study is
complex. "No bright line distinguishes proper and
reasonable payments to parents and children from
payments mat are inappropriate" (p. 214, IOM, 2004).
The ethical issue is at what level might a payment
change from being an incentive (an encouragement) to
participate and become an inducement (the cause or
reason) for participation. Many research organizations
and IRBs do not have written policies or guidelines
regarding the determination of reasonable payment.
Decisions often are made based on the level of
discomfort and burden, costs to participants, and
population characteristics. However, large differences in
payment levels have been found even in multisite studies
in which the same protocol is administered across all
sites.
Grady et al. (2005) performed a survey of practices
for paying research participants in the United States in
Phase 1 to 4 clinical trials and physiologic, behavioral,
and other types of research. Across 467 studies of
varying complexity that included payments, the median
57
-------
payment was $155 (mean $266, range $5 to $2,000). The
basis for dollar amounts was infrequently described,
with 19% of the payments based on time, and 12% based
on the procedures. In a model of payment factors,
studies with some prospect of therapeutic benefit, studies
having at least one invasive procedure, and studies with
greater numbers of clinic visits were significantly
associated with higher dollar amounts. About 9.5% of
the studies offered completion bonuses, and a similar
percentage offered escalating payments for followup
study visits.
5.2.2 Regulations and Guidance Regarding
Payment to Research Participants
There is little specific guidance regarding payments
in Federal human research regulations. The Common
Rule and additional human subjects protections do not
directly address payments to research participants, but
the regulations do discuss providing additional
safeguards for subjects vulnerable to coercion or undue
influence [40 CFR 26.111(b)]. The NIH IRB guidebook
advises IRBs to determine whether the rewards offered
for participation in research constitute undue influence
(U.S. HHS, 1993). According to the IRB guidebook
undue inducement might blind prospective subjects to
risks, impair their ability to exercise proper judgment, or
may cause people to lie or to withhold information that
would make them ineligible to enroll or continue
participation.
The U.S. Food and Drug Administration (FDA) has
provided guidance for investigators and IRBs for clinical
research studies (FDA, 1998). The guidance states that
"payment to research subjects for participation in studies
is not considered a benefit, it is a recruitment incentive."
FDA expects payments to accrue as the study progresses
and not to be contingent on completing the study,
although a "small proportion as an incentive for
completion of the study is acceptable." The guidance is
concerned with the issue of coercion or undue influence,
and it recognizes the IRB as the responsible party for
deciding what is or is not acceptable.
The U.S. Office of Management and Budget (OMB)
develops standards and guidelines for statistical surveys
performed by the Federal government. Under Guideline
2.3.2, OMB states that, while incentives are not typically
used in Federal surveys, agencies may consider use of
respondent incentives if they believe incentives would
be necessary to use for a particular survey to achieve
data of sufficient quality for their intended use (OMB,
2006a). OMB requires that agencies provide a
justification for giving incentives to respondents. Some
of the factors cited by OMB to be addressed include
those particularly relevant to observational human
exposure studies, including unusual reporting burdens
(keeping data logs for extended periods, coordinating
study team visits, participating in a medical examination,
etc.), complex study designs (such as studies requiring
ongoing participation of respondents), and past
experience, especially when there is evidence of attrition
or poor response rates (OMB, 2006b). Although OMB
primarily considers incentives with regard to survey
response rates and data quality, researchers need to
consider payments to participants, including
participation incentives, in the broader context discussed
in this section.
CIOMS also provides guidance and commentary on
this issue in the International Ethical Guidelines for
Biomedical Research Involving Human Subjects (2002).
Guideline 7 and the associated commentary emphasize
that payments to subjects for expenses incurred because
of their participating in a research study are legitimate.
The guideline also allows payment for inconvenience
and time spent, so long as the payments or other direct
benefits are not "so extensive as to induce prospective
subjects to consent to participate in research against their
better judgment."
The approaches of HHS, FDA, and CIOMS above
are consistent in not considering payments to be a
benefit to research participants when considering risks
versus benefits. All of the guidelines recognize the
legitimacy of some recompense, but they all are
concerned with the issue of undue influence. "Payments
or rewards that undermine a person's capacity to
exercise free choice invalidate consent" (CIOMS, 2002),
and voluntariness is a pillar of legitimate informed
consent.
Text Box 5-4 lists some of the concerns about, and
the reasons for, payment of participants.
IRBs have considerable discretion with regard to
payments and consider payments with regard to the
specific circumstances of the research and of the
population being studied. The issue of recompense can
be a difficult but legitimate ethical issue involving
weighing the different ethical principles of justice and
fairness against the concerns about undue influence and
the invalidation of consent. Ethical review committees,
including IRBs, need to consider many factors when
determining when it is appropriate to offer payments to
research participants and the level and form of payments
when they are appropriate. Review committees also
should consider how and when information on payments
is communicated to prospective study participants.
58
-------
Text Box 5-4. Weighing the Ethical Issues
About Payments
Concerns about Payments
Payments may compromise
voluntary participation.
Participants may accept risks
they would not otherwise
accept.
Participants may continue in a
research study beyond a point
they might ordinarily have
withdrawn.
Payments may differentially
encourage research
participation by economically
disadvantaged people.
The offer of payments may
cause guardians or parents to
not act in the best interests of
incompetent persons or children
in their care.
Persons in different
circumstances may view the
same amount of payment quite
differently.
Payments may alter the
composition of the study sample
and potentially could
compromise study integrity.
Reasons for Payments
Recognizing participant
contributions to the
research and knowledge
gained
Providing reimbursement
for direct and indirect
participant costs
Providing reasonable
payment for the time and
effort associated with
participation in research
Providing incentives for
participation in studies
with low risk but no or few
direct benefits
It is the just and fair thing
to do for those who bear
burdens of research on
behalf of society.
5.2.3 Payments When Children or Other
Vulnerable Populations Are Involved
It is essential that special care be taken with regard
to payments when members of vulnerable populations
are included in research studies. Vulnerable populations
may include children and adolescents, those with
cognitive impairments because of medical conditions or
age. economically disadvantaged persons, and prisoners.
These populations often are not capable of making
autonomous, fully informed decisions regarding risks
and benefits, or they may be particularly vulnerable to
undue influence resulting from the offer of a payment for
research participation. In addition, payments made
directly to parents or guardians could alter judgment
regarding the best interests of minor or incompetent
persons in their care.
The ethical concern is that too high a payment may
"undermine free and informed consent by leading
parents to expose their children to unacceptable risks"
(NRC & IOM, 2005). The NRC & IOM committee
recognized that some commentators argued that children
should never be paid, and that parents ought not to be
paid to enroll their children in research. Yet, on balance,
the committee felt that "reimbursement for expenses and
some modest payment for time spent in research
activities is thus justified on the grounds of fairness" (p.
112, NRC & IOM, 2005).
Similarly, the IOM Committee on Clinical Research
Involving Children found that "certain types of
payments to parents or adolescents are usually if not
always acceptable, for example, reimbursement for
reasonable expenses that are necessary for research
participation. The specifics may vary, but examples of
reasonable expenses are costs of transportation to the
research site, parking, lodging, meals, and babysitting.
Oilier payments are never appropriate in pediatric
research, for example, paying parents for the use of their
child in research" (pp. 225-6, IOM, 2004).21
The IOM committee recommends establishing
policies on acceptable and unacceptable types of
payments. They also recommend that the policies
disclose any recompense in a full and open process,22
while not overemphasizing any recompense.
Although the NRC & IOM Committee on Ethical
Issues in Housing-Related Health Hazard Research
Involving Children and the IOM Committee on Clinical
Research Involving Children both concluded that it is
appropriate to reimburse expenses or compensate for
time or inconvenience, neither committee endorsed
incentive payments to parents. In Europe, too, incentive
payments to induce parents to allow their children to
The IOM Recommendation 6.2 states. "In addition to offering small gifts or
payments to parents and children as gestures of appreciation, investigators
may also—if they minimize the potential for undue influence—act ethically to
reduce certain barriers to research participation when they
• reimburse reasonable expenses directly related to a child's participation in
research
• provide reasonable, age-appropriate compensation for children based on
the time involved in research that does not offer the prospect of direct
benefit, and
• offer evening or weekend hours, on-site child care, and other reasonable
accommodations for parental work and family commitments."
" In recommending an open process, the IOM committee chose to reject the
arguments from the American Academy of Pediatrics that "any token payment
to children for participating in research should not be discussed with them
until after research is completed for fear of unduly influencing their decisions
(AAP, 2003).... On balance, the committee agrees that it is best to mention
token or other payments during the permission and assent processes" (p. 215,
IOM. 2004).
The IOM Recommendation 6.1 states. "Institutional review boards, research
institutions, and sponsors of research that includes children and
adolescents should adopt explicit written policies on acceptable and
unacceptable types and amounts of payments related to research
participation. These policies should specify that investigators
• Disclose the amount, the recipient, the timing, and the purpose (e.g.. an
expense reimbursement or a token of appreciation to a child) of any
payments as part of the process of seeking parents" permission, and, as
appropriate, children's assent to research participation;
• Avoid emphasis on payments or descriptions of payments as benefits of
participating in research during the permission or assent procedures; and
• Obtain institutional review board approval for the disclosure of
information about payments in advertisements and in permission and
assent forms and procedures."
59
-------
participate in research are unacceptable. The European
Union requires that clinical trials on minors be
undertaken only if "no incentives or financial
inducements are given except compensation" (European
Parliament, 2001).
Payment for participation of children in research
also is discussed in the literature. Diekema (2005)
emphasizes the need to ensure that payments do not
distort parental decisionmaking and do not tempt parents
to consider other issues than the welfare of their child.
Similarly, Menikoff (2005) suggested that there need to
be relatively robust protections in place to ensure that
families do not change their behaviors to participate in a
study. He suggested that these may include determining
payment as a percentage of a family's income and
developing criteria for documenting that behaviors have
not changed to be eligible for participation in a study. He
suggested that, for a study of pesticides, potential study
participants provide documentation (such as receipts)
mat they routinely have been using a commercial
pesticide service. This may be difficult for potential
participants to do if they do not save receipts, and it
would exclude all potential participants who purchase
products and apply pesticides themselves. This likely
would affect the study objectives and generalizability of
the data collected. A survey of investigators (Iltis et al.,
2006) found that payments were made in 52% of the
pediatric research studies surveyed, and that payment
practices varied, as did the reasons for decisions
regarding payments. They found a range of payment
values separated across cash, gifts, items, vouchers, and
other categories. A survey of IRBs (Weise et al., 2002)
found that payment for participation in research was
allowed by 66% of responding institutions, but that
many IRBs did not have specific policies, and that there
was considerable variability regarding the basis for
decisions on payments in studies with children. The
types of payments included money, certificates, and
bonds with large ranges in the amounts of payments for
approved pediatric research. This research shows a lack
of consistency and the need for guidance and
institutional policies that describe acceptable and
unacceptable payments and the basis for the amount of
any payments.
The NRC & IOM Committee on Ethical Issues in
Housing-Related Health Hazard Research Involving
Children described many of the ethical considerations,
practices, and policies regarding payments (NRC &
IOM, 2005) for research conducted in the participants'
homes, rather than in a clinical facility. The research
setting is similar to the setting of most observational
human exposure studies, and the committee's
commentary and recommendations are also relevant. The
committee notes that it would be unfair to expect
families to make considerable sacrifices to participate in
a time-consuming activity designed to advance
generalizable scientific knowledge, rather than benefit
themselves directly, and that payment for reimbursement
of expenses and modest payment for time spent in
research activities is justified on the grounds of fairness.
But the committee then warns that if payments are too
high, they may distort parents' decisions about enrolling
their children. The committee also found that "how the
payment is made may also result in undue influence. For
example, if payment for a long-term follow-up study is
made in a lump sum and only if the subjects complete
the entire study, then it could constitute an undue
influence to stay in the study. If, on the other hand, the
money is paid weekly, the effect would not constitute an
undue influence."
The NRC & IOM committee recognizes mat the
issue of payment for participation in research is
controversial. They also discuss how "countervailing
ethical guidelines" may complicate the issues even more.
Citing Wendler et al. (2002), the NRC & IOM
committee points out that payments that are trivial for
some families may be substantial for low-income or
disadvantaged families. "Yet to pay economically
disadvantaged families less than more affluent families
for participating in the research is unfair because it
requires similar sacrifices of time and inconvenience
from both" (p. 113, NRC & IOM, 2005). Similar ethical
quandaries can arise in multisite studies with differing
costs for living. If the same payment is used in high-cost
cities as in low-cost areas, the payment may be
inadequate to gain sufficient enrollment in the high-cost
area, whereas the same dollar amount may be "coercive"
in the low-cost area. The NRC & IOM committee notes
that a similar situation can arise when a study enrolls
participants from diverse socioeconomic backgrounds.
There are social justice concerns that poorer people
might incur a disproportionate share of research risk and
burden if payments induce unequal participation rates in
the population. Decisions regarding payment for
research participation will require careful consideration
by IRBs when economically disadvantaged people may
be enrolled. Community advisory boards (CABs) can be
very important in helping researchers and IRBs
determine what is appropriate with regard to payments
within their community.
60
-------
5.2.4 Payments in Observational Human
Exposure Studies
Observational human exposure studies most often
involve minimal risks to study participants and few
direct benefits, but may require considerable time and
burden for participation. Study requirements can include
multiple in-home visits; the burden of wearing personal
air monitors for one or more 24-hour period; preparing
and providing duplicate diet samples; collection of
environmental samples inside and outside the home;
completing questionnaires, food diaries, and time-
activity diaries; and providing urine, blood, saliva, or
hair samples. Monetary payments often have been
included in these studies, with the level of payment
related to the number of study days or visits or the
specific kinds of environmental and biological samples
and information that are collected or provided. Payment
for direct participant costs has been included in some
studies, such as a reasonable payment for providing
researchers with duplicate diet samples.
NERL scientists should review the commentary and
recommendations in the literature before devising a
payment program as part of a research protocol,
especially the two recent National Academies of Science
documents, Ethical Issues in Housing-Related Health
Hazard Research Involving Children (NRC and IOM,
2005) and Ethical Conduct of Clinical Research
Involving Children (IOM, 2004). They should seek
guidance from EPA's HSRRO to determine EPA's latest
policies and guidance in this regard. Input also should be
sought from community representatives to ensure that
any payment is adequate to compensate for expenses and
reward participation, but that the payment is not so high
as to constitute undue influence or coercion in the
community. If the study includes several followup visits
over a long term, NERL researchers should ensure that
payment is made incrementally as the NRC & IOM
committee suggested. NERL scientists also should adopt
the IOM Recommendations 6.1 and 6.2, including
ensuring that any payment should be for appropriate
purposes and age-appropriate, and that the process
should be open and fully disclosed, while not overly
emphasizing payments during the recruiting or informed
consent phases. The final decisions about the ethics of
payments rest with the IRB, which will review, modify
as needed, and approve the research protocol, and with
the EPA HSRRO, who has final authority to approve,
modify, or disapprove all of NERL's human subjects
research efforts.
5.3 Research Rights and Grievance
Procedures
Protecting the research rights of participants and
providing independent access to information regarding
those rights and to grievance procedures is an important
element in developing and maintaining appropriate
participant-investigator relationships. As part of the
informed consent process, the Common Rule requires
"An explanation of whom to contact for answers to
pertinent questions about the research and human
subjects' rights, and whom to contact in the event of a
research-related injury to the subject" [40 CFR
26.116(a)(7)].
Information about the research often best can be
answered by the researcher. However, it may benefit
researchers and participants if information about the
research can be obtained from or confirmed by a trusted
independent person or organization. Participants also
need to know how they can contact someone,
independent from the researcher, who can answer
questions concerning the rights of research participants
and provide information on grievance procedures and
research-related injuries. These questions could be
addressed to the IRB, an ombudsman, an ethics
committee, or other knowledgeable administrative body.
Consent documents are expected to have at least two
names with appropriate telephone contact
information—one that can provide information
regarding the research and another that can provide
information regarding their rights as research
participants. Grievance procedures should be structured
so that grievances reach the approving IRBs and
sponsoring organizations.
5.3.1 Ombudsman
An ombudsman is a neutral independent advocate
for research participants (and their families or guardians,
where applicable). Institutions and IRBs may
recommend or require the use of an ombudsman in
certain types of research studies, particularly those
seeking to study vulnerable populations. Ombudsmen
can fill several roles as participant advocates. They may
be an independent source of information regarding the
study. They may be present during the informed consent
process to ensure that risks, benefits, and study
requirements are communicated correctly and
understood by potential participants or their guardians.
An ombudsman may be used in studies involving
prisoners or military personnel to ensure that there is no
coercion to participate. And the ombudsman may
61
-------
communicate problems or grievances raised by research
participants to the 1KB and sponsoring organization.
5.3.2 Community Advisory Board
CABs can help ensure that participant rights are
considered and addressed during the study design and
can play an important role in monitoring the research
process. Community members may choose to seek
information about the study from the advisory panel, as
an independent entity, before deciding whether to enroll.
Representatives from such advisory boards can be
included in the research team that designs the study (see
Section 2.3). The role of CABs is more fully discussed
in Section 6.
5.4 Creating a Supportive Environment for
Research and Interaction
It is recommended that researchers and institutions
strive to create a supportive environment for research
and interaction with research participants and
communities. At the personal level, this means
researchers building trust with individuals and treating
them with respect. Following the IOM recommendations
about the informed consent process—that it "should be
an on-going, interactive dialogue between research staff
and research participants involving the disclosure and
exchange of relevant information, discussion of that
information, and assessment of the individual's
understanding of the discussion" (Recommendation 4.1,
IOM, 2002)—should go a long way in establishing a
supportive environment with the individual participants.
At the community level, engagement of the community
throughout the design, conduct of the study, and follow-
up will support trust-building and positive interactions.
Developing and providing this kind of support can be
challenging in large-scale studies, and particularly those
that cross communities or are conducted across large
geographic areas. Institutions need to recognize the need
for, and value of creating supportive research
environments by providing adequate funding because
effective interaction takes considerable time and effort.
Many of the factors that create a supportive
environment for research participants are described in
the Report and Recommendations on Public Trust in
Clinical Research for the NIH Director from the
Director's Council of Public Representatives (COPR)
(NIH, 2005). Although the advice from this workshop
was developed in the context of NIH-supported clinical
research, many of the recommendations are applicable to
observational human exposure studies and human
subject research in general. A summary of
recommendations from the report for enhancing public
trust is provided in Appendix D. The recommendations
are focused on the following areas.
• Building trust through community partnerships
• Building relationships with patients (participants)
(True partnerships with patients may not be possible,
but bidirectional relationships must be enhanced.)
• Building partnerships with community providers
• Building trust in scientists
• Building trust in the (EPA) and scientific research.
5.5 Recruitment Strategies
Many strategies are used to select and recruit people
into research studies requiring human participation. The
IRB is responsible for reviewing the selection process to
ensure that it is, above all, equitable. The requirement
for IRB review is stated in 40 CFR 26.11 l(a)3.
Selection of subjects is equitable. In making this
assessment, the IRB should take into account the
purposes of the research and the setting in which the
research will be conducted and should be particularly
cognizant of the special problems of research
involving vulnerable populations, such as children,
prisoners, pregnant women, mentally disabled
persons, or economically or educationally
disadvantaged persons.
The IRB guidebook is an excellent resource for
consideration of concerns and elements for equitable
participant selection (U.S. HHS, 1993). It states that
"Defining the appropriate group of subjects for a
research project involves a variety of factors—
requirements of scientific design, susceptibility to risk,
likelihood of benefit, practicability, and considerations
of fairness." The IRB guidebook raises a number of
points to consider in the process for selection of human
participants (see Text Box 5-5).
Various participant recruitment strategies may be
used depending on the type of research being performed
and the population of interest. This section addresses the
strategies and approaches for identifying and contacting
people and subsequent recruitment into a research study.
Sampling design approaches and issues, such as
statistical issues regarding representative and nonrandom
sampling designs, oversampling of subpopulations, and
environmental justice considerations are part of the
study design process described in Section 2 and are
critical for deciding which recruitment approaches will
be used.
62
-------
Text Box 5-5. IRB Guidebook Issues on
Identifying Subjects
1. Who will bear the burden? Who will reap the benefits?
2. Is there a disproportionate burden on any single group?
3. Is the proposed subject population required or justified?
4. Are there susceptible groups of people who should be excluded
from the research?
5. Are anticipated benefits distributed fairly? Do others have a
greater need to receive any of the anticipated benefits?
6. Are the research burdens distributed fairly?
7. Will any special physiological, psychological, or social
characteristics of the subject group pose special risks for them?
8. Would it be possible to conduct the study with other, less
vulnerable subjects?
9. Has the selection process overprotected potential subjects who
are considered vulnerable (e.g., children, cognitively impaired,
economically or educationally disadvantaged persons, patients
of researchers, seriously ill persons), so that they are denied
opportunities to participate in research?
10. If the subjects are susceptible to pressures, are there
mechanisms to reduce the pressures or minimize their impact?
Some of the common approaches for identifying and
making initial contact with potential participants include,
but are not limited to
• direct telephone or in-person contact with a person
selected through a statistical sampling process to
obtain a representative sample of the population being
studied;
• use of print or other media advertisements, often used
to recruit people in a community with specific
characteristics;
• advertisement or word-of-mouth contacts through
community groups, civic organizations, or other types
of organizations; and
• recruitment at physicians" offices, hospitals, and
clinics or at churches, schools, or other social
institutions, either in person or through the use of
advertisements or study brochures.
CABs can be consulted regarding proposed
approaches for recruitment in community-based
research. All procedures and materials for participant
recruitment are reviewed and approved by the IRB prior
to implementation. Some of the materials prepared for
recruitment might include the following.
• Recruitment scripts—prepared scripts used for in-
person or telephone study information and recruitment
contacts
• Printed materials—brochures, flyers. newspaper
advertisements, letters, and information articles
• Audio/visual materials—radio and television scripts,
video segments, public service announcements
• Internet postings—study announcements and
information, links to study materials, links to related
information.
The IRB reviews all recruitment material to ensure
that it does not adversely affect the informed consent
process, is consistent with the study protocol, and is
likely to result in equitable participant selection. IRBs
will carefully consider how information regarding
payment for participation is presented to potential
participants so as not to create undue influence.
Participant recruitment may be performed directly
by the researcher or staff members of the researcher's
organization, or other individuals or organizations may
be asked to recruit or make initial informational contacts
with potential participants. All persons involved in
recruiting must adhere to the procedures and materials
approved by the IRB. It is recommended that sponsoring
organizations should not pay recruiters on a per-
individual basis to minimize the likelihood that
individual recruiters will put undue pressure on potential
participants to enroll.
5.6 Retention Strategies
Some observational human exposure studies require
only a single visit or a single set of visits with a
participant over a relatively short time period (e.g., 24
hours or 1 week). Other studies may involve repeated
interaction with participants over longer periods of time.
Longitudinal study designs require retention strategies
mat ensure that adequate sample sizes are maintained for
meeting study objectives. It is recommended that
researchers and IRBs evaluate the level of burden in
longitudinal studies and ensure that retention strategies
are not likely to create conditions of coercion or undue
influence.
Some of the common strategies for maintaining high
retention rates in longitudinal studies are listed in Text
Box 5-6.
Text Box 5-6. Common Strategies for Maintaining High
Retention Rates in Longitudinal Studies
• Developing and maintaining a strong study identity
• Building participant trust
• Communicating regularly with participants
• Providing feedback that is of use to participants
• Maintaining confidentiality
• Incorporating active participant tracking mechanisms
• Maintaining reasonable levels of burden
• Providing periodic tokens of appreciation
• Providing reasonable levels of payment at each time point,
sometimes including escalating payments or a higher final
payment for completion of all study activities
63
-------
It is important that strategies that use payments to
encourage retention should be carefully scrutinized
against the possibility that they will result in undue
influence or diminish voluntary participation. Payments
that cover expenses and for time and burden at each visit
have to be reasonable, and researchers and IRBs should
consider whether the cumulative level of payments over
time or the use of escalating payments or final bonus
payments might present undue influence on
decisionmaking regarding participation. Participants
have to feel capable of withdrawing from participation at
any time, and escalating payments or completion
bonuses can impact decisions to withdraw. Withholding
all payment until all study visits are completed or
making payment contingent on completing all activities
is not an acceptable practice in most longitudinal studies
because it can diminish the capacity for voluntary
participation. (See the discussion about payment issues
in long-term studies in Section 5.2.3.)
People are more likely to continue active
participation in longitudinal studies when they believe
that the research is important, and that they are making a
valuable contribution, are receiving regular feedback,
and are treated with courtesy and respect by researchers.
Observational human exposure studies sometimes
involve substantial burdens of time and effort. Over long
periods, this level of burden can reduce retention. It may
be necessary to develop novel methods that reduce
participant time and effort or to focus the study design so
that fewer study procedures are implemented at any time
point. Because the time needed to analyze samples,
verify results, and perform data analyses can be long, it
may be difficult to provide timely feedback to
participants in observational human exposure studies.
Researchers might consider including simple measures
that can provide immediate and useful information of
value to participants to encourage continued
participation. Effective use of these strategies will
reduce the need for higher payments to encourage
retention.
5.7 Ensuring Recruitment or Retention
Methods Will Not Lead to Unacceptable
Risk
Researchers and IRBs need to ensure that the
procedures and materials used to recruit and retain study
participants in observational human exposure studies do
not "undermine free and informed consent by leading
parents to expose their children to unacceptable risks.
(NRC and IOM, 2005)" Payments in these studies
should not be so high that they would cause an undue
inducement for a participant to use a product they would
not normally use or to perform an activity that they
would not normally perform. Not only would this bias
the study results but may lead to higher than normal
levels of exposure. Alternatively, the act of studying one
set of conditions or activities in an observational human
exposure study could lead participants to assume that
those conditions or activities involve substantial risk. In
response, they subsequently may change their activities
in ways that could lead to possibly higher risks. The
potential for such unintentional outcomes is difficult for
researchers to gauge but requires researcher caution in
how information and results are conveyed. However, if
the informed consent process is truly "an on-going,
interactive dialogue . . . involving the disclosure and
exchange of relevant information," then such
misunderstandings should be minimized.
References
AAP (American Academy of Pediatrics Committee on
Environmental Health) (2003). Pediatric and
Environmental Health. (2nd Edition). Elk Grove Village,
IL: American Academy of Pediatrics.
Ackerman TF (1989). An ethical framework for the practice of
paying research subjects. IRB 11(4): 1-4.
CFR (Code of Federal Regulations) (2006). 40 CFR Chapter I
Environmental Protection Agency Part 26 Protection of
Human Subjects. U.S. Code of Federal Regulations.
Available: http://www.access.gpo.gov/nara/cfr/waisidx_06/
40cfr26_06.html [accessed 12 June 2007].
CIOMS (The Council for International Organizations of
Medical Sciences) (2002). International Ethical Guidelines
for Biomedical Research Involving Human Subjects. World
Health Organization. Geneva, Switzerland. Available:
http://www.cioms.cli/frame guidelines nov_2002.htm
[accessed 12 June 2007].
Diekema DS (2005) "Payments for Participation of Children
in Research." Chapter in: Kodish, E, Editor. Ethics and
Research with Children: A Case-Based Approach. New
York, NY: Oxford University Press. 143-160.
Dickert N, Emanuel E, Grady C (2002). Paying research
subjects: an analysis of current policies. Ann Intern Med
136(5):368-373.
Emanuel EJ (2004). Ending concerns about undue
inducement. J'Law MedEthics 32(1): 100-105.
Emanuel, EJ, Wendler D, Grady C (2000). What makes
clinical research ethical? JAMA 283(20): 2701-2711.
64
-------
Erlen JA, Sauder RJ, Mellors MP (1999). Incentives in
research: ethical issues. OrthopNurs 18(2):84-87.
European Parliament (2001). On the approximation of the
laws, regulations, and administrative provisions of the
Member States relating to the implementation of good
clinical practice in the conduct of clinical trials on medicinal
products for human use. Directive 2001/20/EC Article 4(d)
of the European Parliament and of the Council of 4 April
2001.
Flory J, Emanuel E (2004). Interventions to improve research
participants' understanding in informed consent for
research: a systematic review. JAMA 292(13): 1593-1601.
Fry CL, Hitter A, Baldwin S, Bowen KJ, Gardiner P, Holt T,
JenkinsonR, Johnston J (2005). Paying research
participants: a study of current practices in Australia. JMed
Ethics 31:542-547.
Grady, C (2005) Payment of clinical research subjects. J Clin
Invest \ 15:1681-1687.
Grady C, Dickert N, Jawetz T, Gensler G, Emanuel E (2005).
An analysis of U.S. practices of paying research
participants. Contemp Clin Trials 26:365-375.
Iltis AS, DeVader S, Matsuo H (2006). Payments to children
and adolescents enrolled in research: a pilot study.
Pediatrics 118(4): 1546-1552.
IOM (Institute of Medicine) (2004). Ethical Conduct of
Clinical Research Involving Children. Marilyn J. Field and
Richard E. Behrman (eds). Washington, DC: The National
Academies Press.
IOM (Institute of Medicine) (2002). Responsible Research: A
Systems Approach to Protecting Research Participants.
Daniel D. Federman, Katlii E. Hanna, and Laura Lyman
Rodriguez (eds). Washington, DC: The National Academies
Press.
Menikoff J (2005). Of babies, bugs, and bombast: A look
behind the crash-and-burn of the CHEERS pesticide study.
MedRes Law Policy Rep 4(14):586-512.
NCS (National Children's Study) (2007) Research Plan,
Volume 1, Version 1.1, June 20, 2007. Available:
www.nationalcliildrenssmdv.gov/research/researchjlan/upl
oad/Research Plan Volume 1 .pdf [accessed 12 September
2007.].
NIH (National Institutes of Health) (2005). Report and
Recommendations on Public Trust in Clinical Research for
the NIH Director from the Director's Council of Public
Representatives (COPR). National Institutes of Health,
Director's Council of Public Representatives, January 14,
2005. Available: http://copr.nih.gov/reports/public trust.asp
[accessed 12 June 2007].
NRC (National Research Council) (2004) Research Priorities
for Airborne Particulate Matter: IV. Continuing Research
Progress. Washington, DC: The National Academies Press.
Available:
http://books.nap.edu/openbook.php?record_id=10957
[accessed 16 August 2007].
NRC & IOM (National Research Council and Institute of
Medicine) (2005). Ethical Considerations for Research on
Housing-Related Health Hazards Involving Children.
Washington, DC: The National Academies Press. Available:
http://books.nap.edu/catalog.php7record id=l 1450
[accessed 12 June 2007].
OMB (Office of Management and Budget). (2006a). Standards
and Guidelines for Statistical Surveys. Section 2.3, Data
Collection Methodology.
http://www.wliitehouse.gov/omb/inforeg/statpolicy/standard
s_stat_surveys.pdf
OMB (Office of Management and Budget). (2006b).
Questions and Answers When Designing Surveys for
Information Collection. Office of Information and
Regulatory Affairs. January, 2006.
http://www.whitehouse.gov/omb/inforeg/pmc survey guidanc
e 2006.pdf
Paasche-Orlow MK, Taylor HA, Brancati FL (2003).
Readability standards for informed-consent forms as
compared with actual readability. NEnglJMed 348(8):721-
6.
Russell ML, Moralejo DG, Burgess ED. (2000). Paying
research subjects: participants' perspectives. JMed Ethics
26(2): 126-30.
U.S. DHEW (U.S. Department of Health, Education, and
Welfare) (1979). The Belmont Report: Ethical Principles
and Guidelines for the Protection of Human Subjects of
Research. Washington, D.C.: National Commission for the
Protection of Human Subjects of Biomedical and
Behavioral Research. Available:
http://olisr.od.nih.gov/guidelines/belmont.html [accessed
June 2007].
U.S. FDA (U.S. Food and Drug Administration) (1998).
Information Sheets: Guidance for Institutional Review
Boards and Clinical Investigators 1998 Update. U.S. Food
and Drug Administration, Washington, DC. Available:
http://www.fda.gov/oc/olirt/irbs/toc4.html [accessedMay
2007].
U.S. HHS (U.S. Department of Health and Human Services)
(1993). Protecting Human Research Subjects: Institutional
65
-------
Review Board Guidebook. Office for Human Research Weise KL, Smith ML, Maschke KJ, Copeland L (2002).
Protections. Available: National practices regarding payment to research subjects
http://www.hhs.gov/ohrp/irb/irb guidebook.htm [accessed for participating in pediatric research. Pediatrics
12 June 2007]. 110(3):577-582.
VanderWalde A (2005). Undue inducement: the only Wendler D, Rackoff JE, Emanuel EJ, Grady C (2002). The
objection to payment? Am JBioethics 5(5):25-27. ethics of paying for children's participation in research. J
PediatrUl(2):l66-m.
66
-------
SECTION 6
Building and Maintaining Appropriate Community and
Stakeholder Relationships
Community engagement promotes active
community involvement in the processes that shape
research strategies and the conduct of research studies.
In developing this document, NERL held an expert panel
workshop to identify the content and organization of this
document (ERG, 2007). That panel of experts concluded
that the need to engage the community in observational
research was based on the ethical principles of (1)
respect for persons, which manifests itself in both a
respect for the individual and, through respect for the
community, their culture; (2) fairness, resulting in efforts
to assure equity in resources, burden, and benefits; and
(3) beneficence, including "empowering the community
to endure/'
Involving the community in the research effort can
improve the research both scientifically and ethically. In
the document, Ethical Considerations for Research on
Housing-Related Health Hazards Involving Children,
the joint NRC-IOM committee found that community
involvement was a "guiding theme" of their findings
(NRC & IOM, 2005). Because the researchers were
working in the homes and communities of the
participants, they faced issues that were different from a
clinical setting. They were challenged to think about the
fundamental ethical principles in the context of the
research setting and about how those ethical principles
should be interpreted in that setting. "When researchers
discuss a planned study with community representatives,
understand their concerns and needs, and respond to
them, protocols can be strengthened both scientifically
and ethically" (p. xii, NRC & IOM, 2005). Just as was
described in Section 5, where the informed consent
process was described as needing to be "an on-going,
interactive dialogue between research staff and research
participants involving the disclosure and exchange of
relevant information, discussion of that information, and
assessment of the individual's understanding of the
discussion" (Recommendation 4.1, IOM, 2004), so, too,
the process of community involvement should be a
process of effective two-way communication. These
NRC & IOM comments emphasize how critical
effective, bidirectional communication is to the scientific
and ethical foundation of a research study in such a
setting.
EPA has established a public involvement policy to
"improve the content of the Agency's decisions and
enhance the deliberative process" (U.S. EPA, 2003) (see
Text Box 6-1). The policy is focused largely on Agency
decisionmaking processes (e.g., rulemaking, permit
issuance, Superfund remediation, etc.), whereas observa-
tional human exposure studies are intended to provide
data to inform those decisionmaking efforts.
Nonetheless, the Agency policy may be helpful in
planning for community involvement in observational
human exposure studies. The policy is intended to
promote mutual trust and openness between EPA and the
public, to improve the quality of the Agency's actions,
and to promote the public's involvement in the Agency's
mission of promoting human health and the
environment. The policy identifies seven basic steps for
effective public involvement and offers guidance for
implementing public involvement at EPA.
Text Box 6-1. Seven Basic Steps for
Public Involvement at EPA
1. Plan and budget for public involvement activities.
2. Identify the interested and affected public.
3. Consider providing technical or financial assistance to
facilitate involvement.
4. Provide information and outreach to the public.
5. Conduct public consultation and involvement activities.
6. Review and use input, and provide feedback to the public.
7. Evaluate public involvement activities .
67
-------
6.1 Approaches to Community Involvement
Community involvement can take many forms. The
forms of community involvement are not mutually
exclusive, and researchers may use several approaches
for seeking community involvement. The nature and
extent of community involvement reasonably would
depend on the nature of the research itself and the
affected community. In Section 2, the authors discussed
some reasons for involving the community early in the
research planning and scoping process and the benefits
mat community involvement may bring to the research
effort. "Community residents can be involved in the
research process as research staff, through community
consultation and review, membership on community
advisory boards, and involvement in a community-based
participatory research process" if that is used (p. 83,
NRC & IOM, 2005). In addition, IRBs may seek
additional community representation on the 1KB panel.
The Centers for Disease Control and Prevention
(CDC) recommends nine governing principles for
engaging communities in health-related research (see
Text Box 6-2). Health research is a "privileged and
empowered activity in that the researchers have special
access to resources and sensitive information about
people and, through the analysis and presentation of
findings, are able to influence the way people think and
have considerable influence on decisions regarding the
allocation of resources" (CDC, 1997). Regardless of the
approach utilized to involve the community in research,
whether the process employs community-based
participatory research (CBPR) or another community
engagement method, researchers should guide their
interactions with community members using these
underlying principles to promote the aforementioned
ethical principles of respect of persons, fairness, and
beneficence.
One form of community involvement is to include
qualified members of the community on the research
staff. Section 2 advocates community representatives as
part of the research team. Paid research staff members
from the community could serve as valuable consultants
for protocol development and research design, including
how to collect the data, how to recruit and retain
participants, and how to interpret and disseminate the
results. However, including paid research staff from the
community may introduce a conflict of interest among
community members: community representatives may
feel a greater allegiance to the researchers providing the
payment and be less inclined to uphold the interests of
the community. In addition, the community may come to
view the paid research staff from the community as
'"outsiders." To help alleviate these potential issues,
researchers should ensure an equitable distribution of
paid research work among different communities within
the larger community as to not promote a perceived bias
among community members. Additionally, when
possible, researchers should make efforts to provide
payments to community members employed as research
staff through community partner organizations to
prevent conflict of interest issues. Refer to Section 5 for
further discussion on remuneration of community
members as research participants.
Text Box 6-2.
Requirement
Clarity
Knowledge of
the Community
Visibility
Acceptance
Partnership
Respect
Asset Utilization
Flexibility
Commitment
Community Engagement Principles for
Researchers
(From CDC, 1997)
Explanation
Clear communication of the study objectives,
research goals, and the populations or
communities of interest.
Familiarity with the economic conditions,
political structures, demographics, history, past
research experiences, and research perceptions
of the community.
Travel to the community, interact with formal
and informal leadership, and establish
relationships to build trust.
Acknowledge, without judging, the assets and
deficits of the community.
Balanced discussion and shared
decisionmaking among participants concerning
risks, responsibilities, expectations, benefits,
and investment.
Value the diversity of culture, history, beliefs and
opinions within the community for improved
understanding.
Identify and mobilize community assets to
improve scientific credibility of the interpretation
and dissemination of results.
Anticipate changes within the community in
regard to perceived benefits of research and
stakeholder interest and additional time and
resource needs.
Prepare to engage the community beyond the
constraints of the research projects, before and
after, to promote longevity of the relationship for
future research .
Furthermore, institutional and community partners
need to ensure that anyone involved as part of the
research team has the requisite research and social skills.
Researchers often place a greater emphasis on scientific
over social skills and may, in turn, assume that they have
68
-------
the necessary expertise to conduct research in
community settings. Likewise, research staff from the
community should have knowledge of research
fundamentals. Education and training targeted at both
the institutional researcher and the community
researcher may, respectively, improve the competence of
researchers to work with communities and the scientific
literacy of the public.
A recent NERL study, the Detroit Exposure and
Aerosol Research Study (DEARS), included community
members as paid members of the research team. The
community researchers were instrumental in recruiting
study participants from the seven study neighborhoods
in Detroit that required multiple door-to-door visits to
the homes of potential participants to discuss the
DEARS study. The success of DEARS was dependent
on researchers developing strong relations with
community leaders and State and local organizations.
A second approach to community involvement is to
seek community consultation and review. Researchers
may periodically meet with community residents in a
process of "engagement, dialogue, and feedback" (Dula,
1994) to discuss research plans, research progress, and
results. The objective is to seek a dialogue with
community residents to promote co-learning and asset
sharing between the researchers and the community.
Effective communication—open, honest, jargon free—is
imperative to the successful use of this approach.
Effective communication ensures that the community
has a voice in the research process for the expressed
purpose of increasing the community's trust and
engagement in the research, the applicability of the
research to the community's needs, and the benefit of the
research to the community. Authentic community
consultation embraces the "experiential knowledge of
the average citizen" (Leung et al., 2004). When
developing research budgets and timescales, researchers
who plan to incorporate community consultation and
review in the research process should take into account
the additional expenses and time necessary for
community marketing efforts and travel.
Corburn (2002) describes a successful community
participation in an EPA exposure assessment in
Brooklyn, NY. He explains how a shift of focus from
risk assessment to exposure assessment may provide an
opportunity for community engagement to improve the
technical assessment. Listed among the specific factors
integral to the assessment's success are the incorporation
of local, nonexpert information by community groups
during the consultation and review process that was used
to modify the conventional risk assessment process and
the inclusion of community-based organizations on the
scientific research team as contributors to the data
collection, modeling, and interpretation processes, where
they provided data and expertise not available through
traditional research frameworks (Corburn, 2002).
Community advisory boards (CABs) also have been
used as an approach for getting the community involved
in the research effort. A CAB could be formed to serve
as a liaison between participants and researchers. In
particular, CABs can advise the researchers about
community concerns and assist in the development of
materials to explain the study to potential participants.
CABs should be sufficiently large to ensure a diversity
of community views, perspectives, and attitudes.
Representatives from the board may be selected for
participation on the research team. In Section 5, it was
noted by the authors that such a board could function as
an oversight committee in case of any participant
grievances. According to Quinn (2004), the success of
CABs "lies in the ability of the researchers and CABs to
form a true partnership, enabling their different voices to
be heard equally." Creating a CAB may be perceived as
"window dressing" and, in fact, harm relationships with
the community if researchers are not truly open to
working with the CAB.
O'Fallon and Dearry (2002) describe how the Tribal
Efforts Against Lead (TEAL) partnership collaborated
with EPA to clean up and minimize exposure to toxins
from a Superfund site in Ottawa County, OK. The
TEAL project utilized the services of a CAB, which
included representatives from several local tribes, to
ensure that the research would be responsive to the needs
and concerns of the tribal residents. The CAB facilitated
the researchers interactions with the target communities
"by helping the investigators interpret data and distribute
information to the communities" and "developing and
conducting the training" of the community members on
risk reduction strategies. The value of the research to the
TEAL target communities would have been diminished
had a CAB not been formed to assist the researchers
with specific best practices to use when engaging the
tribal members and the appropriate tribal leadership
structures.
Another potential approach to involve the
community is to use a CBPR approach, wherein the
community is actively involved in each step of the
research process, including sharing of decisionmaking
power and resources. This will impact decisions about
study design, study methods, dissemination of findings,
and resulting actions. "Under the principles of
community-based participatory research, research must
69
-------
address the concerns, needs, and priorities of the
communities where it is conducted and lead to actions
and changes that benefit the community" (p. 86, NRC &
IOM, 2005). Information about CBPR approaches can
be found online at the HHS Web sites
http: //www. alirq. gov/clinic/epcsums/cbprsum .htm and
http: //www. ahrq. gov/re search/cbprrole .htm.
Israel et al. (2005a) reviewed the results of CBPR
efforts at six Children's Centers co-funded by EPA and
the National Institute of Environment Health Sciences.
They found that considerable commitment of resources
and time are needed for the approach to be successful,
and the translation of research findings into interventions
and policies is of the utmost importance. Community
partners played little role in defining the research topics
and data analysis, but were vital to disseminating the
findings to the community. Keeler et al. (2002) describe
using CBPR methods to evaluate personal and
community-level exposures to particulate matter among
asthmatic children in Detroit. The research partnership,
Community Action Against Asthma (CAAA), consisted
of representatives from local health organizations,
community environmental advocacy groups, State and
local governmental agencies, and academia. The CAAA
partnership credits community involvement as active
research partners in the research process with the success
of the project to acquire "more relevant exposure data
for the study of children in urban neighborhoods" and to
provide "immediate knowledge and understanding of the
outcomes and results of the combined environmental
health analysis to the communities" (Keeler et al., 2002).
There are, however, several drawbacks to utilizing
CBPR methods that researchers should consider before
developing a CBPR project. CBPR is time consuming
(develop partnership, establish and agree on research
aims and objectives, disseminate results to the
community using appropriate methods, and review
manuscripts and presentations) and not conducive to
situations where rapid decisions are necessary given a
tight timeline. Weighing the research need versus the
community's desire for an intervention is the greatest
source of tension in conducting CBPR (Israel et al.,
2005b).
One additional opportunity for community input
may involve participation on an IRB. IRBs are required
by the Common Rule to have members who are sensitive
to "community attitudes" [40 CFR 26.107(a)]. How they
meet this obligation is totally at their discretion, and
NERL researchers have no influence. There have been a
number of recent articles in the literature about IRBs that
have envisioned a need for more regulatory reform
(Ledford, 2007). Ideally, the IRB should take into
account the views of the community. Quinn (2004)
argues for extending protections now reserved for
individuals to groups (populations and communities)
through CABs. Her argument is that there are "ethical
issues related to research with communities that are
distinctly different from the ethical issues related to
research with individuals." CAB members have to be
educated on human subjects' protections, should
represent their communities honestly, and need to be
willing to interact with researchers on complex research
issues.
Gilbert (2006) goes even further. He suggests
supplementing or even replacing traditional IRBs with
environmental health and community review boards
(EHCRBs). He argues that traditional IRBs are
inadequate for the review of community-based research
because they were developed to address issues related to
individuals involved in research projects, not
communities. He proposes EHCRBs that combine the
fundamental and ethical concept of traditional IRBs with
an expanded ethical construct of dignity, veracity,
sustainability, and justice, with added emphasis on
community. He envisions that an EHCRB would
function as an IRB with the requirements and
responsibilities for review for the protection of human
subjects, plus the additional role for review of
community issues associated with the research project.
Gilbert's recommendation for EHCRBs is consistent
with the recommendations of the authors of the NRC &
IOM report who recommended that "Institutional review
boards that review housing health hazards research
involving children should ensure that those boards have
the necessary expertise to conduct a complete and
adequate review, including expertise on research
involving children and community perspectives" (NRC
& IOM, 2005).
Involving community representatives in the IRB
process is challenging for IRBs, however. One challenge
could be the need to provide sufficient training to
community members about the IRB process and the
regulations governing IRBs. This can be significant if
members sit on an IRB for a limited time to review
specific community-based studies. In some cases, IRBs
may invite community members to participate in the IRB
process as nonvoting members to solicit the community
perspective. This approach, which would be totally at the
discretion of the IRB, might reduce the burden on the
community representative by reducing training
requirements.
70
-------
6.2 Issues in Community Involvement
There are a number of issues that need to be
addressed in any efforts to ensure community
involvement. The Expert Panel that was convened to
advise NERL about scientific and ethical issues in
observational human exposure studies discussed a
number of challenges (ERG, 2007). The topics that the
expert panel identified as issues are discussed below.
6.2.1 Defining "Community"
"Community" refers to a group of people united by a
shared attribute, and the attributes can be wide-ranging,
such as geography, culture, social characteristics, values,
interests, traditions, or experiences (ERG, 2007).
Community can be defined broadly (as a system of
interrelated groups operating to meet the needs of its
members) or more narrowly (as the population from
which study participants are selected). For observational
field studies, the Expert Panel from the workshop
suggested the narrow definition. A narrow definition
allows social and cultural factors to be included but
excludes government agencies, industry, and others who
do not necessarily represent the interests of the
participants (ERG, 2007).
Central to the definition of a community is a sense
of "who is included and who is excluded from
membership" (NRC & IOM, 2005). A person may be a
member of a community by choice, as with voluntary
associations, or by virtue of their innate personal
characteristics, such as age, gender, race, or ethnicity
(NRC & IOM, 2005). As a result, individuals may
belong to multiple communities at any one time. When
initiating community engagement efforts, one should be
aware of these complex associations in deciding which
individuals to work with in the targeted community.
Quandt et al. (2001) discuss a CBPR research
project, Preventing Agricultural Chemical Exposure in
North Carolina Farmworkers, where the process of
defining a community was complicated by language,
ethnic and racial stereotypes, and lack of organization.
Many of the affected farmworkers originated outside the
United States from several different Spanish-speaking
countries and possessed contradicting viewpoints on
research and the utility of community organization.
Moreover, the community organization tailored for this
farmworker demographic did not include enough
members to adequately populate the study. The
researchers utilized multiple approaches, including
community forums, community advisory councils, and
public presentations, to identify a diverse, yet viable,
community within the broader farm worker population.
Through this process of using multiple participatory
strategies to define the community, a sense of
community was nurtured among the farmworkers
collectively (O'Fallen and Dearry, 2002).
Understanding and describing a community (CDC,
1997) involves exploring factors related to
• people (including socioeconomics and demographics,
health status, and cultural and ethnic characteristics),
• location (geographic boundaries),
• commonalities (including shared values, interests, and
motivating forces), and
• power relationships (including formal and informal
lines of authority and influence, stakeholder
relationships, and resource flows).
It is important to distinguish between stakeholders
and the community, but both should be engaged at some
point in the course of a study. Stakeholders are groups or
organizations that may affect, be affected by, or perceive
themselves to be affected by a decision or activity.
Stakeholders may have a direct or indirect interest in the
"matter" of interest. They may include individuals;
environmental, social, or community nongovernment
organizations (NGOs); government entities; businesses;
and industry. Stakeholders include business, industry,
and various levels of government. A critical difference
between the community and stakeholders is that the
community has a right to speak for its own interests, but
stakeholders cannot represent or speak for the
community. Although relationships with stakeholders
can, at times, be confrontational, stakeholders often
provide useful information and expertise. When
stakeholders and the community members overlap in
particular individuals, it is important to distinguish the
role in which the individual is acting (ERG, 2007).
6.2.2 Identifying Who Represents the
Community
To sufficiently represent the community, an
individual has to have not only the right to speak for the
community's interests (a right afforded by legitimate
membership in the group) but also should be able to
describe those interests on behalf of the community.
Identifying those who represent the community is not
simply a matter of identifying the most vocal activists
because those individuals do not necessarily represent
the interests of the entire community. In fact, several
individuals may be necessary to adequately represent the
diversity of viewpoints within a community; in such
cases, a CAB may be appropriate (ERG, 2007). One of
the researcher's first steps should be asking the potential
participants from the community who they see as a
71
-------
legitimate representative (i.e., someone who can speak
for them). Corburn cites an example of a locale in
Brooklyn, NY, that contained individuals with widely
different backgrounds. It was impossible to identify
appropriate spokespeople, or even to define the nature of
the community, without talking with community
members (Corburn, 2007).
The NRC & IOM (2005) Report also discusses the
issue of who can represent the identified community.
Some communities may have a formal governmental
structure and a recognized political authority (e.g.,
Native American tribes). Other communities may have
clearly identifiable leaders (e.g., religious communities),
whereas still other communities have no formal
leadership structure at all. Whether there is a legitimate
political authority or some other hierarchal leadership
structure, the goal is to seek community input as to who
best represents the interests of the community with
regard to the proposed research project, rather than
selecting those who are favorable to the research project.
The NRC & IOM report cautions against the ethically
questionable practice of seeking out population
spokespeople and research participants whose positive
response to a research plan can be predicted in advance
and refers the reader to an article on this topic by Juengst
(2000).
With multiple sources of leadership and authority in
many communities, careful consideration should be
given to what aspect of the community a particular
person will represent, and what efforts may be needed to
ensure that the entire range of views in a community are
obtained. Researchers should consider reaching out to
multiple organizations such as churches, social service
agencies, community-based organizations, and tenant
and other community advocacy groups.
6.2.3 Building Relationships and Trust
A key first step in developing trust is to establish a
relationship with the community before the study. Trust
must be built; it cannot be assumed. This relationship
involves not only listening to community input but
actually taking it into consideration (ERG, 2007). A long
history of research with no direct benefits and no
feedback of results to the community, however, has
contributed to a general mistrust of researchers by
community members (Israel et al., 1998). Moreover, the
recurring abuse of trust in communities is a reality mat
researchers should be aware of when attempting to build
a long-term relationship (Minkler and Wallerstein,
2003). Past ethical failures have created distrust among
some communities and have produced great challenges
for current community organizers. Although it may seem
self-evident, researchers need to remember that ethical
action, during all phases of the research, is necessary for
developing and maintaining the trust of communities
(Perkins and Wandersman, 1990; CDC, 1997).
Developing trust is a difficult and time-consuming
process. Israel et al. (2005b) suggest a number of ways
partners can gain each other's trust. First, partners can
show respect by seriously considering the ideas and
opinions of others. Second, trustworthiness can be
demonstrated by following through with those things
that each partner commits to. Third, partners have to
respect confidentiality. Fourth, they recommend
attending to each other's interests and needs by
participating in activities beyond the specific work of the
partnership.23 A history of prior positive working
relationships is also beneficial (Israel et al., 1998).
Taist cannot be separated from respect. Potential
participants need to see researchers fostering respect for
community members and leaders to gain trust. For
example, meeting with key community leaders and
groups in their surroundings helps to build trust for a
taie partnership. Such meetings provide organizers of
engagement activities with more information about the
community, its concerns, and factors that will facilitate
and constrain participation. Once a successful rapport is
established, the meetings and exchanges with
community members can become an ongoing and
substantive partnership (ERG, 2007).
One mechanism for helping to build trust may be a
contract with the community. A community contract
outlines the roles and expectations of both the researcher
and the community. Living up to these agreements
builds trust with all partners, and the establishment of
the agreement helps reduce misunderstandings.
Contracts or memorandums of understanding that outline
the roles and expectations of the researcher and the
community are discussed in both Minkler and
Wallerstein (2003) and Israel et al. (2005b). An example
outlining expectations in a partnership with tribal
communities is presented in Appendix E of Minkler and
Wallerstein (2003), whereas an example discussing
access to data and authorship issues is presented in
Appendix I of Israel et al. (2005b). An example of a
memorandum of understanding between the University
of Michigan School of Public Health, Detroitefs
Working for Environmental Justice, the Detroit Hispanic
23 For a more detailed description of each of the suggestions
for enhancing trust please see Chapter 3 of Israel et al. (2005).
72
-------
Development Corporation, and the Warren Conner
Development Coalition for a study investigating asthma
is available at http://depts .Washington .edu/ccph/pdf files
/MOUlO.pdf.
Work within communities involves a considerable
investment of researchers' and residents' time. It should
be an ongoing, interactive exchange of information and
ideas between the researchers and the community
members, where voices are both heard and honored.
Trust is fostered when all interested parties feel that they
have influence, and that their input contributes to the
community effort. The collaborations should be
inclusive of the entire community, including those
members with incompatible interests and perceptions. If
participation, influence, and benefits are limited only to
some of the partners, then distrust is likely, and the
potential benefits of community involvement may be
lost. Being inclusive can create some organizing
challenges, but the benefit of effective community
involvement "has the potential to lead to greater
understanding of community perspectives of the risk and
benefits of research, improve informed consent, increase
study enrollment, enhance data validity and quality, and
build trust for research" (NRC & IOM, 2005).
6.2.4 Importance of Language and
Communications with the Community
Even when all partners and community members are
speaking the same language, some terms are not
necessarily understood by all. Communications with
participants should be reviewed by all partners to ensure
that the language used will be appropriate for all
participants. At times, one method to communicate
research findings will not fit all community members
and partners. Even among the partners, understanding
each other's meanings is essential, so that all partners
can move forward with a common understanding (Israel
etal.,2005b).
Furthermore, Minkler and Wallerstein (2003) note
mat "research must be produced, interpreted, and
disseminated to community members in clear, useful,
and respectful language." Researchers, and especially
researchers in a government agency, may have their own
distinct lexicon. Researchers should be careful to avoid
acronyms, jargon, or technical terms that may obscure
the meaning or intimidate participants who are not
familiar with the terms. Communicating in "plain
language" to "explain the research in an honest,
straightforward way" will help build a strong
relationship with the community and the participants and
also help enhance public trust (Recommendation 11,
NIH, 2005).
Culturally sensitive communication is necessary to
developing effective research partnerships with
communities. To develop effective communications,
researchers must understand key aspects of the cultures
influencing the intended audience and build that
understanding into the communication strategy (Tillman,
2002). The symbols, metaphors, visuals (including
clothing, jewelry, and hairstyles), types of actors,
language, and music used in communication materials
all convey culture. Discussions with community
members can assist researchers in identifying messages
and images that resonate across groups or suggesting
situations in which different messages or images are
likely to work best.
As discussed above and in Sections 5 and 7,
communication materials must be tailored for each
individual community and must be written in a language
and at a reading level that will ensure comprehension.
Many IRBs require materials to be prepared at a 5th
grade reading level. Researchers should evaluate the
reading level of all data collection instruments and study
communication materials and should objectively
measure comprehension in pretests prior to use in the
study. Similarly, if translations of materials are required,
certified translating services should be used to ensure
accuracy and comprehension.
6.2.5 Recognizing and Addressing
Cultural Differences
Building and maintaining appropriate community
and stakeholder relationships requires acknowledgment
of the diversity within communities with regard to many
factors, including, but not limited to, race/ethnicity,
religious beliefs, heritage, and lifestyles. Different
groups in the study area may have different cultural
norms and practices. The researchers should take these
issues into consideration as they work in the community.
Community partners can help researchers design the
study to be attentive to the increasing heterogeneity of
community groups (Minkler and Wallerstein, 2003,
Chapter 4) and to the different boundaries of privacy
(crucial when designing sampling strategies) of different
groups (Israel et al., 2005b, Chapter 11).
Vega (1992) provides a thorough discussion of the
theoretical and pragmatic implications of cultural
diversity for community research and explains that
researchers should interact with communities using
methods that promote understanding and demonstrate
sensitivity and competence in working with diverse
73
-------
cultures (e.g., with respect to class, gender, ethnicity,
race, age, and sexual orientation). To aid in this process,
researchers should include sufficient time in their project
timeline to interact and dialogue with the community
before the study begins to understand the cultural issues
that may affect the research. Researchers and the results
of their work are expected to promote a strengthening of
the community; however, it should be recognized mat,
given the heterogeneity and the diverse views within a
community, the study findings potentially may conflict
with the desires of the community or may promote
feelings of anger or distrust among members of the
community with each other. Enlisting the services of a
third-party evaluator/mediator may be useful for
sustaining positive relations between all research
participants and the community at large.
The Research Triangle Park Particulate Matter Panel
Study (Williams et al., 2003), a NERL investigation of
PM and related pollutants involving African-Americans
in North Carolina, demonstrated an effective strategy for
using communication to address cultural differences
between the community and research personnel. Before
beginning the research, the study design included time
and a budget for building collaborations with
organizations having close ties with the African-
American community to establish trust between the
community and research team. Using input from the
community partner organizations, the researchers
developed a systematic communication plan to establish
rapport with the community and to guide interaction
between study participants and the key study personnel.
A well-designed and culturally sensitive communication
plan was integral to the success of the study. Eighty
percent of the original participants completed the four-
season study (Williams et al., 2003).
6.2.6 Honesty, Power Relationships, and
Partnerships
The NRC & IOM report (NRC & IOM, 2005)
describes a relational paradigm that acknowledges that
research is part of a broader societal context, with the
conduct of research often mirroring a system in which
power is unequally and perhaps unfairly distributed. The
trust and mutual commitment required from the
researchers and the community are subject to the overall
power relations in society.24 The Expert Panel convened
to discuss the content of this document recognized that
24 A discussion of the evolution of theories on power relations, including the
contribution of feminism, poststructuralism, and postcolonialism, can be
found in Minker and Wallerstein (2003, Chapter 2).
the researchers have a variety of forms of power that
needed to be understood and acted on ethically (ERG,
2007). One form of power is resources, both funds and
access to resources and decisionmakers. Other forms of
power may be more subtle, including expertise, which
can intimidate or limit a participant's choices. Peer
pressure, fear of intimidation, expectations of benefits
from the research, and power to stigmatize the
community all, whether real or perceived, can influence
the relationship between the researcher and the
community. Many forms of power may be tipped toward
the researcher, but the community often has power in the
fonii of knowledge about the community mat can impact
the quality of the research effort. An ethical balance of
power can lead to benefits for all partners (ERG, 2007).
In describing principles in Methods in Community-
Based Participatory Research for Health, Israel et al.
(2005b) describe CBPR as facilitating "a collaborative,
equitable partnership in all phases of research, involving
an empowering and power-sharing process mat attends
to social inequalities." One way to address the inequities
is to ensure that the roles and responsibilities are
mutually acceptable to all parties. Researchers involved
in CBPR should recognize and address the inequalities,
thereby promoting trust, mutual respect, open
communication, information sharing, collaborative
decisionmaking, and resource sharing.
It is important that researchers recognize that
partnerships with the community are ongoing processes
that need to be monitored and maintained. As a study
progresses, the dynamics of the partnership may change
as roles and responsibilities for the partners change. All
partners need to be willing to make the investment of
time and resources necessary to maintain an effective
partnership.
6.2.7 Building a Lasting Infrastructure
Infrastructure is anything that builds the capacity of
the community by providing its members with skills and
resources. Infrastaicture building ideally occurs
throughout the project and should be included in the
overall plan (ERG, 2007). When involving the
community in the planning process, investigators need to
be forthright regarding funding limitations and research
expectations, such as publication and dissemination of
results. The community needs to be made aware of the
ephemeral nature of funding, even if it results in
apprehension toward involvement.
Researchers should communicate early on those
issues that will become important once the research has
been completed, such as sustainability. Frankness is
74
-------
required to cultivate the community's confidence and
expertise over time. Because so much time and
investment is involved in building an appropriate
relationship with the community, researchers may wish
to continue their relationship with the community even
after the study has ended. Researchers should remain
accessible for technical support related to the subject of
the research. Helping community members identify new
funding opportunities and assisting with the writing of
grant applications are two examples of potential
continued relationships. Many private sponsoring
institutions already recognize the importance of enduring
commitment and have used a variety of approaches,
often involving funding, to ensure that these
relationships are able to continue (ERG, 2007). The
challenge will be for universities and Federal agencies to
be able to establish similar funding mechanisms.
The objective of capacity building is to involve
members of the community in certain roles (e.g.,
performing interventions), training them to perform
some of the functions initially performed by the research
team. Certain research grants specifically support this
type of training. Training can be reciprocal, and allowing
the community to train the researchers (for example, in
cultural sensitivity) not only fosters respect but also can
lead to important new understanding.
Another important step is to formalize the
relationship between the community and the institution
conducting or sponsoring the research, not just between
the community and the individual researcher.
Institutional relationships can survive even if individual
researchers leave. Institutions may be reluctant to build
enduring relationships with communities if they do not
see long-term financial value in this investment.
Researchers may be able to get more support from their
institutions if they can document their successes (ERG,
2007).
6.3 Community Involvement in
Observational Human Exposure Studies
Observational human exposure studies, like those
conducted by NERL, benefit from community
involvement. Scientific rigor apart from ethical
considerations fails as good science. Community
involvement in research ensures mat ethical standards
are a priority in the study design and are pursued by the
researchers throughout the study. The form and extent of
community involvement will vary, depending on the
scope and utility of the research effort. The nature of the
community—the population from which the participants
are selected—often will vary considerably from one
study to the next, ranging from a small group involved in
a pilot study to a randomized, representative sample of
the whole population. As a result, the nature of the
community involvement also will depend on the
particulars of the study. The typical lack of direct benefit
from observational human exposure studies may mean
that many research efforts cannot meet all of the
principles of CBPR. Nonetheless, community
involvement should be included in observational human
exposure research efforts to the fullest extent possible.
As the NRC & IOM committee observes (p. 98, NRC &
IOM, 2005):
Community involvement, though time and resource
intensive, is a necessary and useful component of...
research with the potential to enhance trust and
increase the relevance of research to affected
communities. Thus, attention to the issues raised by
the community and consideration of the most
appropriate method of community involvement for a
given research project is warranted.
NERL researchers also should consider the
recommendations set forth in the NRC & IOM report
(Recommendation 5.1, p. 98, NRC & IOM, 2005) as
they develop their research plans and protocol.
Researchers...should describe in their protocols and
IRB submissions how they have involved and will
continue to involve the affected community in the
research project, justify the lack of such involvement,
and report how they have responded to any
community concerns.
Researchers and the members of the communities in
which they work should recognize, however that the
primary role of the researcher is to be an advocate for the
science, not an advocate for the community. Although
being an advocate for the community is desirable, and in
many projects achievable, it should be recognized that
resource constraints and potential conflicts of interest
may impact the researcher's ability to advocate for the
community, particularly if it involves regulator}'
activities of the Agency.
6.4 Identifying and Interacting with
Other Stakeholders
Like community involvement, stakeholder
involvement in a research study can take many forms.
Also like community involvement, researchers should
engage stakeholders in their studies early in the planning
stages. The relationship between the researchers and
various stakeholders should be maintained during the
study. How this relationship is maintained can be
expected to vary with different stakeholders and may
75
-------
change as the study progresses. Stakeholders can provide
useful information and perspective during the planning
and implementation of observational human exposure
studies.
Stakeholders may include individuals, NGOs,
businesses, industry, and various government entities or
agencies with jurisdiction over or interest in the
community. Stakeholders are a separate entity apart from
the community, although they may conduct business or
operate within the community or have a direct or indirect
interest in the community's activities. Even though they
are not able to speak for the community, stakeholders
may have knowledge of impacts and ideas about how to
interpret and use the results of proposed research studies.
Such knowledge may prove very helpful as part of the
research planning and scoping (ERG, 2007). Including a
variety of stakeholders in the planning process provides
insight that comes from reconciling the disparate
perspectives of different stakeholders.
Observational human exposure studies conducted or
supported by NERL may have many potential
stakeholders, both internally in EPA and outside of the
Agency. Internal stakeholders include the ORD Office of
the Science Advisor, the Office of Science Policy, other
laboratories and centers in ORD that may be interested
in the study and its results, program offices, and the
regional offices. Outside of the Agency, other Federal,
State, and local agencies may be stakeholders. For
example, CDC may be interested in biomonitoring
studies. State agencies will be stakeholders when
research is done in their state. Examples of NGOs that
may be external stakeholders include the Natural
Resources Defense Council, Environmental Defense,
American Lung Association, American Academy of
Pediatrics, American Chemistry Council, and literally
hundreds of other organizations with interests in
environmental or public health issues. Researchers
should identify potential stakeholders and communicate
with them early in the planning stages of a study if they
are determined to be appropriate. Identifying the
appropriate stakeholders who have a legitimate interest
in the study will be done on a study-by-study basis and
should be done in consultation with the research team,
the community representatives, and senior management.
In EPA, the Office of Public Affairs can assist in
identifying contacts in stakeholder groups.
The concept of "stakeholder" has been discussed in
management literature since the 1980s. Mitchell et al.
(1997) have developed an approach for identifying the
relevant stakeholders through an assessment of their
power, legitimacy, and urgency. Such an approach may
be useful for identifying stakeholders to be involved in
the research studies. In describing CBPR, Israel et al.
(2005b) discuss the need to examine the advantages and
disadvantages of extending membership beyond the
"community of identity" at the outset. For example, they
discuss the relative merits of including representatives of
the agricultural industry in a study of farmworkers
because of industry's possible role in policy change and
weigh their inclusion against the concerns that the true
voice of the farmworkers may not be heard under such
conditions. They also describe a possible solution of
creating separate partnership groups. O'Fallon and
Dearry (2002) explain the benefits of including diverse
stakeholders for the dissemination of results.
Successful interactions with stakeholders will
require effective communication strategies and
materials, as is discussed in the following section. After
relevant stakeholders have been identified, they should
be contacted to inform them of the proposed research
study and to determine their interest in the study. If the
stakeholders express an interest in participating, the
research team should develop a plan for interacting with
the stakeholders that includes roles and responsibilities,
activities, and timelines that are mutually agreeable to
the team, community representatives, and the
stakeholders. There needs to be a clear agreement on the
plan with all parties involved. Failure to have such an
agreement may lead to misunderstandings of the roles of
the stakeholders. As planning for the study proceeds, the
plan for stakeholder involvement should be updated to
reflect activities and timelines for longer term
engagement.
When developing relationships with stakeholders,
researchers also should ensure that participation of the
stakeholder in the study, regardless of level of
participation, does not result in actual or perceived
conflicts of interest. This should be addressed in the plan
and agreement for stakeholder involvement.
References
CDC (Centers for Disease Control and Prevention) (1997).
Principles of Community Engagement. ATSDR Committee
on Community Engagement. Available:
http ://www. cdc. gov/phppo/pce/ [Accessed Oct. 2007].
Corburn J (2007). Community knowledge in environmental
health science: co-producing policy expertise. Environ Sci
Policy W(2):l50-I6l.
Corburn J (2002). Environmental justice, local knowledge,
and risk: the discourse of a community-based cumulative
exposure assessment. Environ Mngt 29(4)A51-466.
76
-------
Dula A (1994). African American suspicion of the healthcare
system is justified: What can we do about it? Camb Q
Healthc Ethic 3(3):347-357.
ERG (Eastern Research Group) (2007). Report on the
Workshop to Discuss State-of-the-Science Approaches for
Observational Exposure Measurement Studies. Held in
Durham, North Carolina on November 28-29, 2006. Final
Report: January 25, 2007. (PB2007-108905).
Gilbert SG (2006). Supplementing the traditional institutional
review board with an environmental health and community
review board. Environ Health Per sped 114(10):1626-9.
IOM (Institute of Medicine) (2004). Ethical Conduct of
Clinical Research Involving Children. Marilyn J. Field and
Richard E. Behrman (eds). Washington, DC: The National
Academies Press.
Israel BA, Parker EA, Rowe Z, Salvatore A, Minkler M,
Lopez J, Butz A, Mosley A, Coates L, Lambert G, Potito
PA, Brenner B, Rivera M, Romero H, Thompson B,
Coronado G, Halstead S (2005a). Community-based
participatory research: lessons learned from the Centers for
Children's Environmental Health and Disease Prevention
Research. Environ Health Perspect 113(10):1463-71.
Israel BA, Eng E, Schulz AJ, Parker EA (eds) (2005b).
Methods in Community-Based Participatory Research for
Health. San Francisco, CA: Jossey-Bass.
Israel BA, Schulz AJ, Parker EA, Becker AB (1998). Review
of community-based research: assessing partnership
approaches to improve public health. Annu Rev Public
Health 19:173-202.
JuengstET (2000). Commentary: What "community review"
can and cannot do. J Law Med Ethics 28(l):52-54.
Keeler G, Dvonch T, Yip F, Parker E, Israel B, Marsik F,
Morishita M, Barres J, Robins T, Caldwell W, Sam M
(2002). Assessment of personal and community-level
exposures to paniculate matter among children with asthma
in Detroit, MI, as part of Community Action Against
Asthma. Environ Health Perspec 110(Suppl 2): 173-181.
Ledford H (2007). Human Subject Research: Trial and Error.
Nature 448(7153): 530-532.
Leung MW, Yen IH, Minkler M (2004). Community-based
Participatory Research: a promising approach for increasing
epidemiology's relevance in the 21st century. Int J
Epidemiol 33(3):499-506.
Minkler M, Wallerstein N (eds) (2003). Community-Based
Participatory Research for Health. San Francisco, CA:
Jossey-Bass.
Mitchell RK, Agle BR, Wood DJ (1997). Toward a theory of
stakeholder identification and salience: defining the
principle of who and what really counts. Acad Manage Rev
22(4):853-886.
NIH (National Institutes of Health) (2005). Report and
Recommendations on Public Trust in Clinical Research for
the NIH Director from the Director 's Council of Public
Representatives (COPR). National Institutes of Health,
Director's Council of Public Representatives, January 14,
2005. Available: http ://copr. nih. gov/reports/public trust.asp
[accessed June 2007].
NRC & IOM (National Research Council and Institute of
Medicine) (2005). Ethical Considerations for Research on
Housing-Related Health Hazards Involving Children.
Washington, DC: The National Academies Press. Available:
http://books.nap.edu/catalog.php7record id=l 1450
[accessed 12 June 2007].
O'Fallon LR, Dearry A (2002). Community -based partici-
patory research as a tool to advance environmental health
sciences. Environ Health Perspect 110 Suppl 2:155-9.
Perkins DD, Wandersman A (1990). You'll have to work to
overcome our suspicions: the benefits and pitfalls of
research with community organizations. Soc Policy
Quandt S, Arcury T, Pell A (2001). Something for everyone?
A community and academic partnership to address
farmworker pesticide exposure in North Carolina. Environ
Health Perspect 109 Suppl 3:435-441.
Quinn SC (2004). Ethics in public health research: protecting
human subjects: the role of community advisory boards. Am
J Public Health 94(6):918-22.
Tillman, L. (2002). Culturally sensitive research approaches:
an African-American perspective. Educ Researcher 3 1(9):3-12.
U.S. EPA (U.S. Environmental Protection Agency) (2003).
Public Involvement Policy of the U.S. Environmental
Protection Agency. EPA 233-B-03-022. Available:
http://www.epa. gov/publicinvolvement/pdf/policy2003 .pdf
[Accessed 26 Sep 2007].
Vega WA (1992). Theoretical and pragmatic implications of
cultural diversity for community research. Am J Commun
Psycho! 20(3):37 '5-91
Williams R, Suggs J, Rea A, Leovic K, Vette A, Croghan C,
Sheldon L, Rodes C, Thornburg J, Eijire A, Herbst M,
Sanders W (2003). The Research Triangle Park paniculate
matter panel study: PM mass concentration relationships.
Atmos Environ 37:5349-5363.
77
-------
Additional Information Resources
Blumenthal DS (2006). A community coalition board creates a
set of values for community-based research. Prev Chronic
Dis 3(1): 1-7. Available:
http://www.cdc.gov/pcd/issues/2006/jan/pdf/05_0068.pdf
[accessed 12 June 2007].
Coleman CH, Menikoff JA, Goldner JA, Dubler NN (2005).
The Ethics and Regulation of Research with Human
Subjects. Newark: LexisNexis.
Davies EF (2001). Early and meaningful community
involvement. U.S. Environmental Protection Agency, Office
of Solid Waste and Emergency Response. Washington, DC.
OSWER Directive No. 9230.0-99, October 12. Available:
http ://www. epa. gov/superfund/policv/pdfs/early .pdf
[accessed 12 June 2007].
NACCHO (National Association of County and City Health
Officials). Turning Point: Collaborating for a New Century
in Public Health, see
http://www.naccho.org/topics/iiifrastructure/TumingPoint.cf
m [Specifically, Fourteen Policy Principles for Advancing
Collaborative Activity Among and Between Tribal
Communities and Surrounding Jurisdictions.
http://arcliive.naccho.org/documents/TP-policv-
principles.pdf].
U.S. CDC (Centers for Disease Control and Prevention).
Preventing Chronic Disease [online serial] Available:
http://www.cdc.gov/pcd/.
U.S. EPA (U.S. Environmental Protection Agency) (2005).
Superfund Community Involvement Handbook.
[Specifically, chapters 1 through 3.] EPA 540-K-05-003,
Available:
http://www.epa.gov/superfund/community/cag/pdfs/ci_hand
book.pdf
U.S. EPA (U.S. Environmental Protection Agency) (1999).
Risk Assessment Guidance for Superfund: Volume I -
Human Health Evaluation Manual. Supplement to Part A.
Community Involvement in Superfund Risk Assessments.
EPA/540/R-98/042, OSWER 9285.7-01E-P, PB99-963303.
U.S. EPA (U.S. Environmental Protection Agency) (2000). El
Superfund Hoy Dia: La Estimation de Riesgos: Como
Lograr la Participation de La Communidad. U.S.
Environmental Protection Agency, Office of Solid Waste
and Emergency Response. Washington, DC. EPA 540-K-
99-005, OWSER 9200.2-26K, PB99-963230. Available:
http://www.epa.gov/superfund/community/today/pdfs/risk-
fs-0203.pdf
U.S. EPA (U.S. Environmental Protection Agency) (2001).
Improved Science-Based Environmental Stakeholder
Processes: A Commentary by the EPA Science Advisory
Board. EPA-SAB-EC-COM-01-006. EPA Science Advisory
Board. Washington, DC. Available:
http://yosemite.epa.gov/sab/sabproduct.nsf/CEE3F362FlAl
344E8525718E004EA078/$File/eecmO 1006_report_appna-
e.pdf
U.S. EPA (U.S. Environmental Protection Agency) (2003).
Framework for Cumulative Risk Assessment. Washington,
DC: United States Environmental Protection Agency.
EPA/600/P-02/001F. Available:
http://cfpub2.epa.gov/ncea/cfm/recordisplay.cfm?deid=54944.
U.S. EPA (U.S. Environmental Protection Agency) (2003).
Framework for Assessing Risks of Environmental Exposure
to Children. Washington, DC. Available:
http://cfpub2.epa.gov/ncea/cfm/recordisplay.cfm?deid=22521.
U.S. EPA (U.S. Environmental Protection Agency) (2005).
Guidance on Selecting Age Groups for Monitoring and
Assessing Childhood Exposures to Environmental
Contaminants. Risk Assessment Forum, Washington, D.C.
EPA/630/P-03/003F. Available:
htlp://cfpub.epa.gov/ncea/cfiri/recoidisplay.cfin?deid=146583.
U.S. EPA (U.S. Environmental Protection Agency) (2006).
Considerations for Developing Alternative Health Risk
Assessment Approaches for Addressing Multiple
Chemicals, Exposures, and Effects. External Review Draft.
71 FR 16306.
U.S. EPA (U.S. Environmental Protection Agency) (2006).
Paper on Tribal Issues Related to Tribal Traditional
Lifeways, Risk Assessment, and Health and Well Being:
Documenting What We've Heard. The National EPA-Tribal
Science Council. Available:
http://epa.gov/osp/tribes/tribal/TribalIssues.pdf.
U.S. EPA (U.S. Environmental Protection Agency) (2006).
EPA's Public Involvement Policy. Available:
http://www.epa.gov/piiblicinvolvement/policy2003/indexlitiri.
U.S. Food and Drag Administration (FDA). Exception from
informed consent requirements for emergency research. 21
CFR § 50.24
University' of Kansas Community Tool Box. Available:
http ://ctb .ku. edu/en/tablecontents/.
University of Washington Community-Campus Partners for
Health, Available:
http://depts.washington.edu/ccph/index.htm].
Veazie MA, Teufel-Shone NI, Silverman G, Connolly A,
Warne S, King B, et al. (2001). Building community
capacity in public health: The role of action-oriented
partnerships. JPublic Health Manag Pract 7(2):21-32.
78
-------
U.S. EPA, Office of Environmental Justice
Resources:
Executive Office of the President. (1997). Environmental
Justice: Guidance under the National Environmental Policy
Act. Council on Environmental Quality, pp. 7-17).
Available:
http://epa.gov/compliance/resources/policies/ej/ej guidance
nepa ceql297.pdf
NEJAC (National Environmental Justice Advisory Council)
(2006). Future mechanisms to enhance stakeholder
involvement and engagement to address environmental
justice: A letter report. Available:
http://epa.gov/compliance/resources/pubhcations/ej/nejac/st
akeholder-involv-9-27-06.pdf
NEJAC (National Environmental Justice Advisory Council)
(2004). Environmental justices and federal facilities:
Recommendations for improving stakeholder relations
between federal facilities and environmental justice
communities. Waste and Facility Siting Subcommittee,
Federal Facilities Working Group. Available:
http://www.epa.gOv/compliance/resources/publications/ej/n
eiac/ffwg-final-rpt-102504.pdf
NEJAC (National Environmental Justice Advisory Council)
(2004). Meaningful involvement and fair treatment by tribal
environmental regulatory programs. Indigenous Peoples
Subcommittee. Available:
http://epa.gov/compliance/resources/publications/ej/nejac/ip
s-final-report.pdf
NEJAC (National Environmental Justice Advisory Council)
(2000). Guide on consultation and collaboration with Indian
tribal governments and the public participation of
indigenous groups and tribal members in environmental
decision making. Indigenous Peoples Subcommittee.
Available:
http://www.lm. doe.gov/envjustice/pdf/ips_consultation_gu
ide.pdf
NEJAC (National Environmental Justice Advisory Council)
(2000). NEJAC May 200 Meeting Report. Environmental
Justice and Community-Based Health Model Discussion.
Available:
http ://epa. gov/compliance/resources/publications/ej/nei ac/co
mmunitv-based-health-recom-report.pdf
NEJAC (National Environmental Justice Advisory Council)
(2000). The model plan for public participation. The Public
Participation and Accountability Subcommittee. EPA-
300/K-00-001. Available:
http ://epa. gov/compliance/resources/publications/ej/nej ac/m
odel-public-part-plan.pdf
U.S. EPA (U.S. Environmental Protection Agency). (2008).
The Environmental Justice Strategic Enforcement
Assessment Tool (EJSEAT). Available:
http://www.epa.gov/compliance/resources/policies/ej/ej-
seathtml
U.S. EPA (U.S. Environmental Protection Agency). (2006).
Environmental Justice Collaborative Problem-Solving
Cooperative Agreement Program FACT SHEET. Office of
Environmental Justice. Available:
http://www.epa.gov/compliance/resources/pubhcations/ej/fa
ctsheets/fact-sheet-ej-cps-grants-6-l 3-06.pdf
U.S. EPA (U.S. Environmental Protection Agency). (2004).
Toolkit for assessing potential allegations of environmental
justice. (Section 3.3). Office of Environmental Justice.
EPA-300/R-04-002. Available:
http://epa.gov/compliance/resources/policies/ej/ej-
toolkit.pdf
U.S. EPA (U.S. Environmental Protection Agency). (1995).
The EPA's Environmental Justice Strategy. Available:
http://epa.gov/compliance/resources/policies/ej/ej_strategy_
1995.pdf
79
-------
-------
SECTION 7
Designing and Implementing Strategies for
Effective Communication
Successful implementation of observational human
exposure studies requires effective communications
between the researchers, study participants, community
representatives, community members, stakeholders, and
the public. The previous two sections established the
need for communications that are "on-going, interactive
dialogue...involving the disclosure and exchange of
relevant information, discussion of mat information, and
assessment of the individual's understanding of the
discussion'' (Recommendation 4.1, IOM, 2002). NIH
advocates "plain language" that explains the research "in
an honest, straightforward way" (Recommendation 11,
NIH, 2005). Indeed, strong relationships can be built
with participants, the community, and stakeholders only
if there are clear and effective communications between
the researchers and the community. The previous section
illustrates, also, mat effective communication is
bidirectional; it involves listening as well as "speaking."
The ethical value of respect for persons, including
respect for one another's autonomy and welfare,
demands mat researchers, participants, community
members, and stakeholders strive to establish effective
communications and to foster a relationship of trust and
respect. The researchers should make a commitment to
effective communications and make the appropriate
investment of time and resources to ensure that the
communications are at an appropriate level and are truly
effective. Researchers should regard communications as
intrinsic to the ethical basis for the study.
With the ethical basis for bidirectional
communication assumed as a given, and the need for
open and honest bidirectional communications having
been well established in the previous sections that
discussed relationships between the researchers and the
participant (Section 5) and the researchers and the
community (Section 6), this section discusses strategies
and tools that researchers may find useful in developing
effective communications. The focus in this section is
primarily from the perspective of "getting the word out,"
because that is the aspect of communication most under
the control of the researchers. Nonetheless, effective
communications will be bidirectional and involve
effective listening. Researchers should keep in mind that
it is as important to listen to the participants, community,
and other stakeholders as it is for the researchers to
provide them with information using the approaches
described in this section.
7.1 Communication Strategy and
Implementation Plan
Fundamental to achieving effective communications
are a communications strategy and implementation plan.
In general, the goal of the communication strategy and
plan is to clearly define how effective bidirectional
communications will be achieved in the study. Specific
goals should be developed based on the specifics of the
study design, the study population, the community, and
the stakeholders. The plan will describe who will be
involved in the communications, what communications
are required, and how the communications will be
performed. The communication strategy and
implementation plan should be developed early in the
planning stages of a study. The communication plan,
however, needs to be dynamic, with revisions and
updates occurring throughout the study and in
collaboration with the community and stakeholders.
Text Box 7-1 lists elements mat should be included
in a communication plan. The communication strategy
should be developed based on the goals of the study and
an understanding of the background, education,
attitudes, and opinions of the stakeholders and the
community that will be involved in the many different
aspects of the study from the initial conceptualization to
the final reporting of the study results.
81
-------
Text Box 7-1. Elements in a Communication Plan
• Background information description (overview) of the study,
relevant historical background information, statement of
communication needs, and identification of communication
opportunities and issues
• Purpose and goals of the communication strategy
• List of individuals and groups involved in the communications,
plus relevant demographics and other information to profile the
groups
• Strategy and approach for achieving the goals, including a
statement of the primary message to be conveyed and
descriptions of the communication channels
• Activities and materials to achieve the goals of specific
elements of the plan to be performed
• Timetable
• Roles and responsibilities
• Resources needed (budget)
• Measures of effectiveness
Careful planning is required to develop a
communication plan that will be effective. The research
team has to invest the time and resources necessary to
develop and implement the plan. They also should
recognize mat the communication plan is essential for
conducting the study and is just as important as the study
design, human subjects research protocol, or QAPP. The
observational study, if properly justified as described
earlier, provides a social and scientific value and brings
benefit to society (and perhaps the participants); it
should be a program that the researchers want to discuss
and explore with the public. The communication plan
and strategy provide the researchers with an opportunity
to create effective bidirectional communications with the
participants, community, and stakeholders. They are not
simply a way to "avoid problems" with the community,
stakeholders or the media nor only a plan for reacting to
"negative" feedback.
Recognizing that the communication strategy must
address the bidirectional nature of communications with
the study participants and the community to be effective,
it is important that the communication strategy and
implementation plan be developed in collaboration with
the community and other stakeholders. As discussed in
previous sections, members of the community in which
the study will be performed will provide valuable input
into the development of the communication strategy by
providing information about the community culture,
norms, attitudes, perceptions, etc. They will provide, not
only expertise about the community, but also
experiences and lessons learned about previous
communications activities in the community. It is
important to engage community representatives and
stakeholders early in the process of development of the
communication strategy, the implementation plan, and
the communication tools.
Researchers also may find it helpful to seek
guidance on how to communicate more effectively,
especially because that is not a routine part of their
training or experience. They may consult and learn from
communications specialists in their organization. In
addition, a wide variety1 of resources are available. For
example, the Federal Communicators Network (FCN)
(www.fcn.gov) has prepared a "communicators guide"
that offers advice on how to communicate—in plain
language, in easily digestible "chunks," and in a form
that will be used. They emphasize that "good
communication is difficult because it requires a lot of
effort, time, and patience" (FCN, 2001). Some tips from
the guide to help federal communicators get their point
across are listed in Text Box 7-2.
Text Box 7-2. Tips for Getting Your Point Across
Be prepared.
Be confident.
Stay focused on your conversation and your listener.
Maintain eye contact with your listeners.
Make sure your listeners are following you by asking them for
questions or feedback.
Do not lose your temper or get over-emotional.
Speak slowly and calmly; don't raise your voice.
Speak clearly and concisely.
Get to the point; do not ramble.
Be kind, compassionate, and empathetic.
Be honest; do not play games.
Be assertive but tactful
7.2 Individuals and Groups Involved in the
Communications
An effective communication plan will identify and
involve all of the individuals and relevant groups that
should be included in the communications efforts. When
conducting observational human exposure studies, this
list may be quite long. Although researchers may desire
to limit the number of individuals and groups involved
to keep the effort as simple and focused as possible, they
need to ensure that all potential stakeholders are
identified. The communication plan should identify all
groups, including community organizations and
stakeholders, involved in a study, their roles and
responsibilities in the study, how communications will
be developed with each group, and the timing of the
communications. It is likely that most studies will
involve the individuals, community groups and
stakeholders shown in Text Box 7-3.
82
-------
Text Box 7-3. Individuals and Groups
Involved in Communications
Principal investigator—the researcher with ultimate
responsibility for the study
Research team
Study participants
Third parties associated with study participants (e.g., spouse,
children, landlords)
Community representatives
Community members
Governments (local, State, and Federal)
Study institution management
Study sponsors or funding organization
Organizations with interest in the participants, the community,
or the research question
Stakeholders that may be impacted by the results of the study
The scientific community
Media
The general public
The study participants are a key group involved in
communications during a study and are easily identified.
Similarly, it is generally not difficult to identify the third
parties associated with the study participants, because
these groups must be identified when considering ethical
issues in the study and when developing the human
subjects research protocol. Research teams should ensure
mat the communication strategy includes the strategy
and approach for third-party communications.
As discussed in Section 6, it is critical that
community representatives are identified early in the
scoping and planning phase of the study. It is important
mat researchers are informed about the community in
which they will conduct the study and understand the
unique characteristics and culture of the community and
the potential study participants to develop effective
bidirectional communications. When possible,
researchers should identify other research organizations
who have worked in the community and attempt to
gather information from them on the nature of the
community, who represents the community, and what
communication strategies have been used previously in
the community. Understanding how the community
defines itself or thinks of itself is critical to establishing
effective communications.
Identification of all relevant stakeholder groups may
be more difficult. As defined in Section 6, the term
stakeholder is used here to identity- a person or group
who has a valid interest in an activity or decision, but
who does not speak directly for the community or the
participants. There may be many organizations who
consider themselves as stakeholders that represent the
interests of the community, the participants, or the
research problem. For example, there are many nonprofit
organizations that advocate for the protection of
children's health. When conducting an observational
study involving children, the research team should
identify those groups mat could have an interest in the
study. They need to be identified in the communication
plan, and an approach needs to be developed for
communicating with them about the study. There are
many sources of information on potentially interested
stakeholder groups. This information can be obtained
from the research team based on similar studies, the
participants, the community representatives, sponsoring
organizations, and "umbrella" organizations for various
advocacy groups. The Internet has made identification of
the various stakeholder groups easier and is a source of
information on goals of the groups and contact
information. Approaches for communications with these
and other groups on the list are discussed further in the
following subsections.
Researchers should recognize that identifying and
involving all pertinent community and stakeholder
groups in their observational human exposure studies is
important for effective communications and the potential
success of the study, but that mere is the potential for the
group to become so large that it becomes difficult to
manage. Large groups not only have the potential for
impeding progress on a research effort simply because of
the logistics of the interactions within the group, but,
also, because of the potential for conflict between the
groups. It is essential, therefore, that researchers attempt
to maintain stakeholder groups of manageable size and
be prepared to deal with potential conflict within the
groups. There is extensive literature available on conflict
management, but it is outside of the scope of this
document to recommend specific approaches or
literature references.
7.3 Communications Timetables—When
To Communicate
Communications begin with the initial
conceptualization of the study and continue through the
reporting of the study results and beyond. Even after a
study has ended, followup communications may
continue with the study participants, the community, the
scientific community, and the public. It is beyond the
scope of this document to lay out timetables for
communications in observational human exposure
studies because timing will differ with each study. The
following discussion highlights a few of the issues
associated with the timing of communications to ensure
83
-------
that they are effective. This section also does not discuss
communications among the research team, research
organization, or study sponsors.
Researchers should begin the dialogue with the
community as soon as possible during study
conceptualization and planning. Once the community in
which the study will be performed is identified,
community representatives should be identified and
contacted to discuss the potential study and to get input
on how the study may be designed. As discussed in
Section 6, the observational human exposure studies
discussed in this document are generally not CBPR.
Although the study objectives or hypotheses have been
defined and the general approach has been developed,
the community still can provide valuable input about
their environmental or public health concerns. Again, as
discussed earlier, the planning for the study should be
flexible enough to incorporate community concerns
where feasible. Recognizing that many observational
human exposure studies will not be able to address all of
the community's concerns, it is important that the
communications with the community accurately convey
the value, merit, and benefits from the study that will be
relevant to the community.
Press releases can serve as useful tools for informing
communities about upcoming studies and for identifying
stakeholders. Assuming that community representatives
have been identified early in the development of the
study design and communications strategy, and that
community buy-in for the study has been gained,
researchers can work with community leaders and
community members to develop press releases and other
communication tools. Press releases can serve multiple
purposes. They provide information to potential
community representatives who may not have been
identified by the researchers as potential collaborators in
the study. They provide publicity mat will inform
community members about potential contacts by the
research team (e.g., in a random sample design). They
provide information to public interest and advocacy
groups who may feel that they are stakeholders who
should be involved in the study. Press releases also
provide the transparency for the study and the research
team that is essential for building trust.
Studies also should be announced to stakeholders
and the public (via the media, community interactions,
or other means) well in advance of study
implementation. Large grants expected to have
significant impact in communities often are announced
by EPA at the research institution receiving the grant
and in press releases to the local media. These studies,
therefore, are publicized at a very early stage.
Cooperative agreements, which are another mechanism
by which the government funds some research projects,
are announced in the same way. Cooperative agreements
and studies performed by EPA researchers receive
additional public notice when they are reviewed by the
Office of Management and Budget (OMB) under the
Paperwork Reduction Act. All studies involving
collection of survey information from more than nine
people are reviewed by OMB. This involves submission
of an Information Collection Request (ICR) to OMB,
announcement of the ICR in the Federal Register, and
an opportunity for public comment. A docket is
established specifically to facilitate public comment.
This process results in widespread publication of
upcoming government research studies through scrutiny
by concerned stakeholder groups who routinely review
the announcements in the Federal Register.
Prior to the start of data collection in a community,
relatively large-scale communications may be required.
These communications may involve notifications to
public safety officials about the presence of the research
team in the community, press releases to local media
outlets about the study, and meetings with community
groups to provide details about data collection activities.
It is essential that the research team fosters
continuing bidirectional communications with the
participants, community, and stakeholder groups
throughout the study to maintain transparency, trust, and
interest in the study. This can be aided by providing
project progress reports and interim results to the
participants and the community through community
meetings, mailings, or Web sites. In observational
human exposure studies with repeated measurements
over seasons or years, routinely meeting with
participants can serve as a mechanism for providing
study information to the participants and receiving
feedback from the participants and the community about
the study and their roles in it. Effective bidirectional
communication with participants can be expected to
improve their participation in the study (e.g., in
completing surveys and collecting personal samples) and
for improving retention in longitudinal studies.
However, it is essential that the researcher recognize the
implications of such meetings on privacy and
confidentiality issues for the participants and develop
communications to advise the participants of these
issues.
Community meetings also are effective for
maintaining communications throughout a study. They
provide the opportunity to disseminate information to
84
-------
community representatives and to obtain feedback. They
also provide an opportune setting for news releases to
the mediate maintain or increase interest in the study.
7.4 Communicating at Different Levels
The diversity of interested people and groups often
means that communications materials should be
developed at different levels of scientific literacy. In any
case, the materials should all be written in "plain
language" that is honest and straightforward. Stableford
and Mettger (2007) state that "plain language embodies
clear communication." Many researchers mistakenly
believe that the term means just using simple words, or
worse, "dumbing things down." It actually refers to
communications that engage and are accessible to the
intended audience.
A 1998 Presidential memorandum required mat
"plain language" be used in all governmental
communications with the public. That memorandum
stated, "By using plain language, we send a clear
message about what the Government is doing, what it
requires and what services it offers. Plain language saves
the Government and the private sector time, effort, and
money." Plain language is reader friendly and designed
to increase the participant's understanding of the
communication material. It serves as a means by which
lay audiences can access and understand scientific
information. Researchers desiring to use plain language
in their verbal and written communications with
communities must decide on key messages to include
and delete unnecessary descriptive, bureaucratic, or
jargon-filled language. Researchers should use words
that commonly are understood, rather than difficult
abstract terms and concepts. A friendly, conversational
tone is used to engage the lay audience, rather than a
formal, scholarly tone that distances the community.
Stableford and Mettger (2007) argue that proficiency in
creating appropriate plain language materials is an
acquired skill that requires knowledge and experience.
"It is both an art and a science, requiring the ability to
simultaneously think about the cognitive, emotional, and
visual appeal of the piece as well as applying research-
based strategies to ensure a truly easy-to-read and
understand print material."
The Department of Health and Human Services
(HHS) has developed a Web site that specifically
address the issues related to "plain language" (see
http: //www .health. gov/communication/literacv/plainlang
uage/PlainLanguage.htm). The site includes a list of
references and other helpful resources.
To maintain community engagement through the
research process, it is critical that communications are at
the appropriate level, and that materials are written at a
reading level that is appropriate to the audience. For the
nonscientist, many IRBs and other groups target
materials to be used with participants and communities
at a reading level no higher than the 8th grade to improve
the likelihood of comprehension. In some communities,
however, other factors, like primary languages other than
English, educational disadvantages, etc., may require
communications materials to be written in alternate
languages and at different reading levels. The issue is
comprehension, as was discussed in Section 5.1.2.
Empirical testing of communication methods and
content is essential to ensure comprehension (Health
Canada, 2006).
Researchers also should recognize that in this
information age, dissemination of informational
materials may be rapid and widespread. Therefore, even
documents intended for scientific peers may benefit by
including summary information in an executive
summary or preface mat a lay reader can understand.
7.5 Communications Materials
Researchers need to communicate clearly with the
many groups listed in Section 7.2 to develop their
relationship with the participants in the study, to develop
their partnership with the community, to gain support
from stakeholders, and to inform the public. To achieve
the multiple purposes of communications during a
research study and to communicate with many diverse
groups, a variety of communications materials may need
to be developed. Different materials have different
purposes and different types of information to be
communicated. Because of diversity in interested
individuals and groups with respect to education,
cultures, information needs, etc., the format and content
of communication materials likely will need to be
diverse. Text Box 7-4 list activities and materials that
may be helpful in facilitating communications.
By definition, communication is an exchange of
information. This has to be the primary goal of
communication activities. The accuracy and
completeness of the information transferred is important.
There are many different ways to communicate, the
effectiveness of which varies substantially. The way in
which the information is conveyed is as important as the
information itself. Effective communication should
promote trust and credibility. Peters et al. (1997) found
mat three determinants, namely, (1) knowledge and
expertise, (2) openness and honesty, and (3) concern and
care, were important factors determining perceptions of
85
-------
trust and credibility. Therefore, the approach to
communication in observational human exposure studies
should consider these factors, and communication
materials should be developed with these factors in
mind.
Text Box 7-4. Activities and Materials That
May Be Useful in a Communication Plan
Flyers
Web sites
Brochures
Interviews
Newsletters
Focus groups
Presentations
Direct mailings
Press releases
Questions and answers
Desk statements (government)
Abstracts
Study reports
Talking points
Community meetings
Stakeholder meetings
Technical presentations
Study participant meetings
Scientific meeting presentations
Peer-reviewed scientific journal manuscripts
Final reports describing the total research effort
When developing communication materials, the
researcher should consider the needs of the reader,
listener, or viewer with respect to content, scope, style,
and the level at which the materials are written. There
are many sources of information on design of
informational materials, such as flyers or brochures. For
example, Alderson (1995) provides an example of the
recommended content and style for information leaflets
(mat also may be flyers or brochures) for pediatric
medical research. She suggests that leaflets be provided
to parents of children who will be study participants that
can be read to the children. She recommends mat these
be provided at the time that the parent is being informed
of the study, prior to requesting the informed consent.
The content of the leaflet would include the following
topics.
• Nature and purpose of the research
• Anticipated benefits of the research
• Risks, harms, costs, and inconvenience to the
participant
• Assurance that the participant freely can refuse to
participate in or withdraw from the study
• Details about remuneration
• Names of the project sponsors and the researchers
• Contact information for the researchers
• Respect for privacy and confidentiality
Leaflets and brochures that contain this information
provide a tool for communication with study
participants. However, these materials need to be written
carefully using everyday terms that the average
nonresearcher can understand. The brochure should be
written in a friendly style that conveys the intent of the
researcher to engage the reader as a collaborator on the
study, not as a study "subject," who will be told to do a
series of tasks while participating in the study.
These same leaflets and brochures can be used to
inform other groups that either may be involved or
interested in the study, such as community
representatives, stakeholder organizations, the media,
and the general public. The researcher should ensure that
any brochure developed for the study includes accurate
and complete information that is less likely to be
misinterpreted by anyone who might pick up the
brochure. Brochures and flyers that are used to announce
a study or are used as recruiting tools should be carefully
written in plain language to ensure that there is not a
perception of activities that are unethical. For example,
if flyers announcing a study state that study participants
will be compensated, the flyer needs to ensure that the
compensation is not the focus of the flyer, and that the
payment does not appear to be excessive and coercive
(see Section 5.2). Flyers announcing a study generally
do not include the dollar amounts of payment.
Researchers need to have similar concerns about all
of the communication materials that are developed,
regardless of the type of material, whether it is a direct
mailing, a Web site, a news release, or a set of questions
and answers (Q&As) used to respond to media or
stakeholder inquiries. In developing the communication
materials, the research staff should seek the assistance,
advice, and input of people in their organization and
community groups with experience in developing such
materials. With all of these materials, the researchers
should be concerned with how the materials could be
misinterpreted, and whether mere could be a perception
that the study would not meet the highest scientific and
ethical standards. In this age of rapid communications
and increased accessibility to information, it is essential
that communication materials are well developed,
accurate, and understandable to all audiences that may
read them.
Research study Web sites are especially useful for
communicating information about observational human
exposure studies. Web sites should be developed early in
86
-------
the study to disseminate information to stakeholders and
the community. Additionally, the sites can be set up with
participant-only pages to provide more detailed
information to study participants, including information
on study protocols that require participant assistance
(e.g., protocols for collecting urine samples,
time/activity log entries). However, researchers need to
ensure that Internet access is available to their intended
audience and be prepared to provide alternative
communication tools for those without such access.
The plan for disseminating information from the
study should be developed in the early design phases of
the study and should be included in the study design
document. Sufficient resources, both time and funding,
need to be budgeted for this activity.
7.6 Informing the Study Participants and
Communities
Effective communications require mat all parties,
researchers and participants alike, involved in the
communication understand the content and context of
the information being exchanged. "When researchers
discuss a planned study with community representatives,
understand their concerns and needs, and respond to
them, protocols can be strengthened both scientifically
and ethically" (p. xii, NRC & IOM, 2005).
Comprehension is one of the key pillars of informed
consent, and it means that participants understand the
key elements of the research. The most effective way to
improve comprehension is by talking one-on-one with
study participants.
To accomplish that, the researchers need to make a
commitment to communicating with both the study
participants and the community to inform them of the
study and provide training as appropriate. This can
require a substantial investment of time and resources,
but it is critical to the success of the study.
Working with study participants to inform them
about the study and the scientific basis of the study will
have many benefits. The more educated the participant is
about the purpose of the study and the activities to be
performed during the study, the more likely the
participant will be to develop a beneficial researcher-
participant relationship. By taking the time to infonn the
participant, the researcher demonstrates his or her
commitment to the participant and conveys the
importance or value of their participation in the study. If
the researcher-participant relationship is well developed,
the participant will have a higher level of trust in the
researcher and will be likely to have more interest in the
study and a positive outcome. If such a relationship is
developed and the participant is informed about the
study goals, the participant will take part more readily
and effectively in the specific study activities. For
example, a study participant who understands why
time/activity information is critical to understanding
exposure is likely to do a better job completing a
time/activity log man a participant with no interest in the
outcome of the study. In addition, an informed
participant may have good suggestions for improving the
study and the interactions with the participants and the
community that the researchers should listen to and
adopt. Developing the researcher-participant relationship
and informing the participants also should improve
retention in longitudinal, repeated measures studies
because the participant feels that he or she is
collaborating with the researcher and is not merely a
study "subject."
Similarly, providing information on the research
study to the community should provide significant
benefits in terms of support to the research team and
working with the team to facilitate the study in their
community to address both the scientific issues and the
community's concerns. If community leaders understand
the research problem, the study goals, and the study
activities, they can more effectively articulate the
community's concerns to the researchers and integrate
those issues into the study design. This will enhance
their work with the research team during the design
phase and will enable them to more effectively advise
and assist during the implementation of the study.
7.7 Reporting Study Results to the
Participant and Community
Researchers need to develop the approach for
reporting results to the participants, community,
stakeholders, media, and others during the initial
planning of the study. There are not well-developed
guidelines for when and how to report study results
(Parkin, 2004). In her systematic review of guidelines
and frameworks for reporting study results, Parkin
determined that locating guidance may be difficult and
time consuming for researchers. She found agreement on
the importance of disseminating study results to produce
public health benefits, but there is not a consensus on
when and how results should be reported to either
communities or study participants. Although she did not
identify well-developed guidance documents, she did
identify some common themes. The first was that
researchers are becoming aware of the importance of
systematic planning of the research communications,
planning that needs to be done early in the study.
87
-------
Second, organizations are recognizing the importance of
communicating with communities. And, third, research
professions are recognizing the importance of research
communication and their responsibilities.
There is a large body of literature on processes for
risk communication (e.g., see Covello et al., 1989, 2007;
U.S. HHS, 2002; ASTDR, 2007). Processes for risk
communication are highly relevant to reporting results
from observational human exposure studies, even though
they may not include measurements of health outcomes
or risk assessments.
HHS has prepared a useful document entitled
Communication in a Crisis: Risk Communication
Guidelines for Public Officials, 2002. It is available
online and in hard copy and includes a chapter on
communicating complex, scientific, and technical
information (U.S. HHS, 2002). They recommend using
clear, nontechnical language, avoiding jargon, and
putting technical terms into frames of reference that the
public or other listeners can understand.
Covello and Allen (1988) described seven cardinal
rules for risk communication (Text Box 7-5) mat are still
quite applicable today. More recently, Covello et al.
(2007) have developed a "message mapping" approach
for risk communication. Message mapping is a process
to anticipate the questions likely to be asked after an
incident and to prepare clear and concise answers to the
anticipated questions in advance. The approach builds on
an understanding of current communications practices
(e.g., short messages averaging 27 words, soundbites of
around 9 seconds, the most frequently asked questions
after an incident) and typical human responses to crisis.
The report lays out a series of steps to develop short,
clear key messages to address stakeholder concerns in
advance.
Text Box 7-5. Seven Cardinal Rules of
Risk Communication
(Covello and Allen, 1988)
1. Accept and involve the public as a partner.
2. Plan carefully and evaluate your efforts.
3. Listen to the public's specific concerns.
4. Be honest, frank, and open.
5. Work with other credible sources.
6. Meet the needs of the media.
7. Speak clearly and with compassion.
The report also provides useful approaches for
effectively communicating the messages in times of
crisis. It emphasizes that during a crisis, "people judge
the messenger before the message and they base their
judgment in terms of trust." In times of crises, opinions
about trustworthiness hinge largely on perceptions of
caring and empathy, whereas competence and expertise
are key factors when there is little or no stress. Figure 7-
1 is taken from the Covello et al. (2007) report and
represents the relative importance of various factors in
influencing whether or not people trust a speaker in
times of crisis. Many of the principles and processes for
crisis communications are applicable for communication
of research results from observational human exposure
studies, and the reader should consult the risk
communication literature.
Listening/Caring/Empathy
50%
Honesty/
Openess
15-20%
Dedication/
Commitment
15-20%
Competence/
^ Expertise
15-20%
Figure 7-1. Factors Influencing Trust in Times of Crisis
EPA, 2007
ATSDR (2007) has A Primer on Health Risk
Communication Principles and Practices available
online. Because ATSDR generally responds to
environmental issues identified by individuals or
communities, their guidance focuses on communicating
with individuals and communities that perceive an
imminent or significant health risk because of a problem
in the community. Because ATSDR often enters a
community after a potential problem has been identified,
ATSDR communications are often reactive, by
necessity, rather than proactive.
Health Canada (2006) has recently published The
Strategic Risk Communications Framework. The focus
of the effort is always the stakeholders. Health Canada's
process aims to involve the interested and affected
88
-------
parties at all points in a "dialogue-based'
communication process. (See Text Box 7-6).
Text Box 7-6. Steps in Health Canada's Risk
Communication Framework
1. Identify the issue and its context—define the opportunity and
characterize the situation.
2. Assess the risks and benefits—assess stakeholder perception
of the risks, benefits, and tradeoffs.
3. Identify and analyze options—assess how stakeholders
perceive the options.
4. Select a strategy—develop and pretest strategies, risk
communications plans, and messages.
5. Implement the strategy—implement risk communications.
6. Monitor and evaluate results—evaluate risk communications
effectiveness.
Guidance on risk communication strategies and
practices consistently stresses the importance of
obtaining input and feedback from community
representatives, who can assist in developing approaches
that place the results in relevant contexts for the
community and the participant.
One of the difficulties in reporting results to
participants and communities is timely reporting because
it generally takes a long time to complete both the
chemical and data analyses in large studies. Researchers
desire to report fully validated and analyzed data to
study participants and to the community. But. delay in
reporting data can create a number of difficulties.
Participants may move before they receive results. They
also may lose interest in the study, or more importantly,
lose trust in the researchers and the scientific research
community if they do not receive their results in a timely
manner. Similar problems may occur in the community
as community leaders and representatives change.
Community representatives may have expectations for
data and information that researchers cannot achieve.
Therefore, it is important that researchers clearly
communicate with the participants and the community
about what results will be provided and when they will
be delivered so that expectations do not differ from
"reality."
Reporting study results from observational human
exposure studies can be particularly challenging because
data on exposure concentrations and the factors
impacting exposure may be difficult to relate to a health
outcome that is relevant to the study participant. Health
effects data is often lacking for the concentrations at
which chemicals or their metabolites are measured in
environmental or biological media. This is especially
true for studies of many chemicals for which acceptable
occupational exposure levels have been established, but
for which there are not environmentally relevant
standards for low-level exposures. Williams (2004)
describes an approach for communication using
comparative risk analyses. She describes intrachemical
comparisons, interchemical comparisons, comparisons to
background levels of risk, comparisons to theoretical
risk or safety levels, and risk comparisons to other
actions or activities. Williams also includes an extensive
list of references for guidelines and other information on
risk communication. Readers of this document should
refer to her manuscript to determine which approach
may be applicable to their particular study.
During longitudinal studies with repeated
measurements over months, seasons, or years, it is
important that researchers commit to providing interim
and ongoing results to participants and the community as
the study proceeds. It is important to maintain the
researcher-participant relationship throughout the study.
This can be facilitated by keeping study participants
informed of the study progress and of the interim results
Researchers also should recognize that there may be
potential risks to the study participants, third parties, or
the community because of results generated from a study
(refer to Text Box 7-7). Therefore, providing
information to communities has to be done thoughtfully
and with appropriate preparation. Processes should be
developed that provide participants with the option to
receive, or decline, study results. Researchers should
work with community groups to determine how study
results should be disseminated to the community and
what communications strategies should be used.
Text Box 7-7. Potential Harms of Sharing Research
Results with Participants
(Fernandez, Kodish, and Weijer, 2003)
• Incorrect or harmful decisions based on uncertain or unreliable
results
• Causing distress for those participants who did not benefit
from the research
• Rekindling old memories and emotions, especially in the
setting of serious illness
• Emotional distress among community members
• Possible discrimination in obtaining employment or insurance
for a participant identified by the research to be at high risk of
developing complications
• Financial costs to participants and to researchers
There are a variety of methods for providing study
results to participants and the community. Fact sheets
can be used to describe the study and provide general
study findings to the community and stakeholders.
89
-------
Individualized fact sheets can be used to disseminate
results to the individual participants. Meetings with
study participants have been used to disseminate study
information. Community meetings also can be used to
provide updates on study progress and general results.
Examples of the processes and the materials used for
dissemination of information are included in case studies
described by Israel et al. (2005) and others conducting
CBPR studies.
Overall study results generally are disseminated in
peer-reviewed journal manuscripts and study reports.
The availability of results published in manuscripts and
reports has been greatly enhanced by posting them on
Internet Web sites. For example, all EPA reports are
now available electronically via EPA's National Service
Center for Environmental Publications Web site
(http://www.epa.gov/ncepihom/).
7.8 Reporting Unanticipated Results or
Observations
The previous subsection discussed reporting of
routine results from observational human exposure
studies. The communication plan should include
processes and procedures for the dissemination of the
study results. Additionally, the communication plan
needs to integrate with the data and safety monitoring
and oversight plans for the study and include a plan for
reporting unanticipated results or observations.
Unanticipated results may include measurements of a
chemical at a concentration mat exceeds what is
considered to be an "acceptable" level in environmental
media or biological fluids. Unanticipated observations
might include observation of the use of a chemical not
approved for indoor use, storage of chemicals in
inappropriate containers, storage of chemicals in places
accessible by children, etc. Unanticipated results or
observations may be directly related to the research
question being addressed in the study (e.g.,
measurements of pesticide residues in a home) or
nonstudy hazards (e.g., frayed electrical cords that may
pose a hazard to young children and residences).
Section 4 discusses issues that may affect privacy and
confidentiality. Section 4.3 covers collateral
observations of nonstudy-related hazards, including
those that States may mandate must be reported.
Section 4.5 discusses the need for data and safety
monitoring and oversight, including the development of
plans to report and react to anticipated or unanticipated
adverse events or conditions.
As part of the study implementation plan and the
communication plan, researchers should develop a
protocol for how to identify contaminant measurements
and exposures of "concern" that should be reported to
the study participant as quickly as possible because of
the potential risk associated with the exposure (see
Section 4.5, Data and Safety Monitoring and Oversight,
and also Section 2.7.1, Establishing Criteria and
Standards for Monitoring Scientific and Ethical Issues
During a Study.) The plan needs to include the protocol
for making the determination and the criteria that will be
used as the threshold or "trigger" for reporting. The
plans should describe how the results will be reported to
the participants and what additional action will be
undertaken to assist the participant in reducing their
exposures. The first step in developing the protocol is to
identify what measurement will be used to identify
exposures of concern. In observational human exposure
studies, this will generally be the chemical measurement
in either environmental or biological samples. For
example, measurement of lead concentration in blood
would be an appropriate exposure metric if the research
question being addressed involves lead exposure. The
measurement is relatively simple and can be performed
with a short turnaround time. Similarly, measurements
of chemicals in blood may be appropriate for other
persistent chemicals that have relatively long half-lives
in blood. For nonpersistent chemicals, biomarkers of
exposure measured in urine or saliva may be appropriate
metrics to identify exposures of concern. For some
chemicals (e.g., PM, volatile organic compounds
[VOCs], ozone), biomarkers of exposure either are not
available or difficult to measure or interpret. In these
cases, measurements in environmental media may be the
best exposure metric. Whatever metric is chosen, it is
important that the chemical analyses can be performed
relatively quickly to reduce such exposures as quickly as
possible.
The second, and more difficult, step in developing
the reporting protocol is to determine the level of
concern that triggers reporting of the concentration to the
study participant. For some environmental media, such
as drinking water, EPA (2007) has established maximum
contaminant levels that can be used as triggers for
reporting. For example, if the researcher measures a
level of arsenic in drinking water above 0.010 mg/L, he
or she would be expected to report the level to the study
participant. For other environmental media, such as air,
there are few applicable standards. The National
Ambient Air Quality Standards might be used for the
criteria pollutants. Guidelines for occupational
exposures, such as threshold limit values (TLVs) and
biological exposure indices (BEIs) published by the
90
-------
American Conference of Governmental Industrial
Hygienists (ACGIH, 2008) also may be used. TLVs are
not standards; ACGIH formulates a conclusion on the
level of exposure that the typical worker can experience
without adverse health effects. Many people would
argue that the TLVs are not conservative enough for the
average population, particularly not for vulnerable
lifestages (e.g., children, the elderly) and TLVs are only
for exposure by inhalation. WHO (2005) also publishes
air quality guidelines. These types of guidelines can be
used to advise study participants if their exposures are
high relative to the guidelines. Reporting levels should
be conservative, but not so low that reporting the level to
the participant causes unwarranted concern and stress.
For other environmental media measured in
observational human exposure studies, such as house
dust or surface wipes, the measurement results cannot be
used easily to estimate exposures, and they are a poor
metric if used alone.
An alternative approach to comparison of
measurement results against available guidelines and
standards is the comparison of measurements in
biological fluids to measurement data available from the
National Health and Nutrition Examination Survey
(NHANES). For example, results of measurements of
chemicals or their metabolites in urine or blood can be
compared to different percentiles (e.g., the 95th)
reported in the NHANES national reports (CDC, 2005).
This type of comparison shows mat the participant's
measurements are at the high end of the distribution of
the NHANES data, suggesting that action mav need to
' OO O ./
be taken to mitigate exposures. However, researchers
need to be judicious in the selection of the exposure
metric. Biomarkers in blood and biomarkers in urine can
be very different exposure metrics and may represent
different aspects of the exposure event. A similar
approach could be taken with measurements of
chemicals in environmental media if there are
sufficiently large databases available for comparison.
For many chemicals and many media such databases are
not available. In some NERL studies, one comparison
approach mat has been used is to compare an
individual's environmental media measurements to the
50th, 75th, or 95th percentile concentrations for the
entire study population, so that the participant can
evaluate his or her measurement results relative to those
of the other study participants.
A more complex approach than using simple data
comparisons is to calculate a reporting level defined as a
chemical or metabolite concentration indicative of an
absorbed dose greater than that of a target level (for
example one-tenth) of a lifetime reference dose (RfD)
level. For a pesticide, the absorbed dose could be
estimated from the urinary pesticide metabolite level
using an approach similar to the methodology published
by Fenske et al. (2000). This deterministic approach to
dose estimation allows direct back-calculation of doses
from urinary metabolite concentrations using few
assumptions and is consistent with current pesticide
regulatory procedures for risk assessment. When using
this approach, the research team will need to determine
how conservative the reporting level should be, as there
are no guidelines available for using this approach. If the
concentrations of a metabolite measured in a study
participants' urine level are indicative of elevated
exposures (i.e., above the reporting level), the
researchers would be expected to report the information
to the participants and provide information or local
contacts that could assist in helping the participants
identify sources of exposure and reduce their exposures.
Although this would seem to be a reasonable approach
for some classes of chemicals, the authors are not aware
of reports of the use of this approach in the scientific
literature.
7.9 Anticipating and Responding to
Criticism
As discussed in other parts of this document, in spite
of researchers best intentions, there may be situations
that arise in which people's perceptions of the study
design or implementation plan are not accurate, or their
opinions and beliefs about the ethical issues associated
with a study may not be in agreement with those of the
research team and others involved in the study (e.g., the
peer review panel, the IRB). Just as it is not
unreasonable to expect differences in opinion on
scientific approaches to an observational study, it is not
unreasonable to expect differences of opinion on ethical
approaches. The researchers, therefore, should be
prepared to respond to criticism. The implementation
plan and the communication plan should address how
the research team should anticipate study elements that
may be criticized. During study conceptualization, the
research team should develop a list of potentially
controversial study elements (many of which are
discussed in this document). For each study element, the
research team should describe how the ethical
approaches to the study element were evaluated and
selected. Both the process and the rationale for selection
of a particular approach should be documented. At each
step in the study planning and review process, the
research team should document discussions related to
91
-------
the specific element, considerations that were made,
actions taken, and justification for the actions. Input
from research team members, internal reviewers,
external reviewers, community members, and others
involved in the study should be documented for these
controversial study elements. Similarly, for potentially
controversial study elements, the review and actions by
the 1KB should be documented. All of this information
should be compiled for potential use to prepare a set of
Q&As that can be used by the research team and
sponsoring organization to respond to criticism. When
responding to criticism, establishing trust and credibility
are essential, as discussed previously. The public's
perception of trust and credibility is determined by the
public's perceptions of the researchers' knowledge and
expertise, openness and honesty, and concern and care
(Peters et al., 1997). These factors are important to
consider in developing the information and approach
that will be used to respond to criticism.
There is a large volume of information available on
"crisis communication" that the reader can use to
develop a plan for anticipating and responding to
criticism (e.g., FCN, 2001; ATSDR, 2007; U.S. HHS,
2002). The key is to be proactive and have a plan before
any criticism is raised.
7.10 Responding to the Media, Public
Inquiries, and Other Stakeholders
Like crisis communications, the communication plan
should include detailed plans for how to interact with the
stakeholders, the media, and the public. Standard
approaches have been developed for effective
communications (e.g., the Federal Communicators
Network's Communicators Guide [FCN, 2001]) with the
media and will not be included in this document. A
proactive plan, open and transparent communications.
and easily to comprehend information will ensure
effective communications with stakeholders and the
public.
References
American Conference of Governmental Industrial Hygienists
(ACGffl) (2008). TLV/BEI Resources. Available:
http ://www. ac gih. org/tiv/.
Alderson P (1995). Will you help us with our research? Arch
Dis Child 12(6):54l-42.
Anderson B, Hall B (1995). Parents' perceptions of decision
making for children. JLaw Med Ethics 23(1): 15-9.
ATSDR (Agency for Toxic Substances and Disease Registry)
(2007). A Primer on Health Risk Communication Principles
and Practices [online publication]. Available:
http://www.atsdr.cdc.gov/risk/riskprimer/ [accessed 12 June
2007].
CDC (Centers for Disease Control and Prevention) (2005).
Third National Report on Human Exposure to
Environmental Chemicals. National Center for
Environmental Health, Atlanta, GA, NCEH Pub. No. 05-
0570. Available:
http://www.cdc.gov/exposurereport/pdf/tliirdreport.pdf
[accessed 12 June 2007].
Covello V, Allen F (1988). Seven Cardinal Rules of Risk
Communication. OPA-87-020. U.S. Environmental
Protection Agency, Office of Policy Analysis, Washington,
DC. Distributed by the Pennsylvania Dept. of
Environmental Resources. Available:
littp://www.epa.gov/stakeholders/pdf/risk.pdf [accessed 12
June 2007].
Covello VT, McCallum DB, Pavlova M (eds) (1989). Effective
Risk Communication: the role and responsibility of
Government and Non-Government Organizations. New
York. NY: Plenum Press.
Covello V. Minamyer S, Clayton K (2007). Effective risk and
crisis communication during water security emergencies
summary report of EPA sponsored message mapping.
EPA/600/R-07/027. National Homeland Security Research
Center, Office of Research and Development, U.S.
Environmental Protection Agency, Washington, DC.
Available:
http://www.epa.gov/nlisrc/pubs/reportCrisisCom040207.pdf
FCN (Federal Communicators Network) (2001).
Communicators Guide For Federal, State, Regional, and
Local Communicators by the Federal Communicators
Network. Available:
littp://www.publicforuniinstitute.org/activities/2002/fcn/co
mmguid.pdf [accessed 12 June 2007].
Fenske RA, Kissel JC, Lu C, Kalman DA, Simcox NJ, Allen
EH, Keifer MC (2000) Biologically based pesticide dose
estimates for children in an agricultural community.
Environ Health Persp 108(6):515-20.
Fernandez CV, Kodisli E. Weijer C (2003). Informing study
participants of research results: an ethical imperative. IRB
25(3):12-19.
Health Canada (2006). The Strategic Risk Communications
Framework. Available:
http://www.riskcommunications.gc.ca [accessed 18
September 2007].
IOM (Institute of Medicine) (2002). Responsible Research: A
Systems Approach to Protecting Research Participants.
Daniel D. Federman, Katlii E. Hanna, and Laura Lyman
92
-------
Rodriguez (eds). Washington, DC: The National Academies
Press.
Israel BA, Eng E, Schulz AJ, Parker EA (eds) (2005).
Methods in Community-Based Participatory Research for
Health. San Francisco, CA: Jossey-Bass.
MH (National Institutes of Health) (2005). Report and
Recommendations on Public Trust in Clinical Research for
the NTH Director from the Director's Council of Public
Representatives (COPR). National Institutes of Health,
Director's Council of Public Representatives, January 14,
2005. Available: http://copr.nih.gov/reports/public trust.asp
[accessed June 2007].
Parkin RT (2004). Communications with research participants
and communities: foundations for best practices. J Expo
Anal Environ Epidemiol 14(7):516-23.
Peters RG, Covello VT, McCallumDB (1997). The
determinants of trust and credibility in environmental risk
communication: an empirical study. Risk Anal 17(l):43-54.
Stableford S and Mettger W (2007). Plain language: a
strategic response to the health literacy challenge. J Public
Health Pol 28:71-93.
U.S. HHS (Department of Health and Human Services)
(2002). Communicating in a Crisis: Risk Communication
Guidelines for Public Officials, 2002. Available:
http://www.riskcoinmunication.samlisa.gov/mdex.htm
[accessed 13 June 2007].
U.S. EPA (U.S. Environmental Protection Agency) (2007).
Drinking Water Contaminants. Environmental Protection
Agency. Available:
http://wwvv.epa.gov/safewater/contaminants/index.html
[accessed 13 June 2007].
Williams PR (2004) Health risk communication using
comparative risk analyses. J Expo Anal Environ Epidemiol
14(7):498-515.
World Health Organization (2005). WHO Air Quality
Guidelines Global Update 2005. Available:
http://www.euro.who.int/Document/E87950.pdf [accessed
13 June 2007].
Additional Information Resources
Reckelhoff-Dangel C, Petersen D (2007). Risk Communi-
cation in Action: The Risk Communication Workbook.
EPA/625/R-05/003. National Risk Management Research
Laboratory, Office of Research and Development, U.S.
Environmental Protection Agency, Cincinnati, OH.
Available:
http://www.epa.gov/nrmrl/pubs/625r05003/625r05003.pdf
[accessed 18 September 2007].
U.S. EPA (U.S. Environmental Protection Agency) (2006).
Considerations for Developing Alternative Health Risk
Assessment Approaches for Addressing Multiple Chemicals,
Exposures, and Effects. External Review Draft. 71 FR
16306.
U.S. EPA (U.S. Environmental Protection Agency) (2006).
Paper on Tribal Issues Related to Tribal Traditional
Lifeways, Risk Assessment, and Health and Well Being:
Documenting What We 've Heard. The National EPA-Tribal
Science Council. Available:
http://epa.gov/osp/tribes/tribal/TribalIssues.pdf.
U.S. PHS (Public Health Service) (1995) Risk
communication: working with individuals and communities
to weigh the odds. Prevention Report. Feb/Mar. Available:
http://odphp.osophs.dhlis.gov/pubs/previpt/Arcliives/95fml.
htm.
93
-------
-------
Appendix A
Additional Discussion of Observational and Exposure Terminology and
Examples of Previous NERL Observational Human Exposure Studies
The authors recognize that the words "observational
human exposure studies" may convey a variety of
meanings to different people. The word exposure is
often interpreted differently by different people;
similarly, the word observational may have a variety of
interpretations. The following discussion regards
potential meanings of those terms and describes what the
authors of this document mean when they use the tenii
observational hitman exposure studies.
Exposure
Exposure, as it is used throughout this document, is
a technical term that is defined as the "contact of a
chemical, physical, or biological agent with the outer
boundary of an organism (e.g., a person) (U.S. EPA,
1992). Exposure is quantified as the concentration of the
agent in the medium in contact integrated over the time
duration of that contact."
Exposure is Contact of an Agent with an
Individual
Figure A-1: Source- Exposure- Effect continuum highlighting Exposure K the interaction of the agent
and the percon wtio is exposed.
As the authors use the term, exposure is only the
quantification of the "contact" as defined above. The
word itself carries no connotation of intent, and it is not
equivalent to intentional exposure, dose (dosing), or
intentional dosing. A review of the dictionary definition
of exposure shows that the word has many different
meanings, and shades of meaning, in common, everyday
English. However, when the authors use the word in this
text it is being used in the technical sense that is defined
above.
The authors' understanding of the word exposure is
illustrated in Figure A-1. This graphic is a paradigm that
ORD uses in formulating its human health research
program. The graphic illustrates mat exposure occurs at
the intersection of both the chemical, physical, or
biological agent and the person who is exposed. To
understand exposure, one must understand the
distribution of the "agent" throughout the environment
over time and combine that understanding with
knowledge about the location and activities of people
that bring them into contact with the agent. Knowledge
about the distribution of the agent in space and time
generally involves measurements and data collection
about environmental conditions external to the person.
Knowledge about the person's behaviors requires
collection of personal information or observations of
their location and activities. Understanding exposure,
including the pathways, routes, duration, frequency, and
magnitude of the exposure, requires combing both types
of information.
See the Glossary, Appendix F, for more information
and the definition of additional and related terms.
Observational
Observational human exposure studies, as used in
the context of this document, involve only the collection
of environmental or biological samples and information
for the purpose of quantifying the contact between the
participant and the agent being studied. Observational
human exposure studies are inherently the process of
watching people in context—in their natural
environment, doing routine activities—and collecting the
95
-------
samples and relevant information from them and from
their surroundings to measure and calculate the
exposures that occurred in the same context.
Scientific Study Design: From a scientific study
design perspective, an observational study of the sort
being considered in this document is one where the
researcher does not control the variables but, rather,
observes both the variable and the outcome and tries to
infer the relationship between the variable and the
outcome. This contrasts with a controlled study where
the researcher isolates and controls one or more
variables in a systematic way to assess the impact of
changes in the variables on an outcome measure.
Consider the intervention study examples cited in the
section on regulatory distinctions below (cleaning a
house to reduce exposure in the residence and wearing a
particle mask to reduce exposure to smoke from forest
fires). In those cases, participants would be assigned
either to receive the intervention (treatment) or not (the
control group). The outcomes (exposures) would be
measured and compared between the two groups to
assess the effectiveness of the intervention. Similarly,
drug trials would randomly assign participants to receive
either a test medication or a placebo. The medical
outcome (e.g., pain relief, blood pressure, cholesterol
levels in blood) would then be measured for all
participants, and the outcomes compared between the
treated group (getting the test medication) and the
control group (receiving the placebo).
Controlled studies are often desirable in science
because, when properly designed, they provide
unambiguous estimates of the impact of unique variables
on the outcome (even if the impact of one variable could
normally be overwhelmed by other factors) and
eliminate the need for alternate explanations of the
experimental results (because other factors are held
constant). However, controlled studies are not always
possible. They may be unethical (e.g., to "treat" a group
of women with abortions to test if there is a linkage
between having had an abortion and breast cancer),
impractical (e.g., to follow a cohort large enough to yield
statistically significant results in a test of rare side
effects to a medication), or just impossible to accomplish
(e.g., to measure nutritional levels of a population in the
middle of a war zone).
As a consequence, much medical research is
observational in the scientific sense being considered
here. Public health data represent observations of health
outcomes, but the factors that may have led to or
influenced those outcomes are not under the control of
any researcher. Observational techniques long have been
used in medical studies to infer information about the
impacts of certain factors on health outcomes.
Commonly used techniques include cohort studies, case-
control studies, cross-sectional studies, case reports, case
series, and descriptive studies (NEAC, 2006;
Vandenbroucke et al., 2007).
Observational human exposure studies as used in
this document are considered to be observational from a
scientific study design perspective because the variables
leading to exposure are not controlled by the researchers.
Most of the observational human exposure studies
conducted by NERL to date have been cross-sectional
studies (sometimes repeated several times). While a
particular variable may not be controlled by the
researcher, the study design (e.g., selection of the
population to be studied, location of the research, data
selection to exclude confounding factors) can sometimes
influence the range of values over which a variable may
be observed. For example, NERL's observational human
exposure studies to understand exposure to PM (see
www.epa.gov/heasd/sources/projects completed/pm_pa
nel studies.htm and www.epa.gov/heasd/sources/project
s/a3a_understanding_airshed_sources.htm) have
traditionally excluded homes with smokers from the
study population to avoid cigarette smoke as a
confounding factor in the studies.
Observational Studies in Market Research,
Ethnography, and the Common Vernacular:
Observational market research can involve covert
observation, overt observation, or researcher
participation. Covert observation is said to have a "key
advantage" that "the respondent or consumer is unaware
that they are being observed, allowing their behavior to
be observed naturally" (see
http://www.asiamarketresearch.com/glossary/observatio
nal-research.htm). This means that the subject's behavior
will be natural and uncontaminated by the researcher's
presence. On the other hand, covert observation may be
construed to be deceitful and ethically questionable.
Overt observations involve sampling surveys, polls,
interviews, focus groups, etc. Validity of the data
obtained this way may be shaded by people's natural
tendency to behave or respond ideally when they know
they are being watched. Motorists routinely slow down
when they think they are being observed by the police.
Finally, the researcher may participate in the activity
being observed. Ethnography is a long-term
investigation of a group (often a culture) that is based on
observations made while immersed in and, useually,
participating in that group. "Ethnography provides a
detailed exploration of group activity and may include
96
-------
literature about and/ or by the group." (See
http://writing.colostate.edu/guides/research/observe/com
3al.cfm). One obvious problem with immersion
techniques is that the researcher may lose his or her
objectivity.
In the common language use of the tenii
observational research, it appears to the authors that
some people understand the term to imply that the
observations are both benign and often covert, mat is,
without interaction with the persons being studied. That
is not the case in NERL's observational human exposure
studies. The research often takes place in and around the
homes of the participants. Often, the research involves
asking the participant to answer a questionnaire and to
provide personal samples (e.g., urine, blood). Collection
of survey information and personal urine and blood
samples cannot be done covertly, and collecting a blood
sample is not considered to be benign.
Observational Research in NERL: NERL's
observational human exposure studies entail the
collection of environmental or biological samples, data,
and information from study participants and their
surroundings in their everyday environments, as they go
about their normal activities, for the purpose of
quantifying the contact between the participant and the
agent being studied. As such, the studies are designed to
meet the regulatory definition of observational research
in the CFR. NERL's studies also meet the scientific
definition of an observational study, because the variable
being studied, exposure to some agent, is not controlled
by the NERL scientists. The observations are not covert,
and they may or may not be noninvasive (e.g.,
sometimes blood samples may be collected).
Because NERL's observational human exposure
studies meet the definition of human subjects research as
set forth in the Common Rule, there is also a regulatory
requirement to meet the ethical and scientific standards
set forth in EPA's human subjects regulations and in
Agency rules. The research protocol must be evaluated
and approved by an IRB and by EPA's Human Studies
Research Review Official (HSRRO) before any human
subjects research effort can take place. But, even more
compelling to NERL managers and scientists is the fact
that NERL cannot conduct observational human
exposure studies without the participation of willing
individuals. Indeed, without the research participants,
NERL's human exposure research would be nothing.
This intimate involvement of research subjects in
NERL's research imposes moral and ethical obligations
to deal with participants respectfully and to ensure their
safety, protection, and well-being.
Regulatory Distinction Between Intentional
Exposure and Observational Research
Intentional Exposure: The CFR states, "Research
involving intentional exposure of a human subject
means a study of a substance in which the exposure to
the substance experienced by a human subject
participating in the study would not have occurred but
for the human subject's participation in the study" [40
CFR 26.202(a)]. This definition sets forth two
requirements regarding an intentional exposure: (1) the
exposure has to be to a substance that is being studied,
and (2) at least one aspect of the exposure to the
substance being studied has to be attributable to the
subject's participation in the study.
The first requirement means the objective of the
study must be to understand the impact or exposure of
the substance being studied. In observational human
exposure studies, like those mat NERL conducts, the
research protocol may require the use of isopropyl
alcohol or some other chemical to collect a personal
sample (e.g., to sterilize the puncture site immediately
prior to collection of a personal sample of blood drawn
by a trained phlebotomist). Indeed, it would be both
unethical and bad medical practice not to sterilize the
puncture site before drawing blood. But this requirement
does not constitute intentional exposure because the
research is not the study of isopropyl alcohol but of
some other agent. Incidental exposures to chemicals like
isopropyl alcohol still must be considered from a safety
perspective, pass both scientific and ethical review, and
be approved by the IRB, but such incidental exposures
do not constitute intentional exposure.
The second requirement is that the exposure as
experienced by the subject "would not have occurred but
for the human subject's participation in the study." The
two primary ways in which intentional exposure studies
commonly meet this definition are by the direct
introduction of the study substance into the research
environment under the control of the research protocol
or by scripting the participants' activities in such a way
that their contact with the study substance is determined
by the research. Although either would be sufficient to
meet the regulatory definition, many, if not most,
intentional exposure studies attempt to control both.
Because "exposure is quantified as the concentration of
the agent in the medium in contact integrated over the
time duration of that contact," intentional exposure
studies are usually more scientifically robust if both the
concentration of the agent and the duration of the
exposure are controlled by the research.
97
-------
Observational Research: In the language of the
Code of Federal Regulations, "observational research
means any human research that does not meet the
definition of research involving intentional exposure of
a human subject" (40 CFR 26.302).
By this definition, observational research
encompasses all human research that does not meet the
definition of intentional exposure. As a consequence, it
is a broad regulatory category that includes a variety of
research domains, including human research that does
not involve the study of exposures at all. This regulatory
definition of observational is sufficiently broad,
moreover, mat it encompasses study types that this
document is not intended to address. For example, it
may be possible to design an intervention study that does
not bring about or script in any way the participants'
exposure to a substance, but rather reduces or mitigates
it. (Consider a study to test whether professional
cleaning of the carpets, floors, walls, and other surfaces
in a home might lead to a lower exposure to a residential
contaminant or a study to determine if wearing a particle
mask would reduce an individual's exposure to smoke
from forest fires.) Such a study might meet the
regulatory definition of observational research (Note:
The final decision in regard to whether any EPA study
meets the definition of observational research resides
with the Agency's HSRRO), but it would not meet the
authors' intentions regarding "observational human
exposure studies" as they are defined in this document.
An intervention study, such as described in these two
examples, does not involve observing people's
exposures in their everyday environments, as they go
about their normal activities. In an intervention study,
either the participant's environment (cleaning of the
household surfaces) or their behavior (wearing a particle
mask) has been manipulated by the researcher. Secondly,
the objective of these studies is not to understand
exposures in everyday environments but would use
changes in exposure to test the effectiveness of an
intervention strategy. Such a study is not addressed in
this document.
Observational human exposure studies, as used by
the authors of this document, generally meet the
regulatory definition of observational research. But, not
all studies meeting the CFR definition of observational
research would be considered by the authors to be
observational human exposure studies.
Examples of NERL Observational
Human Exposure Studies
NERL and its predecessor organizations have
conducted observational human exposure studies since
1980. Table A-l lists many of those that NERL has
conducted, supported, or participated in since 1980. The
table gives the name of the study, dates, sample size,
then type of study, a brief explanation of the research,
and NERL's role therein. The table represents a variety
of first-party or second-party research efforts. In many
cases, NERL staff would design, oversee, and, if
possible, participate directly in the study (first-party),
but often contractor support would be needed to
accomplish the field sampling or some of the sample
analysis. In other cases, NERL researchers would solicit
proposals for exposure research to be conducted to
address specific exposure issues and then would fund
researchers in academia or at nonprofit institutions to
design and conduct the research (second-party research).
In a few cases, grants might already be in place with
research institutions, and NERL simply would augment
the pre-existing funding to expand the exposure
component (second-party). Some other cases involve
research efforts initiated by other Federal agencies with
which NERL would collaborate and participate in
directly (first-party).
An examination of the entries in the table shows that
most of the agents being studied were chemicals, often
air pollutants or pesticides. A few of the entries indicate
that other agents, for example, molds and fungi or other
microbes, were the subject of the study. The largest
number of the studies involved small numbers of
participants and often were designed to determine
whether or not a method for collecting exposure-related
data or samples was feasible or burdensome. Most of the
smaller studies used convenience samples. Studies, like
DEARS, CTEPP, NHEXAS, and TEAM, which
involved large numbers of participants, employed
randomized or probability-based approaches for
selecting participants.
Table A-2 shows the types of samples that often
have been collected in NERL's observational human
exposure studies. Some of the samples, like air
concentrations or surface wipes, allow NERL staff to
determine the concentration of the studied chemicals in
the environmental media with which the participants
may come into contact. Other items, like time activity
diaries or videotaping children to measure the amount of
hand-to-mouth activity, allow NERL researchers to
understand how people may come into contact with the
agent and to estimate the duration or frequency of a
potential exposure. Still other measures, like personal
samplers or biological samples, represent an attempt to
determine a time-integrated measure of exposure. NERL
98
-------
staff put all of this information together to estimate an
individual participant's exposure, either through
relatively direct measures of exposure or through
algorithms that combine the media concentrations
measurements and the activity data.
References
CFR (Code of Federal Regulations) (2006). 40 CFR Chapter I
Environmental Protection Agency Part 26 Protection of
Human Subjects. U.S. Code of Federal Regulations.
Available: http://www.access.gpo.gov/nara/cfr/waisidx 067
40cfr26_06.html [accessed 12 June 2007].
NEAC (National Ethics Advisory Committee) (2006). Ethical
Guidelines for Observational Studies: Observational
research audits and related activities. Wellington, New
Zealand: Ministry of Health. Available:
http://www.neac.healtli.govt.nz/moh.nsf/indexcm/neac-
resources-publications-ethicalguidelines.
U.S. EPA (U.S. Environmental Protection Agency) (1992),
Guidelines for Exposure Assessment. Environmental
Protection Agency, Risk Assessment Forum, EPA/600/Z-
92/001. Available:
http://www.epa.gov/nceawwwl/raf/pdfs/exposure.pdf
[accessed 1 May 2008].
Vandenbroucke JP, von Elm E, Altaian DG, Getzsche PC,
Mulrow CD, et al. (2007). Strengthening the Report of
Observation Studies in Epidemiology (STROBE):
explanation and elaboration. PLoSMedicine 4(10): e297.
doi:10.1371/journal.pmed.0040297.
99
-------
Table A-1. Examples of Observational Human Exposure Studies Conducted or Funded by NERL
or Its Predecessor Organizations
Study
Total Exposure
Assessment
Methodology
(TEAM) Studies
Additional
TEAM Studies
Nonoccupationa
I Pesticides
Exposure Study
(NOPES)
Nine-Home
Children's Pilot
Study
Particle Total
Exposure
Assessment
Methodology
(PTEAM) Study
National Human
Activity Pattern
Study (NHAPS)
Date
1980-1984
1987
1986-1988
1990
1990
1992-1994
Size
400
Households
51 in LA area,
17 in NJ,
and 75 in
Baltimore
259
9
178
9,386
Type
Three-stage
stratified
probability
selection,
cross-
sectional
Convenience
sample
Probability
selection,
mainly cross-
sectional,
partly
longitudinal
Convenience
pilot study
to test
methodology
Probability
Probability
(national
random
dialing)
Brief Description
The TEAM study was designed to develop and demonstrate
methods to measure human exposure to toxic substances in
air and drinking water. The goals were to develop methods to
measure individual total exposure (from air, food, and water)
and the resulting body burden of toxic and carcinogenic
chemicals and to apply these methods within a probability-
based sampling framework to estimate exposures and body
burdens of urban populations in several U.S. cities. Air
sampling measured personal exposure to airborne toxic
chemicals in indoor air and in exhaled breath. Related
objectives were to determine the relationships between
personal, indoor, and outdoor, and blood, urine, and exhaled
breath concentrations; determine the variability of VOC
concentrations within a home; and determine seasonal and
multiyear variability. The study was conducted in New Jersey,
North Carolina, and North Dakota in three phases.
A separate VOC TEAM study was carried out in Baltimore,
MD: Baltimore lacks the petroleum and chemical
manufacturing plants present in most of the previous TEAM
study sites and is more representative of many U.S. cities.
Focused studies were also conducted in New Jersey and Los
Angeles to further explore the sources and factors
contributing to personal VOC exposures identified in the
earlier TEAM studies.
First attempt to develop a methodology for measuring the
potential exposure of the general population to common
household pesticides using probability-based sampling;
questionnaire data collection; and personal monitoring of air,
drinking water, food, and dermal contact. Conducted in
Jacksonville, FL, and Springfield and Chicopee, MA, to
capture high- and low-pesticide-use areas, respectively.
A 9-home pilot study was conducted to evaluate monitoring
methods in the field that may be used to assess the potential
exposures of children aged 6 months to 5 years to pesticides
found in the home environment. Newly developed methods
were tested for measuring pesticide residues in indoor air,
carpet dust, outdoor soil, and on children's hands. Information
also was collected on household characteristics, pesticides
used and stored at the residence, and children's activities.
A study of nonsmokers age 1 0 years and older in Riverside,
CA, in which indoor, outdoor, and personal monitoring of
integrated particle mass over 1 2-hour day and night periods
was conducted for PM10 and PM2.5. Polycyclic aromatic
hydrocarbon and phthalate measurements were collected
indoors and outdoors at a subset of 125 homes. A separate
monitoring site provided ongoing 12-hour measurements of
ambient particles throughout the study (48 days). Reference
PM10 and dichotomous samplers were used at this site in
addition to personal and stationary monitors identical to those
used for participant monitoring. Nighttime and daytime recall
time-activity questionnaires, as well as household
questionnaires, were administered to each selected subject
and household, respectively, and meteorological and air
exchange rate data were recorded for each monitored home.
This survey collected 24-hour activity diaries and other
questionnaire data from participants from 48 states. It was
conducted as a next-day telephone recall interview.
NERL Role
EPA designed and
run, contractor
conducted
EPA designed and
run, contractor
conducted
EPA designed and
run, contractor
conducted
EPA designed and
run, contractor
conducted
EPA designed and
run, contractor
conducted
EPA designed and
run, contractor
conducted
100
-------
Table A-1. (Continued)
Study
Agricultural
Health Study
(AHS)
The National
Human
Exposure
Assessment
Survey
(NHEXAS)—
Overview
NHEXAS-
Arizona
NHEXAS-
Maryland
NHEXAS-
Region 5
Minnesota
Children's
Pesticide
Exposure Study
(MNCPES)
Date
1993-
present;
PES 1999-
2003
1995-1998
1995-1998
1995-1996
1995-1997
1997
Size
84 applicators
and a subset
of family
members in
PES
See specifics
in entries
below.
179
Households
80
Households
250
Households
102 Children
Type
AHS survey
is a
longitudinal
study of
pesticide
applicators in
two states.
PES, led by
NERL, was a
cross-
sectional
study.
Probability,
cross-
sectional
Probability,
longitudinal
Probability,
cross-
sectional
Probability,
cross-
sectional
Brief Description
A large-scale, long-term ongoing study of Iowa and North
Carolina pesticide applicators and farm families. This
collaborative effort involving the National Cancer Institute
(NCI), the National Institute of Environmental Health Sciences
(NIEHS), the National Institute for Occupational Safety and
Health (NIOSH), and the U.S. Environmental Protection
Agency (EPA) examines how lifestyle habits, genetic factors,
and agricultural exposures contribute to the risk of disease
using questionnaire survey data from 89,658 participants. This
is the largest, most comprehensive study of agricultural health
ever conducted in the United States. The interagency exposure
assessment study (Pesticide Exposure Study (PES)), led by
NERL, was performed to assess exposures and exposure
classification procedures used in the epidemiological study.
A Federal interagency research effort coordinated by EPA and
consisting of four demonstration studies using probability-
based sampling designs conducted in partnership with other
Federal agencies (CDC, FDA, and NIST), universities, and
research institutions. Household environmental and personal
samples were collected and questionnaires were administered.
Biological media, including blood and urine, also were sampled
and monitored for parent contaminants or their metabolites.
The studies were undertaken to evaluate total human exposure
to multiple chemicals on a community and regional scale with
the following aims: provide a baseline of the normal range of
exposure to chemicals in the general population; identify
subgroups of the general population that are likely to be highly
exposed; and evaluate and improve the accuracy of models
developed to predict exposure to chemicals.
The NHEXAS-Arizona study sampled residences determined
by a population-based probability research design for the total
population of Arizona and measured metals, pesticides, and
volatile organic compounds (VOCs). The study was conducted
by a consortium composed of the University of Arizona,
Battelle Columbus, and the Illinois Institute of Technology.
The study sampled residences of Baltimore and four adjacent
counties and measured metals, pesticides, and polycyclic
aromatic hydrocarbons (PAHs). Data collection occurred up to
6 times over 1 year for each of the participating households.
The study was conducted by Harvard University, Emory
University, Johns Hopkins University, and Westat. The study
investigated temporal variability in multimedia, multipollutant
exposures.
The Region 5 study was conducted in EPA's Region 5 (Ohio,
Michigan, Illinois, Indiana, Wisconsin, and Minnesota), and
included personal exposure, residential concentration, and
biomarker measurements of metals and VOCs. The study was
conducted by the Research Triangle Institute (RTI) and the
Environmental and Occupational Health Sciences Institute
(EOHSI).
Multimedia study of children 3-1 3 years conducted in
Minnesota to evaluate children's pesticide exposure in urban
and rural areas. The study provides exposure, environmental,
and biologic data relating to multipathway exposures of
children for four primary pesticides (chlorpyrifos, malathion,
diazinon, and atrazine), 14 secondary pesticides, and 13
polynuclear aromatic hydrocarbons (PAHs). This study
complements and extends the populations and chemicals
included in the NHEXAS-Region 5 study.
NERL Role
Collaboration with
other agencies;
NERL led a
contractor-
conducted
substudy called
the Pesticide
Exposure Study.
EPA oversaw
general design for
the research, then
solicited, evaluated
and funded the
individual studies
(see below).
Details of research
were established
by recipients
Designed and
conducted by
award recipient
(see above).
Designed and
conducted by
award recipient
(see above).
Designed and
conducted by
award recipient
(see above).
Designed and
conducted by
award recipient
(see above).
101
-------
Table A-1. (Continued)
Study
Participate
Matter (PM)
Panel Studies
Dietary Intake of
Young Children
(DIYC)
Children's
Pesticide Post-
Application
Exposure Study
(CPPAES)
Develop Risk
Assessment and
Define Some
Risk
Management
Options for
Exposure of
Children to Toxic
Mold Using
Stachybotrys
chartarum as an
Example
Children's Total
Exposure to
Persistent
Pesticides and
Other Persistent
Organic
Pollutants
(CTEPP)
Car-Related
Occupational
PM and Air
Toxics Exposure
to Patrolmen
Study (COPP)
Date
1998-2001
1999-2000
1999-2001
2000
2000-2001
2001
Size
200
Individuals
(5 to 63 per
study)
3 Homes
10 Homes
8 cases and
8 controls
257
Households
9
Type
Longitudinal
Convenience
Convenience,
multiday pilot
study to test
methodology
Case control
study
Randomized,
cross-
sectional
Convenience,
longitudinal
Brief Description
A series of longitudinal studies in a number of U.S. cities
conducted by EPA or by organizations sponsored through the
National Exposure Research Laboratory (NERL). A primary
goal was to determine the relationships between personal
exposures to particles and associated gases relative to
stationary outdoor monitor concentrations in high-risk
subpopulations and to identify human activity patterns that
might contribute to personal exposure. The investigators varied
study locations, monitoring seasons, and study populations.
Susceptible subpopulations of interest included chronic
obstructive pulmonary disease (COPD) patients, individuals
with cardiovascular disease, the elderly, asthmatics, and
African Americans with hypertension. Panels of healthy
individuals also were included. Because the elderly is the
subpopulation most sensitive to health effects associated with
PM exposures, the majority of subjects were over age 65.
Performed in the Raleigh, NC, area in homes with children 1 to
3 years old. Homeowners reported either professional or self
applications of diazinon. Goals of the study were to evaluate
methods to measure excess dietary exposures resulting from
food-handling activities by young children during eating and to
assess whether the Children's Dietary Intake Model (CDIM)
accurately represents total dietary exposures of children. Study
resulted in refinements of model parameters for transfer and
activity.
Observational measurement study of exposure to chlorpyrifos
among children 2-5 years of age in urban New Jersey homes
following crack and crevice treatment by a professional
applicator. Study investigated decay of pesticide levels over
time, transfer of pesticide from microenvironmental media to
child, and factors that affect transfer. Dermal transfer evaluated
with surface wipes, hand wipes, dermal wipes, cotton
garments, and videotaping.
Physicians at Case-Western Reserve treating children with
idiopathic pulmonary hemorrhage evaluated the nature of the
fungus found in homes of the afflicted and control children to
determine if Stachybotrys was a potential factor in the illness.
Three of the 8 cases, and none of the controls, had
Stachybotrys strains that potentially were implicated.
The largest children's exposure study undertaken to date. It
examines aggregate exposures of children 1 8 months to 5
years to pollutants commonly found in everyday environments.
The major objectives were to quantify children's aggregate
exposures, apportion exposure pathways, and identify
important exposure media. Participants were recruited from 12
urban and rural counties in North Carolina and Ohio using a
random digit dialing method. Monitoring was performed at both
daycare centers and homes. Samples collected include food,
beverages, indoor air, outdoor air, hand wipes, dust, soil,
transferable residues, floor and surface wipes, and urine. The
samples were analyzed for more than 40 pollutants, including
insecticides, phthalate esters, phenols, polychlorinated
biphenyls, and PAHs.
Scientists monitored air pollutants inside and outside vehicles
of healthy highway patrol officers in North Carolina while
troopers were on patrol for 9-hour shifts and examined
cardiovascular effects. The findings indicated that people
driving in motor vehicles are exposed to PM25 and other
pollutants generated from motor vehicles, and that these
exposures appear to cause cardiovascular changes.
NERL Role
NERL conducted
one panel study
with contractor
support; funded
and collaborated
on others with
Harvard School of
Public Health,
University, of
Washington, and
NYU.
NERL designed,
contractor
conducted
NERL funded the
Environmental and
Occupational
Health Sciences
Institute (EOHSI).
NERL staff
collaborated with
Case-Western
Reserve University
by conducting lab
analysis for
Stachybotrys.
NERL designed,
contractor
conducted
EPA designed and
conducted with
contractor support
102
-------
Table A-1. (Continued)
Study
First National
Environmental
Health Survey of
Child Care
Centers (CCC)
A Pilot Study
Examining
Translocation
Pathways
Following a
Granular
Application of
Diazinon to
Residential
Lawns (PET)
Biological and
Environmental
Monitoring for
Organo-
phosphate and
Pyrethroid
Pesticide
Exposures in
Children Living
in Jacksonville,
FL (JAX)
Exposure
Assessment for
Community-
Acquired
Legionnaires
Disease
Center for the
Health
Assessment of
Mothers and
Children of
Salinas
Quantitative
Exposure
Assessment
Study
(CHAMACOS)
Feasibility of
Macroactivity
Approach To
Assess Dermal
Exposure
(Daycare)
Date
2001
2001
2001
2001
2002
2002
Size
168 Child care
centers
6 Households
9 Households
in NERL
exposure
component
21
20 Children
9 Daycare
centers
Type
Probability-
based
selection on
national scale
Convenience,
multiday pilot
study to test
methodology
for exposure
measure-
ments
Convenience,
cross-
sectional
Case -control
Convenience
Convenience
Brief Description
A collaborative study with the Department of Housing and
Urban Development (HUD) and the Consumer Product Safety
Commission (CPSC) of pesticide use and young children's
(less than 6 years old) potential exposure to pesticides and
other pollutants in institutional childcare centers. This national
study used multistage sampling with clustering. Indoor wipe
and outdoor soil samples were analyzed for pesticides, lead,
and allergens (mold/fungi). Pesticide use practices and
application information were obtained from the commercial
pest control applicators serving the centers.
Observational pilot exposure measurement study of residential
exposures after homeowner had routinely applied granular
formulation diazinon-containing turf treatment to residential
lawns. Study was performed near Raleigh, NC, and was
preceded by a 1-home methodology feasibility study. Purpose
was to evaluate methods for assessing pet-borne transfer,
translocation and exposure pathways, and decay rates.
The objectives of this investigation in Jacksonville, FL, were to
(1) assess organophosphate (OP) and pyrethroid pesticide
exposures in a group of 4- to 6-year-old children from
Jacksonville by measuring the urine metabolite levels,
(2) identify possible sources of these pesticides through
screening measurements and pesticide inventories, and
(3) examine the relationship between environmental and
biological levels. The Duval County Health Department
(DCHD) collected urine samples for CDC from 200 children
visiting six public health clinics in Jacksonville, and collected
environmental screening samples at approximately 25% of
these children's homes. A detailed aggregate exposure
assessment at 9 homes was overseen by NERL and involved
collection of surface wipes, transferable residues, air, duplicate
diet, cotton garment samples, and urine samples. A time-
activity diary of the children's activities was included.
Work with Veteran's Administration hospital staff. Cases of
Legionnaires Disease were evaluated for potential for
exposure from residential drinking water taps. In about 24% of
the tested cases, homes were found to have Legionella
bacteria in water taps at home, compared to their absence in
other cases and in controls
Incidental pesticide exposure measurement study of
farmworkers' children ages 5 to 35 mo. Purpose is the
evaluation of methods for aggregate exposure measurements
and the evaluation of pathways of exposure and important
factors that affect exposure. Measurements include pesticide
distributions in microenvironments where children spend time,
transfer of pesticides from microenvironmental media to child,
and factors that affect transfer.
Study identified daycare centers with previously established
contracts for routine monthly pesticide applications and
conducted screening sampling in each to evaluate the
distributions of transferable pesticide residues on floor surfaces
where children spend time. One daycare was selected for
intensive measurements, and children from different age
groups volunteered to wear full-body cotton suits for short time
periods while their activities were videotaped.
NERL Role
HUD and CPSC
study. NERL
collaborated and
conducted analysis
of molds/fungi and
helped with lead
and pesticide
measurements.
EPA conducted
with contractor
support
Multiagency effort;
NERL was a
participant and led
the 9-home
exposure
component.
NERL funded IAG
with VA to conduct
study.
EPA grant to UC
Berkeley; NERL
augmented the
existing research
effort.
NERL designed,
contractor
conducted
103
-------
Table A-1. (Continued)
Study
Tampa
Asthmatic
Children's Study
(TAGS)
Pilot Study To
Evaluate Data
Collection
Methods for
Young
Children's and
Household
Activities
Detroit Exposure
and Aerosol
Research Study
(DEARS)
Accelerometer
Pilot Study
Pilot Study of
Waterborne
Infections
Date
2002
2004
2004-2007
2004
2005
Size
9 Residences
3 Homes
150
9
1296
Type
Convenience
sample, pilot
study to test
methods.
Convenience,
pilot study to
test
methodology
Randomized
household
selection with
qualification
criteria;
longitudinal
Convenience,
pilot study to
test
methodology
Convenience,
all volunteers
from
community
using public
water supply
Brief Description
Pilot study on methods for measuring personal, indoor
residential, outdoor residential, and ambient combustion-
related products, particulate matter, and air toxics. The study
identified microenvironmental factors affecting penetration of
pollutants into homes and reduction of exposures to pollutants
for asthmatic children (0-5 years of age).
Pilot aggregate exposure study of three homes to assess
burden of alternative exposure sample collection methods.
Recruitment from seven distinct neighborhoods in Detroit
required strong community relations and partnership with State
and local organizations. This study monitored for air pollutants
at the personal level and evaluates how well centrally located
(ambient) monitors represent exposure at the residential and
personal level. Sampling is for 5 days duration in summer and
again in winter.
Nine children <24 months old and their primary caregivers
participated in this study to (a) determine if very young children
will wear an accelerometer for relatively long periods of time
and comply with the protocol for its use, (b) evaluate how well
a caregiver can estimate the activity level of his/her infant or
toddler when completing an exposure-oriented time-activity
diary, and (c) compare accelerometer count output with
caregiver-provided estimates of children's activity level.
Measure antibodies in people's saliva for antibodies to
Cryptosporidium, Noroviruses, Rotaviruses, Helicobacter
pylori, and Toxoplasma gondii before and after installation of
an ultraviolet treatment system for a public water supply. Also
served as test of methodology for detecting people's prior
exposure to infectious agents.
NERL Role
NERL designed,
contractor
conducted
NERL designed,
contractor
conducted
NERL designed
and run; with
contractor support
and collaborators
such as the
University of
Michigan
NERL designed
and run, with
contractor support
EPA planned
research (NERL is
a collaborator),
contractor
conducted
104
-------
Table A-2. Types of Samples Collected in NERL Observational Human Exposure Studies
Samples or Data
Exposure Concern and Typical Type of Analysis
Environmental Media
Air pollutants
Soil
House dust
Surface wipes
Transferable residues
Duplicate diet
Handled food
Air pollutants being inhaled. Collect samples from central site, outside residence, and inside residence to assess
pollutants in various locales. Measure gaseous and particle-bound pollutants.
Estimate track in and subsequent dermal or inhalation contact. Analyze for metals, pesticides, etc.
Dust from carpets and floors that may result in dermal contact or reentrainment and inhalation. Analyze for
pesticides, metals, and tracers of outdoor sources.
Dermal contact, children's hand to mouth, and contact with food and subsequent ingestion. Analyze for metals and
pesticides. Examine eating and food preparation areas too.
Similar to surface wipes, using a surrogate for the transfer from the surface to the skin.
Use to assess ingestion exposures from food. Exact duplicate of amount and items eaten by participant and
analyzed for pesticides, metals, etc. Includes drinking water samples as part of diet or other beverages consumed.
Finger foods like cheese or luncheon meets that have been prepared and processed identically to foods children
might eat to evaluate how much pollution may be removed from surfaces and ingested with the food.
Human Activity Data
Time-activity diaries
Activity loggers
Questionnaires
Videotaping
Household inventories
Researcher observations
Recall diaries to account for all times and activities in a day. Information includes location and activities.
Device used to assess the nature of a person's activities. Portable nephelometers have been used to keep up with
people's activities by showing when they were near PM sources. Accelerometers to measure level of activity of
children at play. CIS and inertial devices to try to measure locations as a function of time of day.
Query participants about things like daily activities for themselves or their children; housing characteristics;
participant characteristics, including occupation, diet, smoking habits, hobbies, etc.; and recent use of pesticides or
other consumer products.
Use videos to measure frequency and duration of mouthing activities in children.
Inventory consumer products in house. Use items to ask about usage frequency and history.
Information about open doors and windows (air exchange), heating and cooking sources, pets, and other activities
that may lead to potential exposures may be observed.
Personal or Biological Samples
Urine
Personal air monitors
Hand wipes
Dermal surrogates
Saliva samples
Blood
Urinary excretion of pollutants and their contaminants give important information about the nature of prior
exposures and their magnitude.
Individual wears samplers on his/her person while going about normal activities to measure pollutants in breathing
zone of individual.
Hand wipes remove contaminants from skin surface. Analyze for pesticides, metals, organic chemicals, and use
values to estimate dermal exposure and hand-to-mouth ingestion of pollutants.
Participant may wear cotton garments or socks as a collector. Clothing is analyzed for pesticides, metals, organic
chemicals, etc. Measured contaminant quantities are used to estimate potential for dermal exposure.
Test for antibodies to infectious agents, suggesting prior exposure and infection by microbial agent.
Some epidemiological studies that NERL scientists have collaborated on have collected blood samples.
105
-------
-------
Appendix B
The Process for Development of This Document:
Description of the Expert Panel Workshop (November 28 and 29, 2006),
the External Peer Review by the HSRB (October 21-24, 2007),
and Public Comment
Expert Panel Workshop
An Expert Panel Workshop was convened in
Durham, NC, on November 28 and 29, 2006. An ad hoc
panel of experts was assembled to discuss issues
associated with the preparation of this document prior to
beginning its first draft. The workshop was coordinated
by ERG, Inc., who was also responsible for compiling
information from the workshop in a final workshop
report, available on the Scientific and Ethical
Approaches for Observational Exposure Studies
(SEAOES) Web site at www.epa.gov/nerl/sots.
The charge to the Expert Panel Workshop members
was as follows.
The panel is asked to consider these issues prior to
the workshop in preparation for discussion during
this workshop meeting:
1. Provide recommendations on the content and
organization of the document.
a. Identify the major scientific and ethical
areas/issues in the design and
implementation of observational human
exposure measurement studies mat should
be considered for inclusion in the document.
b. Identify specific elements in each of these
major areas that should be considered for
inclusion in the document.
c. Provide recommendations on the type and
level of information that should be
considered for inclusion in the document
when describing state-of-the-science
approaches, methods, techniques, or
standards.
d. Provide recommendations on the criteria
that should be considered when evaluating
and identifying the state-of-the-science for
the approaches, methods, techniques, or
standards.
2. Provide recommendations and listings of sources
of information for developing the document
including case studies where available.
3. Identify at least ten specific elements of the
design and implementation of these studies mat
the panel considers to have the most uncertainty
with regard to the "state-of-the-science," discuss
these elements, and provide recommendations on
state-of-the-science approaches for them.
The following individuals were members of the Expert
Panel.
Timothy Buckley (Chair)
Division of Environmental Health Sciences
School of Public Health
Ohio State University
Columbus, OH
Sophie Balk
Attending Pediatrician
Children's Hospital at Montefiore
Professor of Clinical Pediatrics
Albert Einstein College of Medicine
Bronx, NY
David Carpenter
Director, Institute of Health and Environment
University of Albany, SUNY
Rensselaer, NY
107
-------
Giselle Corbie-Smith
Department of Social Medicine
University of North Carolina
Chapel Hill, NC
Alan Fleischman
Senior Advisor
The New York Academy of Medicine
New York, NY
Natalie Freeman
Center for Environmental and Human Toxicology
Department of Physiological Sciences
University of Florida
Gainesville, FL
Loretta Jones
Healthy African American Families
Los Angeles, CA
Bruce Lanphear
Professor of Pediatrics and of Environmental Health
Division of General and Community Pediatrics
Cincinnati Children's Hospital Medical Center
Cincinnati, OH
Michael Lebowitz
Arizona Health Sciences Center
Colleges of Public Health and Medicine
University of Arizona
Tucson, AZ
Jerry Menikoff
Department of History and Philosophy of Medicine
University of Kansas Medical Center
Kansas City, KS
Rebecca Parkin
Associate Dean for Research and Public Health Practice
Professor of Environmental and Occupational Health
School of Public Health and Health Service
George Washington University Medical Center
Washington, DC
Review by the EPA Human Studies Review
Board and Public Comment
The process for developing this document included
the following steps after the Expert Panel Workshop.
• A draft document was written by NERL researchers.
• The draft document was distributed to internal EPA
staff for review and comment (see Acknowledgements
for the list of reviewers).
• The draft document was revised to address internal
reviewer comments; an external review draft
document was prepared.
• The availability of the external review draft document
for public comment was announced in a Federal
Register notice.
• An EPA docket was opened, and the external review
draft document was available for public comment for
45 days.
• The external review draft document also was provided
to EPA's Human Studies Review Board (HSRB) for
review and comment (see
http://www.epa.gov/osa/lisrb/ for information on the
HSRB).
• The review by HSRB was announced in a Federal
Register notice.
• HSRB met October 21-24, 2007, and discussed the
document during the meeting.
• HSRB provided EPA with comments on the document
in their final report of the October meeting.
The charge to the HSRB for the SEAOES document
review was that shown just below.
Draft Document on Scientific and Ethical
Approaches for Observational
Exposure Studies
Charge to the Human Studies Review Board
(October 4,2007)
Observational human exposure studies are
performed to collect information about individuals
and the environment around them in order to better
understand people's exposures. These studies
typically involve measurements of chemicals in the
food people eat, the water they drink, the air they
breathe and dust on the surfaces they touch. In
addition, information about the study participants and
their homes, work environments and activities are
collected, as well as biomonitoring samples. It is
important to understand why and how people are
exposed to chemicals in the environment for EPA to
fulfill its mission to protect human health.
108
-------
EPA scientists and their managers take the
protection of human subjects who participate in their
observational studies very seriously. The steps
needed to ensure protection of the human subjects are
often complex, and the specific actions will vary
depending on the objectives of the study, details
about the participants, and the communities in which
the studies are performed.
This document is intended as a resource and
reference for scientists in EPA's Office of Research
and Development (ORD) National Exposure
Research Laboratory (NERL) as they develop and
implement observational human exposure studies.
The authors recognize that this document may also
prove to be useful to others involved in exposure
science research, but the document is not meant to
represent an official Agency "guidance document"
and should not be used that way. This document does
not provide solutions to the scientific and ethical
issues that will undoubtedly arise as such studies are
undertaken: no document could provide all of the
answers in advance or develop a comprehensive
checklist for all such studies. Rather, this document
attempts to identify the types of issues that will need
to be considered and addressed as NERL researchers
plan and implement observational human exposure
studies. The researchers will need to work with others
- the study team, IRB members, EPA Human
Subjects Research Review Official (HSRRO), the
participants and their community, and other
stakeholders - to identify and address all of the
relevant issues for their particular study in order to
ensure that the specific elements of the study will
safeguard and protect the human research subjects.
Charge to the Human Studies Review Board
The draft document on Scientific and Ethical
Approaches for Obsen>ational Exposure Studies
consists of the following seven sections:
1. Introduction
2. Elements to be Considered in Study
Conceptualization and Planning
3. Ensuring Protection of Vulnerable Groups
4. Privacy, Confidentiality, and Other Concerns
Related to Observational Human Exposure
Measurement Studies
5. Creating an Appropriate Relationship Between
Participant and Investigator
6. Building and Maintaining Appropriate Community
and Stakeholder Relationships
7. Designing and Implementing Strategies for
Effective Communication
The Human Studies Review Board is asked to
address the following questions for each section of
the draft document:
1. One of the goals of the document is identify the
major scientific and ethical areas and issues that
researchers should address in the design and
implementation of observational human exposure
measurement studies, with the emphasis on the
areas requiring ethical considerations. Does each
section identify the major areas and issues where
ethical considerations should be addressed?
2. The document is intended to serve as a reference
and resource of information that researchers can
use in the design and implementation of
observational exposure studies. For each section,
are there additional sources of information mat
should be considered for inclusion?
3. Is the information presented accurately and clearly
in each section?
109
-------
-------
Appendix C
Recommended Content of a Human Subjects Protocol
The Council for International Organizations of Medical
Sciences (CIOMS, 2002) has developed a comprehensive
list of items that they recommend for inclusion in a
human subjects research protocol. Many of the items
that they identify are also useful for observational human
exposure studies.
Items Relevant to Observational Human
Exposure Studies
(1) The title of the study
(2) A summary of the proposed research in lay or
nontechnical language
(3) A clear statement of the justification for the study
(4) The investigators' views of the ethical issues and
considerations raised by the study and, if
appropriate, how it is proposed to deal with them
(5) A summary of previous studies on the research
problem, including unpublished studies known to
the investigators, and information on previously
published research on the topic
(6) A statement that the principles of the Belmont
Report and requirements specified in 40 CFR 26
will be implemented
(7) An account of previous submissions of the protocol
for ethical review and their outcomes
(8) A brief description of the sites where the research is
to be conducted, including information about the
adequacy of facilities for the safe and appropriate
conduct of the research, and relevant demographic
and epidemiological information about the
population to be studied
(9) The names and addresses of the funding
organization, research partners, and collaborators
(10) The names, addresses, institutional affiliations,
qualifications, and experience of the principal
investigator and other investigators
(11) The objectives of the study, its hypotheses or
research questions, its assumptions, and its
variables
(12) A detailed description of the design of the study
(13) The number of research subjects needed to achieve
the study objective, and how this was determined
statistically
(14) The criteria for inclusion or exclusion of potential
subjects and justification for the exclusion of any
groups on the basis of age, sex, social or economic
factors, or other reasons
(15) The justification for involving as research subjects
any persons with limited capacity to consent or
members of vulnerable social groups and a
description of special measures to minimize risks
and discomfort to such subjects
(16) The process of recruitment (e.g., advertisements)
and the steps to be taken to protect privacy and
confidentiality during recruitment
(17) A description and explanation of any and all
interventions
(18) The measurements to be performed in the study,
including environmental and biological sample
collection, and other data and information that will
be collected
(19) If applicable, clinical and other tests involving the
study participants that are to be carried out
(20) The rules or criteria according to which subjects
may be removed from the study or the study may be
terminated
(21) The methods of recording and reporting adverse
events or reactions, and provisions for dealing with
complications
(22) The potential benefits of the research to subjects
and to others
(23) The expected benefits of the research to the
population, including new knowledge that the study
might generate
111
-------
(24) The means proposed to obtain individual informed
consent and the procedure planned to communicate
information to prospective subjects, including the
name and position of the person responsible for
obtaining consent
(25) When a prospective subject is not capable of
informed consent, satisfactory assurance that
permission will be obtained from a duly authorized
person, or, in the case of a child who is sufficiently
mature to understand the implications of informed
consent but has not reached the legal age of
consent, mat knowing agreement, or assent, will be
obtained, as well as the permission of a parent, or a
legal guardian or other duly authorized
representative.
(26) An account of any economic or other compensation
or incentives to prospective subjects to participate,
such as offers of cash payments, gifts, or free
services or facilities, and of any financial
obligations assumed by the subjects, such as
payment for medical services
(27) The plans and procedures and the persons
responsible for communicating to subjects
information arising from the study (on harm or
benefit, for example) or from other research on the
same topic that could affect subjects' willingness to
continue in the study
(28) The plans to inform subjects about the results of the
study
(29) The provisions for protecting the confidentiality of
personal data and respecting the privacy of subjects,
including the precautions that are in place to
prevent disclosure of the results of a subject's
genetic tests to immediate family relatives without
the consent of the subject
(30) Information about how the code, if any, for the
subjects' identity is established; where it will be
kept; and when, how, and by whom it can be broken
in the event of an emergency
(31) Any foreseen further uses of personal data or
biological materials
(32) A description of the plans for statistical analysis of
the study, including plans for interim analyses, if
any, and criteria for prematurely terminating the
study if necessary
(33) A list of the references cited in the protocol
(34) The source and amount of funding of the research,
including the organization that is sponsoring the
research and a detailed account of the sponsor's
financial commitments to the research institution,
the investigators, the research subjects, and, when
relevant, the community
(35) The arrangements for dealing with financial or
other conflicts of interest that might affect the
judgment of investigators or other research
personnel, including informing the institutional
conflict-of-interest committee of such conflicts of
interest; the communication by that committee of
the pertinent details of the information to the ethical
review committee; and the transmission by that
committee to the research subjects of the parts of
the information that it decides should be passed on
to them
(36) The time schedule for completion of the study
(37) Particularly in the case of an industrial sponsor, a
contract stipulating who possesses the right to
publish the results of the study and a mandatory
obligation to prepare with and submit to the
principal investigators the draft of the text reporting
the results
(38) The circumstances in which it might be considered
inappropriate to publish findings, such as when the
findings of any study may present risks to or
stigmatize the interests of a community or
population or of a racially or ethnically defined
group of people
(39) A statement that any proven evidence of
falsification of data will be dealt with in accordance
with the policy of the sponsor to take appropriate
action against such unacceptable procedures
Source: CIOMS (The Council for International Organizations of
Medical Sciences) (2002). International Ethical Guidelines for
Biomedical Research Involving Human Subjects. World Health
Organization. Geneva. Switzerland.
112
-------
Appendix D
Recommendations for Enhancing Public Trust
Although the Report and Recommendations on Public
Trust in Clinical Research for the NIH Director from the
Director's Council of Public Representatives (COPR)
was developed in the context of NIH-supported clinical
research, many of the recommendations are applicable to
observational human exposure studies. A summary of
recommendations from the report is provided below.
Building Trust Through Community
Partnerships
Recommendation 1: Incorporate into the NIH
mission and philosophy that it values the involvement
of the community in research and create language
that expresses this value.
Recommendation 2: Encourage change in the
culture of the scientific community to ensure that
medical research is viewed in the context of a long-
term commitment to the community, not a one-time
research study.
Recommendation 3: Investigate ways to provide
mechanisms that allow for follow-up health care
when a clinical trial or treatment ends.
Building Relationships with Patients
(Participants) (True partnerships with
patients may not be possible, but
bidirectional relationships must be
enhanced.)
Recommendation 4: Educate and reorient the
current research community to the importance of
treating the public as a partner in the research
process.
Recommendation 5: Set the expectation across
the entire research community, NIH funded research
and beyond, that study results and outcomes should
be shared with the research participants and the larger
community promptly and consistently. This will
ensure translational research.
Building Partnerships with Community
Providers
Recommendation 6: Take action to interest
community providers in clinical research and
maintain their involvement.
Recommendation 7: Provide incentives (not just
financial) for primary health care providers and
community specialists to play a role hi clinical trials.
Building Trust in Scientists
Recommendation 8: Engage researchers,
educators, and academic institutions in incorporating
the public's perspective consistently at every level of
training and in both the conduct of clinical research
and the publication of findings from that research.
Recommendation 9: Focus on educational
strategies to help patients and communities better
understand clinical research. This will help scientists
because educating the public will empower and
prepare individuals to be informed partners in the
clinical research process. An informed and trusting
public will enhance research participation.
Building Trust in the NIH and Scientific
Research
Recommendation 10: Continue to develop and
fund efforts to build a national identity for the NIH
based on what NIH does best—research and
education—as a basis for enhancing public trust in
clinical research.
Recommendation 11: Review the role and impact
of Institutional Review Boards and other patient
protections in the clinical research process because
the public views these protections as less effective
than they should be.
Recommendation 12: Document and publish
"best practices" from efforts to reengineer the clinical
research enterprise as soon as the NIH begins to see
results, so that progress in improving public trust in
medical research grows rapidly and steadily.
Source: NIH (National Institutes of Health) (2005). Report and
Recommendations on Public Trust in Clinical Research for the NIH
Director from the Director's Council of Public Representatives
(COPR). National Institutes of Health, Director's Council of Public
Representatives, January 14, 2005. Available:
http://copr.nih.gov/reports/public_trust.asp [accessed 12 June 2007].
113
-------
-------
Appendix E
List of Acronyms and Abbreviations
AAP American Academy of Pediatrics
ACGIH American Conference of Governmental Industrial Hygienists
AHS Agricultural Health Study
ATSDR Agency for Toxic Substances and Disease Registry
BEI biological exposure index
CAB community advisory board
CBPR community-based participatory research
CDC Centers for Disease Control and Prevention
CFR Code of Federal Regulations
CIOMS Council for International Organizations of Medical Sciences
COPR National Institutes of Health Director's Council of Public Representatives
CPSC Consumer Product Safety Commission
CTEPP Children's Total Exposure to Persistent Pesticides and Other Persistent Organic
Pollutants
DEARS Detroit Exposure and Aerosol Research Study
DHEW U.S. Department of Health, Education, and Welfare
DMOC data monitoring and oversight committee
DNA deoxyribonucleic acid
DSMB data safety monitoring board
DSMP data and safety monitoring plan
EHCRB environmental health and community review board
EPA U.S. Environmental Protection Agency
ERG Eastern Research Group
FCN Federal Communicators Network
FDA Food and Drug Administration
HHS U.S. Department of Health and Human Services
HSRB Human Studies Review Board
HSRRO Human Subjects Research Review Official
HUD U.S. Department of Housing and Urban Development
ICR information collection request
IOM Institute of Medicine
IRB institutional review board
NAS National Academy of Sciences
NBAC National Bioethics Advisory Commission
NCI National Cancer Institute
NCS National Children's Study
NEI National Eye Institute
NEJAC National Environmental Justice Advisory Council
NERL National Exposure Research Laboratory
115
-------
NGO Nongovernmental organization
NHANES National Health and Nutrition Examination Survey
NHAPS National Human Activity Pattern Study
NHEXAS The National Human Exposure Assessment Survey
NHLBI National Heart, Lung, and Blood Institute
NHRPAC National Human Research Protections Advisory Committee
NIH National Institutes of Health
NRC National Research Council
OHRP Office for Human Research Protections
OMB Office of Management and Budget
OSMB observational study monitoring board
PM particulate matter
Q&As questions and answers
QAPP quality assurance project plan
RfD reference dose
SEAOES Scientific and Ethical Approaches for Observational Exposure Studies
TEAL Tribal Efforts Against Lead
TEAM Total Exposure Assessment Methodology
TLV threshold limit value
VOC volatile organic compound
WHO World Health Organization
116
-------
Appendix F
Glossary
Agent. A chemical, mineralogical, biological, or
physical entity that may cause deleterious effects in an
organism after the organism is exposed to it
[EPA/600/Z-92/001, May 1992].
Assent. A child's affirmative agreement to participate in
research. Mere failure to object should not, absent
affirmative agreement, be construed as assent [45 CFR
46.402(d)].
Autonomy. The capability and capacity to govern
oneself.
Beneficence. The ethical obligation to maximize
benefits and to minimize harms. This principle gives rise
to norms requiring mat the risks of research be
reasonable in light of the expected benefits, that the
research design be sound, and that the investigators be
competent both to conduct the research and to safeguard
the welfare of the research subjects. Beneficence further
proscribes the deliberate infliction of harm on persons;
this aspect of beneficence is sometimes expressed as a
separate principle, nonmaleficence (do no harm).
Child. A person who has not attained the age of 18
years [40 CFR 26.202(a)]
Collateral observations. Potentially unsafe hazards,
conditions, or situations unrelated to the research study
mat are observed by the research staff
Common Rule. The Common Rule is a short name for
"The Federal Policy for the Protection of Human
Subjects." It was adopted by more than a dozen Federal
departments or agencies in 1991, with EPA adapting it in
Title 40 CFR Part 26 Subpart A.
Community-based participatory research (CBPR).
Collaborative research with a community in which the
community is involved in all phases of the research.
A fundamental concept is that the research aims to
combine knowledge with action and to achieve social
change to improve health outcomes and eliminate health
disparities.
Confidentiality. The keeping safe or not redisclosing
by one of the parties in a confidential relationship
information that originally was disclosed in the
confidential relationship
Environmental justice. The fair treatment and
meaningful involvement of all people regardless of race,
color, national origin, or income with respect to the
development, implementation, and enforcement of
environmental laws, regulations, and policies
Exposure. Contact of a chemical, physical, or
biological agent with the outer boundary of an organism
(e.g., a person). Exposure is quantified as the
concentration of the agent in the medium in contact
integrated over the time duration of mat contact. (The
definition is taken from Guidelines for Exposure
Assessment [EPA/600/Z-92/001, May 1992]).
Exposure concentration. The exposure mass divided
by the contact volume or the exposure mass divided by
the mass of contact volume depending on the medium
Exposure duration. The length of time over which
continuous or intermittent contacts occur between an
agent and a target. For example, if an individual is in
contact with an agent for 10 minutes per day for 300
days over a 1-year time period, the exposure duration is
1 year.
Exposure event. The occurrence of continuous contact
between an agent and a target
Exposure pathway. The course an agent takes from the
source to the target
Exposure route. The way an agent enters a target after
contact (e.g., by ingestion, inhalation, or dermal
absorption)
117
-------
Human subject. A living individual about whom an
investigator (whether professional or student)
conducting research obtains: (1) data through
intervention or interaction with the individual, or (2)
identifiable private information [40 CFR26.102(f)]
Informed consent. A potential participant's
autonomous authorization to participate in the research.
The three pillars of valid informed consent are
(1) information, (2) comprehension, and (3) voluntary
participation.
Institutional review board (IRB). An 1KB established
in accord with and for the purposes expressed in EPA's
Policy for Protection of Subjects in Human Research
conducted and supported by EPA [40 CFR 26.102(g)]
Justice. The ethical obligation to treat each person in
accordance with what is due to him or her. In the ethics
of research involving human subjects, the principle
refers primarily to distributive justice, which requires the
equitable distribution of both the burdens and the
benefits of participation in research. Differences in
distribution of burdens and benefits are justifiable only if
they are based on morally relevant distinctions among
persons.
Minimal risk. The probability and magnitude of harm
or discomfort anticipated in the research are not greater
in and of themselves than those ordinarily encountered
in daily life or during the performance of routine
physical or psychological examinations or tests [40 CFR
26.102(1)].
Nonmaleficence. The proscription of deliberate
infliction of harm on persons
Observational human exposure study. Studies that
involve collection of human exposure data (including
environmental, biological, survey, activity, and various
other forms of data) under real-world field conditions
during normal participant day-to-day activities, with no
additional exposures to the chemical being studied
because of participation in the study. The studies involve
interaction with study participants but do not involve
intervention or manipulation of the factors being studied,
and there is no attempt by the researcher to affect the
outcome.
Observational research. Any human research that does
not meet the definition of research involving intentional
exposure of a human subject [40 CFR 26.302]
Privacy. Control over the extent, timing, and
circumstances of sharing oneself (physically,
behaviorally, or intellectually) with others
Research. A systematic investigation, including
research development, testing, and evaluation, designed
to develop or contribute to generalizable knowledge
Research involving intentional exposure of a human
subject. A study of a substance in which the exposure
to the substance experienced by a human subject
participating in the study would not have occurred but
for the human subject's participation in the study [40
CFR26.202(b)]
Respect for persons. A fundamental ethical value that
is the basis of much of modem bioethical thought and
regulation. The concept incorporates at least two
fundamental ethical considerations, namely (1) respect
for autonomy, which requires that those who are capable
of deliberation about their personal choices should be
treated with respect for their capacity for self-
determination; and (2) protection of persons with
impaired or diminished autonomy, which requires that
those who are dependent or vulnerable be afforded
security against harm or abuse.
Source. The origin of an agent for the purposes of an
exposure assessment
Stakeholder. A person or group who has a valid
interest in an activity, who can affect or is affected by
the activity, and who stands to gain or lose depending on
the decisions implemented
Stressor. Any entity, stimulus, or condition that can
modulate normal functions of the organism or induce an
adverse response (e.g., agent, lack of food, drought)
Vulnerability. A substantial incapacity to protect one's
own interests owing to such impediments as lack of
capability to give informed consent, lack of alternative
means of obtaining medical care or other expensive
necessities, or being a junior or subordinate member of a
hierarchical group. Accordingly, special provision must
be made for the protection of the rights and welfare of
vulnerable persons.
Vulnerable groups. Populations extended additional
human subjects protections, such as children, individuals
with questionable capacity to consent, prisoners, fetuses
and pregnant women, the terminally ill, students and
employees, and comatose patients, etc.
118
-------
-------
SCIENCE
&EPA
PRESORTED STANDARD
POSTAGES FEES PAID
United States
Environmental Protection
Agency
Office of Research and Development (81 OR)
Washington, DC 20460
Official Business
Penalty for Private Use
$300
PERMIT NO. G-35
Recycled/Recyclable Printed on paper that contains a minimum of
50% posiconsumer fiber content processed chlorine free
------- |