&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

-------