Content uploaded by David B Resnik
Author content
All content in this area was uploaded by David B Resnik on Apr 23, 2019
Content may be subject to copyright.
IRB: EthIcs & human REsEaRch Ju ly- august 2016
11
Employees of colleges, universities, medical cen-
ters, or private companies often participate in
research conducted by their employers. Research
with employee participants ranges from low-risk sur-
veys or biological sample collection protocols to riskier
studies, such as clinical trials. Employee participants
may include temporary, full- or part-time workers,
visiting faculty members, or fellows. They may enroll in
studies open to all individuals who meet the inclusion
criteria or in those that specifically target employees,
such as research on factors that affect workplace mo-
rale or health risks related to occupational exposures.1
While enrollment of employees in employer-based
research can yield many benefits for science and society,
it can also give rise to some ethical concerns.2 For
example, in a report published in 1998, the nonprofit
Environmental Working Group concluded that some
pesticide companies had conducted risky pesticide
exposure experiments on their own employees during
the 1980s and 1990s.3 In 2001, Ellen Roche, a labora-
tory technician at Johns Hopkins University (JHU),
died from inhaling hexamethonium in an asthma study
conducted with healthy volunteers. Most of the study
participants were JHU employees and received $365
and a day off work for their participation. An investi-
gation by the United States Office for Human Research
Protections faulted the researchers for not adequately
reviewing the literature on the dangers of inhaling this
drug.4 In 2005, a committee at Seoul University in
South Korea found that Woo Suk Hwang had pub-
lished two articles that included fabricated and falsified
data about the derivation of human embryonic stem
cells by nuclear transfer and that he had obtained eggs
for his research from women working in his laboratory.
Hwang paid the women $1400 each for their eggs and
helped them fill out paperwork for the egg donation.5
Federal regulations governing research with humans
do not single out employees as a special category of
research participants requiring extra protection. How-
ever, due to their susceptibility to coercion or undue
influence in the work environment, employees are a
vulnerable population when recruited to participate
in employer-based research.6 Federal regulations7 and
ethical guidelines8 require that research includes safe-
guards to protect the rights and welfare of vulnerable
participants. A great deal has been written about other
types of vulnerable participants, such as children, men-
tally disabled adults, pregnant women, and prisoners,
but very little has been written about employees.9 This
article will review some of the ethical issues related to
employee participation in employer-based research and
consider some strategies and policies for protecting
their rights and welfare.10
Informed Consent
Perhaps the most important ethical issue regarding
employee participation in employer-based research
is ensuring that employees can make free and informed
choices concerning their participation.11 Research regu-
lations and guidelines require that informed consent
take place under conditions that minimize the potential
for coercion or undue influence.12 They also require
that participants be able to withdraw from research at
any time without penalty or loss of benefits to which
they would otherwise be entitled.13
Coercion involves the use of force, threats, or
intimidation to achieve compliance.14 Coercion could
occur if an employee faces the threat of an adverse
consequence, such as a negative performance appraisal
from a supervisor, if he or she refuses to participate in
research. Coercion need not involve explicit threats,
however; threats in the workplace may be implicit or
subtle. For example, if the supervisors of a research
by DaviD b. Resnik
Employees as Research Participants: Ethical and Policy Issues
Resnik DB. Employees as research participants: Ethical and policy issues. IRB:
Ethics & Human Research 2016;38(4):11-16.
© 2016 The Hastings Center. Permission is required to reprint.
July -A ugust 2016 IRB: EthIcs & humAn REsEARch
12
group announce at a staff meeting that they expect
members of the group to donate biological samples for
their study, a group member might reasonably conclude
that the employers might take action against him or her
for not complying.
Undue influence involves taking advantage of
someone’s weakness or vulnerability to induce him or
her to make a particular decision.15 In discussions of
human research ethics, various commentators have
claimed that excessive financial incentives for partici-
pation in research may constitute undue influence in
some situations.16 While financial incentives offered to
employees may sometimes be a concern in employer-
based research, workplace incentives are likely to be a
far greater concern. For example, if a supervisor offers
an employee a significant employment benefit for par-
ticipating in a research study (such as a high rating on
a performance appraisal), this could constitute undue
influence if it skews the employee’s judgment and leads
him or her to make an ill-advised decision to enroll in
the study.
The greatest potential for coercion or undue influ-
ence occurs when the employee’s direct supervisor
interacts with the employee, since the supervisor may
pressure the employee into enrolling or remaining in a
study when the employee is contemplating withdraw-
ing. The supervisor might also offer employment ben-
efits to an employee who enrolls in an employer-based
study. Even when supervisors do not make explicit
threats or promises during recruitment, the simple act
of approaching a subordinate concerning participation
in a study can create some pressure to enroll.
The potential for coercion or undue influence is also
significant when the employee’s indirect supervisor
interacts with the employee, since the indirect supervi-
sor is above the supervisor in the chain of command
and therefore can influence the supervisor’s actions.
The indirect supervisor may pressure an employee
to participate in one of his or her studies or instruct
the employee’s supervisor to take steps to ensure that
employees participate in the study. Employees who are
enrolled in a study conducted by an indirect supervisor
may be afraid to withdraw from the study because they
are concerned that the indirect supervisor could tell
their supervisor to penalize them for withdrawing.
Coercion or undue influence may occur to a lesser
extent when coworkers interact with an employee,
since they may place pressure on the employee to enroll
in one of their studies or a study conducted by their
supervisor. Although coworkers do not have as much
influence on an employee as the employee’s supervisor,
the employee may be concerned that coworkers will
adopt a negative attitude toward him or her if he or
she does not participate in their research (or withdraws
from a study), which could lead to tensions in the
workplace that could adversely affect the employee.
The lowest potential for coercion or undue influence
happens when independent contractors or employ-
ees outside of an employee’s work group or chain of
command interact with the employee in research, since
these individuals probably have no relationship with
the employee that could affect the decision to enroll in
a study or withdraw.
There are several strategies that investigators, in-
stitutions, and institutional review boards (IRBs) can
Table 1.
Strategies for Protecting Employees in Employer-
Based Research
• Research proposals that plan to enroll employees should
describe measures that will be implemented to protect the
rights and welfare of employee participants.
• Informed consent documents and discussions should
clearly indicate that an employee’s decision concerning
research participation will have no impact on his or her
employment status or benefits.
• Supervisors should not directly recruit employee subordi-
nates for research participation; recruitment by supervisors
should normally occur indirectly, for example, by means of
general announcements or flyers.
• Where feasible, supervisor interactions with subordinate
employee research participants should be restricted.
• If appropriate, an independent party should monitor the
informed consent process when supervisors enroll subordi-
nate employees in research.
• Research staff should receive training on protecting em-
ployees participating in employer-based research.
• Investigators, institutions, and IRBs should implement
strategies to minimize or control for bias, such as develop-
ing appropriate recruitment strategies and survey methods
and assuring employees that their confidentiality will be
protected.
• The IRB should carefully review studies involving employ-
ees as participants in research to ensure that their rights
and welfare are adequately protected.
IRB: EthIcs & human REsEaRch Ju ly- august 2016
13
implement to minimize the potential for coercion and
undue influence when employees participate in employ-
er-based research (Table 1).17 First, research proposals
that include a plan to enroll employees should describe
strategies for avoiding coercion and undue influence.18
The best way to avoid coercion or undue influence is
to limit supervisor-employee interactions related to a
study, since this strategy can preclude opportunities
for coercion or undue influence. Ideally, the employee’s
supervisor should have no contact with the employee
during recruitment for, enrollment in, or the conduct of
research, nor should the supervisor have access to the
employee’s identified data or samples. However, it may
not always be feasible to bar supervisors from inter-
acting with employees during research. For example,
a supervisor may be the only person on the research
team qualified to perform an exam or procedure, or he
or she may need to have access to identified samples
or data in order to conduct research activities. When
employee-supervisor interactions cannot be avoided,
the proposal should describe other steps that will be
taken to avoid coercion or undue influence.
Second, informed consent documents and discus-
sions should assure employees that participation in
research will have no effect on their employment status
or benefits and that they may withdraw at any time
without negative repercussions. If supervisors will
obtain consent from employees, it may be desirable to
have an independent party monitor the consent process
to ensure that employees are able to make free and
informed choices.
Third, investigators should train study staff members
on avoiding coercion or undue influence of employee
research participants. Training could address recruit-
ment, consent, and other aspects of research that
involve the potential for coercion or undue influence.
Fourth, IRBs should review research proposals, con-
sent forms, and other study documents to ensure that
they include adequate provisions for protecting em-
ployees from coercion or undue influence. An IRB may
require consent monitoring, staff training, or any other
provision that it deems necessary to protect employees.
Privacy and Confidentiality
Protecting privacy and confidentiality are also im-
portant ethical considerations related to employee
participation in research. Research regulations and
guidelines require that studies with human partici-
pants include appropriate measures to protect privacy
and confidentiality.19 Privacy is a broader notion than
confidentiality and includes physical privacy as well as
informational privacy. Physical privacy refers to an area
where a person has a reasonable expectation of privacy.
For example, watching someone undress without his
or her permission would be an invasion of physical
privacy but not necessarily informational privacy. Con-
fidentiality refers to the protection of private informa-
tion about an individual. Safeguarding privacy and
confidentiality can be difficult in the workplace because
employees may be tempted to share gossip about peers
who participate in research or use private information
inappropriately (for example, in a performance evalua-
tion).
Protecting privacy and confidentiality is especially
important when employees participate in research for
two reasons. First, protecting privacy and confidential-
ity can help to prevent invasions of privacy or breaches
of confidentiality that can lead to embarrassment,
discrimination, stigma, or other negative effects on the
employee. Although research personnel usually receive
training on procedures and rules for protecting confi-
dentiality, employees may be concerned that supervi-
sors or coworkers will be unusually tempted to breach
confidentiality, given the nature of their relationship.
Second, protecting privacy and confidentiality may
also be necessary to minimize coercion or undue influ-
ence from supervisors or coworkers. For example, if
an IRB determines that the best way to avoid coercion
or undue influence when an employee participates in a
study is to bar the employee’s supervisor from hav-
ing any contact with the employee in the study, then it
would be necessary to implement confidentiality and
privacy protections to prevent the supervisor from
accessing that employee’s private information. The su-
pervisor might be permitted to access only deidentified
samples or data from employee subordinates.
There are several strategies that investigators, insti-
tutions, and IRBs can implement to protect employees’
privacy and confidentiality when they participate in
research (Table 1). First, research proposals that plan to
Ideally, an employee’s supervisor should have no
contact with the employee during recruitment
for, enrollment in, or the conduct of research, nor
should the supervisor have access to the
employee’s identified data or samples.
July -A ugust 2016 IRB: EthIcs & humAn REsEARch
14
enroll employees should describe measures for protect-
ing privacy and confidentiality. Most of these measures
would apply to protecting the privacy or confidentiality
of any participant, regardless of employment status.
Some of these may include limiting access to research
records and samples, storing research records and
samples securely, labeling samples or data with a code
(instead of personal identifiers), and using private areas
for enrollment, testing, sample collection, and other
research activities. As mentioned, a research project
might also bar a supervisor from having access to an
employee’s identified samples or data. Second, informed
consent documents and discussions should inform
employees of measures that will be taken to protect
their privacy and confidentiality. A third strategy would
be for investigators to train research staff members on
protecting the privacy and confidentiality of employee
research participants. The training could address the
temptations for breaching confidentiality that can arise
when employees participate in research and the impor-
tance of protecting confidentiality and privacy. Finally,
the IRB should review research proposals, consent
forms, and other study documents to ensure that they
include adequate protections for employee privacy and
confidentiality. An IRB might require, for example, that
supervisors have access only to deidentified samples or
data from employee participants.
Issues of Bias
Employee participation in employer-based research
may create a potential for bias in research. For
example, if the workforce demographic is significantly
different from that of the general population, over-
recruitment of employees for a study might bias the
results, which could undermine the generalizability
of the findings. Also, employees might give differ-
ent answers to survey questions than members of the
general population due to their employment status. For
example, they might be more hesitant than members of
the general population to self-report illegal or unethical
behaviors because they are concerned that their em-
ployer might have access to this information.20
The potential for bias is an ethical issue in research
with human participants because the risks a study im-
poses on participants (and others) should be reasonable
in relation to the expected benefits to the participants
and society (in other words, the knowledge gained).21
If research is biased, then the benefits of the study may
not be sufficient to justify the imposition of risks.22
Investigators, institutions, and IRBs should imple-
ment strategies to minimize or control for bias, such
as developing appropriate recruitment strategies and
survey methods, and assuring employees that their
confidentiality will be protected.23
Consideration of Policy Options
The remainder of this article will consider policies
that institutions or IRBs have adopted or might
consider adopting to protect employees involved in
employer-based research. Since there are no published
empirical studies on institutional or IRB policies
concerning the protection of employee research par-
ticipants, the discussion below is based on review of
selected policies obtained by a nonsystematic Internet
search conducted on August 25, 2015.
Some institutions prohibit employees from partici-
pating in some types of research that IRBs classify as
more than minimal risk. The federal research regula-
tions define research as posing minimal risk when
“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.”24 JHU’s policy,
for example, prohibits employees from enrolling in
more than minimal risk research conducted by their
supervisors unless the research also offers employees
direct medical benefits.25 The University of Pittsburgh26
and the University of Utah27 prohibit employees from
participating in more than minimal risk research con-
ducted by supervisors, regardless of the potential for
participant benefits.
Although a study’s level of risk is always an impor-
tant ethical consideration related to protecting human
participants, a policy that prohibits employees from
participating in some types of more than minimal risk,
employer-based research is ill-advised. First, adopting
such a policy could prevent employees from obtaining
significant benefits from research. If an employee has a
disease and participation in a clinical study conducted
by the employer offers the employee the best chance at
treatment, prohibiting the employee from participat-
ing in the study could adversely affect the employee’s
welfare. Although the main goal of clinical research
is to develop new knowledge, many studies, such as
phase II or III trials, offer participants potential medi-
cal benefits.28 Even studies that do not offer employees
direct medical benefits may benefit them in other ways
IRB: EthIcs & human REsEaRch Ju ly- august 2016
15
by providing them with information about their health
(such as the results of medical tests or exams), financial
compensation, or a sense of satisfaction in contributing
to the advancement of science or public health.
Second, prohibiting employees from participating in
employer-based research that poses more than minimal
risk restricts their freedom unnecessarily. Employees
might want to be able to decide whether to participate
in such research because, as just suggested, they would
like to contribute to the advancement of human knowl-
edge or earn some extra money if the researcher offers
payment for participating in a study. Employees should
have as much freedom to decide whether to participate
in a study as nonemployees.
Third, the main ethical concerns with involving em-
ployees in employer-based research—coercion, undue
influence, invasion of privacy, or breach of confidenti-
ality—are independent of the overall risks of a study.
The overall risk of a study has little or no effect on
its potential for causing these problems. Coercion or
undue influence could occur during the recruitment of
participants for a clinical trial of a new drug or a mini-
mal risk biological sample collection study. It is impor-
tant to take steps to prevent coercion, undue influence,
and other ethical problems, regardless of the overall
risks associated with a study.
Fourth, “minimal risk” is a vague concept subject
to interpretation. Studies have demonstrated signifi-
cant variation in how IRBs interpret minimal risk.29
A policy that prohibits certain types of minimal risk
research involving employees might encourage an IRB
to classify a study that plans to enroll employees as no
more than minimal risk so that it won’t run afoul of the
policy, which would distort IRB risk assessment. IRBs
should make risk classifications based on the nature of
the research and the study population, not based on the
institutional policy constraints.
Although prohibiting some types of more than
minimal risk research involving employee participants
in employer-based research is ill-advised, one can
understand why some institutions have adopted such a
policy. JHU probably adopted its policy in reaction to
the tragic death of Ellen Roche, which made national
headlines and brought adverse publicity to the institu-
tion. However, there is no evidence that Roche’s status
as a JHU employee had anything to with her death. The
main problems with the study were that the investiga-
tor had not adequately researched the risks of inhal-
ing hexamethonium and had not reported breathing
problems experienced by a prior participant to the IRB.
Additionally, the informed consent document did not
adequately describe the risks of the drug or mention
that it had not been approved by the Food and Drug
Administration.30 Tragic cases seldom lead to good
policy, however.
Some institutions prohibit or discourage supervi-
sors from directly recruiting employees to participate
in their studies. For example, the National Institutes of
Health intramural research program strictly limits the
methods by which supervisors may recruit subordinate
employees to participate in a study; such recruitment in
person or by phone, mail, or email is prohibited. Inves-
tigators may, however, recruit subordinate employees
by means of general announcements or a flyer distrib-
uted in the workplace.31 Duke University,32 JHU,33 the
University of Pittsburgh,34 and the University of Utah35
also prohibit direct recruitment of employees by their
supervisors. Yale University prohibits supervisors from
recruiting subordinate employees unless the study is for
convenience sampling.36 Washington University pro-
hibits direct recruitment of subordinate employees by
supervisors unless they have already agreed to receive
solicitations for research participation by mail.37
Because direct recruitment of employees by supervi-
sors poses a significant risk of coercion or undue influ-
ence, polices that prohibit or strongly discourage this
interaction make sense. Recruitment of employees by
supervisors should occur indirectly by means of general
announcements or flyers. Only independent contractors
or employees who are outside of a potential research
participant’s chain of command or work group should
be involved in direct recruitment of the employee.
Protecting Employees in Employer-Based Research
Employees who participate in employer-based
research are vulnerable participants due to their
susceptibility to coercion or undue influence in the
work environment. Investigators, IRBs, and institutions
should take appropriate measures to protect employees
from coercion, undue influence, invasion of privacy, or
breach of confidentiality when they participate in em-
ployer-based research (Table 1). An employee’s decision
about participation in employer-based research should
have no effect on the employee’s hiring, promotion,
performance evaluation, or other employment benefits.
There should be no penalty for withdrawing from an
employer-based study. Informed consent documents
and discussions should clearly indicate that an employ-
July -A ugust 2016 IRB: EthIcs & humAn REsEARch
16
ee’s decision about research participation will have no
effect on his or her employment status or benefits.
Institutions and IRBs should develop policies that
protect the rights and welfare of employees who
participate in employer-based research. Policies that
prohibit employees from participating in some types of
employer-based research that poses more than minimal
risk are ill-advised because they may prevent employ-
ees from obtaining research benefits and may restrict
their freedom unnecessarily. Such policies also have
little effect on protecting employees from coercion or
undue influence, which are the main risks of employee
participation in employer-based research. Policies that
prohibit investigators from directly recruiting employee
subordinates make sense, because they help protect
employees from coercion or undue influence without
undermining research or employee rights or welfare. To
better understand how institutions and IRBs are pro-
tecting employees involved in employer-based research,
it would be useful to conduct a survey of institutional
or IRB policies on this topic.
n David B. Resnik, JD, PhD, is a bioethicist and the IRB chair at the
National Institute of Environmental Health Sciences, in the National
Institutes of Health.
Acknowledgments
This research was supported by the intramural program
of the National Institute of Environmental Health Sciences
(NIEHS) in the National Institutes of Health (NIH). It
does not represent the views of the NIEHS, NIH, or U.S.
government.
References
1. Schwenzer KJ. Protecting vulnerable participants in clinical
research: Children, pregnant women, prisoners, and employees.
Respiratory Care 2008;53(10):1342-1349.
2. Meyers K. Drug company employees as research participants:
Programs, problems, and ethics. IRB: A Review of Human Subjects
Research 1979;1(8):5-6.
3. Environmental Working Group. The English patients: Hu-
man experiments and pesticide policy. 1998. http://static.ewg.org/
reports/1998/EnglishPatients.pdf?_ga=1.234944519.725809167.142
7468215.
4. Kolata G. Johns Hopkins admits fault in fatal experiments.
New York Times, July 16, 2001:A1.
5. Cho MK, McGee G, Magnus D. Research conduct: Lessons of
the stem cell scandal. Science 2006;311(5761):614-615.
6. Council for International Organizations of Medical Sciences.
International ethical guidelines for biomedical research involving
human participants. 2002. http://www.cioms.ch/publications/lay-
out_guide2002.pdf.
7. Department of Health and Human Services. Protection of Hu-
man Subjects. 45 CFR 46.111(b).
8. Emanuel EJ, Wendler D, Grady C. What makes clinical re-
search ethical? JAMA 2000;283(20):2701-2711.
9. Although this article focuses on the ethical issues of conduct-
ing research on employees, similar issues arise when employee family
members participate in research because of their close relationship to
employees. See ref. 1, Schwenzer 2008.
10. Researchers sometimes participate in their own studies. Al-
though this could sometimes be considered employee participation in
employer-based research, the focus here will be on research involving
other people as participants. For a discussion of issues that arise in
self-experimentation, see Davis JK. Self-experimentation. Account-
ability in Research 2003;10(3):175-187.
11. See ref. 1, Schwenzer 2008.
12. Grady C. Enduring and emerging challenges of informed con-
sent. NEJM 2015;372(9):855-862; see ref. 7, Department of Health
and Human Services 2009.
13. See ref. 7, Department of Health and Human Services 2009.
14. Black’s Law Dictionary, 2nd pocket ed. St. Paul, MN: West
Publishing, 2001. Coercion; pp. 106-107.
15. See ref. 14, Black’s Law Dictionary 2001. Undue influence;
p. 732.
16. Grady C. Payment of clinical research participants. Journal of
Clinical Investigation 2005;115(7):1681-1687.
17. Office for Human Research Protections. IRB Guidebook.
1993. http://www.hhs.gov/ohrp/archive/irb/irb_guidebook.htm. See
ref. 1, Schwenzer 2008.
18. National Institutes of Health, Office of Human Subjects Re-
search Protection. SOP14F: Research involving NIH staff as subjects.
2013. http://ohsr.od.nih.gov/ohsr/public/SOP_14F_v1_6-23-13_508.
pdf.
19. See ref. 7, Department of Health and Human Services 2009;
see ref. 8, Emanuel et al. 2000.
20. Donaldson SI, Grant-Vallone EJ. Understanding self-report
bias in organizational behavior research. Journal of Business and
Psychology 2002;17(2): 245-260.
21. See ref. 7, Department of Health and Human Services 2009;
see ref. 8, Emanuel et al. 2000.
22. See ref. 8, Emanuel et al. 2000.
23. See ref. 1 Schwenzer 2008.
24. See ref. 7, Department of Health and Human Services 2009.
25. Johns Hopkins University, Committee on Participation Poli-
cies for Human Participants Research. 2004. Policy governing the
recruitment and enrollment of employees in research involving hu-
man participants. http://jhuresearch.jhu.edu/Policy_on_Human_Sub-
ject_Recruitment%20-Employees_06.pdf.
26. University of Pittsburgh, Institutional Review Board. Research
involving employees as research participants. 2015. http://www.irb.
pitt.edu/content/research-involving-employees-research-participants.
27. University of Utah. Recruitment and enrollment of employees
in research. 2014. http://irb.utah.edu/_pdf/IGS%20-%20Recruit-
ment%20and%20Enrollment%20of%20Employees1.pdf.
28. Levine RJ. Ethics and Regulation of Clinical Research, 2nd
ed. New Haven: Yale University Press, 1988.
29. Shah S, Whittle A, Wilfond B, et al. How do institutional re-
view boards apply the federal risk and benefit standards for pediatric
research? JAMA 2004; 291(4):476-482.
30. See ref. 4, Kolata 2001.
31. See ref. 18, National Institutes of Health 2013.
32. See ref. 25, Johns Hopkins University 2004.
33. See ref. 26, University of Pittsburgh 2015.
34. Duke University, Human Research Protection Program. Re-
cruitment of students and employees as research participants. 2008.
http://irb.duhs.duke.edu/wysiwyg/downloads/Recruitment_of_Stu-
dents__Employees_Policy.pdf.
35. See ref. 27, University of Utah 2014.
36. Yale University, Institutional Review Board. IRB policy
350: Participation of Yale students, fellows, trainees or employees
in research. 2012. http://www.yale.edu/hrpp/policies/documents/
Policy_350_StudentsEmployees_9-13-12--2.pdf.
37. See Washington University. Ethical review guidelines for
research involving students, employees and/or normal volunteers
as research participants. 2005. http://hrpo.wustl.edu/wp-content/
uploads/2015/01/7-Enrolling-Students-Employees-or-Normal-Volun-
teers.pdf.
