Dearbhail Bracken-Roche Et Al. The Concept of Vulnerability' in Research Ethics (2017)

Bracken-Roche et al.
Health Research Policy and Systems (2017) 15:8

DOI 10.1186/s12961-016-0164-6
RESEARCH Open Access
The concept of ‘vulnerability’ in research

ethics: an in-depth analysis of policies and
guidelines
Dearbhail Bracken-Roche1,2, Emily Bell1, Mary Ellen Macdonald3 and Eric Racine1,2,4,5*
Abstract
Background: The concept of vulnerability has held a central place in research ethics guidance since its introduction in
the United States Belmont Report in 1979. It signals mindfulness for researchers and research ethics boards to the
possibility that some participants may be at higher risk of harm or wrong. Despite its important intended purpose and
widespread use, there is considerable disagreement in the scholarly literature about the meaning and delineation of
vulnerability, stemming from a perceived lack of guidance within research ethics standards. The aim of this study was
to assess the concept of vulnerability as it is employed in major national and international research ethics policies and
guidelines.
Methods: We conducted an in-depth analysis of 11 (five national and six international) research ethics policies
and guidelines, exploring their discussions of the definition, application, normative justification and implications of
vulnerability.
Results: Few policies and guidelines explicitly defined vulnerability, instead relying on implicit assumptions and the
delineation of vulnerable groups and sources of vulnerability. On the whole, we found considerable richness in the
content on vulnerability across policies, but note that this relies heavily on the structure imposed on the data through
our analysis.
Conclusions: Our results underscore a need for policymakers to revisit the guidance on vulnerability in research ethics,
and we propose that a process of stakeholder engagement would well-support this effort.
Keywords: Research ethics, Vulnerable populations, Vulnerability, Research policy, Ethics policy, Research oversight
Background: the function of vulnerability in appropriately balanced, that the proposed strategy for sub-
research ethics guidance and policy ject recruitment is fair, and that voluntary, informed con-
Research on human subjects is thought to be fundamen- sent will be sought from each potential subject [1]. The
tally ethically challenging, requiring ethics standards to concept of vulnerability, which finds it origins in the
guide researchers as well as approval and oversight of re- United States Belmont Report of 1979 [2], plays a central
search proposals from independent committees. Society role in research ethics thinking, drawing attention to situ-
allows researchers to invite individuals to participate in re- ations where these conditions may not be met [1]. Since
search once certain conditions are met, including a re- 1979, the number of legal and non-legal research ethics
search ethics board’s (REB, also known as Institutional policies and guidelines has increased tremendously and,
Review Boards, or IRBs, or Research Ethics Committees, with them, the use and scope of the concept of vulnerabil-
or RECs) determination that risks and benefits are ity or vulnerable populations [2, 3]. However, there is
much scholarly disagreement over the appropriate mean-
* Correspondence: [email protected] ing and application of this concept in research ethics, and
1
Neuroethics Research Unit, Institut de recherches cliniques de Montréal, 110 policymakers are charged with the challenge of navigating
Avenue des Pins Ouest, Montréal, QC H2W 1R7, Canada
2
Biomedical Ethics Unit and Division of Experimental Medicine, McGill
this contentious landscape in the development and refine-
University, Montréal, QC, Canada ment of research guidelines and policies [4]. A growing
Full list of author information is available at the end of the article
© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Bracken-Roche et al. Health Research Policy and Systems (2017) 15:8 Page 2 of 18
body of literature critiques and aims to advance the way identification of specific factors within the research con-
vulnerability is conceptualised and employed in research text and the participants’ personal situation that create
ethics, with major debates regarding foundational ele- possible vulnerabilities [6].
ments of this important ethical concept [5–10]. The foundational debate about the concept of vulner-
ability revolves around its definition, with various pro-
posals made for its delineation in the literature. Arguing
Major debates surrounding the concept of vulnerability in that vulnerability lacks an organising principle, Hurst [5]
research ethics suggests that vulnerable persons are properly conceived
There is widespread agreement that some research par- of as those who have “an identifiably increased likeli-
ticipants may be particularly vulnerable and in need of hood of incurring additional or greater wrong”. This ac-
special protections, yet the concept of vulnerability itself count emphasises that both individual and situational
has been described as ‘vague’ [11], with a lack of consen- factors must be evaluated in defining vulnerability be-
sus in the scholarly literature regarding the concept’s cause being overly focused on individual characteristics
central features. Contrasting accounts have been pro- can obscure features of the research protocol or environ-
posed regarding the justification of vulnerability and ment that may harm participants. Luna [6] argues that
which ethical principles translate into obligations for the vulnerability must be conceived of as ‘relational’, in that
special protection of vulnerable research participants. vulnerabilities can only be discovered by examining an
Some accounts propose a justice-based reason for pro- individual in context, and ‘dynamic’, since one’s vulner-
tection, concerned with the fairness of participant re- ability depends on one’s context. Luna and Vanderpoel
cruitment and of the distribution of research burdens [14] describe layers of vulnerability which arise from inter-
and benefits [12, 13]. Others ground vulnerability in a actions between an individual’s characteristics and their
principle of autonomy or respect for persons, suggesting environment, and which interact with one another to cre-
that persons who cannot provide informed and volun- ate an inextricably context-dependent vulnerability.
tary consent are susceptible to harm because they are To our knowledge, an in-depth analysis of the concept
not able to protect their interests [5, 13]. These ap- of vulnerability as it exists in the policies and guidelines
proaches are not mutually exclusive, but we must at that govern research on human subjects has not been
least be able to identify which ethical principles define to conducted. This is an important gap because, at present,
whom we owe special consideration. In research, there the scholarly literature seeks to advance the concept
are often gaps between the rules intended to govern and without an understanding of the full scope of the regula-
the practices at hand, requiring those tasked with the tory context. Without a clear understanding of the con-
implementation of these rules to interpret and apply ceptualisation and operationalisation of vulnerability in
them in their specific context [3]. An ethical foundation current research ethics, recommendations for its refine-
is needed for these interpretations; otherwise, it is difficult ment risk being disconnected from the range of policy
to understand the intentions of the authoring parties or options. To explore the diversity of options with respect
apply the rule to the situation at hand [3]. In this context, to the enshrinement and application of the concept of
a better understanding of the justifications of vulnerability vulnerability in research ethics guidelines, we conducted
becomes a crucial goal of scholarly work in this area. an in-depth analysis of major national and international
The application of vulnerability and its scope in re- research ethics policies and guidelines.
search has also been a subject of much debate. In par-
ticular, vulnerability has been charged with being both Methods
too broad and too narrow. An overly broad concept cap- Sampling
tures all research participants, creating conceptual con- Inspired by previous research ethics policy analyses [3, 15],
fusion over the meaning of ‘special protections’, while an we compiled a sample of internationally- and nationally-
overly narrow concept may leave some vulnerable partic- adopted research ethics guidelines and policies, focusing on
ipants at risk and without the needed protection [5, 11, Canada (the authors’ own regulatory context) and regions
12, 14]. Practically, a definition of vulnerability must be with similar demographic and legal structures to Canada,
comprehensive enough to capture those in need of add- including Australia, the European Union, the United King-
itional protections without overburdening participants dom, and the United States [15]. We began our search
for whom protection beyond the norm is unnecessary. using a compilation of international human research stan-
Further, it must provide researchers and research ethics dards produced by the Office for Human Research Protec-
boards with the information necessary to identify those tions of the United States Department of Health and
who are vulnerable, as well as what they might be vul- Human Services [16]. Additionally, we performed second-
nerable to. There are compelling arguments against nar- ary searches of the references of any included guidelines
row definitions of vulnerable groups that support the and policies for relevant, non-duplicated documents.
Our primary goal was to build a sample of guidelines reviewed by other authors (EB, ER, MEM) and consensus
and policies that discussed or referenced vulnerability in was achieved through team discussions.
general health research. As such, we excluded those in The results of our comparative data analysis are pre-
which vulnerability was not explicitly discussed (e.g. the sented in tables, with direct excerpts from the guidelines
Nuremberg code) as well as those focused on specific and policies provided where possible. Two codes (groups
areas of or issues within research (e.g. paediatric research, and sources of vulnerability, and implications) included
genetic research), put forward by professional organisa- more data than others and thus the text has been con-
tions, or published as working papers, drafts, commentar- densed (i.e. direct citations are not provided). To ensure
ies, or otherwise less broadly adopted documents. Our fidelity to the data, one author (DBR) condensed this
final sample included 11 guidelines and policies, six of text and another (ER) reviewed it to ensure accurate rep-
which are internationally-adopted (i.e. across multiple resentation of the guidelines and policies.
countries) [17–22] and five of which are nationally-
adopted (i.e. within individual countries) [23–27]. All doc- Intra-policy holistic analysis
uments were downloaded and saved for data extraction. After the inter-policy comparative analysis, we examined
See Table 1 for an overview of our sample and the key the conceptualisation and operationalisation of vulnerabil-
characteristics of included policies. ity within each policy. Building on the structure developed
in our comparative analysis, we assessed the logical
consistency between the four content areas of vulnerabil-
Inter-policy component analysis ity. More specifically, we analyzed each policy and guide-
This stage of analysis consisted of an inter-policy analysis, line in isolation to examine (1) which major content areas
allowing us to capture and explore patterns in the data are lacking, (2) whether the four content areas (definitions,
across our sample. Given our specific interest in under- justifications, groups and sources, and implications) are con-
standing how guidelines and policies employ the concept sistent (e.g. in their meaning) with one another, and (3) what
of vulnerability, each document was word-searched for overall impression a guideline or policy user might have
the term ‘vulnerab’. Using this truncated keyword allowed about the concept of vulnerability within the document.
us to identify all uses of the terms ‘vulnerability’ or ‘vul-
nerable’. We read the broader sections of text surrounding Results
the key terms to facilitate a contextual understanding of Inter-policy comparative analysis
how the notion of vulnerability was used. Defining vulnerability
We employed a content analysis strategy, developing an All policies in our sample reference vulnerability and/or
initial coding guide deductively and refining it inductively. vulnerable subjects, but only three out of eleven expli-
We hypothesised, based on the literature (as described in citly define these terms (Table 2). Of these, the Council
the Background section), that research ethics guidance on for International Organizations of Medical Sciences
vulnerability should include at least the following basic con- (CIOMS) and the Tri-Council Policy Statement: Ethical
tent: (1) a definition of vulnerability, (2) a discussion of the Conduct for Research Involving Humans (TCPS2) guide-
sources or circumstances from which vulnerability can arise lines define vulnerability itself, while the International
and/or identification of groups likely to be in those Conference on Harmonisation, Good Clinical Practice
circumstances, (3) an explanation of the ethical justification (ICH GCP) instead provides a definition of vulnerable
of the concept to aid in its application. A preliminary subjects. These definitions share similar structures, all de-
coding guide was created to capture these content areas. fining vulnerability or vulnerable subjects and identifying
This preliminary guide was applied to a subset of the sam- paradigmatic sources (or causes) of vulnerability. The ICH
ple (n = 3). Through this ‘piloting’ stage, the coding scheme GCP definition focuses on issues of consent, where a lack
was refined inductively by three authors (DBR, EB, ER) to of voluntariness in a subject’s decision to participate estab-
ensure other major areas of content were captured. This re- lishes their vulnerability. The CIOMS and TCPS2 guide-
sulted in the addition of a fourth content category, ‘implica- lines employ broader language, both stating that
tions of vulnerability’, which captures responses to vulnerability arises from a subject’s lack of ability to
vulnerable participants laid out within the guidelines and protect their own interests. Both identify sources of vul-
policies. Definitions and rules for the application of each nerability located within the subject (e.g. a lack of
code were developed to ensure rigor and thoroughness. decision-making capacity) and in their environment (e.g. a
Throughout the coding process, three authors (DBR, EB, lack of access to medical care). Only the definition pro-
ER) engaged in open discussions in order to account for vided by the TCPS2 makes explicit reference to another
any biases of the primary coder (DBR) and to ensure the central ethical concept – that of autonomy. This reference
full depth of the data would be represented through this suggests an important link between vulnerability and au-
analytic strategy. Once final coding was complete, it was tonomy, though this connection is not further explained.
Table 1 Key characteristics of guideline and policy sample

Guideline/Policy Date Adopted Abbreviation Status Intended Users Guiding Ethical Framework/
Principles
Declaration of Helsinki 2013 Intl Declaration A statement of ethical Primarily physicians; others Articles of the Declaration
of Helsinki principles proposing how involved in medical research itself are intended as
physicians should act in with human subjects are guiding ethical principles for
research; not legally binding encouraged to adopt its research
principles
Council for International 2002 Intl CIOMS A guidance document CIOMS member bodies, Cites three guiding ethical
Organizations of Medical intended to guide the which include international principles: respect for
Sciences, International effective application of the and national biomedical persons, beneficence, and
Ethical Guidelines for Declaration of Helsinki’s organisations (e.g. World justice
Biomedical Research ethical principles in Medical Association)
Involving Human Subjects research, especially in low-
resource countries; not
legally binding
UNESCO Universal 2005 Intl UNESCO A universal framework of Addressed to States, but Articles of the UNESCO
Declaration on Bioethics Declaration principles to guide States in also provides guidance for Declaration itself are
and Human Rights formulating legislation and individuals, groups, intended as guiding
policies, as well as to guide communities and bioethical principles
the actions of individuals, corporations, public and
groups, communities, private
institutions and
corporations, public and
private
Directive of 4 April 2001 N 2001 EU EU Clinical A legislative act that EU member states Not explicitly provided;
°2001/20/EC Trials establishes specific states that “[t]he accepted
Directive provisions for good clinical basis for the conduct of
practice in clinical trials; EU clinical trials in humans is
Member States must meet founded in the protection of
these provisions though the human rights and the dignity
Directive does not legislate of the human being… as for
how instance reflected in the 1996
version of the Helsinki
Declaration” [20]
Regulation of 16 April 2014 EU EU Clinical A binding legislative act EU member states Not explicitly provided
2014 N°536/2014 Trials applying to all clinical trials
Regulation conducted in the EU
International Conference 1996 US, EU, ICH GCP An ethical and scientific Targeted at those involved Not explicitly provided;
on Harmonisation, Good JP, AUS, quality standard for in the generation of clinical states that “clinical trials
Clinical Practice CA designing, conducting, trial data intended to be should be conducted in
recording, and reporting submitted to regulatory according with the ethical
human subject research authorities, especially in CA, principles that have their
trials; serves as a unified the EU, JPN, and US; can origin in the Declaration of
standard for CA, the EU, also be used by others Helsinki” ([22], Art. 2.1)
JPN, and US to facilitate the involved in clinical
mutual acceptance of investigations “that may
clinical data by the have an impact of the safety
regulatory authorities in and well-being of human
these jurisdictions subjects” [22]
National Statement on 2007 AUS Australian Must be used to inform the Researchers, members of Describes four guiding
Ethical Conduct in Human National design, ethical review, and ethical review bodies, and values and principles:
Research Statement conduct of human research those involved in research research merit and integrity,
funded by or taking place governance, as well as justice, beneficence, and
under the auspices of the potential research respect
bodies that have developed participants
the Statement (i.e. National
Health and Medical
Research Council, Australian
Research Council, Australian
Vice-Chancellors’
Committee)
Tri-Council Policy 2014 CA TCPS2 To be eligible to receive All those involved in the Sets out three core
Statement, 2nd edition and administer research conduct and review of principles: respect for
funds from the federal research funded by the persons, concern for
research agencies federal research agencies, welfare, and justice
Table 1 Key characteristics of guideline and policy sample (Continued)

responsible for this policy e.g. institutions, researchers,
(i.e. Canadian Institutes of ethics review boards, etc.
Health Research, Natural
Sciences and Engineering
Research Council of Canada,
Social Sciences and
Humanities Research
Council), institutions must
agree to comply with it;
while not required to do so,
other organisations and
entities are encouraged to
adopt this Policy to guide
the ethical aspects of the
design, review and conduct
of research involving
humans
Research Governance 2005 UK UK Research Sets out a framework of Intended for all those who Not explicitly provided
Framework for Health and Governance principles, requirements, design research studies,
Social Care, 2nd edition Framework and standards for the participate in research, host
governance of research in research in their
health and social care and organisation, fund research
applies to all research proposals or infrastructure,
relating to the manage research, and
responsibilities of the undertake research
Secretary of State for Health
The Belmont Report 1979 US Belmont A statement of basic ethical Those involved in the Lays out three basic ethical
Report principles and guidelines review and conduct of principles: respect for
intended to assist in research persons, beneficence, and
resolving the ethical justice
problems that surround the
conduct of research created
by the National Commission
for the Protection of Human
Subjects of Biomedical and
Behavioral Research for the
Department of Health,
Education, and Welfare
Title 45 Code of Federal 1991 US Common Serves as a federal policy for Those involved in the Not explicitly provided, but
Regulations, Part 46 Rule human subjects research, review and conduct of the Regulations were
and applies to all research research associated with the created on the basis of the
conducted or supported by federal agencies by which Belmont Report
or affiliated with the federal the Common Rule has been
agencies by which is has adopted
been adopted
Intl international, EU European Union, AUS Australia, CA Canada, JP Japan, UK United Kingdom, US United States
The definition provided by the TCPS2 is distinct from thought of as existing on a spectrum, or as a feature
the others because it explicitly states that vulnerability is that can vary between circumstances.
context-dependent, and is experienced “to different
degrees and at different times, depending on [an individ- Ethical justifications for the concept of vulnerability
ual’s or group's] circumstances” ([24], p. 210) . However, Many guidelines and policies (CIOMS, UNESCO
qualifying language employed in other policies implicitly Declaration, Declaration of Helsinki, Australian National
suggests a similar view that vulnerability exists on a Statement, TCPS2, Belmont Report) provide explicit ethical
spectrum or as a matter of degree (Table 2). The Declar- argumentation relating to vulnerability and/or vulnerable
ation of Helsinki, Australian National Statement, and subjects. There is significant overlap across the sample be-
Belmont Report, for example, discuss participants who tween the principles from which obligations or consider-
are “particularly vulnerable” [17], “more-than-usually ations relating to vulnerability arise (see Table 3 for an
vulnerable” [23], or “especially vulnerable” [26], re- overview). In all cases where guiding ethical principles are
spectively. Unlike the TCPS2, no other guidelines in provided by a policy or guideline, vulnerability-related con-
our sample state explicitly that vulnerability should be cerns are discussed in the application of each principle.
Table 2 Content regarding definitions of vulnerability and detailing the use of qualifying language
Policy/ Explicit definition of vulnerability or vulnerable subjects Use of qualifying languagea
Guideline
Intl
Intl Declaration of – • Some groups and individuals are “particularly vulnerable”
Helsinki ([17], Art. 19)
CIOMS “‘Vulnerability’ refers to a substantial incapacity to protect one’s • Persons with serious, potentially disabling or life-
own interests owing to such impediments as lack of capability to threatening diseases are “highly vulnerable” ([18], p. 65)
give informed consent, lack of alternative means of obtaining • Selection of the “least vulnerable” subjects required for
medical care or other expensive necessities, or being a junior or research ([18], p. 18)
subordinate member of a hierarchical group” ([18], p. 18)
UNESCO – • Certain individuals and groups are of “special
Declaration vulnerability” ([19], Art. 8)
EU Clinical Trials – –
Directive
Clinical Trials – –
Regulation
US, ICH GCP Glossary defines vulnerable subjects as “[i]ndividuals whose –
EU, JP, willingness to volunteer in a clinical trial may be unduly influenced
AUS, CA by the expectation, whether justified or not, of benefits associated
with participation, or of a retaliatory response from senior members
of a hierarchy in case of refusal to participate” ([23], Art. 1.61)
National
AUS National – • Where “potential participants [in dependent or unequal
Statement relationships] are especially vulnerable” special measures
may be required ([23], p. 53)
• Neonates in intensive care have a “unique developmental
vulnerability” ([23], p. 56)
• People with a cognitive impairment, intellectual
disability, or mental illness have “distinctive vulnerabilities
as research participants” and are “more-than-usually
vulnerable to various forms of discomfort or stress” ([23], p. 58)
CA TCPS2 “Vulnerability – A diminished ability to fully safeguard one’s own • Participants, researchers, and research ethics board
interests in the context of a specific research project. This may be members may be rendered “more vulnerable” during
caused by limited decision-making capacity or limited access to publicly declared emergencies ([24], p. 90)
social goods, such as rights, opportunities and power. Individuals • “The least organisationally developed communities are the
or groups may experience vulnerability to different degrees and most vulnerable to exploitation” ([24], p. 130)
at different times, depending on their circumstances. See also ‘Au- • Participants may be “in highly vulnerable circumstances”
tonomy’” ([24], p. 210) because of social or legal stigmatisation ([24], p. 141)
UK Research – –
Governance
Framework
US Belmont – • “Also, inducements that would ordinarily be acceptable
Report may become undue influences if the subject is especially
vulnerable” ([26], Part C.1)
Common – –
Rule
a
Qualifying language captures nuances about degrees or types of vulnerability
Intl international, EU European Union, AUS Australia, CA Canada, JP Japan, UK United Kingdom, US United States, CIOMS Council for International Organizations of
Medical Sciences, TCPS2 Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans, ICH GCP International Conference on Harmonisation, Good
Clinical Practice
The normative status of the concept of vulnerability is concerns or obligations related to vulnerability are them-
inconsistent across policies and guidelines. In certain selves characterised as fundamental principles. Specific-
cases (CIOMS, Australian National Statement, TCPS2, ally, principles 19 and 20 of the Declaration of Helsinki
Belmont Report), obligations towards vulnerable re- focus on vulnerability, with 19 stating that “[s]ome groups
search participants arise from the application of other and individuals are particularly vulnerable and may have
fundamental principles. For example, in the TCPS2, obli- an increased likelihood of being wronged or of incurring
gations towards those in circumstances of vulnerability additional harm” [17]. Similarly, Article 8 of the UNESCO
are entailed by the policy’s core principles of Respect for Declaration promotes “[r]espect for human vulnerability
Persons, Concern for Welfare, and Justice. In others, and personal integrity” [19] as a principle in and of itself.
Table 3 Content on the ethical justification of vulnerability and its normative status in each policy
Policy/Guideline Justification for vulnerability Normative status of vulnerability
Intl
ICH GCP – Consideration for ethics review
Intl
CIOMS The protection of dependent or vulnerable persons and populations is described itself as a Fundamental principle/
principle; additionally, concerns relating to vulnerability are grounded in both the principles application of other principles
of respect for persons and justice
UNESCO Respect for human vulnerability and personal integrity is itself a fundamental principle in Fundamental principle
Declaration this framework
Declaration of Concerns related to vulnerability are themselves principles in this framework Fundamental principle
Helsinki
Clinical Trials – Consideration for ethics review
EU Directive
Clinical Trials – Consideration for ethics review
Regulation
National
Australian National Considerations related to vulnerability are discussed in relation to the principles of principles Application of other principles
AUS Statement of respect for persons, research merit and integrity, justice, and beneficence
TCPS2 The principles of respect for persons, justice (fairness and equity), and concern for welfare all Application of other principles
CA entail special obligations regarding vulnerability
Research – Consideration for ethics review
UK Governance
Framework
Belmont Report The principles of respect for persons, beneficence, and justice all entail special obligations Application of other principles
US relating to vulnerability
Common Rule – Consideration for ethics review
Intl international, EU European Union, AUS Australia, CA Canada, JP Japan, UK United Kingdom, US United States, CIOMS Council for International
Organizations of Medical Sciences, TCPS2 Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans, ICH GCP International Conference
on Harmonisation, Good Clinical Practice
The CIOMS guidelines are a unique case in our sample vulnerable. Vulnerable groups identified in our sample
because they characterise vulnerability as both a principle are captured in Table 4, along with the corresponding
and as a consideration derived from other principles. In explanations of why a subject group is considered vul-
the introduction to the CIOMS guidelines, issues of nerable or what they are vulnerable to, when these de-
human rights are described as relating to two principles, tails are available. Notably, while the EU Clinical Trials
one of which is the “protection of dependent or vulnerable Directive and Clinical Trials Regulation, as well as the
persons and populations” ([18], p. 11), while the principle United Kingdom Research Governance Framework, all
of Respect for Persons is described as entailing “at least identify vulnerable subject groups, none of these policies
two fundamental ethical considerations”, including “pro- provide any supporting explanation. Further, only four
tection of persons with impaired or diminished autonomy, policies (CIOMS, Australian National Statement, TCPS2,
which requires that those who are dependent or vulnerable and the Common Rule) provide any explanations of
be afforded security against harm or abuse” ([18], p. 17). what certain identified groups are vulnerable to.
In the remaining guidelines (ICH GCP, EU Clinical Across the sample, a great number of groups are iden-
Trials Directive, EU Clinical Trials Regulation, United tified as vulnerable. Counting only those broad groups
Kingdom Research Governance Framework, Common identified in our table (i.e. excluding the examples of
Rule), vulnerability is not explicitly discussed in relation subgroups discussed in the footnote to Table 4), 32
to any ethical principles, nor is it described as a guiding groups were identified; when these subgroups are in-
principle itself. In these cases, concerns relating to vul- cluded, the total number of groups identified as vulner-
nerable persons seem to serve the role of consideration able expands to 51. Groups most frequently identified
for ethics review or ethical research with no explicit are children, minors or young people (discussed in seven
ethical status. policies), prisoners (discussed in five policies), as well as
persons with mental health issues, patients in emergency
Identifying vulnerable groups and individuals settings, and certain ethnocultural, racial or ethnic mi-
All guidelines and policies in the sample provide means nority groups (each discussed in four policies). Concerns
through which vulnerability can be identified. The ma- for the vulnerability of children centre around consent,
jority identify subject groups who are likely to be with both the CIOMS and TCPS2 guidelines positing a
Table 4 Vulnerable groups identified in our sample, as well as explanations for this designation, where available
Vulnerable Group (Mentioned in) Explanation
Grouped by social status or situation
Prisoners (CIOMS, ICH GCP, Aus. National Vulnerable because:
Statement, TCPS2, Common Rule) • Historically considered vulnerable and “have, at times,
been treated unfairly and inequitably in research, or have
been excluded from research opportunities”a ([24], p. 8)
• Explanation unclear [18, 22]
Vulnerable to:
• Coercion or undue influence [27]
Certain ethnic, racial minority, or ethnocultural Vulnerable because:
groups (CIOMS, ICH GCP, TCPS2, Belmont Report) • Historically considered vulnerable and “have, at times,
been treated unfairly and inequitably in research, or
have been excluded from research opportunities”a ([24], p. 8)
• May continually be sought as research subjects due to ready
availability and administrative convenience; have a dependent
status and, frequently, compromised capacity for free consent;
are easy to manipulate as a result of their illness or socioeconomic
condition b[26]
Patients in emergency settings, prospective Vulnerable because:
participants for emergency research (CIOMS, • Their incapacity to make decisions creates vulnerable
Clinical Trials Regulation, ICH GCP, TCPS2) circumstances [24]
• No explanation [21]
Subordinate members of hierarchies or relationshipsc Vulnerable because:
(CIOMS, ICH GCP, Aus. National Statement) • Voluntary consent may be compromised by expectations
of benefit or repercussions from superiors [18, 22]
• Pre-existing relationships may compromise the voluntariness
of consent because they typically involve unequal status, where
one party has influence or authority over the other [23]
Vulnerable to:
• Being over-researched [18, 23]
Economically disadvantaged persons Vulnerable because:
(Belmont Report, Common Rule) • Dependent status, impaired capacity to consent, easy to
manipulate as a result of their illness [26]
Vulnerable to:
Homeless persons (CIOMS, ICH GCP) • Explanation unclear [18, 22]
Institutionalized persons (TCPS2, Belmont Report) Vulnerable because:
• Historically considered vulnerable and “have, at times, been treated
unfairly and inequitably in research, or have been excluded from
research opportunities”a ([24], p. 8)
• Their ability to fully safeguard their own interests in research may
be limited, and their situation may compromise the voluntariness
of consent in other ways [24]
• May continually be sought as research subjects due to ready availability
and administrative convenience; have a dependent status and, frequently,
compromised capacity for free consent; are easy to manipulate as a result
of their illness or socioeconomic conditionb [26]
Nomads (CIOMS, ICH GCP) • Explanation unclear [18, 22]
Persons in nursing homes (CIOMS, ICH GCP) • Explanation unclear [18, 22]
Persons lacking political or social power (CIOMS) • Explanation unclear [18]
Refugees or displaced persons (CIOMS, ICH GCP) • Explanation unclear [18, 22]
Women (CIOMS, TCPS2) Vulnerable to:
• In some parts of the world, they may be vulnerable to neglect or harm
in research “because of their social conditioning to submit to authority,
to ask no questions, and to tolerate pain and suffering” ([18], p. 73)
Vulnerable because:
• Historically considered vulnerable and “have, at times, been treated
unfairly and inequitably in research, or have been excluded from
research opportunities”a ([24], p. 8)
Table 4 Vulnerable groups identified in our sample, as well as explanations for this designation, where available (Continued)
Countries or communities with limited resources (CIOMS) Vulnerable to:
• Exploitation by sponsors and investigators who are relatively wealthy [18]
Educationally disadvantaged persons (Common Rule) Vulnerable to:
Members of communities unfamiliar with modern • Explanation unclear [18]
medical concepts (CIOMS)
Neonates in intensive care (Aus. National Statement) Vulnerable because:
• Developmental vulnerability (potential for long-range impacts on
health and development) [23]
Patients in terminal care (Aus. National Statement) Vulnerable to:
• Unrealistic expectations of benefit [23]
Participants and researchers in research that uncovers Vulnerable because:
illegal activities (Aus. National Statement) • Vulnerability may arise because of discovery of participants’
illegal activity [23]
Those with diminished capacity for self-determination (TCPS2) • Historically vulnerable and “have, at times, been treated unfairly
and inequitably in research, or have been excluded from research
opportunities”a ([24], p. 8)
The least organizationally developed communities (TCPS2) Vulnerable to:
• Exploitation [24]
Grouped by patient/participant condition
Children, minors, or young people (CIOMS, Clinical Vulnerable because:
Trials Directive, Clinical Trials Regulation, Aus. National • Limited freedom or capacity to consent [18, 24]
Statement, TCPS2, Common Rule) • Vulnerability arising from developmental stage [24]
• No explanation [20, 21]
• Explanation unclear [23]
Vulnerable to:
Persons with mental illness or mental health problems Vulnerable because:
(Clinical Trials Regulation, Aus. National Statement, • Historically considered vulnerable and “have, at times,
TCPS2, UK Research Governance Framework) been treated unfairly and inequitably in research, or
have been excluded from research opportunities”a ([24], p. 8)
• Unclear [21, 25]
Vulnerable to:
• Various forms of discomfort and stress [23]
Elderly persons (CIOMS, Clinical Trials Regulation, TCPS2) Vulnerable because:
• Likely to acquire “vulnerability-defining” traits (e.g., institutionalization,
dementia) ([18], p. 65]
• Historically considered a group in vulnerable circumstances “have, at
times, been treated unfairly and inequitably in research, or have been
excluded from research opportunities”a ([24], p. 8)
Persons with limited (or no) freedom or capacity Vulnerable because:
to consent (CIOMS, Clinical Trials Regulation, ICH GCP) • Relatively (or absolutely) incapable of protecting their own interests [18]
• Explanation unclear [22]
Vulnerable to:
• Exploitation for financial gain by guardians [18]
Pregnant or breastfeeding women (Clinical Trials Vulnerable to:
Regulation, Common Rule) • Coercion or undue influence [27]
Adults with learning difficulties (UK Research • No explanation [25]
Governance Framework)
Handicapped persons (Common Rule) • No explanation [27]
Mentally disabled persons (Common Rule) Vulnerable to:
Persons who have serious, potentially disabling or Vulnerable because:
life-threatening diseases (CIOMS) • May be treated with drugs or other therapies with unproven
safety and efficacy [18]
Table 4 Vulnerable groups identified in our sample, as well as explanations for this designation, where available (Continued)
Very sick persons (Belmont Report) Vulnerable because:
• May continually be sought as research subjects due to ready
availability and administrative convenience; have a dependent status
and, frequently, compromised capacity for free consent; are easy to
manipulate as a result of their illness or socioeconomic conditionb [26]
People suffering from multiple chronic conditions (Clinical Trials • No explanation [21]
Regulation)
Persons with a cognitive impairment or intellectual disability Vulnerable to:
(Aus. National Statement) • Various forms of discomfort and stress [23]
a
It is not clear whether the TCPS2 intends these groups it refers to as having been historically in vulnerable circumstances as still at risk of this. Given that this is
mentioned but not negated, we included these groups in our table
b
The Belmont Report lists a number of vulnerable groups and a series of explanations of their vulnerability. It is unclear whether certain groups were intended to
be linked to certain explanations, so all have been included
c
Within this category, specific subject groups are provided as examples. For the CIOMS these are “medical and nursing students, subordinate hospital and
laboratory personnel, employees of pharmaceutical companies, and members of the armed forces or police” ([18], p. 65). The ICH GCP adds pharmacy and dental
students and persons kept in detention to this list [18]. The Australian National Statement lists “carers and people with chronic conditions or disabilities, including
long-term hospital patients, involuntary patients, or people in residential care or supported acumination; health care professionals and their patients or clients;
teachers and their students; prison authorities and prisoners; governmental authorities and refugees; employers or supervisors and employees (including
members of the Police and Defence forces); service-providers (government or private) and especially vulnerable communities to whom the service is provided”
([23], p. 53).
The table is grouped by category, and organized by the number of times a group is mentioned in the policies and guidelines
vulnerability arising from their limited freedom or cap- The EU Clinical Trials Directive and Clinical Trials Regula-
acity to consent and the Common Rule emphasising tion and United Kingdom Research Governance Frame-
children’s vulnerability to coercion or undue influence. work provide no explanation or justification for any of the
The Australian National Statement similarly positions groups they designate as vulnerable, and while the Com-
the vulnerability of young people relative to capacity and mon Rule specifies that it is concerned with vulnerability to
consent, though it is unclear how this policy conceives coercion or undue influence, it does not address 'handi-
of the relationship between these concepts. It outlines capped persons' in this explanation despite also identifying
various scenarios regarding the vulnerability of young them as a vulnerable subject group. The CIOMS and ICH
people: in some cases, young people may be able to GCP guidelines, on the other hand, provide definitions of
understand information but their “relative immaturity vulnerable subjects and explanations for some vulnerable
means they remain vulnerable” ([23], p. 50); in other groups. However, both of these policies include categories
cases they may be “mature enough to understand and of 'other vulnerable groups' and fail to provide any connec-
consent [though] not vulnerable through immaturity in tion between these other groups and their overarching def-
ways that warrant additional consent” ([23], p. 50); and inition of vulnerability. As such, it is unclear whether they
in yet other cases, young people may be “mature enough are designated as vulnerable on some other unstated
to understand the relevant information and to give con- grounds.
sent, although vulnerable because of immaturity in other Some policies and guidelines identify sources or
respects” ([23], p. 50). The ‘other respects’ in which im- circumstances of vulnerability independently, i.e. without
maturity can render young people vulnerable are not any relation or association to a specific vulnerable group.
made explicit, leaving the designation of vulnerability For example, neither the Declaration of Helsinki nor the
open to interpretation in this case. Other policies employ UNESCO Declaration identifies any particular subject
similarly open-ended strategies, the CIOMS guidelines groups as vulnerable. Instead, they identify characteris-
most explicitly by listing vulnerable groups and sources of tics of vulnerable participants or key sources of vulner-
vulnerability, and adding that “[t]o the extent that these ability (Table 5). It is important to note that, while the
and other classes of people have attributes resembling TCPS2 does identify certain groups as likely to be in vul-
those of classes identified as vulnerable, the need for spe- nerable circumstances, it qualifies any such labels,
cial protection of their rights and welfare should be emphasising that “[i]ndividuals should not automatically
reviewed and applied, where relevant” ([18], p. 65). be considered vulnerable simply because of assumptions
There is little overlap between the explanations provided made about the vulnerability of the group to which they
by policies and guidelines for other frequently-identified belong” ([24], p. 54).
vulnerable groups, and there was a lack of explanation from
at least two of them for prisoners, patients in emergency Implications of vulnerability in research
settings, and ethnocultural and racial minorities. For over All policies in our sample identify practical implications
half of the groups identified across our sample, an explan- of vulnerability in research, i.e. responses to vulnerability
ation of their vulnerability was unclear or lacking entirely. in the design and review of research and to vulnerable
Table 5 Sources of vulnerability identified independently from puts forth numerous measures intended to promote the
vulnerable groups agency of those in vulnerable circumstances. For example,
Policy/Guideline Sources of vulnerability it is suggested that they should be afforded opportunities to
Declaration of An increased likelihood of being wronged or of influence research and that research ought to enhance vul-
Helsinki incurring harm nerable persons’ capacity for participation. Furthermore,
UNESCO Persons may be rendered vulnerable by disease or the TCPS2 guidance states more broadly that vulnerable
Declaration disability or other personal, societal or environmental groups may need or desire special measures to ensure their
conditions
safety, suggesting a role for participants in the design and
TCPS2 Persons may be in vulnerable circumstances because implementation of their protections.
of social or legal stigmatisation associated with their
activity or identity In addition to conditions and restrictions for research
TCPS2 Tri-Council Policy Statement: Ethical Conduct for Research involvement, the process of informed consent is a major
Involving Humans area of focus in the policies and guidelines. Here in par-
ticular there is an emphasis on the provision of mean-
participants themselves. A wide range of implications ingful support to enable vulnerable persons to offer a
were identified, some directed explicitly towards REBs fully informed consent to research. Mechanisms of sup-
and/or investigators but the majority formulated more port include ensuring adequate time and an appropriate
broadly with no specific group targeted. Further, these environment (CIOMS), as well as ensuring that informa-
implications span the research process, from consider- tion is fully explained and understood (United Kingdom
ations important in the design of research to actions that Research Governance Framework). Additionally, the
must be taken when vulnerable persons are participating Australian National Statement uniquely suggested that
in research (Table 6). participants be given the option of using a participant
A majority of policies and guidelines identify implica- advocate within the consent process.
tions relating to restrictions for research with vulnerable
groups or individuals, but these entail both negative and Holistic policy analysis
positive duties. Overall, these policies and guidelines Of the 11 policies and guidelines in our sample, only
propose that the involvement of vulnerable groups in re- two, the CIOMS guidelines and TCPS2, meet our cri-
search ought to be restricted to some extent; vulnerable teria for a full conceptualisation of vulnerability, ad-
persons ought to be involved only when the research dressing all content areas (Table 7). In this section, we
cannot be carried out with persons who are less vulner- present the results of our intra-policy analysis of vulner-
able and special justification is required for their involve- ability with a narrative about each policy statement, ad-
ment. However, when these persons are involved in dressing (1) which major content areas are lacking, (2)
research, additional actions are required, such as the de- whether the content areas are consistent (i.e. in their
sign of research that is responsive to their needs or pri- meaning) with one another, and (3) what overall impres-
orities and the provision of benefits relevant to their sion a guideline or policy user might have about the
group/subject population. Across our sample, a common concept of vulnerability within the document.
underlying assumption seems to be that vulnerable
groups can and should be involved in research, but that
International policies and guidelines
additional measures are required to ensure this involve-
ment occurs in an ethical manner. In fact, several pol-
icies (CIOMS, EU Clinical Trials Directive, Australian Declaration of Helsinki The Declaration conveys a
National Statement, and TCPS2) assert that vulnerable harm/wrong-based conceptualisation of vulnerability
groups have a right to participate in research and access that is internally coherent due to its broad language. It
its benefits, and while the others do not identify such an does not identify what these wrongs or harms might
entitlement, none go so far as to state that the outright consist of and, because concern for vulnerability is pre-
exclusion of vulnerable groups from research best serves sented as a fundamental principle, interpretation cannot
to protect them. be guided by other ethical principles. Implications of
The implications of vulnerability all tend towards careful vulnerability focus on the need for responsive research,
inclusion rather than outright exclusion of vulnerable special justification for involving vulnerable persons, and
groups from research. However, there is more variability re- to-group benefits, suggesting these harms include the
garding the extent to which these protections afford agency unfair distribution of the risks and benefits of research.
to vulnerable subjects. The majority specify considerations
and actions for researchers and REBs, with few explicitly CIOMS These guidelines present an autonomy-based
identifying the desires of these individuals as relevant in the conceptualisation of vulnerability that is comprehensive
application of these measures. The TCPS2 in particular in scope but lacks internal clarity in its discussion of
Table 6 Implications of vulnerability, grouped by theme Table 6 Implications of vulnerability, grouped by theme
Restrictions for research with vulnerable groups Policy/Guideline (Continued)
or individuals
Special attention or regard should be paid UNESCO Declaration;
When research is carried out with vulnerable Declaration of Helsinki; to vulnerable communities, groups, or persons TCPS2
participants it should be responsive to the CIOMS
needs, conditions, or priorities of the Researchers and REBs should recognise and TCPS2
vulnerable group involved address changes in participants’ circumstances
that may impact their vulnerability
Vulnerable subjects should be involved in CIOMS
research only when it cannot be carried out
with less vulnerable subjects Research ethics board composition
Special justification is required for involving CIOMS; REBs reviewing research with vulnerable Common Rule;
vulnerable groups in research and Belmont Report subjects should include members with EU Clinical Trials
appropriateness ought to be demonstrated expertise on these populations Regulation
Children should not be included in early-phase CIOMS Community members on REBs ought to TCPS2
research until therapeutic effects have been reflect participant’s perspectives, particularly
shown in adults important when participants are vulnerable
and/or risks are high
Opportunities to participate in and influence TCPS2
research affecting their welfare should not be
withheld from vulnerable groups Assessing harms, risks and benefits
Members of vulnerable groups are entitled to CIOMS For those gauging the severity of harm in Australian National
access the benefits of research research, the vulnerability of a population will Statement
be relevant
Children must be involved in studies of EU Clinical Trials
medicinal products likely to be of value to Directive The existence of vulnerable circumstances TCPS2
them may require greater effort to minimise risks/
maximise benefits to participants
People with a cognitive impairment, Australian National
intellectual disability, or mental illness are Statement Care must be taken to ensure the risks and Australian National
entitled to participate in research, which need burdens of proposed research with persons Statement
not be limited to their particular impairment, with a cognitive impairment, intellectual
disability, or illness disability, or mental illness are justified by
potential benefits
Research with communities vulnerable to TCPS2
exploitation should strive to enhance capacity
for participation Recruitment practices
Patients receiving high-risk clinical care should TCPS2 The vulnerability of persons in unequal, National Statement
not be inappropriately included in or excluded dependent relationships must be taken into
from research account when considering recruiting these
persons
Risk to vulnerable subjects is justified when it CIOMS
arises from interventions that will provide a
direct health benefit, or when it will benefit Process of informed consent
the subject’s population group Consent may need to be re-confirmed in re- National Statement
search where participants are vulnerable
Special protections and obligations The method of consent in qualitative National Statement;
Individuals and groups of special UNESCO Declaration research depends, in part, on the vulnerability TCPS2
vulnerability should be protected of the research participant; the method must
be tailored for their protection
Special ethical obligations exist towards TCPS2
vulnerable subjects When requirements of free, informed, TCPS2
ongoing consent cannot be met, vulnerable
Vulnerable subjects should receive special/ Declaration of Helsinki participants ought to be involved in decision-
specific protections making, i.e. obtaining assent, asking about
their feelings regarding participation
Groups or individuals in vulnerable TCPS2
circumstances may need or desire special Clinician-researchers must take care not to TCPS2
measures to ensure their safety in a specific overplay the benefits of research participation
research project to vulnerable patients, who may be misled to
enter research with false hope
Vulnerable subjects should be afforded CIOMS
security against harm or abuse Inducements that may not be excessive or Belmont Report
inappropriate for other participants may be
Special (or additional) protections for the CIOMS; Common Rule
undue influences if the subject is especially
rights and welfare of vulnerable subjects
vulnerable
should be applied
Care should be taken in the informed CIOMS
consent process to ensure that women
Attention and consideration
vulnerable to coercion have adequate time
Special attention should be paid to trials ICH GCP and a proper environment in which to take
involving vulnerable subjects decisions
Table 6 Implications of vulnerability, grouped by theme vulnerable groups. While the provided definition fo-
(Continued) cuses on vulnerability as stemming from an incapacity
Care should be taken in the informed UK Research to protect one’s own interests owing to both individual
consent process for adults with mental health Governance Framework and environmental features, vulnerability is also expli-
problems or learning difficulties to ensure that citly linked to justice-based concerns about the distri-
information is provided in the appropriate
format and that the roles and responsibilities bution of the risks and benefits of research. The
of those involved are clearly explained and identified implications of vulnerability thus correspond
understood to concerns relating to participants’ ability to provide
Additional consent from a parent or National Statement free and informed consent and relating to the appropri-
guardian may be required for young people
who are vulnerable through immaturity in
ateness of involving vulnerable participants in research.
ways that warrant this There is a lack of clarity and consistency, however, in
Researchers should invite participants in National Statement the discussion of vulnerable groups. The CIOMS
dependent or unequal relationships to discuss guidelines distinguishes between three types of vulner-
their participation with someone who can able groups – those who are “conventionally considered
support them in making their decision;
especially vulnerable participants in these vulnerable”, those who are vulnerable due to social
circumstances should be offered participant pressures (i.e. persons in dependent relationship with
advocates researchers, such as students or pharmaceutical em-
ployees), and “other groups or classes” ([18], p. 64) for
Debriefing whom no explanation is provided and who do not, on
REBs must assess risks and benefits of TCPS2 their face, bear a significant resemblance to these other
debriefing participants and whether debriefing
plan is appropriate for participants, especially
groups.
when they are vulnerable
REB research ethics board, CIOMS Council for International Organizations of UNESCO Declaration The Declaration is concerned
Medical Sciences, TCPS2 Tri-Council Policy Statement: Ethical Conduct for with both a general “human vulnerability” and a more
Research Involving Humans, ICH GCP International Conference on Harmonisation,
Good Clinical Practice
particular “special vulnerability” ([19], Art. 8), neither of
which are defined. Its identification of personal, societal
and environmental conditions as sources of vulnerability
suggests a concept with wide-ranging concerns. The
Table 7 Major content areas of vulnerability addressed within each policy/guideline

Policy/Guideline Definition: What is Groups/Sources: Who is Justifications: What ethical Implications: How should we respond
vulnerability? vulnerable and why? concern(s) does vulnerability reflect? to vulnerability in research?
International
Declaration of – X X X
Helsinki
CIOMS X X X X
UNESCO – X X X
Declaration
EU Clinical Trials – X – X
Directive
EU Clinical Trials – X – X
Regulation
ICH GCP X X – X
National
Australian National – X X X
Statement
TCPS2 X X X X
UK Research – X – X
Governance
Framework
Belmont Report – X X X
Common Rule – X – X
CIOMS Council for International Organizations of Medical Sciences, TCPS2 Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans, ICH GCP
International Conference on Harmonisation, Good Clinical Practice
Declaration identifies respecting the personal integrity of vulnerable. The guidelines do not provide an ethical
vulnerable groups as a key implication, suggesting that framework to facilitate interpretation of the concept of
vulnerability may consist, at least in part, of risks to one’s vulnerability.
personal integrity. Since concerns relating to vulnerability
are presented as fundamental principles, their interpret-
National policies and guidelines
ation cannot be guided by other ethical principles.
Australian National Statement The National State-
EU Clinical Trials Directive The Clinical Trials Direct- ment suggests a comprehensive conceptualisation of vul-
ive conveys a primarily consent-based vulnerability, with nerability relating to concerns about consent, fair
children as the focus of its vulnerability-related regula- involvement in research, and a balance of risks and ben-
tions. The implications it identifies focus on obtaining efits to participants. It favours a group-specific approach
proxy consent and assent, but also on the need to avoid fi- to vulnerability, where this concept is discussed largely
nancial inducements for participation, suggesting a con- in reference to specific groups. General statements about
cern for a risk of exploitation. Other implications include vulnerability suggest that it is an important factor when
a need to perform research with children in which group considering the appropriate method of consent. While
benefits will be obtained and ensuring the interests of the vulnerability is not defined, explanations for the vulner-
patient prevail over those of society. As such, in addition ability of all identified groups are provided and are dis-
to concerns relating to consent, the Directive implicitly re- cussed with reference to the Statement’s guiding ethical
lates vulnerability to concerns with the distribution of the principles. Interestingly, explanations of the vulnerability
benefits and burdens of research. The Directive does not of identified groups the principles from which obliga-
provide an ethics framework, so interpretation of this tions to those groups stem do not always line up. In
guidance cannot be guided by ethical principles. some cases, the relationship is clear; the vulnerability of
young persons originates in their lack of ability to pro-
EU Clinical Trials Regulation The Clinical Trials Regu- vide consent and is linked to respect for persons, and
lation conveys a mixed concept of vulnerability, concerned the vulnerability of neonates in intensive care originates
both with issues of consent and increased health risks. in the risks of long-term harms and is linked to benefi-
While vulnerability is not defined and no explanation for cence. However, while persons in pre-existing/dependent
the vulnerability of listed groups is provided, they can be relationships with researchers are said to face issues pro-
grouped by those assumed to face issues of consent in re- viding voluntary consent, the key implication relating to
search (people affected by mental health disorders, mi- this group is grounded in the principle of justice (i.e. en-
nors, and incapacitated subjects) and those who may be at suring they are not over-researched). Similarly, while
greater physical (i.e. health) risks in research (frail or older persons with terminal illness are said to be vulnerable to
people, people suffering from chronic conditions, and unrealistic expectations of benefit (i.e. may have a com-
pregnant or breastfeeding women). The implications iden- promised ability to consent), the key response to this is
tified do not fall along this consent/health risk distinction, to balance the benefits and burdens of research and is
however, with a need for research to improve treatments a grounded in beneficence.
key implication for frail or older people, people with
chronic conditions, and people affected by mental health TCPS2 The TCPS2 presents an autonomy-based con-
disorders, and the need for special expertise in research ceptualisation of vulnerability that is comprehensive in
ethics review identified as a specific consideration for mi- scope. The provided definition of vulnerability states
nors, incapacitated subjects, and pregnant or breastfeed- that it stems from a diminished ability to protect one’s
ing women. No ethical framework is provided in the own interests caused by both individual (e.g. lack of
Regulation to facilitate interpretation of this guidance. decision-making capacity) and environmental (e.g. lack
of access to social goods) factors. Importantly, vulner-
ICH GCP These guidelines present a consent-based ability is said to be context-specific and dynamic, dis-
concept of vulnerability that lacks internal clarity due to couraging assumptions of vulnerability based on group
its broad scope of vulnerable groups. Vulnerable subjects membership. However, the policy still relies on the
are defined as those whose ability to provide voluntary identification of groups likely to be vulnerable, as well as
consent may be compromised by social pressures, and the identification of circumstances that can create
the first category of groups listed is clearly linked to this vulnerability for a participant. While the definition of
definition. However, it is not clear how the wide range of vulnerability itself is implicitly linked to the principle of
'other vulnerable groups' relates to this definition or autonomy, obligations towards participants in vulnerable
which characteristics are thought to render them circumstances are more comprehensive and are
grounded in the principles of respect for persons, con- [6, 28, 29]. For example, the policies and guidelines iden-
cern for welfare and justice. tify sources of vulnerability that are both individual and
situational [29]; vulnerability can stem from a lack of cap-
United Kingdom Research Governance Framework acity or from one’s health status, but also from social pres-
The framework conveys a consent-based conceptualisa- sures that may impact one’s ability to make a free and
tion of vulnerability that is narrow in scope, labelling informed decision, consistent with some scholarly per-
adults who may have issues with understanding and spectives [30]. Responding to vulnerability requires cau-
decision-making as vulnerable. Consistent with this, the tion and special consideration on the part of researchers
implications of vulnerability focus on providing partici- and REBs but, ultimately, the implications identified in
pants with the necessary support in the informed con- our study suggest that participant vulnerability need not
sent process. Since no ethical framework or principles signal a need for exclusion from research. The few explicit
are discussed relative to vulnerability, these cannot be definitions in our sample define vulnerability as a defi-
used to facilitate interpretation of the guidance. ciency of the participant, as an inability to protect one’s in-
terests in research. The majority of other guidelines and
Belmont Report The Report conveys a consent-based policies implicitly convey a similar conceptualisation of
conceptualisation of vulnerability that lacks clarity in the vulnerability as a deficiency in a participant’s ability to
features of vulnerability it aims to target. It is assumed provide voluntary informed consent. Accordingly, even
that vulnerable subjects have a “dependent status and fre- though there is some diversity and richness in policies, it
quently compromised capacity for free consent” ([26], Part tends to be scattered across multiple policies and relies on
C. 3), which seems to form the basis of their vulnerability. implicit assumptions about the definition and nature of
Special considerations about vulnerable subjects are dis- vulnerability. Indeed, a significant analytic effort was re-
cussed in reference to respect for persons (ordinary in- quired to bring structure to the data and yield the guid-
ducements may be come undue influences for vulnerable ance captured in this paper. We further discuss how our
populations), beneficence (special justification is required findings relate to (1) previous critiques found in the schol-
for research with vulnerable subjects) and justice (vulner- arly literature and (2) the role of stakeholder engagement
able subjects must be protected from over-recruitment to in the process of refining the concept of vulnerability in
research). research ethics policies and guidance.
Common Rule The Common Rule conveys a consent- Previous critiques from the scholarly literature
based conceptualisation of vulnerability that lacks internal Within the scholarly literature several critiques of vul-
clarity regarding its scope. A number of groups are identi- nerability in research ethics guidelines have been voiced.
fied as vulnerable, including handicapped persons, but First, concerns have been raised that the manner in
while the other groups are said to be vulnerable to coer- which vulnerability is defined and operationalised in re-
cion or undue influence, no explanation is provided for search ethics governance stereotypes and reinforces
handicapped persons. Similarly, the implications of vul- stigma about whole categories of individuals [9, 12, 31].
nerability include concern for equitable subject selection Our results reinforce these concerns, as the reliance on
and the provision of additional safeguards, but handi- listing groups of vulnerable persons is rampant. This la-
capped persons are never associated with these protec- belling [6] or sub-population [30] approach does little to
tions. Without a definition of vulnerability, it is not clear bring attention to the importance of context and of
what special vulnerability handicapped persons may be assessing the characteristics of individual research par-
faced with in research. ticipants beyond their membership in a group [5, 6, 9].
It is important to note that research protocols create
Discussion groups through sampling “regardless of whether the sam-
The objective of this analysis was to describe the concept ple is drawn from a naturally occurring community” ([9],
of vulnerability in research ethics policies and guidelines, p. 2221). Understanding this point underscores the fact
and to assess how it is conceptualised and operationalised. that group membership in this context may not well
All policies and guidelines employed the concept of vul- capture the various relevant aspects (and potential vul-
nerability but very few define it. Instead, vulnerability is nerabilities) in an individual participant’s situation. This
most frequently discussed in terms of vulnerable groups, may result in inappropriate and ineffective protections
with some attention given to the sources of vulnerability, being applied in some protocols. Group listings may also
and the implications of conducting research with vulner- cause confusion due to the broadness of some labels
able participants. In many respects the policies come out, (e.g. persons with mental illness or mental health prob-
on the whole, as richer and more complex than some lems). Furthermore, it seems that the designation of
scholarly analyses of the concept of vulnerability suggest some groups as vulnerable may be based on
assumptions not supported by evidence (e.g. the designa- to other pre-existing ethical concepts, but if these concerns
tion of pregnant women as vulnerable to coercion or undue would be otherwise missed, the concept would then be
influence in the Common Rule). proven to have a vital practical function in research ethics.
Another major concern has been that vulnerability, as A few authors have made explicit claims to that effect. For
conceived of in the guidelines, focuses overwhelmingly example, Kipnis argues that vulnerability stems from im-
on a lack of ability to consent [10], blinding researchers pairments to one’s ability to provide voluntary informed
and REBs to other relevant types of vulnerability, relat- consent. He identifies six types of vulnerability which all
ing, for example, to an increased risk of exploitation [32] signal potential issues with a participant’s ability to consent:
or a lack of basic rights [33]. While vulnerability is rarely cognitive, deferential and medical vulnerability, all of which
defined, the majority of policies and guidelines convey relate to characteristics of the participant themselves, and
implicitly that vulnerability is fundamentally an inability juridic, allocational and infrastructural vulnerability, all of
to provide free and informed consent. However, the im- which relate to factors in the participant’s environment
plications of vulnerability often move beyond consent, [30]. These categories help bring attention to more specific
addressing issues of fair subject selection and favourable aspects that generate vulnerability. Luna argues that vulner-
risk benefit assessments. In addition to providing explicit ability, when conceived of as dynamic, flexible and inessen-
definitions for what, exactly, is meant when the term tial, can serve as “a fine grain tool to analyze, interpret, and
‘vulnerability’ is used, the clarity and usability of policies evaluate the research situation” ([6], p. 130). She proposes
and guidelines could be improved by ensuring that these that vulnerability be conceived of through the metaphor of
definitions clearly relate to the concerns with which vul- layers, in which different layers of vulnerability can operate
nerability is associated. and interact within a given participant’s circumstances.
Though they recognise both individual and contextual Luna’s account of vulnerability thus provides researchers
sources of vulnerability, all policies and guidelines con- and REBs with a conceptual tool with which to examine a
veyed that vulnerability is a personal characteristic. Even research participant’s circumstances, identify potential vul-
the TCPS2, with its notable emphasis on vulnerability as nerabilities (e.g. relating to capacity or social pressure in the
a context-dependent feature, ultimately defines it as a consent process), and develop targeted strategies for their
person’s inability to protect their own interests in re- remediation.
search. In contrast, a growing body of scholarly literature In spite of these more sophisticated proposals, there is
converges around the notion that vulnerability is a rela- a dearth of empirical evidence on the functioning of re-
tional feature, borne of power asymmetries between par- search ethics committees and outcomes of research eth-
ticipants and research staff, investigators and institutions ics policies and, to our knowledge, few studies have
[10, 31, 34]. Adopting such a view in research ethics guide- examined the impact or understandings of the concept
lines may better serve participants, encouraging measures of vulnerability based on research ethics guidelines and/
that would empower and promote their agency in the re- or more elaborate scholarly accounts. Empirical evidence
search context [34]. Furthermore, the focus on research has shown that an understanding of vulnerability in the
participants neglects how research environments (e.g. the context of research cannot be assumed to be universal –
existence of conflicts of interest) can actively contribute to in a study with Russian and Romanian research ethics
disempowering research participants/research subjects and trainees, Loue and Loff [36] found that, at the initiation
thus create the need for remediation that does not neces- of their training, their existing understanding of vulner-
sarily concern the research participant per se [6, 10]. ability varied considerably from conceptualisations in
the international guidelines. A study by Sengupta et al.
A need for evidence and stakeholder engagement to [37] gathered researchers’ perspectives on vulnerability
refine research ethics policies and guidance on in HIV/AIDS clinical trials and on the Common Rule
vulnerability guidance related to vulnerability and found that re-
Research ethics guidelines and policies typically stress the searchers assessed vulnerability in relation to situational
importance of vulnerability. However, it has been argued factors that can render participants vulnerable, and that
that vulnerability is not a substantive ethical concept in it- they emphasised the need to assess vulnerability on a
self, serving only as a marker of other research ethics con- case-by-case basis (i.e. rather than relying on a group-
cerns already captured by existing concepts such as harm based strategy). Taken together, these studies underscore
or consent [35]. This is certainly an important conceptual the need for policymakers to clearly delineate and define
concern, but what may be of greater relevance in the realm the concerns vulnerability is intended to encompass, and
of policy development is the degree to which the concept to assess the alignment of these views with those of re-
of vulnerability is a useful, effective tool for those designing, search stakeholders. Further, there is a need to assess the
reviewing and conducting research [5, 6]. It may be the case outcomes of vulnerability-related guidance and policy
that vulnerability merely serves to signal concerns relating and to understand whether protections are actually
effective and their impact on vulnerable participants DBR developed the coding strategy and conducted the data analysis and EB
themselves. For example, there are crucial questions and ER reviewed. DBR drafted the manuscript and ER served as the primary
reviewer and editor, with EB and MEM providing feedback. DBR drafted the
about the actual usability and impact of such guidelines final version of the manuscript and all authors agreed on the final version.
as well as the potential need for mid-level guidance be-
tween general guidelines and the actual analyses of REBs Competing interests
The authors have no competing interests to declare.
[38]. It has been suggested that more elaborate, on-the-
ground guidance on vulnerability would be beneficial to Consent for publication
help REBs direct their attention to the most pertinent Not applicable.
concerns [30]. In the process of developing such guid-
Ethics approval and consent to participate
ance, the voices of those concerned by the application of Not applicable.
what sometimes appears as a label of vulnerability could
be instrumental in moving forward and avoiding the per- Author details
1
Neuroethics Research Unit, Institut de recherches cliniques de Montréal, 110
petuation of stereotyping or stigmatising accounts of Avenue des Pins Ouest, Montréal, QC H2W 1R7, Canada. 2Biomedical Ethics
vulnerability [34]. In this endeavour, the perspectives of Unit and Division of Experimental Medicine, McGill University, Montréal, QC,
researchers and REBs, but also of research participants, Canada. 3Faculty of Dentistry, Oral Health and Society Research Unit, McGill
University, 2001 McGill College, Suite 500, Montréal, QC H3A 1G1, Canada.
who seem to have been largely left out of the develop- 4
Department of Medicine and Department of Social and Preventive
ment of research ethics guidelines, could be investigated. Medicine, Université de Montréal, Montréal, QC, Canada. 5Department of
Neurology and Neurosurgery, McGill University, Montréal, QC, Canada.
Conclusion Received: 26 October 2016 Accepted: 14 December 2016
Our in-depth analysis of human research ethics guidelines
and policies allowed us to analyse different perspectives on
the concept of vulnerability, including the definitions, justi- References
1. Coleman CH. Vulnerability as a regulatory category in human subject
fications, sources, and implications of vulnerability for re- research. J Law Med Ethics. 2009;37:12–8.
searchers and REBs. In some respects, this synthetic 2. ten Have H. Respect for human vulnerability: the emergence of a new
account yielded a richer perspective than sometimes admit- principle in bioethics. J Bioeth Inq. 2015;12:395–408.
3. Eriksson S, Höglund AT, Helgesson G. Do ethical guidelines give guidance?
ted in scholarly literature. At the same time, there are con- A critical examination of eight ethics regulations. Camb Q Healthc Ethics.
ceptual gaps within individual guidelines and policies that 2008;17:15–29.
require the attention of those charged with their develop- 4. Solomon SR. Protecting and respecting the vulnerable: existing regulations
or further protections? Theor Med Bioeth. 2013;34:17–28.
ment. This lack of clarity could diminish the usability of the 5. Hurst SA. Vulnerability in research and health care; describing the elephant
guidance put forth in policies and therefore undermine its in the room? Bioethics. 2008;22:191–202.
impact on research practices. Policymakers should revisit 6. Luna F. Elucidating the concept of vulnerability: layers not labels. Int J Fem
Approaches Bioeth. 2009;2:121–39.
the concept of vulnerability to ensure each of its key com- 7. Lange MM, Rogers W, Dodds S. Vulnerability in research ethics: a way
ponents is spelled out, and that these components are in- forward. Bioethics. 2013;27:333–40.
ternally consistent (i.e. within individual guidelines). 8. Rogers W. Vulnerability and bioethics. In: Mackenzie C, Rogers W, Dodds S,
editors. Vulnerability: new essays in ethics and feminist philosophy. New
Practically-oriented refinement of vulnerability could be fa- York: Oxford University Press; 2014. p. 60–87.
cilitated by engaging research stakeholders and examining 9. DuBois JM, Beskow L, Campbell J, Dugosh K, Festinger D, Hartz S, James R,
the concrete impact of guidance and policy related to Lidz C. Restoring balance: a consensus statement on the protection of
vulnerable research participants. Am J Public Health. 2012;102:2220–5.
vulnerability. 10. Bell E, Racine E, Chiasson P, Dufourcq-Brana M, Dunn LB, Fins JJ, et al.
Beyond consent in research: revisiting vulnerability in deep brain
Acknowledgements stimulation for psychiatric disorders. Camb Q Healthc Ethics. 2014;23:361–8.
The authors would like to thank Michael McDonald for his feedback on an 11. Schroeder D, Gefenas E. Vulnerability: too vague and too broad? Camb Q
earlier version of this manuscript, as well as the members of the Neuroethics Healthc Ethics. 2009;18:113–21.
Research Unit for their feedback throughout the conduct of this research. 12. Levine C, Faden R, Grady C, Hammerschmidt D, Eckenwiler L, Sugarman J,
et al. The limitations of “vulnerability” as a protection for human research
Funding participants. Am J Bioethics. 2004;4:44–9.
Funding for this study was provided by a CIHR grant held by EB, ER and 13. Nickel PJ. Vulnerable populations in research: the case of the seriously ill.
MEM, an FRQ-S career award held by ER, and graduate scholarships from the Theor Med Bioeth. 2006;27:245–64.
IRCM and McGill University held by DBR. No funding body played any role in 14. Luna F, Vanderpoel S. Not the usual suspects: addressing layers of
the design of the study or collection, analysis, interpretation of the data, or vulnerability. Bioethics. 2013;27:325–32.
in the writing of the manuscript. 15. Zimmerman E, Racine E. Ethical issues in the translation of social
neuroscience: a policy analysis of current guidelines for public dialogue in
Availability of data and materials human research. Account Res. 2012;19:27–46.
The datasets used and/or analysed during the current study are available 16. International Compilation of Human Research Standards. Office for Human
from the corresponding author on reasonable request. Research Protections, Rockville. 2015. http://www.hhs.gov/ohrp/
international/compilation-human-research-standards/. Accessed 1 May 2015.
Authors’ contributions 17. World Medical Association. World Medical Association Declaration of
EB and ER developed the idea of conducting a review of research ethics Helsinki: ethical principles for medical research involving human subjects.
policies and guidelines. DBR developed the search strategy and inclusion Fortaleza, Brazil: 2013. http://www.wma.net/en/30publications/10policies/
and exclusion criteria and gathered the sample with input from EB and ER. b3/. Accessed 7 January 2015.
18. Council for International Organizations of Medical Sciences (CIOMS).

International Ethical Guidelines for Biomedical Research Involving Human
Subjects. Geneva: CIOMS. 2002. http://www.cioms.ch/publications/
guidelines/guidelines_nov_2002_blurb.htm. Accessed 7 January 2015.
19. United Nations Educational, Scientific and Cultural Organization. Universal
Declaration on Bioethics and Human Rights. UNESCO: 2005. http://portal.
unesco.org/en/ev.php-URL_ID=31058&URL_DO=DO_TOPIC&URL_SECTION=
201.html. Accessed 7 January 2015.
20. Directive 2001/20/EC of the European Parliament and of the Council of 4 April
2001 on the approximation of the laws, regulations and administrative
provisions of the Member States relating to the implementation of good
clinical practice in the conduct of clinical trials on medicinal products for
human use. Official J L 121:34-44. https://ec.europa.eu/health/sites/health/files/
files/eudralex/vol-1/dir_2001_20/dir_2001_20_en.pdf. Accessed 21 July 2015.
21. Regulation (EU) No 536/2014 of the European Parliament and of the
Council of 16 April 2014 on clinical trials on medicinal products for human
use, and repealing Directive 2001/20/EC. Official J L 158:1-76. https://ec.
europa.eu/health/sites/health/files/files/eudralex/vol-1/reg_2014_536/reg_
2014_536_en.pdf. Accessed 21 July 2015.
22. ICH Steering Committee. ICH Harmonized Tripartite Guideline. In Guideline
for Good Clinical Practice E6. 1996.
23. The National Health and Medical Research Council, the Australian Research
Council and the Australian Vice-Chancellors' Committee. National Statement
on Ethical Conduct in Human Research (Updated March 2014). Canberra,
Australia: Australian Government; 2007.
24. Canadian Institutes of Health Research, Natural Sciences and Engineering
Research Council of Canada, and Social Sciences and Humanities Research
Council of Canada. Tri-Council Policy Statement: ethical conduct for
research involving humans, December 2014. http://www.pre.ethics.gc.ca/
pdf/eng/tcps2-2014/TCPS_2_FINAL_Web.pdf. Accessed 6 January 2015.
25. Department of Health. Research Governance Framework for Health and
Social Care, second edition. COI; 2005. https://www.gov.uk/government/
uploads/system/uploads/attachment_data/file/139565/dh_4122427.pdf.
Accessed 6 January 2015.
26. The National Commission for the Protection of Human Subjects of
Biomedical and Behavioral Research. The Belmont Report: ethical principles
and guidelines for the protection of human subjects of research.
Washington D.C.: Department of Health, Education, and Welfare; 1979.
27. Federal Policy for the Protection of Human Subjects, 45 C.F.R. § 46 (2001).
https://www.hhs.gov/ohrp/regulations-and-policy/regulations/45-cfr-46/
index.html. Accessed 7 January 2015.
28. Kipnis K. Seven vulnerabilities in the pediatric research subject. Theor Med
Bioeth. 2003;24:107–20.
29. Rogers W, Ballantyne A. Special populations: vulnerability and protection.
RECIIS. 2008;2:S30–40. doi:10.3395/reciis.v2.Sup1.207en.
30. Kipnis K. Vulnerability in research subjects: a bioethical taxonomy. In:
National Bioethics Advisory Commission, editor. Ethical and policy issues in
research involving human participants. Bethesda: National Bioethics
Advisory Commission; 2001. p. G1–G13.
31. Schrems BM. Informed consent, vulnerability and the risks of group-specific
attribution. Nurs Ethics. 2014;21:829–43.
32. Macklin R. Bioethics, vulnerability, and protection. Bioethics. 2003;17:472–86.
33. Zion D, Gillam L, Loff B. The Declaration of Helsinki, CIOMS and the ethics of
research on vulnerable populations. Nat Med. 2000;6:615–7.
34. Bracken-Roche D, Bell E, Racine E. The “vulnerability” of psychiatric research
participants: why this research ethics concept needs to be revisited. Can J
Psychiat. 2016;61:335–9.
35. Wrigley A. An eliminativist approach to vulnerability. Bioethics. 2015;29:478–87. Submit your next manuscript to BioMed Central
36. Loue S, Loff B. Is there a universal understanding of vulnerability?
Experiences with Russian and Romanian trainees in research ethics. J Empir and we will help you at every step:
Res Hum Res. 2013;8:17–27.
• We accept pre-submission inquiries
37. Sengupta S, Lo B, Strauss RP, Eron J, Gifford AL. How researchers define
vulnerable populations in HIV/AIDS clinical trials. AIDS and Behav. 2010; • Our selector tool helps you to find the most relevant journal
14:1313–9. • We provide round the clock customer support
38. Deslauriers C, Bell E, Palmour N, Pike B, Doyon J, Racine E. Perspectives of
• Convenient online submission
Canadian researchers on ethics review of neuroimaging research. J Empir
Res Hum Res. 2010;5:49–66. • Thorough peer review
• Inclusion in PubMed and all major indexing services
• Maximum visibility for your research
Submit your manuscript at

www.biomedcentral.com/submit

Dearbhail Bracken-Roche Et Al. The Concept of Vulnerability' in Research Ethics (2017)

Uploaded by

Copyright:

Available Formats

Dearbhail Bracken-Roche Et Al. The Concept of Vulnerability' in Research Ethics (2017)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Dearbhail Bracken-Roche Et Al. The Concept of Vulnerability' in Research Ethics (2017)

Uploaded by

Copyright:

Available Formats

Bracken-Roche et al.

Health Research Policy and Systems (2017) 15:8

RESEARCH Open Access

The concept of ‘vulnerability’ in research

Table 1 Key characteristics of guideline and policy sample

Table 1 Key characteristics of guideline and policy sample (Continued)

Table 7 Major content areas of vulnerability addressed within each policy/guideline

18. Council for International Organizations of Medical Sciences (CIOMS).

Submit your manuscript at

You might also like