Keywords

2.1 Introduction

The main purpose of this chapter is to explain how some groups of vulnerable people are left behind in research unnecessarily. This can happen when regulatory and governance mechanisms, including research ethics codes and research ethics committees, serve to exclude or restrict the participation of vulnerable persons in ways that could be avoided. The chapter is informed by findings from an empirical investigation into the experiences and opinions of researchers who work with vulnerable individuals, as well as a review of research ethics codes and guidelines to find out what they have to say about the inclusion or exclusion of vulnerable persons in research.

2.2 The Regulation and Governance of Research

In Carol Levine’s often repeated statement, research ethics was “born in scandal and reared in protectionism” (Levine 1988: 167), and its evolution has followed a pattern of crisis and response (Marshall 2002). Early medical experiments by physicians and biomedical scientists led to great advances in healthcare such as the discovery of a vaccine for smallpox in 1796. However, as science advanced, cases came to light of the use of vulnerable individuals as human guinea pigs, including enslaved black women (Wall 2006), orphaned children and prisoners (Lederer 1995).

Revelations about the exploitation of people in the name of science motivated the development of the first ethics codes and governance mechanisms in research. The Nuremberg Code was formulated in 1947 as a direct response to the abhorrent medical experiments by Nazi and Japanese doctors during the Second World War (Shuster 1998), and in 1964 the World Medical Association published the Declaration of Helsinki (WMA 2013). This declaration is a statement of 37 principles for medical research involving human participants, human materials and human data, and remains the most influential ethics code globally. Despite its enduring influence, the Declaration of Helsinki has not brought exploitation to an end, and the development of research ethics codes and guidance has continued. For instance, revelations about dishonesty and the withholding of treatment in the notorious Tuskegee Syphilis Study (1932–1972) (Heller 1972) prompted the development of the Belmont Report in the US in 1978 (Beauchamp 2008).

Nowadays, there is a plethora of ethics codes for professional organisations, businesses, government agencies and other institutions around the world with around 4000 examples held in the Ethics Codes Collection (ECC), the world’s largest online collection of codes of ethics and ethical guidelines (CSEP n.d.). Many of these codes are devoted to or include matters of research ethics. Different professional associations, government agencies and universities have adopted specific codes, rules and policies for ethics governance in research (Resnik 2020).

In addition to research ethics codes, research governance mechanisms typically include a system of ethics approval for individual studies, with the development of these systems also prompted by ethics scandals. In 1966, the United States medical profession “was hit with a bombshell” (Stark 2016: 2374) when Henry Beecher published an article exposing 22 unethical studies in biomedical research. Beecher (1966) revealed that many participants never had risks explained to them, nor did they know that they were participants in an experiment. His revelations about unethical research prompted the implementation of federal rules governing the conduct of human experimentation in the United States (Harkness et al. 2001). At that point, ethics-related regulations were introduced for nearly all American-funded research around the world (Stark 2016). In the United Kingdom, where researchers were in receipt of American funding, ad hoc committees were introduced hastily, and in May 1967, the newly established Committee on the Ethical Supervision of Clinical Investigations in Institutions (set up by the Royal College of Physicians) met for the first time. The recommendations from the meeting included that each hospital authority had “a responsibility to ensure that all clinical investigations carried out within its hospital or institution are ethical and conducted with the optimum technical skill and precautions for safety” (Hedgecoe 2009: 336).

Today, research ethics committees (RECs)—also referred to as institutional review boards (IRBs) or ethics review boards (ERBs)—and REC review are commonly regarded as synonymous with research ethics. However, research governance in its entirety consists of a complex network of ethics codes and guidance, funder requirements and legal requirements, which can differ according to discipline, country and level of international collaboration, as well as individual institutional policies and systems of oversight (see Fig. 2.1). Together, these requirements, guidance documents and systems of oversight are designed to help maintain appropriate standards in research ethics and research integrity.

Fig. 2.1
figure 1

Research governance overview

Full size image

Figure 2.1 provides an overview insight into the research governance landscape that a researcher must navigate, which is particularly complex when the research being conducted crosses borders and disciplines and involves people who are considered vulnerable.

In the broadest sense, research ethics includes all issues of a moral nature that are associated with the planning, conduct, dissemination and impacts of research, while research integrity has a more demarcated meaning, specifically concerned with research quality and forms of research misconduct like fabrication, falsification and plagiarism (Poff 2014).

As the focus of this book is on vulnerable research participants and vulnerable groups that might take part in research, the next section reports on researchers’ opinions and their experiences of ethics approval systems when seeking to conduct research with vulnerable individuals.

2.3 Research Ethics Approval Systems and Vulnerability

In recent years research ethics approval processes have been the subject of a fair amount of criticism, with many researchers expressing frustration about their experiences of seeking ethics approval. For instance, ethics review has been described as a dreaded bureaucracy (Pickersgill 2012) and as a hurdle to overcome and a power game (Roets 2017), with researchers feeling they have to “play the game” in order to obtain approval (Vermeylen and Clark 2017). Frustrations have been ascribed to inconsistency between different ethics committees (Guta et al. 2013) and REC overreach in commenting on issues viewed as beyond the scope of ethics review (Gunnison and Helfgott 2021). Other concerns relate to a perceived lack of transparency in the ethics review process, with some researchers believing that ethics committees can “behave like county sheriffs—with no rhyme or reason in the assessment decisions made between boards” (Calfano 2016: 304). In addition, it is reported that many research ethics committees have “routinised” the ethics review process, leading to a checklist-based approach to assessing the sometimes complex ethical implications of research projects (Cox et al. 2020). Consequently, it is maintained, researchers go through the motions of ticking the “right boxes” or “saying the right thing” rather than thinking through the relevant ethical principles (Dyer and Demeritt 2009).

Against the backdrop of these criticisms, we investigated the experiences and opinions of researchers about governance measures that apply to the inclusion (or exclusion) of vulnerable populations in research. Their insights are essential for analysing how systems and processes work in practice and where changes might be needed. We know that research ethics approval mechanisms generally incorporate a protectionist approach (see Sect. 2.4), but can they also provide an enabling environment for the ethical inclusion of vulnerable populations? To inform this perspective on research governance we spoke with researchers around the world who are highly experienced in working with vulnerable populations. What follows is a brief overview of how we conducted this investigation and what the researchers had to say about the ethics approval process.

2.3.1 Investigating the Experiences and Perspectives of Researchers Who Work with Vulnerable Populations

2.3.1.1 The Method

Researchers with experience of working with vulnerable populations were identified via reputation, existing networks or snowballing (recommendations from other researchers). We wanted to include people with wide-ranging experience, who worked with different groups, across different fields and in different countries. It was important to us that at least half of the final number should have experience of working with vulnerable populations in low- and middle-income countries (LMICs).

Initial contact was made via an email, with an information sheet attached, which invited interested researchers to consent via Microsoft Forms. The method of interaction with the researchers and data collection was experimental to some extent, because we were seeking to explore ways of working that minimised power imbalances. Unlike traditional qualitative interviews, this investigation employed peer-to-peer conversations, similar to “participatory conversations” that involve interactive dialogue between the researcher and other(s) (Swain and King 2022), but also resembling “conversations with a purpose” as used in anthropology (Burgess 1988). The approach taken to the conversation was informal: a discussion between peers rather than a researcher–interviewee interaction, but with a specific purpose in mind.

Conversations took place online via Microsoft Teams and lasted around 45 min. To try to reduce the processing of personal data, an audio recording was made, but this was not transcribed verbatim as would be routine for interview research. Instead, notes were made from the recording to summarise the primary findings. After this, the recording was destroyed. For one researcher, who did not want to be audio-recorded, notes were made during the online conversation. These measures helped to minimise the processing of personal data. Data analysis was undertaken in an iterative manner. For this chapter, only the information related to experiences and opinions about ethics approval were extracted and themed.

2.3.1.2 The Findings

Conversations were held with 22 highly experienced researchers who had collective experience of working across a wide range of fields with persons from the following groups:

• Refugees

• Pregnant refugees

• Asylum seekers

• Undocumented migrants

• Evacuees from armed conflicts

• Homeless people

• People with mental health concerns

• Domestic abuse survivors

• People with cognitive disorders

• Unconscious patients

• Prisoners

• Young offenders

• Sex workers

• People with drug dependency/‌substance abuse

• People with post-traumatic stress disorder (PTSD)

• Members of minoritised ethnic groups

• People living in extreme poverty

• People with stigmatised conditions like HIV, TB and syphilis

• People who engage in activities that are against the law in their country, like men who have sex with men

• Indigenous communities

• People with multiple traumas

Additionally, several researchers mentioned that intersectional, complex vulnerabilities were the norm for participants in their projects. Of the 22 researchers, eight conducted their research in high-income countries (HICs)—four of these with refugees and asylum seekers—nine conducted their research in LMICs and five had experience of working in both HICs and LMICs.

Most of the researchers we spoke with employed social science/‌qualitative approaches, and many spanned disciplines (for instance, midwifery and social science, or nursing and anthropology). The type of research they undertook did not necessarily reflect the field in which they had originally been trained. While most were involved in health or wellbeing-related research, only three of the 22 were exclusively involved in clinical research.

2.3.1.3 Challenges

Insights from the researchers revealed that most had encountered challenges with the research ethics approval process, although experiences and opinions varied widely. The five most significant challenges are shown in Table 2.1 and explained further below.

Table 2.1 Perceived challenges when seeking ethics approval for research involving vulnerable individuals
Full size table

Different Priorities The researchers we spoke with noted a divergence in priorities between themselves and the REC members when RECs demanded compliance with institutional requirements while the researchers wanted to prioritise a community’s wishes and perspectives: for instance, when RECs insisted that legal information about data processing be added to the information sheets in a certain format while the researcher knew that legalistic language would raise fears and could act as a barrier with specific populations (e.g. undocumented migrants), or required all information to be written on the information sheets even when this was likely to be overwhelming or not understood. In general, there were concerns about the application of generic institutional ways of protecting participants, rather than a nuanced approach. Often RECs had established ways of doing things, the researchers noted, and they could struggle when people wanted to do things differently.

Subjectivity in the Process Most of the researchers we spoke with had experience of seeking approvals from a number of different RECs. It was a common observation that RECs fared differently in terms of level of scrutiny, the time taken to review, the questions asked and the rigour with which proposals were reviewed. It was also noted that review depended largely upon which particular members of the committee were looking at the application, because people from different backgrounds would bring their own beliefs with them. Mostly, researchers regarded the quality of feedback from their RECs as good, but sometimes there were requirements and remarks that they saw as nonsensical: for instance, asking for information about women being trafficked, when the research had nothing to do with trafficking.

On the subject of vulnerability, the researchers noted that some REC members were more risk-averse than others, meaning that ethics approval for the inclusion of vulnerable individuals in research was harder to obtain from those members. This might depend upon the familiarity of the REC members with the location, population or specific circumstances of the study concerned. Two researchers commented that they had observed a big difference between African RECs and those in the UK, Europe and the US. They noted that RECs in these HICs were more ready to label approaches as unethical.

Time Taken Some of the researchers had experienced months of delays associated with ethics approval, leading to obvious frustration. They ascribed this to a number of factors:

  • delays early in their research careers, before they became used to the ethics approval systems

  • approval being sought from a number of different committees

  • approval being sought from an under-resourced committee

  • disagreements/negotiations between the researcher and the REC regarding certain processes (such as informed consent processes).

Difficulties Understanding the Methods or Approach A common observation was that RECs can push back if the proposed research methodology does not fit in with their understanding of what approach should be taken in the research and how the study should be conducted. This seems to be a greater problem for co-created, participatory forms of research than for other approaches because it is more difficult to provide a clearly defined protocol at the proposal stage. For participatory research, the proposal has to be developed and/or refined in conjunction with the research participants, and this often happens in the early phases of the study, after ethics approval. It can also be a problem in research with Indigenous populations—a problem that some of the researchers associated with colonialist perspectives and a lack of appreciation for Indigenous research methods and knowledge creation.

Several had experienced condescending attitudes, with REC members asking, “What’s the point of this research?” or making pejorative statements like “This is not research!” Among those researchers who engaged in participatory research, there was the perception that ethics approval faced more challenges than other types of research, especially with RECs that were unfamiliar with the approach.

Paternalism and Risk-Averse Attitudes A common theme to emerge in conversation concerned the risk-averse and sometimes paternalistic attitude of some REC members. RECs are concerned with protection: when they spot the inclusion of a “vulnerable” group, they want to know why it is included, the risks involved and the special protective measures to be taken. According to the researchers in this study, these demands sometimes go too far. For instance, RECs can be concerned about traumatisation during an interview and automatically take the stance that the participants will be upset. Interestingly, researchers in this study, who worked with highly vulnerable individuals, had not experienced this as an issue. In fact, the opposite had been observed, namely, that people were able to report their traumas without becoming overly distressed and appreciated someone taking the time to listen to them. If these people were to be excluded from research for paternalistic reasons, they said, they would effectively be silenced.

The unjust exclusion of people perceived as vulnerable was viewed as paternalistic, as was the tendency to lay down too many conditions or ask for unnecessary changes to the protocol. It could seem as if there was no confidence in the capacity of the vulnerable to make wise decisions. Researchers who had experience of both HIC and LMIC RECs had observed paternalism only from HIC RECs, even where the research was to be based in an LMIC: thus, a type of remote paternalism. Getting the balance right between protection and inclusion was generally regarded as a challenge (see also Chap. 1), but not an insurmountable one.

2.3.1.4 Benefits

Insights from the researchers also brought to light benefits associated with going through the research ethics approval process. While most had experienced some challenges, a few had only positive things to say about the REC approval process. Indeed, most of the researchers expressed deep-felt appreciation for the feedback and support they received from experienced RECs. The primary benefits included:

  • the REC’s careful consideration of every detail of the study

  • a heightened understanding of the research and its justification through having to explain it to the REC

  • increased confidence that the ethical issues of the research were being addressed

  • increased awareness of the ethical and legal implications of the research

  • learning through doing.

The last two benefits enumerated above—increased awareness of, and learning about, research ethics—appeared to be extremely important to the group of researchers who expressed appreciation for the research ethics approval process. Most had had very little, if any, formal training in research ethics, and the process of applying for ethics approval had played a significant role in their education and development.

Another common theme was that applying for ethics approval had become easier over time as their careers progressed. This appeared to be thanks to two complementary factors: the researchers becoming more familiar with the process and requirements, and their RECs becoming more familiar with their research.

2.3.1.5 Research Ethics Codes

During the conversations, all researchers were asked which research ethics codes or guidelines they were familiar with and applied when undertaking research with vulnerable individuals. Their responses were surprising. More than half of the researchers (13) replied that they did not consult research ethics codes or guidelines beyond those of their institution.

This aligns with our finding that most had learned “on the job”, developing their skills and understanding while applying for ethics approval, as well as in the field. Aside from a range of national and/or professional codes, the most commonly mentioned codes and guidelines were:

  • Declaration of Helsinki—World Medical Association (WMA 2013) (5)

  • International Ethical Guidelines for Health-related Research Involving Humans—Council for International Organizations of Medical Sciences (CIOMS 2016) (3)

  • Good Clinical Practice Guideline—International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH 2023) (2).

As a statement of ethical principles for medical research involving humans, the Declaration of Helsinki is the most widely used research ethics guidance in the world; the CIOMS guidance is also used internationally but provides much more breadth and depth; and, as the name suggests, the ICH requirements apply specifically to the pharmaceutical industry. It is noticeable that these three documents focus heavily on (clinical) medical research. This was surprising, given that the disciplines of most the researchers were non-clinical (see Sect. 2.3.1.2), but also predictable, given the alleged “medical dominance within research ethics committees” (Humphreys et al. 2014).

The overall impression from our conversations was that the majority of researchers did not use ethics codes or guidance regularly. This might also be true of some of those who did mention specific codes. For instance, of the three researchers who used CIOMS, one stated that it was hard to use regularly because there were too many nuances, and another noted that they did not refer to the guidelines very often because they had been working in the field for so long.

In the next section, we turn to research ethics codes and guidelines, and what they have to say about the inclusion of vulnerable persons in research.

2.4 Vulnerability and Research Ethics Codes and Guidelines

Historically, the inclusion in clinical studies of individuals perceived as being vulnerable has been the primary motivation and focus of concern for the development of protectionist codes and guidelines (Levine 1988). In this respect, the protection of vulnerable individuals can be regarded as inherent within ethics codes for research with human participants. But debate is ongoing, especially regarding these questions:

  • Who should be deemed especially vulnerable in research?

  • To what precisely are they vulnerable?

  • How can this vulnerability be addressed?

These questions are important because ethics codes and processes that aim to protect vulnerable populations can also lead to the exclusion of certain individuals if they are wrongly labelled as vulnerable, or if the researchers do not understand how to mitigate their vulnerability in research.

For our investigation, a search was undertaken to see what current research ethics codes and guidelines have to say about who should be deemed vulnerable, to what they are vulnerable in the context of research, and how this might be addressed. Not all ethics codes and guidelines mention vulnerability, so the first step was to identify those that do. The search was limited to English documents and official English translations of guidelines and documents of the following databases:

  • Ethics Codes Collection (n.d.)

  • Office of Human Health Protections (US Department of Health and Human Services): International Compilation of Human Research Standards (HHS 2022)

  • UNESCO Digital Library: Standard-setting instruments (UNESCO n.d.)

  • World Medical Association: Medical Ethics (WMA n.d.)

The following search terms or parameters were used: vulnerability; vulnerable; vulner*; high-risk; high risk; “groups at risk”.

In total, 57 current versions of ethics codes and guidelines were identified as relevant. Table 2.3 provides an overview of the numbers and types of authors or owners of the codes and guidelines.

Table 2.3 Numbers and owners of relevant research ethics codes and guidelines
Full size table

Once the relevant codes and guidelines had been identified, their guidance on the involvement of vulnerable people in research was extracted and analysed. There is a very wide variation in the extent to which vulnerability is addressed in the codes and guidelines, ranging from a single sentence to whole chapters. The following provides a summary of our findings.

2.4.1 Who is Vulnerable?

Most codes and guidelines do not offer a precise definition of vulnerability or of who is to be considered vulnerable, but there are certain commonalities. The most striking, regarding who is vulnerable, is the suggestion that vulnerability can stem from belonging to a certain group. When vulnerability is mentioned, it is commonly accompanied by examples of the sorts of groups to which it applies. For instance, the International Council for Harmonisation (ICH) Good Clinical Practice Guideline (ICH 2023) defines “vulnerable participants” as follows:

Individuals whose willingness to volunteer in a clinical trial may be unduly influenced 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. Examples are members of a group with a hierarchical structure, such as medical, pharmacy, dental and nursing students; subordinate hospital and laboratory personnel; employees of the pharmaceutical industry; members of the armed forces and persons kept in detention. Other vulnerable participants may include persons in nursing homes, unemployed or impoverished persons, patients in emergency situations, ethnic minority groups, homeless persons, nomads, refugees, minors and those incapable of giving consent.

The groups most commonly mentioned are set out in Fig. 2.2, together with the number of documents (out of 57) in which they are named.

Fig. 2.2
figure 2

Mentions of vulnerable groups across 57 research ethics codes and guidelines

Full size image

More than half (32/57) of the codes and guidelines we examined give examples of groups that are vulnerable, but as we can see from the ICH guideline (ICH 2023), these groups are often specified in the context of participation in clinical trials. Ethics codes and guidelines originated in the biomedical domain, so it is hardly surprising that opinions about who is vulnerable in research have been influenced strongly by clinical research ethics. However, as explained in Chap. 1, this book is primarily concerned with adults who are able to provide informed consent but whose decision-making power may be constrained by other factors.

A problem with the naming of particular groups, even as examples, is the implication that people are to be considered vulnerable simply by virtue of belonging to a certain group. This can render the term unhelpful and patronising. For example, the labelling of impoverished people as vulnerable in resource-poor regions means that virtually the entire population is to be considered vulnerable. Nowadays, there is (at least implicit) acknowledgement across many codes and guidelines that determining who is vulnerable requires a more nuanced approach because vulnerability can stem from a range of factors—for instance, power differentials between researchers and participants, capacity to understand and unrealistic expectations.

The British Association for Counselling and Psychotherapy (BACP) Ethical Guidelines for Research in the Counselling Professions (Mitchels 2019) highlights three primary lines of vulnerability that are broadly acknowledged, in that “participants may be particularly vulnerable, for example, as a result of their innate characteristics, the context of their life or the research or the type of research undertaken” (Mitchels 2019: 61). This adds a third dimension to the definition of vulnerability we are using in this book, namely:

To be vulnerable means to face a significant probability of incurring an identifiable harm while substantially lacking ability and/or means to protect oneself. (Schroeder and Gefenas 2009: 117)

People with innate characteristics that make them vulnerable in research are likely to be those who are unable to provide informed consent because of, for instance, severe cognitive disabilities. These are people who lack the ability to protect themselves from harm and exploitation in research. When the context of a person’s life makes them vulnerable in research, this may be because they live in severe poverty, reside in a refugee camp without humanitarian protection, or work in a strict hierarchy—the military, for instance. These are people who might lack the means to protect themselves from harm and exploitation in research.

Where the research or the type of research places people in a vulnerable position, this might include high-risk clinical studies. For instance, human challenge studies for COVID-19 could be called high-risk clinical studies as “wildtype SARS-CoV-2 can cause severe disease and death and there is no rescue therapy” (Weijer 2024). This third dimension to vulnerability highlighted by the BACP would make it possible for anybody to be vulnerable to harm or exploitation, depending on the study they enter.

There is also evidence of a shift away from reliance upon the naming of groups in the biomedical domain, for instance in the fourth version of the CIOMS (2016) guidelines:

A traditional approach to vulnerability in research has been to label entire classes of individuals as vulnerable. The account of vulnerability in this Guideline seeks to avoid considering members of entire classes of individuals as vulnerable. (CIOMS 2016: 57)

Similarly, the Canadian Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans (known as TCPS2) cautions that:

individuals should not automatically be considered vulnerable simply because of assumptions made about the vulnerability of the group to which they belong. Their particular circumstances shall be considered in the context of the proposed research project. (CIHR et al. 2022: 73)

Hence, while the labelling of specific groups as vulnerable in research—a tendency that has its roots in biomedical research—continues in many research ethics codes and guidelines, there are a growing number of calls for a more individualised approach.

2.4.2 To What Are They Vulnerable?

To be useful in research ethics, the question “Who is vulnerable?” needs to be supplemented with the question “What they are vulnerable to?” Many ethics codes and guidelines adopt a very similar, if not identical, description of vulnerability to that of the ICH guidelines (2023), including national guidelines from Bulgaria, India, Malaysia, Nigeria, Philippines, Saudi Arabia, Sierra Leone, Singapore, Tanzania and Zimbabwe. This description concerns the voluntariness of participation in research when the potential participants are people who might be unduly influenced “by the expectation … of benefits” or of a “retaliatory response from senior members of a hierarchy in case of refusal to participate” (ICH 2023). These are essentially consent-based matters, as are the vulnerability-related concerns about the inclusion of people who lack capacity to consent and the inclusion of minors in research. Hence, what they are vulnerable to is taking part in research without having given the proper consent.

Undoubtedly, vulnerability as described in the codes and guidelines that were analysed focuses overwhelmingly on whether or not free and informed consent can be given (Bracken-Roche et al. 2017). In fact, some prominent ethics codes do not refer to vulnerability at all and make the inability to provide informed consent the only criterion for special protective considerations. For instance, the Council of Europe’s Convention on Human Rights and Biomedicine, also called the Oviedo Convention, makes specific protective provisions only for “persons not able to consent to research” (Council of Europe 1997: art. 17).

There is scant information about why groups like pregnant women and impoverished people are included as examples of vulnerable populations, when they clearly have the cognitive abilities to provide informed consent. Addressing this issue, Coleman (2009) helpfully distinguishes between three types of vulnerability in research: consent-based, risk-based and justice-based.

2.4.2.1 Consent-Based Vulnerabilities

Most groups that are mentioned in research ethics codes and guidelines are vulnerable to consent-based violations that can be associated with innate characteristics (e.g., those with severe cognitive dysfunctions) or with their circumstances, which might impact upon the voluntariness of their consent (e.g. the existence of strong power differentials between them and the researchers). This conveys the message that informed consent cannot be obtained from certain groups. However, there is a significant difference between people who are intrinsically unable to provide consent and people who possess full cognitive capabilities but are in a situation that makes them potentially vulnerable to undue influence. Hence, we refine Coleman’s distinction, splitting consent-based vulnerabilities into two types: intrinsic consent-based vulnerability (e.g. babies, or adults with severe cognitive impairment) and extrinsic consent-based vulnerability (e.g. prisoners or students) (Rogers and Ballantyne 2008; Schroeder and Gefenas 2009). This makes it possible to apply a more nuanced approach to consent-related matters for inclusion in research.

Importantly, we are not arguing for refining or changing the approach to the inclusion of those who cannot provide informed consent because they lack the ability to do so. We are arguing for changing the approach to the inclusion of some of those who are adults with the cognitive abilities to provide consent, but may lack the means to protect themselves from harm or exploitation (see our recommendations in Chap. 6).

2.4.2.2 Risk-Based Vulnerabilities

In Coleman’s (2009) second category, vulnerability stems from being at increased risk of mental or physical harm. For instance, the exclusion of pregnant women from clinical studies is likely based on the health risks to those women and their unborn foetuses, but there are also many other reasons why people might be at increased risk of harm from to their involvement in research. The Australian National Statement (NHMRC 2023) provides an extensive list of the sorts of harm to which a research participant might be vulnerable, as shown in Fig. 2.3.

Fig. 2.3
figure 3

Source Figure produced using Australian National Statement (2023) available under CC-BY 4.0

Potential types of harm.

Full size image

2.4.2.3 Justice-Based Vulnerabilities

Coleman’s (2009) third category of vulnerability, justice-based vulnerability, refers to unfair inclusion or exclusion from research. This type of vulnerability is the least frequently mentioned across the codes and guidelines, but some describe it at length. For instance, CIOMS (2016) notes:

In the past, groups considered vulnerable were excluded from participation in research because it was considered the most expedient way of protecting those groups (for example, children, women of reproductive age, pregnant women) … This has resulted in a serious injustice. Since information about the management of diseases is considered a benefit to society, it is unjust to intentionally deprive specific groups of that benefit. (CIOMS 2016: 8)

Inclusion in research can also be unfair in circumstances where there is no, or limited, potential for benefit, for instance pharmaceutical research in regions where the local population would have no access to the resultant medication:

Inequity is created when particular groups fail to receive fair benefits of research or when groups, or their data or their biological materials, are excluded from research arbitrarily or for reasons unrelated to the research question. (CIHR et al. 2022: 9)

We need to understand clearly what it is that people are vulnerable to in order to identify how it should be addressed. This is outlined in the next section.

2.4.3 How Should Vulnerability Be Addressed in Research?

One way to deal with vulnerability in research would be to exclude everybody to whom a given vulnerability might apply. However, none of the codes and guidelines examined proposes that vulnerable persons should always be excluded from research. Indeed, some codes explicitly warn against this, such as the Norwegian Guidelines for Research Ethics in the Social Sciences and the Humanities (NNESH 2022), which states that “excessive protection of weak and vulnerable groups is inappropriate. This might result in their perspectives being excluded in research, and society might not gain knowledge about important topics” (NNESH 2022: 29). Likewise, the Canadian Code of Ethics for Psychologists (CPA 2017) stipulates that research activities should be designed in such a way that they “do not unfairly exclude those who are vulnerable or might be disadvantaged” (CPA 2017: 13).

However, most codes and guidelines offer little more than generalised guidance about how to ensure ethical inclusion for vulnerable participants. The Declaration of Helsinki states: “All vulnerable groups and individuals should receive specifically considered protection” (WMA 2013: art. 19), and the Malawian Framework of Guidelines for Research in the Social Sciences and Humanities advises that when vulnerable persons are involved, “researchers … shall be required to obtain extra protections or safeguards for their safety and welfare” (NCST 2011: 8). Precisely what these extra protections or safeguards might be is largely left up to the researcher. Special protections, to the extent that they are further elucidated, generally include measures to promote voluntary decision-making and to limit the risk of harm, with researchers and RECs both having a role to play in ensuring that these special measures are put in place.

An overriding message across the codes is that inclusion should be restricted, often along the lines of the Declaration of Helsinki requirement:

Medical research with a vulnerable group is only justified if the research is responsive to the health needs or priorities of this group and the research cannot be carried out in a non-vulnerable group. (WMA 2013: art. 20)

Of course, one could argue that the first part of this requirement, as a matter of fairness, should apply to all research, as in the TRUST Code: “Local relevance of research is essential and should be determined in collaboration with local partners. Research that is not relevant in the location where it is undertaken imposes burdens without benefits” (TRUST 2018: art. 1). This speaks to a form of justice-based vulnerability (Coleman’s third criterion), or unfair inclusion in research. The second part of the requirement (that only research that cannot be undertaken with a non-vulnerable group may be conducted with a vulnerable group) is aimed primarily at Coleman’s risk-based vulnerabilities. For instance, clinical investigations into certain diseases that have a high likelihood of being fatal in children (e.g. Fanconi anaemia) might involve a risk of harm to the participants, but it is impossible to undertake all of the necessary research with another group. This proviso is intended to prevent unnecessary harm. Nevertheless, certain vulnerable groups have been excluded from research because of this proviso, even though there is no scientific reason for it. For example, people living with HIV are routinely excluded from lymphoma clinical trials even though they are at increased risk of developing the disease (Venturelli et al. 2015). Seeking to avoid a risk-based vulnerability can create a justice-based vulnerability (unfair exclusion).

While most research ethics codes and guidelines have something to say about the inclusion of vulnerable people in research, in general they promote the same two messages: first, that most vulnerabilities are associated with voluntariness, and second, that certain groups should be awarded more protection than others. Beyond exclusion, precisely how this protection should be afforded is unclear. A minority of longer and more detailed codes and guidelines (e.g. CIOMS and TCPS2) warn against blanket exclusions and offer more nuanced advice, but these codes are not applied universally, nor would it be practical to recommend that all research ethics codes be made longer and more detailed. Furthermore, we know from our conversations with researchers that many, if not most, rely upon their own institution’s requirements and processes rather than appealing directly to research ethics codes. In the last section of this chapter, we consider the implications of the findings from our review of the codes and our conversations with researchers. We analyse further what we have learned and look at how this might inform research governance to help reduce exclusion and promote ethical inclusion,

2.5 Promoting the Ethical Inclusion of Vulnerable People in Research

Often cited as the cornerstone of research ethics (Xu et al. 2020), the requirement for informed consent in research was recognised as early as the nineteenth century, when the development of clinical medicine was accompanied by an increased demand for experimentation on humans, who were often included without their permission (Vollmann and Winau 1996). Consequently, an enduring message throughout the history of ethics codes has been that people who, for whatever reason, are unable to provide informed consent must be awarded special protection because they are at increased risk of being exploited or harmed in research. Yet, one of the most striking points to emerge from our conversations with researchers was that it could seem to them as if REC members had no confidence in the capacity of adult vulnerable individuals to make reasonable decisions about their involvement in research. In their efforts to protect research participants, RECs can stray into paternalism, defined by Dworkin (1972: 65) as “the interference with a person’s liberty of action justified by reasons referring exclusively to the welfare, good, happiness, needs, interests or values of the person being coerced”.

Of course, there are times when a blanket approach to protection is warranted for entire groups, such as babies and young children, or for adults with severe cognitive impairment. However, this protectionist stance is often applied to groups of people who do have the capacity to provide consent (such as prisoners). So why does this happen? To answer that question, it is helpful to remember that research ethics codes originated in the biomedical domain and were intended to prevent the kinds of abuses exposed by Beecher (1966), in which people might be subjected to a high risk of harm in research, even without their knowledge or consent. The resolve to protect people from exploitation in research runs deep, and it is widely acknowledged that RECs remain dominated by the ethics requirements that were originally set down for biomedical and experimental research (Pelek et al. 2023). These requirements and processes have been extended to non-clinical research involving human participants, including humanities and social science studies, even though research in these fields does not generally pose equivalent risks (Doyle and Buckley 2017).

The ways in which vulnerability is described in codes and guidelines, with repeated reference to certain groups, has the inevitable impact of ringing protectionist alarm bells when people from these groups are being considered for inclusion in research. This raises the question of whether it might be best to avoid the naming of groups altogether, and to address vulnerability as a matter that pertains purely to individuals rather than to groups. However, Coleman (2009) claims that it would be a mistake to characterise vulnerability as either an entirely individual or an entirely group-based phenomenon. In Coleman’s view, consent- and risk-based vulnerabilities make more sense when conceptualised as individual issues, but justice-based vulnerabilities are more likely to concern groups or communities.

But this is not the message that is conveyed by most codes and guidelines. Most do not distinguish between individual and group vulnerabilities, nor do they distinguish between different types of vulnerability. In our view, the blanket labelling of groups as vulnerable is only relevant in two scenarios:

  • where there is consent-based vulnerability, and this vulnerability is of an intrinsic nature (e.g. including babies or adults with severe cognitive impairment), hence where there is a lack of ability to consent

  • where there is justice-based vulnerability that has implications for entire groups.

As previously mentioned in this chapter, justice-based vulnerabilities can manifest themselves in either unfair inclusion or unfair exclusion. Unfair inclusion can occur when research is undertaken with a group to whom the research is not relevant. For instance, Kenyan prisoners were going to be enrolled in a medical study undertaken by HIC researchers for the sole benefit of the HIC’s military at home. Only the vigilance of a local Kenyan REC prevented this research (Chatfield et al. 2021). This is why the Declaration of Helsinki adds to its requirements on vulnerability the stipulation that any group involved in research “should stand to benefit from the knowledge, practices or interventions that result from the research” (WMA 2013: art. 20). There must be a fit with local health needs and priorities plus access to the results. Similarly, CIOMS guidance (2016: 63) refers to the potential for “group vulnerability” in, for instance, some resource-limited countries or communities. These types of situations often bring to light justice-based vulnerabilities that require actions at the community level, for example agreements for benefit sharing (Schroeder 2007).

Other justice-based vulnerabilities take the form of unfair exclusion. This might be due to a paternalistic stance on the part of the RECs—and paternalism is not a rare phenomenon being “quite common and present wherever global compassion has become institutionalized” (Barnett 2012: 487). Given this tendency, it is extremely important for RECs to be aware that protectionist measures intended to address vulnerability in research can result in unfair exclusion. The protectionist exclusion of vulnerable people from research can result in serious injustice and may also be contrary to the wishes of individuals or groups (Friesen et al. 2017).

While unfair exclusions can be motivated by a desire to protect, they can also be related to matters of convenience. For instance, Charles et al. (2016) analysed research applications involving UK prisoners, augmented by a survey of UK researchers and REC members. They found that “pragmatic concerns regarding the perceived burden of including prisoners [in research] are far more prominent in motivating their exclusion than ethical concerns” (Charles et al. 2016: 1). In other words, obtaining REC approval for research with prisoners can be considered too burdensome.

Likewise, De Poli and Oyebode (2023) found that for qualitative, participatory and collaborative research, perceived or expected barriers at the ethics approval stage could deter researchers from involving vulnerable groups:

At worst, it could undermine the role of qualitative, participatory, and collaborative research in promoting research inclusivity and social justice, and in answering research questions that no other research methods could address. (De Poli and Oyebode 2023: 3)

The findings from our conversations with researchers echo the challenges associated with seeking ethics approval that are articulated in the published academic discourse, as noted above. However, certain issues are likely to be even more tricky to navigate when the research involves vulnerable individuals. Among the researchers there were obvious frustrations associated with these challenges, which might be perceived as hindrances to research involving vulnerable individuals, especially for researchers with less experience.

The routine identifying of groups as vulnerable can blind researchers and RECs on occasions when there is a need for a more nuanced approach. Treating vulnerability as a group-based phenomenon only makes sense when it applies to all members of a group (e.g. babies), and that is often not the case for the many people who are excluded from research because of a particular label (e.g. prisoners, Indigenous populations, sex workers).

Most of the researchers we spoke to regard the ethics approval process as a major point of learning and do not consult research ethics codes and guidelines beyond those of their institutions. This highlights the importance of the role that RECs play in research ethics. Well-informed RECs with a clear understanding of who can be vulnerable in research and how this vulnerability might be addressed can help to ensure that unnecessary exclusion is avoided. This might also contribute to the minimisation of convenience exclusions.