Vulnerability: On a Welcome Paradigm Shift in the 2024 Declaration of Helsinki

The 2024 version of the Declaration of Helsinki [1] provoked a flurry of commentaries from the international research community on a host of different issues [2-6]. These included the broadened scope of the revised Declaration, its emphasis on integrity, the inclusion of the notion of meaningful engagement, the importance of ethical research in public health emergencies, research ethics committees’ (RECs) independence and the importance of providing them with adequate resources. It should also be noted that, with regard to some very controversial issues such as placebo control and post-trial benefits, the Declaration is not vastly different from its 2013 version [4]. However, some significant changes have arguably not been sufficiently considered in the published commentaries. Here, I will focus on paragraphs 19 and 20 of DoH 2024 that refer to vulnerability and suggest that these changes usher in a welcome paradigm shift regarding the concept of vulnerability.

The importance of certain changes to language in the revised version of the DoH has been noted. For example, the term “research subject” was replaced by “participant” [3, 6]. Thus far, no scholarship has addressed the linguistic dimension surrounding the Declaration's revision of the notion of vulnerability. Yet in paragraph 19, there is indeed a crucial change in language related to this concept. A new description of vulnerabilities is introduced: individuals, groups, and communities are in a “situation of more vulnerability.” In other words, the 2024 DoH leaves behind the essentialist conception of vulnerability–that individuals and groups are particularly vulnerable (Paragraph 19 (DoH 2013)). Note that the term “vulnerable populations” has been avoided, as now different situations may render these groups or communities more vulnerable. Furthermore, this is dynamic and contextual and may change.

This addition of a more dynamic conception of vulnerability has important consequences for research ethics. Perhaps most significantly, it means not labeling a priori with a categorical or essential property or following a predetermined list of people already considered vulnerable. However, some commentators fail to appreciate this difference and still use the previous terminology from 2013 [5, 7] as the 2024 version still maintains that factors may be fixed. But the 2024 DoH expands and refers to “a situation of more vulnerability as research participants, due to factors that may be fixed or contextual and dynamic,”¹ thus introducing a different way of thinking about the concept that captures the various ways in which vulnerabilities are expressed. This non-essentialist vision avoids stereotyping and labeling individuals, groups, and communities as vulnerable. As when a label is applied, it fixes the content and is quite difficult to remove; but if we think there are circumstances or characteristics that are dynamic, they may change. The context can also be modified.

The essentialist approach makes two assumptions. On one hand, it assumes a baseline standard for a default paradigmatic participant: a mature, reasonably educated, literate, self-supporting person. This is an overly idealized vision of research participants. Second, it assumes that identifying vulnerabilities in populations means identifying populations failing to meet the characteristics of this paradigm. (In addition, it may also include subpopulations or persons that may not be vulnerable.)

The alternative concept of vulnerability instead allows for a complex description of vulnerabilities. That is, there might be several vulnerabilities that may overlap, compounding a participant's vulnerability. They might be multiple and different, some related to the lack of informed consent, to socio-economic conditions, age, gender, and other characteristics. Vulnerabilities may co-exist, and they can be minimized or eradicated (when possible) one by one. I have discussed this extensively elsewhere [8, 9].² The idea of multiple vulnerabilities allows us to understand what is called the cascading effect of some situations of vulnerability, where exacerbating factors worsen existing vulnerabilities or generate new ones. They can lead to a domino effect. Their amplifiable nature makes them very harmful [9]. This paradigm shift in the understanding of vulnerabilities allows us to think outside the box and carefully analyze participants’ or groups’ situations and context. For example, during the pandemic, physicians in many Low- and Middle-Income Countries—normally considered a non-vulnerable population—lacked access to adequate protective equipment (PPE) and were thus over-exposed to the virus, so they were subject to a situation of vulnerability. Thus, the context may introduce new and unpredictable vulnerabilities [10].

This document speaks of different “characteristics,” “factors” that may coexist and that are context dependent. What are those characteristics or factors that the CIOMS-WHO Guidelines refer to? Situations of more vulnerability due to different factors or “layers of vulnerability.”

The second sentence of paragraph 19 tackles one of the problems of the essentialist and “check-mode-reasoning” some researchers or RECs use. For such a mentality, vulnerability and “vulnerable populations” are synonymous with exclusion from research. This attitude has been quite harmful. In several cases, extreme protectiveness has paradoxically led to a lack of protections. This was the case with pregnant women. Because of their consistent exclusion from research for decades, there is very little proven data on medication used during pregnancy. More than 80% of pregnant patients are routinely prescribed therapies that have not been studied in pregnancy [12]. Pregnant persons get ill and individuals with illnesses also get pregnant. Clinical research that is inclusive of such people is needed [13, 14].

The DoH 2024 rightly notes that when individuals, groups, and communities have distinctive health needs, exclusion can potentially perpetuate or exacerbate disparities. Situations of vulnerability are not synonymous with ethically-sound arguments for immediate exclusion. Researchers and RECs need to consider each situation, and if research ought to be undertaken, then adequate safeguards or empowerment of research participants should be implemented. This explains why a list of vulnerable populations does not work. It also shows the importance of deliberation and why “the harms of exclusion must be considered and weighed against the harms of inclusion” (DoH 2024 Paragraph 19).

Apart from the new, last condition, this paragraph has not changed greatly from the previous version. And it is also in line with Guideline 15 of CIOMS, as it speaks of specific protections that should be in place to safeguard the rights and welfare of these particular individuals or groups.

To conclude, this revisited version of the DoH introduces a different paradigm of the concept of vulnerability that reflects in a better way the complexities of the real world. It brings flexibility and dynamism to the analysis. It indirectly asks for more deliberation by warning of the harms of exclusion while also specifying clear conditions for protection.