According to the NPRM, “the only vulnerability that needs to be considered is vulnerability to coercion or undue influence, and not other types of vulnerability.” It therefore replaces all standalone uses of “vulnerable” with “vulnerable to coercion or undue influence.” This change is justified on the basis that it will “provide greater consistency and clarity in IRB consideration of vulnerability of subject populations in research activities and appropriate protections,” where the vulnerable populations in question are “children, prisoners, pregnant women, physically or mentally disabled persons, or economically or educationally disadvantaged persons.”
Two provisions of the Common Rule (§107.a and §111.a.3) currently discuss vulnerability without further specification, whereas one (§111.b) discusses vulnerability to coercion and undue influence. Deleting §111.b’s reference to coercion and undue influence, however, would achieve consistency while making even fewer changes than the NPRM proposes. The proposed revisions, then, rest on improved clarity rather than improved consistency.
I doubt that narrowing vulnerability by adding the terms “coercion” and (in particular) “undue influence” adds much clarity. Rather, these changes may reduce protections against research flaws other than coercion and undue influence without offering counterbalancing advantages for the research enterprise.
While the NPRM suggests greater definition of several terms, it does not propose defining coercion or undue influence. The FDA has indicated in guidance, however, that it adopts the Belmont Report’s interpretation of those terms:
Coercion occurs when an overt threat of harm is intentionally presented by one person to another in order to obtain compliance. Undue influence, by contrast, occurs through an offer of an excessive, unwarranted, inappropriate or improper reward or other overture in order to obtain compliance.
Coercion is the clearer of these two, but many agree that true coercion of subjects by researchers is vanishingly rare. Adding undue influence to the mix, meanwhile, multiplies rather than reduces unclarity, by replacing one unclear term, vulnerable, with four: excessive, unwarranted, inappropriate, and improper. Research indicates that confusions surrounding both coercion and undue influence are widespread.
Focusing exclusively on coercion and undue influence not only fails to improve clarity, but also ignores other forms of vulnerability. Coercion and undue influence threaten informed consent: a subject vulnerable to coercion or undue influence is in danger of participating in research she has not informedly consented to. But, on one prominent account, subjects are entitled not only to informed consent but also to research that has social value, is scientifically valid, selects subjects fairly, has a favorable risk-benefit ratio, is independently reviewed, and respects subjects. The biggest threats to these goals are not coercion or undue influence, but rather disrespect, discriminatory subject selection, excessive risk, and exploitation, to name just a few. (I would go further: such dangers are more common than coercion and more serious than undue influence.)
In light of these other dangers, as Samia Hurst has observed, “it is not clear that the sort of thing we mean by vulnerability should refer strictly to being at risk of giving faulty consent.” For instance, revising §111.a.3—which concerns equitable subject selection—to focus solely on vulnerability to coercion and undue influence leaves pregnant women and disabled persons less protected against unjust exclusion from clinical trials. Similarly, revising §107.a—which concerns IRB membership—to ignore other forms of vulnerability fails to ensure that IRB composition aims at protecting educationally and economically disadvantaged people from disrespect and exploitation during the research process.
Ultimately, the NPRM’s exclusive focus on vulnerability to coercion or undue influence is only one of many options. Alternatives include:
- Retaining the current Common Rule’s concern about vulnerability broadly construed
- Focusing instead on vulnerability to some other research-related danger(s), such as the danger of being wronged by research.
- Abandoning the concept of vulnerability altogether.
I am not yet convinced that the NPRM is correct to abandon (1). But even for those convinced that the treatment of vulnerability in research ethics should change, shifting to (2) or (3) may be preferable to the NPRM’s approach.