The critique of vulnerability offered by Levine et al.(2004) affirms recent discussions about the disutility of this imputed characteristic of individuals and groups for protecting research subjects. Being “too broad,” vulnerability stereotypes whole categories of individuals, and everyone might be considered vulnerable. Being “too narrow,” vulnerability’s focus on group characteristics diverts attention from features of the research project and its environment that might affect subjects. We agree with these assessments. The application of a label of “vulnerability” to particular groups resembles the disfavored legal category of “status crimes” like vagrancy or homelessness. Such an externally imposed label is static and thought to inhere in the person; moreover, besides being potentially misleading, it is highly likely to exacerbate stigma and dependency. However, in our view, the solution Levine et al.(2004) offer, that research involving vulnerable individuals or groups be targeted for “special scrutiny,” misses a fundamental point. The reason vulnerability is so inadequate is that it is not merely a subject label, nor is it merely a trigger for heightened scrutiny of particular types of research. Vulnerability is, by definition, relational. One is always vulnerable to someone’s or something’s influence. By virtue of being relational, vulnerability to influence is potentially bidirectional in all clinical research relationships, including the primary relationship of interest here: that between researcher and subject. Vulnerability and influence are produced to varying degrees for both subjects and researchers in research relationships. Applying vulnerability to only one party contributes to conceptual confusion and undermines constructive application of the term.

To define and assess vulnerability to influence, we propose a relational model. This model does not assume vulnerability. Instead, it considers influence as inevitable in research relationships and seeks to determine whether, in a given relationship, influence is so powerful as to be deemed “undue.”