The Continued Complexities of Paying Research Participants

This article refers to:

“Paid to Endure”: Paid Research Participation, Passivity, and the Goods of Work

How Payment for Research Participation Can Be Coercive

Paying research participants is a widespread, long-standing, ethically acceptable, and perennially fraught practice. Although data are limited, payment is offered to participants in many research studies with amounts ranging from a few dollars to several thousands of dollars. Payment can facilitate recruitment to clinical trials and enables research participation for persons in a variety of studies by replacing some of their otherwise lost wages or reimbursing them for expenditures. Payment may be essential to motivate and compensate healthy phase 1 trial participants, as it is usually their primary reason for joining these trials (Grady et al. 2017). IRBs, sometimes uncomfortable with payment, generally keep amounts of payment low in an effort to comply with ethical and regulatory directives to minimize the possibility of undue inducement and coercion (Gelinas et al. 2018, US FDA 2018). Commentators continue to justify payment across many types of studies and have called for paying higher amounts in order to mitigate exploitation or make research participation fairer (Gelinas et al. 2018; Lamkin and Elliott 2018; Largent and Lynch 2017)

Joining this fray, two articles in this issue address paying research participants. Millum and Garnett, acknowledging that institutional review boards (IRBs) and others continue to worry about payment coercing participants despite convincing theoretical arguments to the contrary, introduce a new way to understand payment as coercion. Malmqvist, while agreeing that a work model offers protection for research participation, argues that the work performed by healthy volunteers in pharmaceutical research is uniquely passive, hindering the ability of participants to satisfy the positive goods of “just” work. The authors of both articles ultimately conclude, for different reasons, that the amount of money paid to research participants should be increased.

Millum and Garnett distinguish coercion as subjection from coercion as consent-undermining, agreeing with others that offers of money are not consent-undermining coercion (Millum and Garnett 2019). They describe someone as a victim of coercion as subjection when she is subjected to a foreign will (not her own), resulting in a sort of unfreedom, a negative effect on her well-being, and a partial obstacle to her interests. They distinguish undue influence (which they describe as irrationality) and exploitation (unfairness) from coercion as subjection (unfreedom), although recognize significant overlap. For them, payment to research subjects is correctly understood as coercion as subjection when three conditions are met: (1) The participant joins the study because getting paid for research is her only way to avoid something unacceptably bad (such as poverty), (2) the researcher has intentionally gotten the participant to enroll “by helping to make it true that enrolling is the participant’s only way of avoiding continued dire poverty” (Millum and Garnett 2019, 25), and (3) the participant’s reasons and motivations for enrolling differ from those of the researcher.

It would be challenging to determine when Millum and Garnett’s three conditions that result in coercion as subjection apply. They themselves offer examples of when condition 1 or condition 3 might not apply or would be difficult to determine. Taking the first condition, it may be unusual and extremely difficult for a researcher or IRB to discover whether enrolling in research is the only way to make money in a given context. Even when participating in research might be the best or the easiest way to acquire money, it will rarely be the only way. Millum and Garnett’s preferred response for researchers and sponsors to avoid coercion as subjection, namely, increasing payment in order to amplify benefits, might actually contribute to making research participation the single best (or only) way to obtain a certain amount of money. Does that increase the risk of coercion as subjection rather than minimize it?

For me, however, the toughest criterion to satisfy is the second one—what does a researcher do to help make it true that offering money for study participation is the only way for a participant to get money? Or the only way to avoid poverty? One might imagine the researcher selecting an isolated island of desperately poor people where other opportunities for transaction do not exist, and somehow making sure that the researcher is the only one allowed in. Short of that, would researchers or sponsors have to deliberately attempt to discourage other types of transactions or divert resources, cutting off any other possible money-earning opportunities for participants? This seems almost implausible to me. Offering an option that doesn’t exist is not the same as cutting off other options, or “helping to make it true that enrolling” is the only option. Similarly, if an employer offered employment to persons who desperately need money in a location where jobs are scarce or nonexistent and an individual accepts this job for financial reasons, is the employer guilty of coercing the employee and subjecting her to a foreign will? Or is there something special about research?

Introducing a new type of “coercion” into the debate about payment to research subjects is unlikely to help inform policy or practice and may simply further muddy the waters. IRBs and researchers would need to differentiate and respond to concerns about two types of coercion, as well as to concerns about undue inducement and exploitation, and there is already considerable confusion about these concepts. Many understand undue inducement as a threat to voluntariness and rational decision making. Yet defining undue inducement has proven elusive, as has delineating circumstances under which an offer of money becomes “undue” instead of a mere inducement. Some have described payment enticing poor participants to enroll in research as an example of undue inducement (Largent et al. 2013), others as a problem of disproportionate distribution of risks and benefits (Halpern et al. 2004). Undue inducement has been described as interfering with rational judgement and consent (Lamkin and Elliott 2018; Largent et al. 2013), but also as inducing someone to do something to which they are averse (Dickert and Grady 2008; Grant and Sugarman 2004). Although unclear, the type of unfreedom that Millum and Garnett invoke seems a lot like using money to induce someone to do something to which they are averse—subjecting them to a foreign will—because of the appeal of money.

To further complicate matters, the authors note that coercion as subjection is not always impermissible, even if wrong. Sometimes, for example, the “badness” of coercion is outweighed by other benefits or by the social value of the study, making it morally permissible. Since the proposed conditions, responses to, and permissibility of coercion as subjection are context dependent, and there are conceptual intersections with undue inducement and exploitation, I am not sure how much insight it adds regarding how and when to pay research participants.

In his article, Malmqvist also argues for more generous payment to research participants but for different reasons (Malmqvist 2019). His focus is a small, but interesting, group of research participants: repeat healthy volunteers who participate as a way to generate income. He claims that these participants cannot satisfy the positive goods of work through participation in research, so giving them more financial security is necessary to help them achieve these positive goods in their non-research activities. Appealing to an account by Gheaus and Herzog of the moral adequacy of work as a matter of distributive justice, justice requires labor markets and individual jobs to be organized in such a way to not only protect workers from harm, coercion, or exploitation (the negative aspects of work), but also to have “opportunities to realize a set of positive goods, namely excellence, social contribution, community and social recognition” (Malmqvist 2019, 15). Malmqvist contends though that because of the “unique passivity” of research participation, these healthy volunteers cannot achieve any of these positive goods by participating in research. He advocates instead for more financial security and more free time for healthy research volunteers so that they perhaps could achieve such positive goods in their “nonwork” activities. Financial security and predictability certainly seem desirable, and the world would be more just if a majority of workers had both. But is research participation really uniquely passive and different from other work activities? And does even relying on a partial income from research participation—at best a minor proportion of research participants—differ in some way from relying on a partial income from many other types of activities? As Fisher and colleagues have pointed out, those already disadvantaged by social and economic inequalities are overrepresented among healthy phase 1 volunteers (Walker et al. 2018).

Malmqvist’s insistence on unique passivity seems to me to ignore some of the more active aspects of research, as well as passive aspects of some other forms of work. The dictionary definition of passivity is a state of not reacting, not participating, being acted upon, or being inert. Research volunteers do indeed ingest experimental substances and contribute blood and bodily tissues, but they also actively participate by adhering to study requirements, and monitoring and reporting side effects, and they often engage in computer simulations, group discussions, surveys, exercise tests, and other activities that require active participation. Malmqvist says research participants are “paid to endure.” Depending on how one understands passivity, other jobs such as collecting tolls, modeling, working on an assembly line, being a doorman, and others could be described as more passive than active, jobs in which persons may be paid to endure. Furthermore, research volunteers arguably are making a social contribution by helping to generate knowledge useful for health. Many healthy volunteers say that contributing to medical science or helping future patients is part of their motivation; some paid research participants are health professional students or laboratory workers, who well understand the value of their contributions. Other workers, especially the self-employed or those employed in today’s growing gig economy, do not have predictable schedules or financial security, and are thus by Malmqvist’s analysis also unlikely able to realize the positive goods of work through their jobs. One solution might be to legislate better conditions for all or to reduce social and economic inequalities, but it is hard for me to understand why research participation stands apart.

I support the notion that research participants ought to be paid more often and more than they often are. I also support compensation for research-related injuries, good conditions for research participation, better protections for workers in all kinds of jobs, and universal health care coverage. These would increase fairness and respect, and be consistent with principles of justice. Unfortunately, I am not convinced that distinguishing coercion as subjection or research participation as uniquely passive helps move us in that direction or does much to resolve ongoing controversies about paying research participants.

Disclaimer

Views expressed are those of the author and do not necessarily reflect those of the Clinical Center, the National Institutes of Health, or the Department of Health and Human Services. ▪