High pay: When is it too much of a good thing?
High sum may actually increase vigilance
Large compensation to subjects for their participation in a study is considered a red flag by many IRBs, who worry that it could provide undue inducement to join a study without considering its risks.
But a recent study of participants' attitudes about compensation amounts suggests that large sums may actually do the opposite.1 Subjects appeared to see them as a signal that they should examine the risks of a study more intently, says Cynthia E. Cryder, PhD, MS, an assistant professor of marketing at Washington University in St. Louis, MO.
"People are more vigilant when they are offered these high amounts of compensation," Cryder says. "And in one of our experiments, when we told people that their compensation amount was linked to the amount of risk in study, people were even more vigilant."
Cryder, who studies decision making, teamed with specialists in research ethics and decisional research and with a clinical researcher to look at how people infer information from incentives that are offered to them in a study.
They used an online survey to ask participants to evaluate a "health and cognition study" involving the use of transcranial magnetic stimulation (TMS), a procedure that uses magnetic fields to stimulate nerve cells in the brain. TMS was chosen because it was more intrusive than many common procedures and was newer, with more uncertainty about risks.
One group asked to evaluate the study was told the compensation for participation was $25, while another group was told it would be $1,000. After reading about the study, participants were asked to rate the riskiness, first by using a Likert scale (1-7, with 1 being "not at all risky" and 7 being "very risky") and then by comparing the relative risk of the procedure to other activities such as flying in a plane, talking on a cell phone or obtaining a body piercing.
Participants rated the riskiness of the TMS procedure to be significantly higher when they believed the compensation was $1,000 than when it was presented as $25.
In a second experiment, participants actually believed they were being recruited for either the TMS study or for a second study that would require a simple blood draw, Cryder says. For each type of procedure, participants were offered $25, $100, or $1,000 as compensation.
After looking at a Web page showing the main study details, participants could navigate through other pages explaining the study further, including details of risks, contraindications and side effects. At any point in the session, the participant could stop viewing study details by clicking a button marked "make a decision about participating."
For both the TMS study and the blood draw study, participants clicked more pages for information and spent more time viewing study information when the compensation amounts were larger. In the case of the TMS study, for example, participants spent 3½ minutes viewing information when compensation was $1,000, compared to one minute when it was only $25.
Higher compensation specifically increased the likelihood that participants would read the information about contraindications.
While participants viewed more information in the TMS study than in the blood draw study, they were more likely to agree to participate in the blood draw, which Cryder and her colleagues believes shows increased vigilance because of the higher compensation, as opposed to simply increased interest.
In a third experiment, which only looked at the TMS study, participants were told that the compensation amount (either $25 or $1,000) was "appropriate" for the risk level of the study. In that experiment, participants were even more likely to see the higher compensated study as more risky.
Cryder says it's significant that this result held across socioeconomic lines. Two populations were recruited, one from the Web site of the New York Times science blog and one from various Pittsburgh neighborhoods using a mobile lab with Internet access.
"One of the biggest things we were looking for was (the effect of) socioeconomic status," Cryder says. "One of the big concerns with high levels of compensation is that it is especially dangerous for poor people."
But she says the results showed that both groups behaved similarly in their assessments of the risks of the studies.
Cryder says IRBs can glean two important points from the study:
-Don't worry so much about high compensation. "I don't know if IRBs should encourage high compensation, necessarily, but they should be less worried about it, for sure," Cryder says. "Because people are more vigilant when they are offered these high amounts of compensation."
-Subjects may assume that low compensation equals low risk, even if that's not necessarily true.
"If IRBs allow compensation to [equal] risk, they should let participants know," Cryder says. "If not, they should also let participants know that the amount of money they are being offered is not related to the risk in the study. Because participants naturally assume otherwise, that's what our study showed."
Despite her study's findings, Cryder doesn't think the issue of whether to allow high compensation for research participation will be resolved anytime soon.
"There are still differences of opinion among the experts, and I'm sure there are differences of opinion among IRBs," she says. "I think it is controversial and I think it will continue to be controversial."
- Cryder CE, London AJ, Volpp, KG, et al. Informative inducement: Study payment as a signal of risk. Soc Sci Med 2009 (Epub).