Penn Med researchers find people lie to participate in paid studies

Researchers from the Perelman School of Medicine have found that paying people to participate in research studies can lead to more people lying about their eligibility, leading to misleading results. The study has raised questions about the practices of several research groups at Penn that pay their subjects.

The study was led by Holly Fernandez Lynch, a professor of Medical Ethics and Health Policy in the Perelman School of Medicine, and published in JAMA Network Open in January 2019. Lynch said lying in studies can not only lead to misleading results, but also create health risks for other participants.

“It could be dangerous for people to enroll in a study, for example, with a criterion of non-smokers because of some drug interactions studied in the research would be dangerous for smokers,” she said. 

To test whether people lie, the Penn researchers prepared an online survey about attitudes toward vaccines and gave a $5 to $20 reward to participants who completed it. Participants in the experimental group were only eligible to take the survey if they reported that they received a flu shot in the past three months.

Control group participants were also asked whether they had received a flu shot, but if they reported that they had not, they were still eligible to take part in the study. The researchers found that 23 percent more people in the experimental group said they had received a flu shot compared to the control group. This large difference suggested that some people in the experimental group were lying so they could receive money for completing the survey. 

Lynch was joined by Medical Ethics Division Chief Steven Joffe, Biostatistics professor Dawei Xie, and Medical Ethics and Health Policy professors Emily Largent and Harsha Thirumurthy. 

There was no evidence, however, that higher payments were associated with higher frequency of deception. Joffe said the research team was surprised by this finding.

“We thought that maybe paying people five dollars would induce some people to deceive us, and then paying them ten dollars would get some of those that are not willing to deceive for five,” Joffe said. 

This research has implications for the many research groups at Penn that currently pay participants. The Wharton Behavioral Lab, a popular Penn research program that students often participate in, typically pays people $10 an hour.

Wharton Behavioral Lab Faculty Director and Marketing professor J. Wesley Hutchinson said the lab is not currently planning to change its policies in light of Lynch's research. 

"[Payment] could be an incentive for people to lie, but I don’t think it is a huge problem for the accuracy of the data that we collect for studies,” he said.

To discourage deception, Hutchinson added, the Behavioral Lab pays a "show-up fee" to participants who sign up for studies and are later found to be ineligible. This fee is equal to half the normal pay.

College sophomore Harrison Tandy said he regularly participates in Wharton Behavioral Lab surveys. While Tandy said he has never lied in these surveys, he added that “[WBL] needs to find a way to counteract lying and make sure that the data generated is useful.”

Lynch cautioned that while her study suggests monetary compensation is correlated with deception, it is also important for people to be compensated for their participation in the research, as participation has potential risks and takes time and effort.

To deal with false claims, Lynch suggested that researchers put more objective measures in place to determine whether participants are eligible rather than relying on self-reports. For example, instead of asking participants whether they smoke, researchers could fact-check with a blood test.

Hutchinson said that the Wharton Behavioral Lab does some fact-checking, such as verifying subjects' photo IDs and looking up names and birth dates to make sure people are not taking surveys more than once. 

Jason Karlawish, a professor in the Medical School who tests experimental drugs for Alzheimer's disease, said he looks at patients' medical records to determine clinical trial eligibility rather than relying on their reports. He added that he is adamant about paying his participants.

“Researchers have an obligation to pay people to be a research subject because they have to be rewarded for their work, times, and efforts,” Karlawish added.