Taking stock of financial disclosures: Most want to know, won't drop out

More express negative views of stock interests held by PIs

Knowing that an investigator has a financial interest in a clinical trial doesn't substantially affect people's willingness to participate, according to a study that surveyed 470 people about their reaction to informed consent documents that detailed such interests. At the same time, participants expressed stronger concerns about situations in which an investigator has an equity, or stock, interest in the product being tested — so much so that a few immediately said they would refuse to participate in such a study.

"There were a small number of people, 5%, who just said, 'Read no further, I wouldn't participate, because that equity relationship sounds a little bit greasy to me,'" says lead author Kevin Weinfurt, PhD, associate professor of psychiatry and behavioral sciences at Duke University School of Medicine.

The study, published in a recent issue of the American Heart Journal, is the latest in a series of studies that are part of a five-year, $3 million project known as the Conflict of Interest Notification Study (COINS).1 The project's purpose is to provide guidance about how best to disclose financial conflicts of interest to potential research participants.

"Even though many professional groups and sets of guidelines call for the disclosure of financial interests in research, we really don't know who should do it, how to do it, when to do it, where to do it and what effects it's going to have on the research enterprise or on potential research participants," says Jeremy Sugarman, MD, MPH, MA, Harvey M. Meyerhoff Professor of Bioethics and Medicine at Johns Hopkins University, Baltimore, MD. Sugarman heads up the COINS project.

Model consent language

Previously, the COINS team has conducted focus groups with patients to discuss conflicts of interest and developed model language to describe various kinds of conflicts to patients. In a study published earlier this year, that model language was used in a study of diabetes and asthma patients. The Internet-based study gave patients informed consent documents for hypothetical clinical trials, with different groups reading about different types of financial conflicts, Sugarman says.

One group read a document explaining that the investigator would receive a per-capita payment for enrolling patients to offset the costs of the study. Another group read a document stating that the investigator held stock in the company whose product was being studied. In each case, there was language saying that an IRB had reviewed the financial relationship and determined that it did not pose a risk to participants.

Sugarman says people assigned to receive the document about stock ownership were less likely to participate and less trusting in research in general than those who received the other documents. But he says questions were raised about the way the study was conducted.

"One of the critiques we had was 'You did this as an Internet study, but real research doesn't happen that way. Real research happens by way of someone talking to someone about the research,'" Sugarman says.

"So we repeated the study, but this time it was cardiology patients and we did it in a way that people actually might get enrolled in research," he says. "They got the informed consent document and reviewed it and then the research coordinator reviewed the essential points with them and asked if they thought they would participate in research."

Spontaneous responses

Again, the participants were more troubled by the equity or stock interest scenario. There were three times as many clearly negative comments in the equity group as in the per capita payments group. A total of 11 patients spontaneously said they would not participate in the study — 10 of them from the equity group.

"The thing that was really revealing to us in this study was for the first time, we were able to find out what people's spontaneous reactions were to this information during the consent process," Weinfurt says. "We tried our best to do a full consent process with these patients and record everything they said. We were able to say exactly how many people expressed negative reactions to the disclosures."

But overall, in both disclosure groups, patients were not substantially less likely to participate in the hypothetical study than a third group that received no financial disclosure at all. In fact, Weinfurt says, a far bigger influence on patients' willingness to participate in the study was the pre-existing level of trust that they had in research and researchers.

"We measured that before we even took them through the simulated consent process," he says. "That sort of basic trust they had — their trust in the specific investigator, the institution — was a very consistent predictor of their willingness to participate."

IRBs: Determine risks

What does this information mean for IRBs? Weinfurt says the results clearly showed that participants wanted to have the financial disclosure, even if it didn't dissuade them from enrolling.

"A lot of them say they want to know these things and they would feel angry if something happened and they found out later that this information was withheld from them," he says.

But Weinfurt says that disclosure to potential participants doesn't absolve the IRB from determining for itself the risks the conflicts pose. For one thing, conversations with participants have shown that even with the informed consent language, some patients still don't completely understand the complex financial relationships involved, he says.

"That's why in our recommended consent language, we have language saying 'This relationship has been reviewed by the conflict of interest committee and the institutional review board and judged to present little risk to you,'" he says.

"The important thing there is that the primary risk determination should be made by the institution and that our work has shown that subjects are not in the greatest position to make that risk determination."

Weinfurt says he's heard from IRB chairs that they're using the model consent language the COINS group recommends since it was published in a previous article. "We're very gratified by that," he says.

Sugarman says that as the COINS project draws to a close, his group is planning a policy workshop in Washington, DC, gathering IRBs and other stakeholders to help shape policy guidance on disclosing conflicts of interest.

Reference

  1. Weinfurt KP, Hall MA, Friedman JY, et al. Effects of disclosing financial interests on participation in medical research: A randomized vignette trial. Am Heart J 2008; 156(4):689-97.