"Members of the same club": challenges and decisions faced by US IRBs in identifying and managing conflicts of interest.

unlabelledConflicts of interest (COIs) in research have received increasing attention, but many questions arise about how Institutional Review Boards (IRBs) view and approach these.

methodsI conducted in-depth interviews of 2 hours each with 46 US IRB chairs, administrators, and members, exploring COI and other issues related to research integrity. I contacted leaders of 60 IRBs (every fourth one among the top 240 institutions by NIH funding), and interviewed IRB leaders from 34 of these institutions (response rate = 55%). Data were analyzed using standard qualitative methods, informed by Grounded Theory.

resultsIRBs confront financial and non-financial COIs of PIs, institutions, and IRBs themselves. IRB members may seek to help, or compete with, principal investigators (PIs). Non-financial COI also often appear to be "indirect financial" conflicts based on gain (or loss) not to oneself, but to one's colleagues or larger institution. IRBs faced challenges identifying and managing these COI, and often felt that they could be more effective. IRBs' management of their own potential COI vary, and conflicted members may observe, participate, and/or vote in discussions. Individual IRB members frequently judge for themselves whether to recuse themselves. Challenges arise in addressing these issues, since institutions and PIs need funding, financial information is considered confidential, and COI can be unconscious.

conclusionsThis study, the first to explore qualitatively how IRBs confront COIs and probe how IRBs confront non-financial COIs, suggests that IRBs face several types of financial and non-financial COIs, involving themselves, PIs, and institutions, and respond varyingly. These data have critical implications for practice and policy. Disclosure of indirect and non-financial COIs to subjects may not be feasible, partly since IRBs, not PIs, are conflicted. Needs exist to consider guidelines and clarifications concerning when and how, in protocol reviews, IRB members should recuse themselves from participating, observing, and/or voting.