Invited Peer Commentary: Min-Fu Tsan: Improving the Quality and Performance of Institutional Review Boards in the U.S. Through Performance Measurements.

Abstract

Dr Tsan’s invited commentary “Improving the quality and performance of institutional review boards in the U.S. through performance measurements” should be commended for two important reasons. First, the article is based on a review of the performance measurement data literature on IRBs, thus grounding the discussion in the empirical evidence. There is a welcome trend in the IRB literature to see a focus on empirical research on IRBs (when the question being asked is empirical in nature), led by JERHRE and continued by this article. I welcome work such as this, and I would hope our field would as well. Second, the author should be commended for focusing not merely on turn-around time, but on other measures of quality as well. For too long, the single measure of IRB quality has been turn-around time. While this is certainly important, and we ought not lose sight of it, seeking to reduce turn-around time whenever possible, it is not the sole measure of quality for IRBs. The Coast IRB scandal, in which a commercial IRB was exposed by the Government Accountability Office for approving fake, risky studies with fictitious PIs, is a clear reminder that turn-around time alone cannot account for the quality of an IRB (Dove, 2009). Regardless of turn-around time, Coast IRB was far from a quality IRB in part because it failed to meet some of the very quality recommendations made in this article. Yet, turn-around time is important, and not simply to approve studies faster. It matters, and is related to the IRB mission, as it facilitates adherence to the Belmont ethical principles. If IRBs do not work to improve the time it takes to review and approve studies, we fail to live up to respect for persons and justice. If we believe that prospective research participants have the right to decide for themselves whether or not to participate in research, as respect for persons requires, delays in the review and approval of research will mean that some individuals will miss that chance because the research was approved too late. So too for justice, in that if we believe that research requires a fair distribution of the benefits and burdens of research, delays will slow efforts to address healthcare and research inequalities through research. Yet, we must emphasize that the IRB is tasked with reviewing and approving research, not simply expediting to reduce turnaround time. That is an essential point made by the author of this article, focusing on how the quality and efficiency of IRBs can be improved. The importance of looking to measures of quality beyond mere turn-around time relates to the question of what the role of an IRB is, or what is its mission. While many would say it is to protect research participants, including OHRP, FDA, and the regulations that govern research, I have argued elsewhere that this view is short-sighted and over focuses on non-maleficence, perhaps at the expense of other Belmont principles such as respect for persons and justice (Spellecy & May, 2012). When an IRB focuses too narrowly on protection research participants, it may disapprove or place restrictions on research that ought to be left to participants to decide for themselves whether or not they are acceptable. So too, an IRB might disapprove research or make it so burdensome in the name of protecting vulnerable populations that research necessary to improve the lives and communities of vulnerable populations is not done, further exacerbating injustices. An IRB that seeks to facilitate ethical research, on the other hand, does not merely approve research, or simply facilitate research, but rather ensures that it meets the regulations and adheres to ethical principles. Tsan’s article is a step towards improving the quality and performance of IRBs and helping them better facilitate ethical research. The author’s suggestion, based on performance measurement data, that primary reviewers for fullboard protocols should assess each approval criterion, whether or not it has been satisfied, and note any suggested changes is an example. While it might seem burdensome and unnecessary to do so when the reviewer has no concerns with a given criterion for approval, implementing such changes could improve the quality and performance of IRBs in two ways. As the author notes, we don’t know if