Executive Director's Column

The Problem with IRBs

A frenzy is building over the behavior of many Institutional Review Boards (IRBs).

By Steven Breckler, PhD

A frenzy is building over the behavior of many Institutional Review Boards (IRBs). Increasingly, we hear horror stories about IRBs that are imposing incredible burdens on researchers, creating bureaucratic nightmares, and otherwise hindering the progress of research.

Many people ask what APA is doing about this. The Science Directorate already supports an office for research ethics and regulation (http://www.apa.org/science/research.html). It is an area that we have targeted for growth at APA, mainly in response to concerns in the research community that many IRBs are creating inappropriate difficulty for researchers.

One area in which APA is devoting attention is developing guidance for researchers and for IRBs. For example, the federal regulations specify that expedited review procedures apply for certain kinds of research involving no more than "minimal risk." The problem, of course, is that IRBs interpret "minimal risk" in different ways -- there is no uniformity or strong guidance on this. APA convened a workshop last spring, bringing together experts from research ethics, IRBs, and federal agencies to build some consensus on how to guide IRBs in defining "minimal risk." The result will be a procedure that IRBs can use in making "minimal risk" determinations. This particular problem is only one aspect of the current concern with IRBs, and clearly similar and continuing efforts will be needed to make the system work better.

On November 17, the Center for Advanced Study at the University of Illinois released a White Paper on Improving the System for Protecting Human Subjects: Counteracting IRB "Mission Creep". The White Paper (available at www.cas.uiuc.edu) summarizes the work of a group of ethics and IRB experts, and offers several recommendations for improving the work of IRBs. One of the problems, according to the report, is that many IRBs are taking on responsibilities that interfere with their main mission of protecting research participants.

For me, the most important recommendation of the report is a call for empirical research on the problem - we know very little about how IRBs around the country operate, what they demand, what procedures they classify as minimal risk, and so on. As scientists, we need to be approaching the problem systematically and empirically. We hear sensational stories, but we need to be careful to ground the vivid, attention-grabbing examples against the true base rates. Plenty of evidence suggests that IRBs have crept into areas where they were not intended to go. But we really have little reliable data on the extent or origin of the problem. For example, some have commented that the problem is not with the federal regulations, but rather with the interpretation and application of those regulations. We need to understand the nature of the problem, because some "solutions" may cause even bigger problems.

It is quite remarkable how little we really know about the structure and function of IRBs. APA's own Gerald Koocher has done some important work in this area, and he and collaborator Patricia Keith-Spiegel developed a tool for assessing how investigators perceive IRB characteristics (available at www.ethicsresearch.com). The so-called IRB RAT allows investigators to rate their IRBs on such dimensions as procedural justice, interpersonal justice, impartiality, and competence. With the accumulation of data, we will be in a much stronger position to take corrective action.

Some point the finger of blame at University administrations, who are accused of using IRBs to protect their institutions from lawsuits and bad press. If this is the case, it represents a major failure of Universities to support and enable the research of their faculty. Everyone wants to make sure that human participants in research are treated with respect, beneficence, and justice. When they work as intended, IRBs help researchers achieve this goal. When they get saddled with other responsibilities, IRBs can no longer be ensured of achieving the goal, and they impose burdens that get in the way of research progress.
For the social and behavioral sciences, Institutional Review Boards represent an important infrastructure. We all need to play a part in making them work, and work well. We need to take ownership of IRBs, and we need to make University administrations understand that the success of their faculty depend on well-functioning IRBs.