Allergy medicines may make it more difficult to breath for some people. Erectile dysfunction drugs can cause loss of vision or a stuffy nose. But don't worry about the stuffy nose. You can just take an allergy medication for that. Sounds depressing. But there's a medication for that too. And that can be a problem because anti-depressants may lead some people to have increased levels of anxiety and thoughts of suicide.

How in the world did researchers get people to volunteer to be guinea pigs for studies of these drugs before they were approved? More important, how is it that researchers were allowed to expose people to these potentially harmful chemicals? It is this latter question that Laura Stark discusses in her book Behind Closed Doors: IRBs and the Making of Ethical Research. As a biomedical researcher who has had research proposals reviewed by Institutional Review Boards (IRBs) at several public and private institutions over the last 20 years, I certainly understand how frustrating the process can be. On the other hand, as the current chair of my college's IRB, I am now in the unenviable position of having to answer the myriad emails I receive each week from frustrated researchers. So it was with great interest that I read Stark's book, hoping to gain some additional perspective on the process that I could use to improve the workings of my IRB.

Stark provides a historical perspective on how our current system of group review was first conceived in the early years of the National Institutes of Health (NIH). Researchers, many of them physicians, believed they didn't need anyone else to weigh in on how a study should be conducted or a patient treated. But many of the study participants weren't patients. Many were healthy volunteers used to establish "normal" responses. So medical ethics did not always apply. Stark makes clear that many of these physician-researchers were pulled kicking and screaming into the new system—a system in which a group of their peers needed to give their approval before studies could be conducted. This new system was not born from moral obligation but rather from the fear of litigation. Moreover, as the NIH began to fund more research outside the NIH campus, at universities and hospitals, political maneuvering helped solidify the model of group review as normative across the country, and supposedly further insulated NIH from litigation. Stark's historical review ends with the creation of the National Research Act in 1974, which was birthed out of the civil unrest and increased social consciousness of the 1960s. This legislation mandated a priori review of research involving humans by institutional review boards. Stark's analysis makes clear that our current institutionalized review procedures are the offspring of a 20-year-old maneuver intended to shield the NIH and those researchers working on the NIH campus from litigation.

My greatest disappointment with Stark's analysis is that she uses a very broad brush to paint a picture of IRBs in action. For example, she asserts that while everyone on the board is an expert, not everyone's expertise is equally applied. But this criticism misses the point. While everyone brings expertise in some area(s) to the meeting, not everyone's expertise can be brought to bear during the review of a given study. For example, my expertise lies in biomedical research. I have little experience in the type of ethnographic research Stark uses. Thus I might rely on members of our board from psychology and political science, for example, to help me understand a particular project's design and methods—and perhaps assuage some of my ill-informed concerns.

It was also disappointing to me that little space was given to how investigators interact with the IRB and their responsibility within the process. For example, Stark ends the first section of the book by stating that IRBs "suggest (read require) changes that a researcher could make to the proposal that would result in the board's approval." This implies that communication is one-directional, that IRBs mandate and investigators comply, that the IRB exerts all the power in this relationship. But Stark's account doesn't fit with my experience as a researcher and IRB member. While our IRB does indeed require changes to protocols (most often with regard to the wording of the informed consent document), we more commonly ask for clarification or offer suggestions that we feel would improve the overall study design or data analysis plan. As a researcher I have often responded to IRB suggestions, not by changing my study, but by informing the IRB why my study design is adequate or even preferable to their suggested change. My experience has been that the principal investigator plays a pivotal role in IRB deliberations. Such an approach, where researcher and IRB work together toward the common goal of excellent ethical research, results in better proposals and shorter approval times, as opposed to the extended back-and-forth and ill will engendered by an adversarial relationship.

Stark provides some insightful comments on the use of documents to remove the focus from any one individual board member, allowing the IRB to speak with one voice regarding deliberations. This could be seen as resulting in a loss of personal accountability between IRB members and researchers. The advantage, as Stark points out, is that it "liberates IRB members to advocate more incendiary actions than they would otherwise have suggested, since individual members cannot be held accountable." Thus, while I grudgingly fill out and file the mandated paperwork, it is these mandated processes and the accompanying paperwork that allow for all perspectives and concerns to be heard during the review process. This is the strength of the current IRB review process: that behind closed doors all voices are heard, no views are silenced, and in the end the most persuasive arguments usually win the day, leading to better, more ethical research.

Brian Hunt is associate professor of Applied Health Science at Wheaton College and chair of its IRB

Copyright © 2012 Books & Culture. Click for reprint information.