- Shopping Bag ( 0 items )
ScienceBehind Closed Doors makes an important contribution to our understanding of IRBs and the ethical regulation of research.
— Charles W. Lidz
Although the subject of federally mandated Institutional Review Boards (IRBs) has been extensively debated, we actually do not know much about what takes place when they convene. The story of how IRBs work today is a story about their past as well as their present, and Behind Closed Doors is the first book to meld firsthand observations of IRB meetings with the history of how rules for the treatment of human subjects were formalized in the ...
Although the subject of federally mandated Institutional Review Boards (IRBs) has been extensively debated, we actually do not know much about what takes place when they convene. The story of how IRBs work today is a story about their past as well as their present, and Behind Closed Doors is the first book to meld firsthand observations of IRB meetings with the history of how rules for the treatment of human subjects were formalized in the United States in the decades after World War II.
Drawing on extensive archival sources, Laura Stark reconstructs the daily lives of scientists, lawyers, administrators, and research subjects working—and “warring”—on the campus of the National Institutes of Health, where they first wrote the rules for the treatment of human subjects. Stark argues that the model of group deliberation that gradually crystallized during this period reflected contemporary legal and medical conceptions of what it meant to be human, what political rights human subjects deserved, and which stakeholders were best suited to decide. She then explains how the historical contingencies that shaped rules for the treatment of human subjects in the postwar era guide decision making today—within hospitals, universities, health departments, and other institutions in the United States and across the globe. Meticulously researched and gracefully argued, Behind Closed Doors will be essential reading for sociologists and historians of science and medicine, as well as policy makers and IRB administrators.
— Charles W. Lidz
— Nathan Emmerich
— Norman M. Goldfarb
“Behind Closed Doors makes an important contribution to our understanding of IRBs and the ethical regulation of research.”
— Charles W. Lidz
“[T]his is one of the most important books concerned with the governance of research ethics, particularly in the social sciences, to have appeared in recent times. It deserves to be widely read by social scientists, applied ethicists who seek to comment on research ethics in the social and natural sciences and, perhaps most importantly, the academic and non-academic bureaucrats who are involved with the ethical governance of academic research.”
— Nathan Emmerich
“Stark’s book combines lively and entertaining writing about the historical development of IRBs with careful research on their current operation. All in all, her account provides a good resource for both lay and expert readers.”
The Imperative of Consensus
Most of the time board members were not confrontational. Their discussions of research proposals were brief, and their demeanor toward each other was civil. Occasionally, though, they got red-faced. Witness this exchange about reporting suspected child abuse, which unfolded during a late-autumn meeting at Greenly State:
JOHN (FACULTY, ANTHROPOLOGY): But, that's what worries me. You got two choices here. One is you stop doing this research or you set people up so you can catch them, which destroys research, too. And so you get a few people off the street that way. one time around and in the process you end research on this area.
EDWARD (FACULTY, ARCHITECTURE): Oh yeah, John, that's pretty cool. A friend of mine's child was killed /
JOHN: No, I /
EDWARD: Wait, wait, let me finish. You know, a friend of mine's child was killed because he was shook. Because he was shaken. You know, two years old and they found her dead. I don't want to go too far into this, but you know, I loved that person. The research be damned. You know, yeah, it's hard to do research, but if you suspect child abuse then you have to say it.
Consensus among group members is an achievement, not a foregone conclusion. As the exchange between John and Edward makes plain, board members disagreed with their colleagues at times—more placidly, in most cases, but nonetheless pointedly.
IRBs get their authority from administrative law. Open the Code of Federal Regulation, and you will read that the authority to approve, disapprove, or request modification of protocols rests with "the IRB." It is worth noticing who is missing from the federal regulations. There is no mention of individual IRB members, who have no power. This is not to say that individual opinions do not matter. Quite to the contrary: individual members are selected because of their particular knowledge in a specific area. However, because the regulations grant power to the expert body—not to individual experts—it is "the IRB" that has to produce a decision. In other words, administrative law empowers the declarative body to do things that a piece of it—say, edward—could not accomplish on its own. Like our human bodies, declarative bodies are more than the sum of their parts.
If each board member has an equal claim to expertise, then how do they reach a decision? one option would be to vote, which IRB members do. But IRB votes have an uncanny likeness to each other. Almost without exception, the outcomes that I observed were unanimous, and individual members never voted against the group. To signal their disagreement, board members in a few instances abstained from voting altogether. It was the most damning act of resistance they mustered.
The work of achieving unanimity happened before the vote—when members tuned their individual voices into a chorus that their audience (including researchers, study participants, and the university legal team) could hear as one voice. The appearance of unanimity among members consolidates these individuals into the regulatory actor that is empowered by the law. Consensus creates the impression of a strong, incontrovertible decision and erases the underlying disagreements of individuals who have no authority. IRB documents tend to report a consensual decision because individual members have no authority to act on their own. Thus the pervasiveness of group consensus and the strictly symbolic function of voting are themselves products of the social organization of expert bodies as laid out in federal regulations.
This begs the question: if everyone on the board is an expert with an equally valid claim to his opinion, then how do declarative bodies reach a consensus? one fundamental way in which individual IRB members built a group decision about how rules applied in specific cases was to persuade other members that they had the best backing for their opinion compared to colleagues on the board. The group deliberations that I observed were essentially referenda on who had the most salient knowledge and thus whose advice should be followed.
I use the concept of warranting to explain how IRB members reached consensus on group recommendations during their closed-door deliberations. Warrants are reasons or justifications that people give for their views. In the sections below, I describe three types of warrants that group members used during deliberations. This set of warrants is not exhaustive, to be sure, but I lay out for your inspection three of the most common and persuasive warrants. IRBs also had decision-making habits, that they used when they processed studies that felt familiar and routine (see chapter 2). However, when the proper course of action was either ambiguous or contested, board members reached consensus by deciding whose recommendation was based on the most relevant knowledge, which all members could endorse as the view of "the IRB."
What the warranting process shows at the broadest level is that the actual process of decision making among knowledge experts is quite different from its appearance in formal records. Meeting minutes essentially freeze a moment in time, in particular, the very end of deliberations. The minutes do not allow readers to see what took place beforehand to move all of the elements into position so that "the IRB" could render a decision. Understanding the differences between the process of decision making and the final product of a decision—the letters of judgment sent to researchers, for example—is especially important in studies of governance because IRB members, unlike elected officials, are appointed. Knowledge experts often do not have a clear, tangible set of constituents whose views they are called to articulate and who hold them accountable. Adding to this, declarative bodies often interpret regulations in physical locations that are difficult for most people to access. This chapter gives a chance to step inside one such place.
What Is Warranting?
Members of declarative bodies use warranting as a tool for reaching shared decisions. When warranting their statements, people are implicitly answering the question "Why?" or "On account of what?" When I analyzed my field notes and transcripts of IRB meetings, I coded as warrants the statements that IRB members made to account for their recommendations on how the group should proceed with a study.
Warrants have three distinctive features. First, although they are given verbally, warrants are often implicit in a speaker's language. People tend to use warrants when they are challenged or pushed to justify themselves. Because IRB meetings are organized around making choices, as opposed to talking for its own sake, board members have to account for their views quite frequently.
Second, warrants are context specific, which means that different types of justifications are persuasive in different settings. As sociologist Robert Wuthnow has written, people do not justify their views haphazardly, but "choose ideas that seem legitimate in relation to the norms present in their social context." Because of this, warrants "are likely to reflect the norms of which social structure is composed in different contexts." The implication is that the warrants used in IRB meetings tell us how group members understand their setting, by showing us what people are willing to say in a context that many people regard as a bureaucratic social space.
Finally, the success or failure of an individual's warrant is determined by his listeners and whether they accept or reject his justification. This means that an IRB member's warrant is convincing only if fellow board members treat it as such. The study of warrants, as a result, is a pragmatic approach to analyzing group deliberations.
In interviews, many board members described a division of labor among IRB members, both regarding the issues on which they could make judgments and regarding the topics on which colleagues should intervene. In general, they agreed that each member was valued for her or his recommendations in the area of her or his professional training. For example, Dr. Cain, a faculty member who was trained as a physiologist, described himself as being someone who judged, in particular, the quality of the science that investigators proposed. Compared to his colleagues on the Greenly IRB, he explained, "I'm looking at the value of the science whereas I think somebody like nathan, being an ethicist, he's really looking at more / He's not looking at the science. He's looking at the ethical value or implications of the work." In keeping with this division of labor, Dr. Cain cautioned that community members should not be undervalued, describing them as "some of the most important people there," because they took care of certain review tasks that other IRB members neglected. "They keep the scientists and the medical folks honest [because] their perspective is not to drive or improve the science," he explained, "but is really to protect people. I feel I have that obligation as well, but I think theirs is much more weighted in that direction, so I think they're very important." Like Dr. Cain, other board members described what can be thought of as role-specific norms of participation in IRB meetings during interviews. Still, it is important to note that Dr. Cain felt that he, too, could usefully weigh in on the evaluations that properly belonged to community members.
Community members also saw a close affinity between formally designated roles and appropriate types of recommendations. Reverend Quinn, a community member serving on the Sander IRB, felt that she had brought to the board "a lot of knowledge of what the patient's world was like.... In a general sense, I knew what their suffering was like, what their anxieties were." She also felt she had brought "a sense of the values of the community that I serve, the wider neighborhood, and my church." When I asked whether she could predict the concerns that her colleagues would raise, she tied her colleagues' concerns to their professional training:
A couple of the doctors I know look at things like blood chemistry, requirements for pharmaceuticals, and the synergistic effects of things, and I know they're really on the ball with that. And I know that for instance, Kitty, if there's an issue of nursing practice and logistics of how privacy is dealt with in the nursing setting, she's really on top of that stuff. You know, and when she speaks, she seems to be usually right on target with it. In that sense, I think these are people who are very vigilant about their area of expertise. So the whole works together pretty well.
Reverend Quinn's description of board members' complementary roles sounded much like Dr. Cain's account. Also like Dr. Cain, she reserved for herself some latitude to weigh in on issues outside of what she described as her role-specific task. She felt she could judge the quality of research, as she indicated when she lamented that she had seen proposals with "pretty questionable scientific merit be passed through." And by her description, she "cut her teeth" on chemotherapy protocols while serving on other IRBs, so she knew "a lot about the basics of drugs and the things you look for in drug protocols."
IRB members had a shared sense of their individual jobs on the board, as well as the jobs of their colleagues. At the same time, they reserved the right to make recommendations on issues that extended beyond their official area of specialization. It was not obvious to Reverend Quinn that she should limit herself to discussions about community values, Dr. Cain about scientific merit, Nathan about ethics, or Kitty about nursing practice. There was flexibility in board members' sense of the areas in which they and their colleagues could usefully contribute to deliberations. This flexibility was evident in board meetings, in which board members gave their views on topics that they were formally qualified to judge—and on other topics as well.
Claiming Expertise: Three Types of Warrants
The IRB members whom I observed reached consensus by siding with the colleague they thought had the best justification, or warrant, for his or her recommendation. Under conditions of uncertainty or dispute, in other words, board members arrived at decisions by persuading each other with what appeared to them to be good evidence for their views. The warrants that IRB members used took three forms, which I call matters of fact, private experience, and professional experience.
Matters of Fact
IRB members often warranted their views with "objective" findings, often in the form of numbers, such as statistics. Such seemingly objective evidence is authoritative because it is not associated with any one person and instead can take a material form, for example, in a diagram, table, or image. I call warrants based on seemingly disembodied, objective evidence matters of fact.
This kind of evidence offers promise for those hopeful about a democratic decision-making process. Scholars have shown how nonspecialists can harness the seemingly universal authority of objective evidence for their own purposes. This is possible, they argue, because objective evidence carries the persuasive power of expert knowledge and yet is accessible outside of expert communities. Sociologists Harry Collins and Robert evans have pointed out, for example, that today the Internet provides a way for people without formal training to access information traditionally sequestered among experts in a given field. They use the term "primary source knowledge" to describe information that comes from credentialed experts but is widely available to people without formal training in the field. The process through which nonspecialists make effective use of certified knowledge has been cast as a positive move toward the democratization of decision making about science.
What I mean to designate by matters of fact is the kind of knowledge that is shared but comes from specialists: the information that one could learn by studying academic books, reading scholarly journals, or hearing lectures in a given field. This feature of matters of fact made it possible for community members, such as Reverend Quinn, to use statistics to warrant their views. An example of this happened when board members reevaluated a previously approved vaccine trial because the investigator had reported a Serious Adverse event. (The U.S. Food and Drug Administration requires that researchers report to their IRB when a study participant unexpectedly experiences a problem, which may or may not have been caused by the studies.) A woman enrolled in the vaccine trial had gotten pregnant (against study protocol) and then had a miscarriage. The issue the board was considering was whether to make the investigator change the protocol so that all women enrolled in the study had to use a semipermanent form of birth control, such as a hormone injection, rather than a barrier method, such as diaphragms and condoms.
Reverend Quinn argued against more stringent birth-control requirements for women and warranted her opinion using a matter of fact:
REVEREND QUINN: Condoms are very effective. They fail in [5 to 15 percent of cases /
OWEN (EMERITUS, PHARMACOLOGY): If used!]
REVEREND QUINN: If used incorrectly. This patient / This subject reports that she used condoms and had a failure. It happens. It happens. And so, I wouldn't want to raise the bar and say it's not an effective enough birth control. I think (it's fine).
As I coded my field notes and meeting transcripts, I looked, in particular, for moments when IRB members asserted an opinion, and I then tracked how IRB members linked their opinions to accounts of why they held that view. In the exchange about the Serious Adverse event, Reverend Quinn offered her recommendation: the study should not change because condoms are effective. Then she substantiated it with a warrant: condom failure rates. It is not important whether the evidence that Reverend Quinn used is strictly accurate. What is important is the type of evidence she used—in this case, disembodied, seemingly objective quantitative measures, or matters of fact. The type of evidence she used is important to note because in the end the board acted on the view of Reverend Quinn, whose role on the board, by her own description, was to articulate the sensibilities of research subjects, not to weigh in on birth control statistics. Board members did not require the investigator to make women on the protocol use a semipermanent form of birth control.
Excerpted from Behind Closed Doors by LAURA STARK Copyright © 2012 by The University of Chicago. Excerpted by permission of The University of Chicago Press. All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
Introduction: Behind Closed Doors
Part 1. IRBs in Action
Chapter 1. Everyone’s an Expert? Warrants for Expertise
Chapter 2. Local Precedents
Chapter 3. Documents and Deliberations: An Anticipatory Perspective
Part 2. Setting IRBs in Motion in Cold War America
Chapter 4. An Ethics of Place
Chapter 5. The Many Forms of Consent
Chapter 6. Deflecting Responsibility
Conclusion: The Making of Ethical Research
Appendix: Ethnographic Methods