Much of the academic medicine community’s confusion over institutional review board (IRB) composition and representation stems from the regulations themselves, which date back to 1981, when the U.S. Department of Health and Human Services and Food and Drug Administration (FDA) aligned their regulations.1,2 The intent of these regulations is twofold—to ensure that IRBs are properly composed (i.e., that the makeup of members conforms to regulatory requirements) to review research proposals, and to garner respect from the community for the decisions that they make. Specifically, the regulations stipulate:
Each IRB shall have at least five members, with varying backgrounds to promote complete and adequate review of research activities commonly conducted by the institution. The IRB shall be sufficiently qualified through the experience and expertise of its members, and the diversity of the members, including consideration of race, gender, and cultural backgrounds and sensitivity to such issues as community attitudes, to promote respect for its advice and counsel in safeguarding the rights and welfare of human subjects.1,2
Defining Nonscientific and Nonaffiliated IRB Members
Klitzman’s research, presented in this issue of Academic Medicine, focuses on the role of the nonscientific and nonaffiliated members who serve on IRBs.3 As he points out, the regulations state that the IRB must “include at least one member whose primary concerns are in nonscientific areas” and “at least one member who is not otherwise affiliated with the institution.”1 Furthermore, and relevant to this discussion, the regulations state that the IRB must “include at least one member whose primary concerns are in scientific areas.”1 However, the regulations do not require that IRB members be classified as scientists and nonscientists. They also do not specify what type of training or experience qualifies one’s primary concerns as scientific or nonscientific.
Compounding the confusion about who represents which concerns on the IRB is the fact that the FDA and the Office for Human Research Protections (OHRP) have their own interpretations of the regulations, which are not always consistent with the everyday practice of scientific and nonscientific disciplines.
The FDA and OHRP have interpreted the regulations, for example, to mean that all physicians have concerns that are primarily scientific. This determination ignores the reality that there is wide variation among medical schools in providing scientific training for physician IRB members. Although a few schools offer substantial training in the scientific method, most offer little or insufficient training.4 In addition, under the FDA and OHRP classifications, all physicians are scientists even if they do not practice medicine and if they, themselves, do not perceive their concerns as primarily scientific. The same holds true for nurses. Yet, certain types of doctors, including those who hold PhDs and are specifically trained in the scientific method, are sometimes considered nonscientists by the FDA and OHRP.
Similar confusion surrounds the selection of nonaffiliated members. In this case, we should first consider the intent of the requirement that an IRB have one member who is otherwise unaffiliated with the institution. The purpose of this requirement is to prevent IRBs from inadvertently engaging in “groupthink.” The rationale—a strong one—is that IRBs composed entirely of like individuals from a single institution would be more inclined to represent the institution’s interests and less likely to take a more critical, outsider’s view toward proposed research. Here, again, the regulations do not stipulate that nonaffiliated members have any particular skills, expertise, or experience. As a result, as Klitzman found, nonaffiliated members may and do represent either scientific or nonscientific concerns.
The same may be said for nonscientific members. Although nonscientists do not have concerns that are primarily scientific, they are not required to represent the perspective of the human subjects or the communities to which the human subjects belong. Nonscientific members often work in other areas of the institution, such as the budget or human resources departments, with no knowledge of what is required of human research subjects in a study.
The term community member, however, gets closer to the intention of the regulations—that an IRB have expertise and experience in community attitudes. However, the term can be misinterpreted to mean someone who represents a specific community of research subjects, such as a risk group like injection drug users, a disease group like patients with alpha-1 antitrypsin deficiency, or an ethnic group like Hmong.
Recognizing the Role of the Public
Currently, membership on most IRBs reflects the increased emphasis on the role that the public plays in ensuring that research is ethically sound. In a 2001 report, “Preserving Public Trust: Accreditation and Human Research Participant Protection Programs,” the Institute of Medicine (IOM) maintained that, for research to be ethically justified, it must have scientific merit and be acceptable to human subjects and to the communities those subjects represent.5 The IOM also recommended that the operations of IRBs directly involve research participants as one means of preventing human subjects and their communities from being harmed or offended by the results of research studies. Because these participants could have the greatest influence if they were a part of the IRB review process, serving on an IRB is the natural vehicle for public participation. But to what extent should the public be involved? And how should representatives of the public be selected?
In an attempt to answer these and other questions, over the past decade, several organizations have tried to establish guidelines for appropriate representation of individuals other than scientists on IRBs. The National Bioethics Advisory Commission called for 25% of IRB members to represent the perspectives of the human subject, nonscientific, and nonaffiliated members—differentiating the three type of members and the skills, expertise, and perspective each brings to the IRB.6 The Association for the Accreditation of Human Research Protection Programs (AAHRPP) was even more specific. They required that, in addition to fulfilling the regulatory requirements for composition and quorum, IRBs must have members who represent the general perspective of the research participants, and these individuals must attend meetings.7
Currently, an IRB can be in full compliance with the federal regulations1 even if it does not have a single member who represents the concerns of human subjects or reviews proposed research studies from the subjects’ perspective. However, that IRB would not be in compliance with AAHRPP standards. Instead, AAHRPP would view it as lacking some of the expertise required to make ethical decisions. And, the IRB would be out of step with the current practice of most IRBs.
Scientific expertise is necessary but not sufficient to assess research protocols. Just as essential is understanding the views of research subjects (i.e., having served as a subject in a study or working in a profession, such as law or social work, which serves populations likely to participate in research). In fact, although there are other mechanisms in place to ensure that a proposed study is scientifically sound, there are no other means to assess whether the study is justifiable from an ethical standpoint. The IRB’s responsibility, then, is to recognize that the two perspectives are intertwined and to find the right balance between scientific expertise and public input. Appropriate representation of the public on the IRB helps achieve that balance.
Recommendations for the Future
Klitzman’s research looks at the roles of nonaffiliated and nonscientific IRB members by relying on in-depth interviews primarily of IRB chairs, directors, and administrators—not with nonaffiliated and nonscientific IRB members. Interviewees referred to nonaffiliated and nonscientific IRB members as community members and were confused both about the roles of these members on the IRB and their contributions to the review process because they traditionally lack scientific skills. Although Klitzman’s findings shed light on some of the challenges facing IRBs regarding composition, they are limited in revealing how well IRB members understand their own roles and those of their fellow members. To learn more about these perspectives, future research should include questions for scientific members about how they perceive their own roles on the IRB. Researchers also should interview nonaffiliated and nonscientific members to determine (1) if they, too, are confused about their roles, and (2) if that confusion extends to their perception of scientific members.
In our view, the structure of Klitzman’s research is almost as revealing as the findings. By interviewing primarily nonscientific and nonaffiliated members, Klitzman followed the tradition of labeling IRB members (as scientific, nonaffiliated, etc.) instead of focusing on what they, as individuals, can contribute to the research review process. We recommend moving away from this system of labeling members and returning to the intent of the federal regulations—ensuring that the composition of the IRB results in an ethical review of the proposed research.
As already noted, the primary goal of the membership requirements is to ensure that proposed research has scientific merit and is ethically sound. Every IRB appointment, and every member’s qualifications, should be evaluated in light of this goal and the requirement that the IRB meet it. For example, if an individual’s primary concerns are in scientific areas, identify the individual as such, and do the same for those whose primary concerns are nonscientific. If the IRB needs individuals whose primary concerns are in nonscientific areas to represent the perspectives of human subjects, then it should recruit current or former research subjects for these positions. Similarly, if the IRB needs to have a certain population represented, such as parents of chronically ill children, then it should select such an individual.
Recruiting appropriate members is just the first step, though. Training and support for all members—those whose primary concerns are in scientific areas, those who are not affiliated with the institution, those whose primary concerns are in nonscientific areas, those who represent the general perspectives of the community, and those who represent the perspectives of research subjects—are essential to provide each with the necessary background and confidence to hold his or her own during IRB discussions. In addition, the roles of all IRB members should be clearly defined, and their responsibilities should be comparable. It is unacceptable, for example, to limit the duties of members who represent the perspectives of subjects to reviewing consent documents. It is equally unacceptable to preclude members whose concerns are primarily scientific from reviewing consent documents on the grounds that scientists do not share the layperson’s perspective. The intent of having multiple perspectives represented is not to parcel out duties according to qualifications but to have all members benefit from diverse viewpoints. Scientists have something to contribute in reviewing consent documents, just as nonscientists add value to reviewing the study methods and protocols.
We should not allow prejudices and assumptions about the different types of IRB members to undermine the review process. The expectation should be that all IRB members contribute to the research review, and the academic medicine environment should foster this expectation. Moreover, it is the duty of the IRB chair and IRB staff to ensure that all members, particularly those without scientific training, feel welcomed and respected and that their opinions are valued. Such an environment would help reduce the tensions that Klitzman notes among IRB members and dispel the feeling, among members whose concerns are primarily in nonscientific areas, that they are unempowered. Even more important, such an environment would strengthen both the review process and the quality of the research.
Other disclosures: None.
Ethical approval: Not applicable.
Disclaimer: Dr. Speers is president and CEO of the Association for the Accreditation of Human Research Protection Programs.