Type of Institution: Large Academic Medical Center
Type of IRB: Biomedical
Type of Case: Ethical/IRB Operational
Author: Chairman of the IRB
Subject: IRB Members and Personal Conflicts of Interest
From the Editors:
The ability of an IRB to provide effective review of research protocols that come before it depends in no small way upon the membership of the committee and the manner in which the members interpret and thus adhere to the Federal rules which govern IRB operations. All members of an IRB - affiliated and non-affiliated, scientific and non-scientific – bring their personal beliefs and biases to the table when reviewing research and oft times consider the merits of research, at least in part, through the lens of these beliefs. This case illustrates the operational issues that may arise when IRB members permit their strongly held personal beliefs to affect their objectivity in considering research protocols for approval. This case differs from the others in that the interesting and significant activities of the case occurred, not at the IRB meeting itself, but post hoc, in the IRB administrative offices, and in response to the occurrence at the IRB meeting.
This institution has a busy and well-recognized program in Reproductive Endocrinology and Infertility that frequently brings protocols before the IRB to test new techniques to enhance the efficiency of Assisted Reproductive Technology (ART) procedures such as In-Vitro Fertilization (IVF) and Intra-cytoplasmic Sperm Injection (ICSI). This study planned to assess a new device to select the best sperm for selection for use in ICSI/IVF. At a convened IRB meeting, this protocol was presented for consideration and a vote and of nine members of the IRB, eight were present thus making a quorum. After a thorough discussion of the protocol, those present at the meeting voted, four for approval, three opposed, and one member abstained. As is policy for this IRB, individuals who abstain are asked to explain their rationale for their vote which is then recorded in the meeting minutes. In this case, the abstaining member stated that use of ART to enhance fertility in infertile couples was against her religious beliefs and therefore she could not vote to approve such a study. However, she decided to abstain rather than vote against the study since, specific ART issues aside, the study did have merit by enhancing fertility in otherwise infertile couples for which she might otherwise have voted yea. Her abstention, then, was her approach to balancing her religious beliefs against the other merits of the research. Since the majority of those members present did not vote yea, the protocol was not approved.
The following considerations occurred among the senior IRB staffers at this institution, including the Chairs of the three IRB’s, as a consequence of the occurrence at the IRB meeting. The specific procedural question is whether the IRB member who abstained due to her strongly held religious beliefs should have been permitted to participate in discussing or voting on this matter at all, since her position on this protocol had been determined prior to hearing any discussion of the study. While seemingly straightforward, this question raises a variety of issues regarding IRB membership, IRB voting procedures, and ‘personal’ conflicts of interest.
- How is the current IRB voting mechanism affected by personal beliefs?
In the United States, IRB’s vote according to Robert’s Rules of Order which provides members with three standard procedural options each time they vote. Members may vote yea (yes), nay (no) or may abstain (neither yes nor no). There is, however, a fourth and important voting option which is for a member not to vote but rather to recuse oneself or be recused from the vote by the Chair.
The Common Rule assigns no particular significance to abstention and in fact, is not mentioned in the Common Rule. It’s clear that no research may be approved at a convened IRB meeting unless a majority of the membership is present (quorum), at least one member attending the meeting is a non-scientist and the study is approved by “...a majority of those members present at the meeting (45 CFR 46.108(b)).” All that matters is the number of votes in support of a motion in relation to the total number of voters present. An abstention, therefore, has the same procedural effect as a nay vote, even though those abstaining may wish to convey a neutral feeling rather than a negative one. Abstention could be characterized as, “the voting individual’s ambivalence about the measure or mild disapproval that does not rise to the level of active opposition.1” Abstention might also be used to reflect a member’s ignorance of the issue. Frequently, for example, IRB members abstain from voting on approving minutes from a previous meeting if they had not been present at that meeting. Robert’s Rules does acknowledge the existence of these kinds of voting rules based on the number of members present as in the case in the Common Rule, but characterizes these rules as undesirable for the very reason that an abstention counts as a negative vote, and denies members the right to maintain a neutral position by abstaining.
The only circumstance specifically addressed in the Common Rule in which an individual should recuse oneself or be recused by the Chair, is for a conflict of interest. The word “recuse” is not actually used in the Common Rule, but implied in 45 CFR 46.107:
“No IRB may have a member participate in the IRB’s initial or continuing review of any project in which the member has a conflict of interest, except to provide information requested by the IRB. 2"
Under the Common Rule, self-recusal does not specifically mean physically removing oneself from the meeting, although most IRB’s have that requirement. If recusal for conflict of interest does include leaving the room, however, there is the risk that the quorum could be lost and no vote on that sudy could be undertaken. While recusing an IRB member from a discussion or a vote due to a conflict of interest places the quorum at risk, it is nonetheless critical that the member physically leave the room as their continued presence may inappropriately influence the discussion or the vote nonverbally through body and facial language or even intimidation-by-presence. On the other hand, when the member leaves the room, there is now one fewer vote in opposition, but also one fewer vote in toto.
The Conflict of Interest Policy at this institution states the following:
“If a conflict is disclosed, when the study is reviewed, the member with the conflict is required to leave the meeting for the deliberation and voting. If assigned as a reviewer for a study in which the member has a conflict, the member is responsible for contacting the IRB Office so that the Chair may assign another reviewer.”
Clearly the COI policy demands recusal for any IRB member with a conflict regarding any study that comes before the IRB. In this case, the motion would have passed had the abstaining member been recused from the vote and excused from the room since that individual no longer would have counted towards the quorum and the voting makeup would have been changed to 4 in favor and 3 opposed. Her personal opinion, therefore, had a direct outcome on the vote and whether or not the research would be conducted at this institution.
- How do we expand the definition of ‘conflict of interest’ to include personal conflicts of interest? How do we observe the impact of personal conflicts of interest on day-to-day IRB operations?
Within the Common Rule, 45 CFR 46.107(e) refers to a “conflicting interest” without actually defining the term in the Definitions section, 45 CFR 46.102. OHRP guidance on conflict of interest is centered entirely on the issue of financial conflict of interest and provides no discussion of any ‘personal’ conflicts and their potential impact on IRB operations. Clearly the OHRP guidance is insufficient since the IRB must obviously deal with conflicts of interest at many more levels than purely financial. It is this missing category of personal conflict that is at issue in this case.
McDonald3 defines a conflict of interest as a situation where a person “ … has a private or personal interest sufficient to appear to influence the objective exercise of his or her official duties.” McDonald goes on to list 3 key elements:
- a private or personal interest
- the conflict of the private interest with an “official duty”
- an interference in a specific way with objective professional judgment.
Personal conflicts of interest can be subtle and unexpected, particularly when they exist among the scientists on the IRB. However, such personal conflicts, for scientists and non-scientists alike, may not be easily recognized by the Chair or other IRB members, as such discussions may simply be taken to represent vociferous protection of the rights of study subjects, when in fact they may be part of a personal and unrecognized agenda. Certainly the IRB Chair or any member who suspects such a covert bias should question the member regarding his/her ability to objectively perform the duties consistent with IRB membership. Institutional IRB conflict of interest policies generally enable the Chairman of the IRB to recuse any member from a discussion or vote whom the Chair believes has a real or apparent conflict of interest regarding a particular protocol. This may apply even if the conflict is ‘personal’ and not financial or related to one of the COI categories more commonly encountered and which are thus more readily mitigated. Conflict of interest policies are explicitly delineated within the SOP’s (Standard Operating Procedures) at each institution and may or may not include language that pertains to ‘personal’ conflicts of interest.
The COI policies at this institution define a conflict of interest as
“...a situation in which an individual (or someone in his/her immediate family) has a significant financial, professional or personal interest in the approval or outcome of a study and the interest could affect decisions related to their the design, conduct or reporting of the research or adversely affect the rights and welfare of research subjects.”
The impact in this case is obvious. Had the Chair been aware of this member’s strongly held personal beliefs regarding ART prior to presentation of this study to the committee, this individual would likely have been recused from both the discussion and the vote in order to mitigate the impact of her beliefs on the operations of the IRB at that meeting. The outcome, as noted previously, would have been a 4-3 vote in favor of study approval.
- How does the diversity of the IRB membership, as mandated by the Common Rule, potentially affect IRB operations? How do ‘personal’ conflicts of interest differ by the varying role of each IRB member?
The Common Rule provides guidelines for IRB membership in 45 CFR 46.107:
“…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.”
The importance of members’ expertise is further delineated in the Common Rule as it is explicit about mandating experts be part of the committee whenever studies are presented that involve vulnerable subjects such as fetuses, pregnant women, children and prisoners. IRB’s are advised to include members “...whose primary concerns are in scientific areas...” and at least one member who is a non-scientist. Further guidance speaks to including persons as IRB members affiliated with the institution as well as “...at least one member who is not...affiliated with the institution and who is not part of the immediate family of someone who is affiliated with the institution (the ‘community’ member).” This prior section segues directly into the Common Rule’s brief discussion of conflicts of interest by stating, without defining this in the Definitions section that, “No IRB may have a member participate in the IRB’s...review of any project in which the member has a conflicting interest...”
Clearly, one responsibility of each IRB member is to bring to the committee some form of expertise. ‘Expertise,’ however, is not something acquired or utilized in a vacuum, but rather is the net result of a person’s educational, employment, and research experiences, as well as one’s personal life experiences, beliefs and biases. IRB members’ review of and vote on each study also depends in no small way on the role each member plays on the panel. For example, in reviewing a protocol, an IRB member who is a physician and an assistant professor of medicine at a large academic medical center must balance his/her scientific ‘hat’ with his/her institutional ‘hat’ with his/her personal beliefs, but ultimately must conclude, in his/her opinion, how best to balance the protection of human subjects with the ability to conduct the research. In contrast, a community member who is a non-scientist but reviewing the same study, must balance his/her understanding of ‘community standards’ against his/her own personal beliefs. While each member of the panel is strongly encouraged to consider each study based solely on the merits of the research without regard to personal considerations, the impact of members’ ‘human-ness’ on their deliberations is unavoidable and largely dependent on the personal ethics of each member in their commitment to consider each study as objectively as possible. Each of these members, scientist as well as non-scientist, brings strongly held personal beliefs to the discussion, but must each either find a way to minimize the impact of these beliefs on their decision-making or must recuse themselves from deliberations. It is imperative that each member recognizes their biases and consciously examines the potential for these to impact their perception of the research before them.
Such personal biases, when not overt, may present differently, depending on the role of the member.
A scientific member’s strongly held personal beliefs may easily be masked by the member’s uncompromising or perhaps unreasonable stance on a particular study. He/she may seem to unduly highlight concerns over the power calculations, holes in the hypotheses, adverse event reporting or other technical issues that often result in non-scientist members deferring to the opinions of the scientist. The possible protections in such a case are the awareness by the Chair and by the other IRB members of the possibility that even a scientific member may espouse strong personal biases that may not be overt and may influence his/her decision-making, but be masked by a member’s approach to study deliberations. Additional protections are afforded by the structure of the IRB in that it is likely that a number of other voting members on the IRB will not share the same biases. There is also an expectation, by dint of serving on an IRB, that each member is committed to reviewing research based solely on the objective merits of the study and to sublimate his/her strongly held personal beliefs accordingly. Alternatively, should a scientific member be unable to successfully accomplish this, that member should discuss his/her concerns with the Chair and be recused from the discussion and the vote.
One section of the Common Rule language about IRB membership relates to the mandate to attain diversity of members of the IRB,
“… including consideration of race, gender, and cultural backgrounds and sensitivity to such issues as community attitudes…”
There is, however, no discussion concerning which members have responsibility for considering “...cultural backgrounds...” and “...community attitudes...” ”Community" in this context is a broad concept which embraces gender, race, religion, attitudes towards science, attitudes towards privacy, and attitudes towards research. The Common Rule does not delegate community considerations specifically to any particular constituency within the IRB, and particularly not to any individual member, thus theoretically making the IRB collectively responsible to address “...cultural background...” and “...community attitudes...” That said, IRB’s expect community members to deliberate and vote according to their knowledge of and connection to the community. As a consequence, there is a tendency in many IRB’s to consider the non-affiliated (community) members of the panel as sole representatives of the community and thus keepers of the cultural and community yardsticks. This creates two inseparable problems related to this case.
- It may be true that non-affiliated members possess some special knowledge or expertise in some aspect of their community, making their presence on the IRB invaluable. But when such members contribute to the IRB’s deliberations of a study, are they representing beliefs generally held in the community or are they actually speaking as individuals and contributing based on their own personal beliefs and biases?
- The tendency to value community members as arbiters of community standards may lead IRB’s to unduly defer to these members in study deliberations. When community members are truly knowledgeable concerning community norms and contribute such information to IRB discussions, they are providing an important service to the IRB and the community. However, if the community member, as in this case, voted according to some strongly held personal beliefs not representative of the community and other members of the IRB deferred to her opinion in some way, the IRB failed in its duties to manage a conflict of interest, albeit a COI somewhat outside that usually considered by IRB’s.
One way for the IRB to protect against a personal agenda being masked as a community norm is to rely on all IRB members to bear equally the cultural and community responsibility and not defer to the non-affiliated members without due consideration. In so doing, the IRB can establish whether or not a particular attitude is commonly shared among all members of the panel and so create an IRB-based ‘community norm’ while at the same time considering the input from the community members. The litmus test for community members should be whether or not their comments represent the community norms, and if so, that person’s opinions should be considered in the IRB deliberations. However, if one’s opinions clearly represent strongly held personal beliefs that are widely divergent from community norms or if that individual states an inability to contribute objectively to the discussion due to such beliefs, then the final protection for the IRB is to recuse that individual from further discussion and from the vote.
- How should IRB’s manage members’ personal conflicts of interest that arise from their religious beliefs, ethical values, personal family experiences, personal ambitions, and other deeply held personal opinions and beliefs?
One possible way to determine the existence of a personal conflict that may impact on IRB proceedings is to estimate the level of commitment of the IRB member to the issue. Conflict of interest, hence recusal, implies substantial concerns over the matter at hand, or perhaps even active involvement. On the other hand, abstention reflects neutrality or feelings not sufficient to rise to the level of objection. Does the member have active concerns over the matter at hand which would clearly represent a conflict of interest or does he/she simply not care and believes this personal neutrality is not likely to interfere with his/her ability to judge the research objectively? Does the member’s opinions fall into what Robert’s Rules describes as “...an interest not common to the other members...” which suggests the presence of a personal COI. In this case, 4 of 8 members voted yea and 3 voted nay, having nothing to do with their personal beliefs regarding ART. Clearly 7 of 8 members in attendance had no strong personal opposition to ART as part of a research protocol.
It seems clear how best to handle IRB members who have professional or employment related conflicts, for instance members of the clergy. One would expect members of the clergy or those otherwise employed by certain Christian denominations to maintain a specific negative attitude towards ART as a condition of employment or their office as a member of the clergy. The conflict of interest would be clear and the individual would be recused almost automatically. But what about the IRB member who is not employed by the church, is not a member of the clergy, has no direct benefit, and is constrained only by his/her personal religious conviction? Does stating, “Assisted reproductive technology is interdicted by my religion,” signal the same automatic conflict of interest as saying “I am holding office in the church”?
Whether occult conflicts of interest exist is a complex matter; the IRB must be aware such personal conflicts may present in subtle ways and be on guard for their possibility. When suspected, fair-minded individuals may reasonably differ in the conclusion they reach concerning particular research, but it is hoped, as always, that all IRB members will ultimately use their good judgment and commitment to their duties as IRB members to add their objective voice to the discussion.
The IRB administrators all agreed that this member had a significant personal conflict of interest and should not have been permitted to participate in the discussion or the vote. She should have been encouraged to discuss her beliefs with the Chair prior to the meeting and been recused from this part of the meeting. No concrete alterations in the policies and procedures occurred as a consequence of these events as the SOP’s already include a detailed discussion of conflicts of interest along with formal COI policies. However, the issues raised by this case prompted the IRB administration at this institution to add the discussion of personal conflicts of interest to the agenda for the semi-annual educational seminar.
Comment 1: “Community” members and conflict
Comment 2: Recusal without “conflict of interest”