Trudo Lemmens, Lic. Lur., LLM, Clarke Institute of Psychiatry and Joint Centre for Bioethics, University of Toronto
About two years ago, the late Dr. Benjamin Freedman invited me to participate in a study on the ethics and regulation of commercial medical research. Our attention was drawn to contract research organizations, the prototypes of commercial pharmaceutical research, which are visible through local newspaper advertisements that invite healthy volunteers to participate in clinical trials. We received some troubling details about these clinical trials. In some trials, healthy volunteers were sometimes exposed to significant risks in exchange for significant amounts of money, but the informed consent process seemed flawed. We were startled by the fact that some consent forms contained anti-liability clauses stating that, in cases of adverse effect, research subjects would receive appropriate medical treatment but the company would not be liable for further bodily damages, yet the Quebec Civil Code explicitly states that no person can "exclude or limit his liability for bodily or moral injury caused to another".
We were also concerned by clinical trials that clearly attracted people by offering considerable financial rewards. This practice is prohibited – be it with some room for manipulation or interpretation – by the Quebec Civil Code and runs counter to Medical Research Council (MRC) and other international guidelines, codes, and regulations.
We decided to start with an analysis of the review process for these clinical trials. We found that most studies in the commercial sector were reviewed by private research ethics boards (REBs). We were startled by the total lack of governmental oversight for these REBs. Neither the Health Protection Branch (HPB), nor other official agencies could provide appropriate information on the number of private REBs, membership, functioning, etc. This lack of oversight is even more of a concern now, in light of some recommendations made by the HPB’s Clinical Trials Working Group for the review of Phase I clinical trials.
With the support of the National Council on Bioethics in Human Research (NCBHR), we surveyed private REBs in Canada and the USA to compare the functioning of private and institutional REBs. Our survey was based on a similar survey of institutional REBs, the results of which were published in NCBHR Communiqué.
I had hoped to present the survey results at this conference, but the analysis of results has been delayed. I will refer to some of my ideas and impressions from answers to the limited number of questionnaires that we have received to date. This research, based on these limited survey results and my discussions with people in private REBs, is both reassuring and troubling.
I am reassured by the underlying reasons for establishing private REBs, the integrity of their members, and their functioning. Private REBs were established in reaction to the growing commercialization of medical research. They fulfill a clear demand and need from industry. Without them, many clinical trials could not be reviewed appropriately. Institutional Review Boards (IRBs) are understaffed and have no interest in reviewing studies from outside their institution.
The primary objective of many private REBs is to enable appropriate, timely review of clinical trials and to facilitate medical research. Commercial interests do not top the list of most private REBs.
Moreover, the composition and functioning of these REBs often reflect higher standards of adherence to the rules of Research Ethics Codes and Guidelines than in many institutional REBs. Nearly all of the surveys that I have received indicate that private REBs meet in person, unlike some IRBs, and that they have an appropriate representation in terms of expertise, community representation, gender equality, and minority-group representation, which is often lacking in IRBs.
I have learned that private REBs are capable of acting independently and resisting commercial and financial pressure from pharmaceutical companies and contract research organizations; they reject studies, notwithstanding commercial coercion. The frank admission of this pressure has convinced me even more that a fundamental problem remains. I believe that it is in the interest of research subjects, of medical research, and private REBs to discuss this issue and urge more governmental regulation.
Intrinsic conflicts of interest in private REBs
Conflicts of interest in private REBs should be viewed as a paradigm, indicating the need for more, rather than less, governmental oversight. They illustrate the need for tighter rules on private and institutional REBs, if appropriate solutions for the fundamental conflict of interest in private REBs can be found.
There are two separate forms of private REBs. One is the in-house REB, which is connected to a contract research organization or pharmaceutical company that conducts studies; the second is the "for-hire", independent, commercial REB.
Why focus on conflicts of interest in private REBs? Two major types of rules govern any type of regulation: procedural and substantive. Through procedural rules, legislators or self-regulatory agencies can establish a system of review and licensing. Procedural rules are ordinarily, if not always, accompanied by a set of substantive rules. Substantive rules specify what is allowed and what is forbidden. As applied to research ethics review, substantive rules specify what research is acceptable, while procedural rules specify how to and who can decide whether a research study is acceptable.
In modern research ethics codes, procedural rules dominate. They typically provide details about the constitution and composition of REBs, record-keeping, appeal procedures, and more. They are generally vague about how to weigh risk and benefits and they call for significant value judgments on the part of REB members.
For example, the U.S. Food & Drug Administration (FDA) regulations, which are very concrete on the procedural side, prescribe that research procedures must be consistent with sound research design and should not unnecessarily expose patients to risks. Risks must be reasonable in relation to anticipated benefits. The selection of subjects has to be equitable, and coercion or undue influence must be avoided.
There is no gold standard on which IRBs can rely when reviewing research protocols. Appropriate IRB review requires a fair exercise of intelligence and discretion.
Procedural rules, such as those dealing with the composition of IRBs, are important in research ethics review, precisely because there is so much reliance on fairness and intelligence in IRB review. IRBs should be composed of people with varied backgrounds and expertise, who are sufficiently detached from a given protocol that they can be trusted to weigh risks and benefits fairly. Conflict of interest (CI) rules are essential to safeguard this principle. The problem is that current research ethics guidelines and the way in which they are interpreted are too vague to be helpful.
CI rules in ethics codes and regulations
The World Medical Association Declaration of Helsinki clearly stresses the need for independence in IRBs. S.3 of the Declaration states:
"The design and performance of each experimental procedure involving human subjects should be clearly formulated in an experimental protocol which should be transmitted for consideration, comment and guidance to a specially appointed committee independent of the investigator and the sponsor..."
The notion of dependence clearly prohibits attempts to influence review through employment relationships and the payment of review boards.
Surprisingly, some later research ethics codes totally lack references to this problem. Even when a useful rule exists, regulatory agencies have often lacked initiative in this context and do not respect the spirit of the Helsinki rule. For example, the FDA prescribes that "No IRB may have a member ... who has a conflicting interest", without saying more. According to FDA officials, this provision only prohibits researchers from participating in a review of their research. FDA data on site visits of IRBs suggest that it has only found CI in such cases.
The most specific reference to potential financial CI within REBs comes from the Canadian Tri-Council’s Code of Ethical Conduct for Research in Humans. The Tri-Council is a working group of three federal research and granting agencies: the Social Sciences and Humanities Research Council (SSHRC), the Medical Research Council of Canada (MRC), and the Natural Sciences and Engineering Research Council (NSERC).
The most recent version of the Tri-Council Code has specific references to potential for CIs, even for commercial REBs. There again, the reference to undo or excessive honoraria are very vague and cannot be clearly interpreted.
What is a conflict of interest?
The problem is that it is difficult to define CI. The concept is used in many different circumstances. Those who try to define it offer only general principles.
Dennis Thompson suggests that "a conflict of interest is a set of conditions in which professional judgment concerning a primary interest ... tends to be unduly influenced by a secondary interest." Following this definition, we must specify the primary obligation of REBs to determine when its goals risk being affected by CI.
IRBs clearly have a public, protective role. "The purpose of IRB review", according to FDA rules, "is to assure that appropriate steps are taken to protect the rights and welfare of humans participating as subjects in the research."
We believe that, in many respects, IRBs function like administrative licensing boards or tribunals. As IRBs, these tribunals often have important, public-policy goals. They are given much discretion in the implementation of their policies, and their decisions may have a major impact on the activities that they regulate.
For that reason, administrative law on CI can inspire us in this debate. The independence and impartiality of judges and administrators are major principles of judicial and administrative review. It is expressed in the adagium "nemo judex in causa sua" (no one can judge their own case). In American law, absence of bias and impartiality are requirements of due process and protected as basic constitutional principles.
The law of bias generally distinguishes between bias as a result of:
- a pecuniary interest;
- personal involvement of the decision-maker; and
- alleged prejudgment of the merits of a particular case.
The law has recognized that a certain prejudgment does not necessarily disqualify someone from judging a case. In these cases, a party invoking bias must indicate that a prejudgment is likely to influence the decision-maker.
Financial interests are judged severely. When reviewers or judges have financial interests, they are disqualified if a clear potential for personal loss exists and if the financial interest is not too remote. Some British and Commonwealth courts hold that an interest as shareholder or ratepayer is sufficient to identify bias.
How does this apply to private IRBs?
A financial CI is imbedded in the way that private IRBs are organized. In proprietary IRBs, the presence of a direct financial interest is real. In for-hire IRBs, financial interests are more remote and depend on future actions of those who submit protocols for review, i.e., the client research organization or pharmaceutical company.
In proprietary IRBs, the same company that establishes the IRB submits its research protocols for review. Here, two CI are a cause for concern. Individual IRB members may be recruited from company personnel. They have an employer-employee relationship with the institution that submits its protocols for review. Direct or indirect, conscious or unconscious pressure may exist not to reject protocols submitted by the employer. While this situation exists within institutional IRBs, pressure within private companies, who are not subjected to the same public and academic scrutiny, may be higher. Employee protection from corporate sanction may be less reliable.
The economic survival of pharmaceutical testing companies depends directly on protocols under review. In a worst-case scenario, systematic rejection of research protocols leads de facto to a loss of employment. Moreover, employees may have financial interests as shareholders. Every protocol rejection affects their profit margin. Their interest is clearly more than "minimal".
In the case of for-hire IRBs, CI depends on the attitude of companies who submit protocols for review. In an ideal world, the workload of for-hire IRBs should not diminish if the IRB is known to do a thorough review. In fact, serious review should be a marketable item. But is it too suspicious or critical to suggest that a corporation may abandon a private IRB that often rejects its protocols, if other IRBs, known to be more lenient, are available? For-hire IRBs, which depend on external contracts, may suffer the consequences of frequent protocol rejection. In fact, they may depend on a few lucrative contracts with large corporations. Their financial gains may be directly affected by their critical attitude.
Outside the field of medical research, it is a general principle that reviewers or judges should not be directly paid by only one party in a review. If this fundamental rule exists elsewhere, why not in medical research?
Is conflict of interest avoidable? Disclosure of CI to research subjects or patients is often proposed as a solution. However, IRBs are supposed to control CI among researchers. It seems odd, not to say impossible, to explain to research subjects that the committee that is supposed to protect them has a CI.
It is difficult to avoid or correct CI in proprietary IRBs. The danger that secondary interests, e.g., keeping a job, financial gain, or promotion, will have an impact on the appropriate protection of research subjects are too serious.
The only way to avoid CI would be to establish a system of accredited reviewers who would follow clear guidelines on research review and who could be held accountable for violations of their professional code. The paradigm example is the accredited accountant, who is paid by the company but adheres to strict professional rules. The problem is that IRB membership is virtually unrestricted, no obligatory educational programs exist, and above all, a clear, reliable research code, containing substantial rules that must be respected, is lacking. In this context, it becomes difficult to control how professional research ethics reviewers do their job.
For-hire IRBs are independent contractors, but the only way to guarantee that they will not be tempted by client-friendly politics is to implement a system that prohibits IRB shopping, i.e., forum shopping. By creating an administrative structure involving exclusive, mandatory jurisdiction, accreditation, and control, this CI could be avoided. Doug Kinsella* has already tried to implement such a system for provincial IRBs. Contract research organizations (CROs) and others would be required to pay a license fee for submitting protocols. They would have no direct financial or other link with IRB members.
The credibility and integrity of IRB review is affected, if not undermined, by inherent problems of conflict of interest. It is time for regulatory agencies to deal with the way that private protocols are reviewed. Private IRB review is undermined by structural problems of CI. The integrity of reviewers is not sufficient to remedy this problem. Such a system would not be a novelty on the international scene.
It is important to view the development of private REBs in context. Some federal agencies are suffering from what can be diagnosed as "deregulitis". This highly infectious disease is particularly dangerous when it affects one of the core duties of responsible government: protecting the health and well-being of its citizens. A Clinical Trials Working Group of Canada’s Health Protection Branch (HPB) has recently proposed to shift the onus of determining the safety and toxicity of new drugs in Phase I trials from the Drugs Directorate to private REBs. HPB would no longer be involved in this assessment, and private REBs would have to collaborate with industry to obtain the necessary data. This proposal runs counter to international developments and would deprive Canadian citizens of an essential step in assuring the safety of Phase I trials. Neither institutional nor private REBs have the expertise and capacity to do such review. Moreover, it removes yet another essential barrier between those who conduct clinical trials and those who review them. This proposal does not serve research subjects and by this token, does not serve medical research.
As Jay Katz stated in the recent Presidential Report on Human Radiation Experiments: "Society can no longer afford to leave the balancing of individual rights against scientific progress to the low-visibility decision-making of IRBs with regulations that are porous and invite abuse. ...[It] should be forced to make these difficult moral choices in bright sunlight and through a regulatory process that constantly strives to articulate, confront, and delimit [human and societal] costs."
This conference and the process of implementing and debating the new Tri-Council Code are but a small step in this direction. Both systems of institutional and private REB review must be submitted to appropriate regulation and cannot be left to local improvisation.
Jack Corman, Institutional Review Board Services: I am the president and founder of an independent (for-hire) REB. We were founded in 1993 and are one of the older boards. I am troubled by some of your comments. Our board can satisfy your concerns and has demonstrably done so, even to the Office for Protection from Research Risks (OPRR). We have, in fact, satisfied them for an NIH-funded study. However, there is a bigger issue – one that represents a bias that has been expressed in other circumstances within the private and public sector. The public sector often feels that it has a right to the moral high ground. I do not think that you can legislate or regulate integrity. It comes from within the individual, whose integrity, competency, and goodwill matter.
We have two principles when we review protocols: to consider the value of research and how to protect the patient.
Trudo Lemmens: I am reassured by the integrity of private REB reviewers, but that is not the issue. I am not satisfied by the argument that integrity cannot be regulated and that we have to rely on people’s integrity, because, if that was true, you would not need a review process.
How do you convince the public and how do you assure the best protection of humans in medical research? How do you assure the independence of review boards? How do you establish a system with no constraints and pressures, financial or otherwise, to accept or reject protocols? That is the essential question.
Can you achieve that goal through the private sector? Perhaps, but I would argue that we need a different system, e.g., one of mandatory jurisdiction that forbids forum shopping in private and institutional REBs. How do we develop a system that assures integrity and protects research subjects?
Jack Corman: Our board has flourished because we have demonstrated those factors. What about ISO 9000 and ISO 9002 international standards? Anybody involved with that process knows that it depends on a set of well-defined criteria. But the manual does not count; how it is implemented is crucial. That process should be transparent. Any board should have that facility, i.e., they should be able to demonstrate it to any interested parties.
Gordon Crelinsten, Royal Victoria Hospital/NCBHR: Private, for-hire REBs are often viewed as fast, efficient, and permissive, because their main task is primarily to ensure that research protocols comply with already established regulations, guidelines, or particular codes of conduct. We have also heard that institutionally based REBs are often viewed as slow, costly, resource-challenged, and often confrontational, because their main focus is often not only the protection of research subjects but the interpretation of guidelines with respect to local and individual concepts of ethical behaviour. We have heard that industry is interested in funding greater research efforts and that institutions are not only competing for but coveting industry money. We have seen that some national research funding councils have cozied up to industry.
Do you foresee that private or proprietary REBs, because of their success and efficiency, will eventually be accepted by IRBs as appropriate fora for ethics review and that IRBs, because of resource challenges, will come to accept the decisions of private REBs as acceptable for research in public institutions?
Trudo Lemmens: I would be worried by this development, because institutions have an obligation to assure the appropriate review and protection of research subjects at their institutions. I would firmly urge institutions not to do so, because they cannot shift their responsibilities to other REBs. I do not anticipate this occurrence in university hospitals, but it may happen in more low-profile hospitals where necessary funds are unavailable. Because we see more financial involvement by pharmaceutical industry and institutions, it is essential that the government intervenes to establish a system of appropriate REB review.
Henry Dinsdale, NCBHR: The past-chair of the REB at the University of Rochester in upstate New York told me that, a few years ago, they were so short-staffed that they passed the majority of protocols to an independent REB for review.
Heather Baker, Phoenix International Life Sciences Inc.: Two observations have surprised me in this discussion: the lack of understanding of how and what a CRO IRB can and is doing; and how it is regulated.
Trudo Lemmens: One essential element in matters of conflict of interest is public trust and perception. The golden rule is to ask, "How do we explain a situation? How is a situation perceived?" Most people who are outside of the research community are troubled by the way that research review is structured in Canada, particularly commercial CI. We have an obligation to clarify issues and find potential solutions.