3
Back to Basics: Scientific, Conflict of Interest, and Ethical Review of Research Protocols

Protecting the rights and welfare of research participants is based on respecting the relevant ethical principles that underlie such protection, including the principal of beneficence. The Belmont Report: Ethical Principles and Guidelines for the Protection of Human Subjects of Research states that “persons are treated in an ethical manner not only by respecting their decisions and protecting them from harm, but also by making efforts to secure their well being” (National Commission, 1979, p.2). In the context of research, individuals are sometimes placed at risk, and such risks should be carefully weighed against potential benefits, either to the individual or to society. The principle of beneficence incorporates the rules of “do no harm” and “maximize possible benefits and minimize possible harms” (National Commission, 1979, p. 6). The best means for assessing risks and benefits is through independent review of the proposed research by individuals who have no direct vested interest in its outcome.

THE NEED FOR RESEARCH ETHICS REVIEW BOARDS

As detailed in this chapter, Institutional Review Boards (IRBs) should be reshaped and reformed to serve the role for which they were originally intended—ensuring participant protection through the careful ethical review of research protocols. As such, IRBs are the cornerstone of a system in which other entities, such as research sponsors, also have obligations to protect research participants, for example, by ensuring that investigators and research staff have completed necessary education requirements. How-



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 70
Responsible Research: A Systems Approach to Protecting Research Participants 3 Back to Basics: Scientific, Conflict of Interest, and Ethical Review of Research Protocols Protecting the rights and welfare of research participants is based on respecting the relevant ethical principles that underlie such protection, including the principal of beneficence. The Belmont Report: Ethical Principles and Guidelines for the Protection of Human Subjects of Research states that “persons are treated in an ethical manner not only by respecting their decisions and protecting them from harm, but also by making efforts to secure their well being” (National Commission, 1979, p.2). In the context of research, individuals are sometimes placed at risk, and such risks should be carefully weighed against potential benefits, either to the individual or to society. The principle of beneficence incorporates the rules of “do no harm” and “maximize possible benefits and minimize possible harms” (National Commission, 1979, p. 6). The best means for assessing risks and benefits is through independent review of the proposed research by individuals who have no direct vested interest in its outcome. THE NEED FOR RESEARCH ETHICS REVIEW BOARDS As detailed in this chapter, Institutional Review Boards (IRBs) should be reshaped and reformed to serve the role for which they were originally intended—ensuring participant protection through the careful ethical review of research protocols. As such, IRBs are the cornerstone of a system in which other entities, such as research sponsors, also have obligations to protect research participants, for example, by ensuring that investigators and research staff have completed necessary education requirements. How-

OCR for page 70
Responsible Research: A Systems Approach to Protecting Research Participants ever, IRBs have come to shoulder an increasing share of these tasks in ways that were not originally intended. Because IRBs should be constituted to carry out their obligation to focus primarily on the ethical aspects of human protection issues, the committee suggests that protection programs consider changing the name of these boards to reflect this recommended fundamental shift in emphasis. Recommendation 3.1: The Institutional Review Board (IRB), as the principal representative of the interests of potential research participants, should focus its full committee deliberations and oversight primarily on the ethical aspects of protection issues. To reflect this role, IRBs should be appropriately renamed within research organizations’ internal documents that define institutional structure and policies. The committee suggests the name “Research Ethics Review Board (Research ERB).” Changing the name of an entity can be a useful way to signal important substantive change. In the United Kingdom and elsewhere, ethics review is carried out by what are typically called “Ethics Committees” or “Ethics Review Committees.” The International Conference on Harmonisation uses the terminology “Independent Ethics Committee.” In each case, the objective of these reviews is to ensure the ethical conduct of research and that participants’ interests are fully recognized, represented, and protected. The committee therefore recommends moving away from the term “Institutional Review Board,” which conflates institutional interests with those of participants and which may cause at least the appearance of an institutional conflict of interest. Admittedly, the term IRB is now firmly embedded in the regulations and literature, and is likely to continue to be used despite its imperfect reflection of the function that the board is designed to serve. However, many research organizations in this country have given these bodies different names (“Committee on the Protection of the Rights of Human Subjects” or “Committee on Clinical Investigations”)1 that more accurately describe their appropriate functions, while empowering them to carry out the functions that have been assigned to IRBs by the applicable federal regulations. This committee urges all research organizations (as well as free-standing IRBs) to signal their commitment to reform by changing the name of the bodies serving the functions of IRBs to “Research Ethics Review Board” 1   In addition to these designations in use at the Duke University Medical School and the Albert Einstein College of Medicine, respectively, Johns Hopkins University also uses the terms “Joint Committee on Clinical Investigation” and “Committee on Human Research” within their medical school and school for public health.

OCR for page 70
Responsible Research: A Systems Approach to Protecting Research Participants (“Research ERB”). The modifier “research” is intended to distinguish the body from the ethics review board commonly found in many hospitals, which is charged with providing advice and consultation on difficult issues related to health care delivery. Throughout the remainder of this report, the term “Research ERB” will be used when describing the idealized system of protections. The term “IRB” will be used when describing aspects of the current system. The Research ERB should refer issues of institutional interest (e.g., risk management concerns, resolution of institutional or investigator conflict of interest) to the institution’s management and/or compliance office (see Chapter 6). THREE-PRONGED REVIEW A central tenet in the protection of research participants is the independent review of research protocols to assess their scientific merit and ethical acceptability. It is also critical to consider whether conflicts of interest on the part of the investigator, the Research ERB, or the institution place research participants at undue risk. Thus, every protocol requires an autonomous analysis of several interrelated factors (scientific merit, ethical design, and potential financial conflicts of interest)2 before it is deemed appropriate for investigators to enroll participants. Evaluation of each factor requires a specific knowledge base for sufficient assessment, both in depth and breadth. Thus, it is unrealistic to expect a single group of individuals to possess the requisite skills to competently carry out the many tasks needed to protect the rights and welfare of research participants. Recommendation 3.2: Research organizations and research sponsors should ensure that Human Research Participant Protection Programs utilize distinct mechanisms for the initial, focused reviews of scientific and financial conflicts of interest. These reviews should precede and inform the comprehensive ethical review of research studies by the Research Ethics Review Board (Research ERB) through summaries of the relevant findings submitted to the Research ERB for full board consideration. 2   The committee acknowledges that nonfinancial conflicts of interest also pose potential threats to the integrity of study conduct. Some possible conflicts of this nature are discussed in Chapter 4, and the need for the research community to rigorously pursue policies to oversee and manage such potential conflicts is discussed in Chapter 6. In light of the need for further development of the policy discussion in this arena, the committee has largely concentrated its discussion with respect to the protocol review process on financial conflicts of interest.

OCR for page 70
Responsible Research: A Systems Approach to Protecting Research Participants In the United States, ethics review of federally funded research and of some sponsor-funded research (specifically, clinical trials of products subject to Food and Drug Administration [FDA] regulation) is generally conducted by IRBs, as specified in federal regulations.3 The degree to which these bodies should and can explore the scientific merits of protocols has been a matter for impassioned debate since review of protocols by nonscientists was first suggested in the 1960s and 1970s (Levine, 1986; Moreno, 2001). Today, some observers assert that most IRBs as currently constituted do not routinely or sufficiently review scientific concerns, such as justification for sample size, eligibility criteria, or the qualifications of the investigator (Bohaychuk et al., 1998). Others believe that ethics and science review cannot and should not be separated because they are intrinsically tied (Freedman, 1987). In the United States, not all human research is routinely subjected to both scientific and ethical review. In some cases, particularly when no federal funds are used to sponsor the research, scientific review may not occur or may be conducted only cursorily by a group internal to the research organization. In other cases, ethics review may not occur if the research is not subject to the various federal regulations governing the conduct of research with human participants.4 As stated in Chapter 2, the committee believes that all research involving human participants, regardless of site or funding source, should be subject to an independent review and a common system of protections (Recommendation 2.1). The protection program should ensure that each research protocol receives objective scientific review, relying on input from content experts. These experts may be found, for example, within local scientific departments, external academic institutions, pharmaceutical companies, federally organized peer-review groups (i.e., the National Institutes of Health [NIH] or National Science Foundation [NSF] study sections), or FDA Review Divisions, and they should be sufficiently insulated from the interests of the investigator or the protocol. Program procedures should clearly articulate mechanisms for documenting such insulation and for the transmittal of the findings to the relevant Research ERB. The program should be subject to external audits (by FDA, NIH, or accreditation bodies) that verify, among other things, the appropriate degree of insulation of these functions and their operations. Although some IRBs rigorously consider the scientific merits of proposed research, the extent to which they are aware of or consider potential financial conflicts of interest is not clear. In addition to scientific and ethical considerations, it is essential to ensure that potential financial conflicts of interest involving the investigator or the institution are identified, managed, 3   45 CFR 46.103, 21 CFR 56.103. 4   45 CFR 46, Subpart A, 21 CFR 50 and 56.

OCR for page 70
Responsible Research: A Systems Approach to Protecting Research Participants and, if possible, eliminated. This determination is especially critical if the conflicts pose possible risks to research participants.5 In its 2001 report, the National Bioethics Advisory Commission (NBAC) suggested that IRB review of research studies is “one method for identifying and dealing with conflicts of interest that might face investigators” (2001b, p.58). However, NBAC concluded that “IRB review alone…is not sufficient to manage conflicts of interest, because the options available to IRBs to eliminate such conflicts are limited” (2001b, p.59). Indeed, given the burdens faced by IRBs in terms of work volume, expecting these boards, especially in large research organizations, to assume primary responsibility for such reviews is unrealistic (see Chapter 6). Potential financial conflicts of interest of the investigator, Research ERB members, or the institution should be assessed by the organization’s relevant conflict of interest oversight mechanism (Recommendation 6.5) and communicated to the Research ERB. As described in Chapter 6, the conflict of interest oversight body should determine whether financial conflicts should be disclosed, managed, or are so great that they compromise the safety or integrity of the proposed research. The conflict of interest body should communicate to the Research ERB its determination of potential conflicts relevant to protecting the rights and welfare of research participants, the rationale for its determination, and any recommended conflict management plan. Such communications could be verbal or achieved by providing the Research ERB chair or administrator a copy of the conflict of interest committee’s final determination, if that conclusion suggests a conflict that poses greater than minimal risk to potential research participants. Some institutions “cross-fertilize” various review committees to maintain communication and promote awareness of the relevant issues within each committee’s purview (Dretchen, 2001). The Research ERB should use this information to determine if and how participant protection could be negatively affected, whether the recommended conflict management plan is sufficient to ensure participant protection, what information pertaining to any conflict should be disclosed to participants through the informed consent process, and whether ongoing review is required in the event that the research goes forward. By ensuring that properly constituted bodies review protocols for scientific merit and freedom from conflicts of interest, the Research ERB should be able to focus its efforts on assessing whether the protocol meets the ethical requirements as stated in the Belmont Report (National Commission, 1979) and the federal regulations.6 Each program should ensure that 5   For FDA-regulated products, the responsibility for managing investigator conflicts lies with the study sponsor (21 CFR 54). 6   45 CFR 46.111, 21 CFR 56.111.

OCR for page 70
Responsible Research: A Systems Approach to Protecting Research Participants FIGURE 3.1 The Confluence of the Research Review Process these reviews occur in accordance with established standards and are subject to a system of internal checks and balances. Despite the need for the three distinctive reviews (science, ethics, financial conflicts of interest), their interrelated nature and their underlying considerations requires that a single body be vested with the explicit authority and accountability for the final determination regarding the ethical acceptability of research protocols. The committee believes that this body should be the newly designed Research ERB (Figure 3.1). Therefore, the focused reviews of scientific merit and the evaluation of potential financial conflicts of interest should feed into the ethics review process for each protocol, and the Research ERB should have the ultimate authority regarding participant enrollment. ENSURING DISCRETE SCIENTIFIC REVIEW OF PROTOCOLS Scientific and ethical reviews of research protocols are both essential because each considers different sets of questions and therefore each can yield different determinations. For example, a proposed study might be deemed scientifically sound and intellectually intriguing and yet pose significant or even intolerable risk of harm to participants. Conversely, the former Office of Protection from Research Risks (OPRR) wrote that “a proposal without scientific merit can on the surface appear to be ethically

OCR for page 70
Responsible Research: A Systems Approach to Protecting Research Participants acceptable, but the fact that it will not produce new or usable data does not justify the use of human participants regardless of the level of risk” (1993). Furthermore, a scientifically meritorious study might be ethically conducted in one cultural context but not in another (NBAC, 2001a). Unquestionably, there are areas in which the two sets of considerations intersect, for example, in determining if inclusion/exclusion criteria are wisely chosen on scientific grounds and properly justified ethically. Each review process is likely to consider aspects of the other, but in general, greater benefit can result through separate, focused reviews. For these reasons, programs should not rely solely on one review mechanism (e.g., the Research ERB) to conduct all aspects of ethical and scientific review of protocols.7 When a Research ERB is called on to conduct the exclusive scientific review of a protocol, two primary problems can arise: 1) it can be distracted from intensive review of the ethical issues due to lack of time, or 2) it may lack the scientific expertise necessary to adequately assess the technical merit of a proposal (OIG, 1998a). An additional complication arises if the only resource available to the Research ERB to answer technical questions is the principal investigator (PI) who submitted the protocol— obviously not a disinterested party. In this case, the Research ERB would lack an appropriately independent technical resource who could address challenging scientific questions pertaining to the given protocol. The Human Research Participant Protection Program (HRPPP) should also ensure that any expert scientific panel is free of significant conflicts of interest. Elements of Scientific Review All protocols involving human participants should undergo an independent and rigorous scientific review to assess scientific quality, the importance of the research to increase knowledge, and the appropriateness of the study methodology to answer a precisely articulated scientific and, in some cases, clinical question. For example, the design of clinical trials should be based on sound statistical principles and methodologies, including sample size, use of controls, randomization, population stratification, stopping rules, and the feasibility of relating endpoints to objectives. Ensuring that the chosen study design minimizes bias and generates data that will answer the scientific question requires some understanding of the research process and the area under study (Spilker, 1991). These issues are pivotal to a successful study and 7   The committee recognizes that there may be situations in which adequate content and ethics expertise can be assembled within one body, such as at a small institute with a focused research portfolio or a protection program formed to oversee a specific line of research. However, even in these instances, the science and ethics review functions should be understood to be distinct activities.

OCR for page 70
Responsible Research: A Systems Approach to Protecting Research Participants Box 3.1 Elements of Scientific Review Importance and novelty of the scientific question Strength of the scientific design and methodology Feasibility of the research as designed Appropriateness of the statistical analysis plan Estimate of the probability of meeting the enrollment goals Need for, and structure of, a Data and Safety Monitoring Board/Data Monitoring Committee (DSMB/DMC) Assessment of the thoroughness of the proponent’s evaluation of the relevant literature and previous studies, if available Strength of the qualifications of the investigator to carry out the protocol and the facilities available to him or her Appropriateness of the inclusion/exclusion criteria Dissemination plan (to enrolled participants and through formal publication) should be evaluated by a mechanism that is distinct from the ethical review process before participants are enrolled (see Box 3.1). This mechanism should ensure that adequate technical expertise in the evaluation of the proposal occurs, which could be accomplished by using a separate committee(s) based on expertise, by using a subcommittee of the Research ERB augmented by others in the institution with specific expertise, or by using outside experts. The result of the scientific review should include the elements shown in Box 3.1; the result should be provided to the Research ERB. There will always be some level of overlap between scientific and ethical reviews. Research ERB review should continue to include some consideration of a protocol’s scientific merit; however, delegating the in-depth scientific review to an upstream mechanism should facilitate the Research ERB’s more focused consideration of the ethical elements of particular protocols. One advantage of ensuring a distinct scientific review mechanism is the opportunity to identify protocols that are not yet suitable for Research ERB consideration. Such protocols should be returned with suggestions for revision to the author of the proposed research. This would help ethics review meetings maximize their time to focus on a thorough deliberation of the ethical considerations of fully developed, scientifically sound protocols. In rare and controversial cases, however (e.g., proposals to conduct embryonic stem cell research or xenotransplantation research), it may be appropriate to pursue the ethical consideration of a protocol before, or in conjunction with, the evaluation of its scientific merits.

OCR for page 70
Responsible Research: A Systems Approach to Protecting Research Participants Mechanisms for Scientific Review It is the responsibility of the research organization directly overseeing the conduct of the research to ensure an adequate process for the scientific review of protocols. As with other components within protection programs, the scientific review process should be accountable to the highest authority within the research organization, and failure to conduct an independent, nonconflicted scientific review should be met with sanctions by that authority. Therefore, a mechanism for periodic audit of the scientific review process should be established. A variety of mechanisms can be used to ensure independent scientific review. In fact, most protocols currently undergo some level of scientific review through existing mechanisms. Measures should therefore be taken to ensure that protocols not currently subject to technical review are funneled into existing or newly created mechanisms for this purpose (Box 3.2). Research sponsored by established industrial entities typically undergoes rigorous scientific review (FDA, 2001c; Spilker, 2001). For example, protocols designed by pharmaceutical companies often go through numerous and prolonged iterations before ultimately being approved by an internal oversight committee of physicians and scientists. In large pharmaceutical companies, scientific review committees typically comprise medical directors, clinical scientists, safety managers, and regulatory affairs professionals, and these committees report to a high-level clinical executive within the organization. In addition, they often rely on other experts in areas such as pharmacology, toxicology, and pharmaceutical development for additional support. The committees are generally part of the clinical organization of the sponsor (e.g., in the Chief Executive Officer’s office), rather than being placed outside it. This process is not entirely devoid of conflicts of interest, and care should be taken by companies to appropriately insulate their scientific review committees, perhaps by providing a charter that governs the committee’s operations and allows the committee to be audited. The charter should also indicate the qualifications required of scientific review committee members. If protocols involve investigational drugs, devices, or biologics, they must be submitted to FDA for regulatory review, comment, and approval; they can be rejected by the agency on scientific or safety grounds. FDA reviewers are also trained scientists and physicians versed in the pertinent therapeutic area and intimately familiar with issues of inclusion/exclusion, appropriate endpoints, and safety issues. Thus, comments provided to sponsors by FDA reviewers should be made available to the Research ERB to inform the final comprehensive assessment of a protocol. Federally funded biomedical, social, and behavioral research protocols are typically subjected to a scientific peer review process by the funding

OCR for page 70
Responsible Research: A Systems Approach to Protecting Research Participants Box 3.2 Possible Mechanisms for Scientific Review Academic-based research: A separate, internal committee with requisite expertise (such as departmental or GCRC [General Clinical Research Center] unit); a subcommittee of the Research ERB, perhaps augmented with others from the institution with appropriate expertise; external committees of experts (i.e., federal peer review mechanisms). Outside expert consultants should always be considered as a resource to remove the perception or the reality of conflicts of interest and to ensure a sound scientific review. Funding for scientific review should be assumed by the academic institution, which should also include auditing of the scientific review process within its overall quality assurance (QA) activities. Written assurance of the scientific review should be provided to the Research ERB. Industry-sponsored research (for FDA-regulated products): Company Protocol Review Committee independent of the author(s) of the research protocol; applicable FDA Review Division. Both should provide written assurances to the Research ERB. Privately sponsored research (not for FDA-regulated products): Protocol Review Committee (e.g., leaders in the applicable field) independent of the author of the research protocol and external to the research sponsor. The committee should provide written assurance to the Research ERB. May be funded by the research sponsor, but should operate at arm’s length and according to a charter. Federally funded research (NIH or equivalent): Protocol Review Committee independent of the author of the research protocol. Written assurance to the Research ERB should be provided (such as grant “pink sheets”). May be funded by the research sponsor, but should operate at arm’s length and according to a charter. NIH or equivalent agency should have an audit mechanism to verify adequacy of the scientific review process. Locally sponsored research (e.g., a university department using unrestricted grants): Departmental Protocol Review Committee independent of the author of the research protocol. Written assurance should be provided to the Research ERB. Should be subject to audit by institutional-level body. agency. The guiding principles for the initial review of research project grant applications submitted to NIH are based on the Public Health Service Scientific Peer Review Regulations, which state that peer review groups are to make recommendations concerning the scientific merit of applications. The specific criteria used to assess the merit of research project grant applications vary with types of applications reviewed. However, the review by the scientific panel is expected to reflect existing codes adopted by disciplines relevant to the research or the collective standards of the professions

OCR for page 70
Responsible Research: A Systems Approach to Protecting Research Participants represented by the membership. In addition, the evaluation is to take into consideration the investigator’s response to six points relevant to the protection of human participants, ranging from the inclusion criteria to the protection of confidentiality to the minimization of risks and to obtaining informed consent (NIH CSR, 2001, 2002). No awards are made until all expressed concerns about human participants have been resolved to the satisfaction of NIH. At NSF, proposals are assigned to the appropriate NSF program for acknowledgment. If they meet NSF requirements, they are submitted for review. All proposals are reviewed by a scientist, engineer, or educator serving as an NSF program officer and usually by 3 to 10 other individuals outside the organization who are experts in the fields represented by the proposal. Proposers are invited to suggest names of those they believe are especially well qualified to participate in the review and/or those they would prefer not participate. These suggestions may serve as one element of the reviewer selection process, at the program officer’s discretion. Program officers may obtain comments from assembled review panels or from site visits before recommending final action. Senior NSF staff further review recommendations for awards (NSF, 2001). Scientific review of a protocol should be particularly rigorous at the local level if the study will not be submitted for federal funding and/or will not be subjected to a peer review process similar to that of NIH or NSF. Scientific beliefs and biases, as well as competing interests by reviewers and the relationship of reviewers to investigators or chairs of departments, can affect the outcome of the review and should be considered in the selection of a scientific review mechanism. The responsibility for scientific review should not be left solely to a department chair, as he or she may lack sufficient time. The committee recognizes that it will not be practical or appropriate to subject every protocol to rigorous external peer review (e.g., a student-led research project in the social sciences). However, even in such cases, some level of internal scientific review should occur under the auspices of the departmental faculty based on a documented process that can be audited by an institution-level body. Finally, when commercial Research ERBs are called upon to review research protocols, their standard operating procedures should provide the mechanisms to ensure that scientific review of proposed research occurs and that their primary function remains focused on the ethics review and the integration of the scientific and financial conflict of interest review elements pertinent to the research. Currently, most protocols reviewed by independent IRBs undergo intensive scrutiny by the same group for both scientific merit and ethical safeguards, with reliance on external content experts as necessary. For multisite studies, the protocol may be submitted to the IRB directly from the sponsor, often before investigator selection.

OCR for page 70
Responsible Research: A Systems Approach to Protecting Research Participants FIGURE 3.2 Content Knowledge Intrinsic to Human Research Ethics The requisite expertise, or “specialized knowledge,” which the committee believes should be available to (or preferably found within) the membership of a Research ERB can be broken down into various levels of familiarity with the topics listed above. The depth of knowledge necessary in the particular areas correlates to the position of the issue within the pyramid, moving from the base to the top tier. The committee agrees that the composition of the Research ERB should reflect its ultimate function as the ethical review body for human research protocols. Science and ethics expertise should be appropriately balanced with each other and with noninstitutional member perspectives. Review boards could better avoid conflicts by including a greater proportion of individuals who are unaffiliated with any of the institutions/firms involved in the research study.

OCR for page 70
Responsible Research: A Systems Approach to Protecting Research Participants Research ERB Voting System Modern IRBs have tended to become larger as members have been recruited to reflect the broad range of scientific expertise needed to conduct informed scientific review of a diverse array of research protocols (OIG, 1998a). Consequently, some deliberations have the potential to be dominated by a scientific voice, leading to a corresponding reduction in ethics-centered reflection and the possible marginalization of the perspectives of nonscientific members (Cho and Billings, 1997; Peckman, 2001). Under the committee’s proposed refocusing of the ethics review, responsibility for the initial in-depth scientific review would lie elsewhere, enabling the Research ERB to focus on its primary functions of providing ethical review of proposed research and integrating the scientific and financial conflict of interest reviews. Consistent with this mission, the most important roles of the board chair are to constitute a proper and balanced membership, facilitate open discussion, and promote consensus. In fact, it is not required or always the case that board chairs themselves be trained physicians or scientists (Jacobs, 2001).26 Recommendation 3.6: A Research Ethics Review Board’s deliberative process should aim for consensus. If consensus cannot be achieved, approval of a protocol should require favorable votes by three-quarters of the voting members. If the membership of the Research ERB is modified to reflect the increased number of unaffiliated and non-scientific perspectives as suggested in Recommendation 3.5, but a simple majority voting procedure remains in place, no actual mechanism or basis for change will have been established and only the appearance of change will have been created. Such a scenario would tend to divide the deliberative body, encourage power blocks and leave the less powerful perspectives in the minority. Therefore, an alternative to standard parliamentary process is needed. Seeking consensus can be a way of acknowledging, exploring, and managing conflict and is a process that would facilitate the full expression of the minority perspective(s) and provide the basis for substantive discussion and debate. It also requires chairs who are skilled in techniques of mediation and conflict resolution. The process of facilitating consensus requires skills that include listing the players/parties who represent competing interests, characterizing their stakes and goals, identifying and maxi- 26   In New Zealand, the chair of the health research ethics board must be a “lay person.” Similarly, the United Kingdom requires that either the chair or vice chair be a lay person (Bastian, 1994).

OCR for page 70
Responsible Research: A Systems Approach to Protecting Research Participants mizing the options that might be employed to accommodate competing interests, and reaching agreement on this new solution. The committee acknowledges that sometimes mediation and the search for consensus will fail and that at some point a vote may need to be taken. The committee therefore recommends that the Research ERB adopt super-majority rules stating that no protocol will be approved unless it has the affirmative endorsement of at least three-quarters of the voting members present. A vote of unanimity would be unworkable and would effectively give a veto to a single dissenting committee member, while allowing a simple majority to approve a protocol in the face of substantial minority opinion to the contrary poses too much risk of suppressing respectable and responsible ethical opinion. Research ERB Member Education The need to educate board members has been discussed by every major national body charged with the review of human research protections, with each clearly stating the need for additional resources and educational programs (ACHRE, 1995; National Commission, 1978; NBAC, 2001b; President’s Commission, 1983). In addition, a 1998 survey documented strong support for increased board member education (Bell et al., 1998). Training programs at the national, regional, or institutional level could provide education to Research ERB members about the required core body of knowledge (Box 3.4 expands this point with a biomedical focus). Although the specific content will vary based upon the portfolio of the Research ERB, the basic elements will remain constant. Currently, Public Responsibility in Medicine and Research (PRIM&R) offers a course known as “IRB 101” to train board members. PRIM&R, in cooperation with OHRP, also has produced a CD-ROM version of “IRB 101.” NIH offers a computer-based training course27 designed for NIH board members but also accessible to the public. In addition, some universities include ethics training in research design and history of science courses. The inclusion of the perspectives of individuals not affiliated with a particular research organization in all decision-making and oversight bodies, particularly the Research ERB, is a vital component of any protection system. Specialized training for these members can help maximize their contribution to the process. Such training could include the following: 27   Available at ohsr.od.nih.gov/irb_cbt.

OCR for page 70
Responsible Research: A Systems Approach to Protecting Research Participants Box 3.4 Basic Knowledge for Research ERB Members Background: Ethical perspectives on the research enterprise, history of human research, current structure and funding of research, identifying who conducts and pays for research, setting the research agenda, methodological issues, current regulatory structure and the Common Rule. Foundation Knowledge: An introduction to clinical medicine, basic science, epidemiology, introduction to ethical principles, concepts, and issues, and skills, such as reading scientific literature and research protocols. Methodological Issues: Identifying the elements of research design, thinking about statistical power without statistical or mathematical training, the use and abuse of the placebo model, randomization before or after consent, and adequate use of animal models and competent persons before investigations with vulnerable or incapacitated persons. Difficult Design Questions: Questions of placebo arms in general, in psychiatry protocols, and as required by the FDA for review when effective treatment exists; testing of “me too” drugs when effective treatment exists and when the desire is for a more or a less expensive treatment; stratification of subjects and treatments; use of children, fetuses, or prisoners, whose involvement in research is governed by a specific sub-set of federal regulations; investigations with incapacitated persons whose use in research is subject to state law and regulation and to institutional policy in the absence of clear federal policy. Conflicts of Interest: How to think about the relationship of institutions, investigators, and participants in light of the monetary and status benefits that institutions and investigators can expect from the research, and how to manage and minimize possible conflicts. Cultural Competency: How to understand the design issues that will either encourage or discourage participation by persons of color and other vulnerable populations that have a history of being suspicious of the research enterprise. Reviewing Research Proposals—The Intellectual Calculus: Determine if a proposal presents an important question; identify who is included and excluded and why—women, children, racial minorities, prisoners, and other vulnerable populations; the risk-benefit calculus; informed consent; conflicts of interest; confidentiality; monitoring and review. Overarching Issues: International research, genetic and stem cell research, AIDS research, research with children, research in the name of national security. The Responsibilities of Research ERB Members: Ethics review, protocol assessment, research participant advocacy, accountability, community perspective.

OCR for page 70
Responsible Research: A Systems Approach to Protecting Research Participants A detailed description of the process of research, the identity and roles of all who are involved, and the components of a research study; A description of the process within a specific institution, including scientific review and conflict of interest review; Rules of scientific ethics (and broader theories of ethics as well). It is preferable that these training programs be designed, funded, and owned by consumer or community oriented nonprofit organizations. However, collaborations of such organizations with scientific trade associations or research organizations might also be sufficient to provide these programs. Several programs developed by consumer groups provide intriguing models for supporting and training unaffiliated members on Research ERBs or any other research oversight committee, such as programs developed for medical specialty boards by the Citizen Advocacy Center or those targeted to the research setting by Project LEAD (Leadership, Education, and Advocacy Development) and the National Alliance for the Mentally Ill (CAC, 1994; Cowdry, 2001; Dickersin et al., 2001; Hinestrosa, 2001; NAMI, 2001; Swankin, 2001). Certification of Research ERB Professionals Certification for Research ERB members (and investigators) has been a problematic issue, in part because they are predominantly scientists and traditionally belong to subspecialty professional organizations. However, IRB administrators are now being certified by the Council for Certification of IRB Professionals; through the National Association of IRB Managers; and as Certified IRB Professionals, through the Applied Research Ethics National Association in conjunction with the Professional Testing Corporation. The committee supports these efforts because they encourage the development of professional staff who can facilitate the ethics review function of the Research ERB. ORGANIZING AND INTEGRATING THE REVIEW PROCESSES In this chapter, the committee has proposed that separate and independent scientific review and conflict of interest review mechanisms should be available to the Research ERB, and that relevant findings from those reviews should be communicated to the Research ERB to inform its comprehensive ethical assessment. It is the responsibility of the program to ensure that this process occurs. Under this new vision, Research ERBs would review the research protocol, the written and signed findings of the scientific review and financial conflict of interest review bodies, and the oral

OCR for page 70
Responsible Research: A Systems Approach to Protecting Research Participants reports given by members of these bodies as appropriate. In the final stage, the Research ERB would integrate this information and determine the acceptability of the proposed research. A critical factor in making this three-pronged review process successful is to ensure that it is not unduly extended, complicated, or confused by these improvements. Participant interests are not served if research protocols are unnecessarily delayed. Therefore, in establishing review mechanisms, it is important to remain responsive to incoming applications by aligning submission and meeting dates for the preliminary reviews with the Research ERB schedule. Furthermore, communication between investigators and the review structure should continue to originate with the Research ERB office, except in instances of identified problems at the initial review levels, in order to prevent duplication of effort and avoid confusion resulting from overlapping and possibly contradictory messages to investigators about a protocol’s status. Likewise, efforts should be made to align and consolidate other administrative procedures within the HRPPP—e.g., the creation of one master application for the submission of protocols to all appropriate review bodies.28 If the functional integration of review activities is to successfully improve protection, it must be mirrored by a similar integration of administrative activities. Review of Multisite Studies In addition to ensuring coordination and communication between the various review groups, special consideration should be given to protocols conducted at more than one site, and possibilities for alternative review mechanisms should be considered. Recommendation 3.7: The review of multisite trials should be streamlined, as allowed by current regulations. One primary scientific review committee and one primary Research Ethics Review Board should assume the lead review functions, with their determinations subject to acceptance by the local committees and boards at participating sites. The rapid growth of multisite studies, particularly in the clinical trials arena, brings a new complexity that challenges the ability of review committees to meet their responsibilities efficiently and effectively. Multisite protocol review has become a cumbersome and labor-intensive process, because in most cases each research organization’s board considers the 28   For example, Children’s Hospital Regional Medical Center in Seattle has devised a combined electronic application for the purposes of the General Clinical Research Center Advisory Committee and the IRB. In this instance, the electronic format will enable investigators to complete relevant information and avoid irrelevant questions, through a series of drop boxes. The form will be posted at www.seattlechildrens.org/research after October 2002.

OCR for page 70
Responsible Research: A Systems Approach to Protecting Research Participants same protocol; performs the same risk assessment; examines the same, or similar, consent form; and later reviews the same, often voluminous, set of adverse event reports. In addition, multisite review can introduce considerable variability into the approvals and/or required modifications to study design or disclosure language, which actually detracts from participant protections. At best, it is an inefficient use of time and money for the review boards, investigators, and sponsors. The time involved in these reviews could be better spent attending to the participant protection needs of other studies or, in the case of the investigator, interacting with participants. In general, the review of multisite studies by each organization participating in a study does not significantly increase the level of protection provided to research participants or enhance the scientific design of the protocol. In fact, this repetitive review can be detrimental to the protection of the other research participants within a program, as the review board is unable to provide the needed time, resources, and expertise to thoroughly evaluate other protocols. Current federal regulations and guidance for IRBs contain provisions for the sharing of oversight responsibility with institutions in which regular collaboration takes place—including the ceding of authority for the reviews.29 In 2001, NBAC recommended that alternatives be considered for multisite review, including the use of central or lead review boards (2001b). In a multisite project, the sites could designate an independent Scientific Review Committee (SRC) and Research ERB or could designate one of the SRCs and Research ERBs as the “primary SRC and Research ERB.” The organization that has primary responsibility for obtaining the assurances should also assume the responsibility for acting decisively should violations occur, including termination of the study or the site and/or reporting violations and violators to authorities. In FDA-regulated trials, it should not be assumed that the industry sponsor has primary responsibility for the program; it would be preferable for the medical institutions involved to share that responsibility, since they are most directly and closely involved with the research participants. In addition, determinations regarding potential financial conflicts of interest should be forwarded to the lead Research ERB by the appropriate entity (i.e., the party responsible for the oversight of an investigator’s role in a project). Sponsors and federal regulators should encourage such collaboration and centralization, and local committees should reserve the right to refuse the primary review body’s determination for serious safety concerns and unique local requirements. Under this system, the opportunity and responsibility for locally appropriate oversight would continue to be intrinsic to protocol review, and 29   45 CFR 46.114, 21 CFR 56.114.

OCR for page 70
Responsible Research: A Systems Approach to Protecting Research Participants some institutions might insist on being involved in the review process for reasonable instances of indemnification and confidentiality. The committee believes, however, that flexible approaches might provide helpful alternatives to reduce redundancy that does not enhance participant protections. Effective use of communication tools, such as the use of protocol Web sites that would enable local Research ERBs to follow the lead Research ERB’s progress, will be fundamental to successful collaboration. Sharing Responsibilities: Alternative Review Models Models exist in the United States for sharing program responsibilities among collaborating institutions and their Research ERBs—for example, MACRO and BRANY.30 Similarly, the United Kingdom relies on regional committees for review of multisite research, and Denmark handles multisite studies by assigning the review responsibility to a lead committee (Alberti, 2000; Holm, 2001). These approaches can reduce duplicative workload and assure that reviews take place in settings that can bring to bear the appropriate scientific and ethical expertise. For example, complex protocols may involve consulting with biostatisticians, epidemiologists, and clinical specialists who might not be available at some individual sites. The ability to distribute costs could also place a regional program in a better position to provide the resources and infrastructure needed for various functions, such as maintaining qualified monitors for higher-risk research involving human participants. Furthermore, by ceding certain responsibilities to a regional unit, local programs could direct their efforts and resources to the remaining single-site studies for which they are responsible. This could be particularly useful to research organizations that have few resources, including small academic centers and community hospitals. In addition, regional or centralized review could provide a cost-effective alternative to smaller institutions and study sites that cannot afford to maintain a sufficiently comprehensive program onsite. Such organizations, for example, may find it difficult to sustain Research ERBs with the associated increased costs for training, monitoring, and, should it become a standard of practice in the community, accreditation preparation and the asso- 30   MACRO is a collaboration between Baylor College of Medicine, University of Alabama at Birmingham, the University of Pennsylvania School of Medicine, Vanderbilt University, and Washington University School of Medicine. See ccs.wustl.edu/macro/aboutmacro.htm for more information. BRANY focuses on sites near New York City. The Alliance serves more than 100 sites in multiple states, from New Jersey to Hawaii. For more information, see www.brany.com.

OCR for page 70
Responsible Research: A Systems Approach to Protecting Research Participants ciated fees. Possible examples include smaller hospitals, community-based organizations, local health agencies, and CROs. Regional programs/Research ERBs offer a solution to another set of problems encountered by boards today—the difficulty of recruiting and maintaining active and involved members. Many recent reports have noted the problems of heavy workloads, lack of professional or personal reward for committing the time needed to do the work, problems of ascertaining and managing conflict of interest within small organizations that may have a limited choice of potential board members, and a host of similar problems (IRB teleconference;31 Levine, 2001a,b; Moreno et al., 1998; OIG 1998a). Of course, regionalized approaches would need to include an equitable fee or honorarium structure to cover the expected services of board members and of expert consultant reviews needed in specialized topic areas and would need to extend liability insurance coverage to all involved in the process (see Chapter 6). Utilizing regional Research ERBs would also alleviate the potential conflict of interest inherent to the use of internal review boards. To encourage institutions conducting or sponsoring research to develop shared arrangements or use regional programs and to make Research ERBs more comfortable with ceding authority (but not ultimate accountability), adequate communication between local and regional programs will need to be assured. Research organizations should be kept abreast of the status and progress of studies, adverse events, media coverage, and other matters relevant to the organization, while keeping the reporting burden reasonable for the regional programs. Obviously, such a scenario could draw on the experience of the independent IRB review model. RESOURCE NEEDS The most important aspects of the review process depend on the experience and commitment of those who conduct the reviews. The processes of protecting human participants, which require skill, learning, wisdom, and sufficient practical experience to make complex judgments, are not tasks that can be assigned exclusively to junior faculty members or unskilled researchers. Time and focus are needed to accomplish these tasks, and if their difficult and demanding nature is to be taken seriously, the members—and not just the administrative staff—of scientific review groups and Research ERBs should be compensated for their efforts. This compensation may be monetary, may support academic promotion, or may provide release time from other duties. If the ethical review of research involving human participants is to be adequately conducted, appropriate resources 31   See Appendix A, “Methods,” for more details about the teleconference.

OCR for page 70
Responsible Research: A Systems Approach to Protecting Research Participants should be committed to the process. These considerations apply equally to those involved in scientific review, financial conflict of interest review, and ethics review of proposed research. Need for a Common Body of Knowledge Recommendation 3.8: The Office for Human Research Protections and the National Human Research Protections Advisory Committee should convene conferences and/or establish working groups to develop and disseminate best practices, case presentations, and conference proceedings for use by local protection programs and their Research Ethics Review Boards. Intellectual and educational resources are as important as financial resources in ensuring protections. The current protection system emphasizes local control and archiving of collective wisdom in the decision-making process for approving human research protocols. A more systematic literature and “case presentation” approach is needed for educational purposes and to promote consistency and high-quality decision making within Research ERBs. The lack of consistency among boards regarding, for example, the interpretation of the definition of minimal risk or investigator education requirements, can lead to contradictory—and often unproductive—directives to investigators that detract from time interacting with participants and overseeing research staff (investigator teleconference).32 Steps should be taken to improve the consistency of review board practices and regulatory interpretations through the dissemination of best practice guidelines; consensus conferences involving board chairs, investigators, ethicists and participant groups; and eventually, the building of a rich database of case dispositions analogous to case law in the legal arena. Such activities could also be useful in the exploration of complex ethical situations, such as research with children. The National Human Research Protections Advisory Committee (NHRPAC)33 has begun to address this void within the research review community through working groups focused on social and behavioral sciences and research with children.34 However, their work alone cannot be expected to meet this need. The availability of such materials would assist Research ERBs and other review bodies in making difficult decisions by aggregating institu- 32   See Appendix A, “Methods,” for more information. 33   As this report went to press, it was reported that DHHS had disbanded NHRPAC (Weiss, 2002). Committee discussion and recommendations directed to NHRPAC should now be directed to any future advisory body constituted to address participant protection issues. 34   See ohrp.osophs.dhhs.gov/nhrpac/nhrpac.htm for further information about the National Human Research Protections Advisory Committee.

OCR for page 70
Responsible Research: A Systems Approach to Protecting Research Participants tional experiences and creating precedent cases. Although it is anticipated that the continued development and implementation of accreditation programs will contribute to the dissemination of best practices, this mechanism cannot substitute for the recommended deliberations at the national level. OHRP and NHRPAC, with input from the FDA to ensure that issues particular to drug, biologic and device research are incorporated, should initiate such efforts forthwith. SUMMARY The adequate review of research to ensure that human research participants are protected involves the evaluation of several factors that require specific expertise. The IRB structure should be redefined and renamed the “Research ERB” to assert its mandate to conduct ethics reviews on protocols on behalf of those who will eventually be enrolled in studies. However, the newly configured Research ERB cannot be expected to carry out all of the specialized tasks required of a comprehensive protection program. Therefore, when appropriate to the research risks and context, scientific and financial conflicts of interest review should occur through distinct mechanisms that feed into and inform the Research ERB’s comprehensive ethical review of research. In order to carry out the responsibilities that should be under the purview of the Research ERB, all members should have core knowledge regarding human research ethics. In addition, at least 25 percent of the body’s membership should have no affiliation with the institution, not be trained as scientists, and be able to represent the local community and/or participant perspectives. To ensure that no study goes forward if a substantial portion of the Research ERB objects, no protocol should be approved absent three-quarters of the voting members’ agreement. One of the primary responsibilities of the Research ERB is the review of consent forms and the informed consent process. The board should ensure that the consent forms convey information relevant to the participant’s decision about whether to enroll and limit, or preferably delete, any language that would serve only to protect the institution. Robust, ethical review by the Research ERB could be enhanced by better employing risk-stratification, allowing boards to deal with minimal risk studies efficiently and devote more attention to high-risk studies. The review of multisite protocols could be streamlined by designating lead scientific review and ethics review committees, whose judgments would be subject to acceptance by the local review boards (this would include the sharing of conflict of interest determinations with the lead Research ERB). A common body of knowledge and guidance should be developed at the national level to assist review boards in their deliberations and promote consistency among the research oversight community.