Click for next page ( 166


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 165
System Issues THE U.S. SYSTEM for human research participant protection in- volves many components researchers, participants, research institutions, institutional review boards (IRBs), regulatory agen- cies, funding agencies, statistical agencies, professional associations, and others. The system is dynamic; it evolves as social, economic, and cultural changes affect various system components, and as they in turn respond sometimes reactively, and sometimes with forethought and care about how to improve the operation of the system. In the future as in the past the system should have two goals: first and foremost, the protection of the rights and welfare of research vol- unteers and, second, the facilitation of ethically responsible research that may result in useful knowledge about humans and human soci- eties. The different actors in the system at times have different per- spectives on how best to achieve these goals. Our report addresses some of these differing perspectives on the appropriate procedures for informed voluntary consent, methods and policies for protecting con- fidentiality, assessment of risk, harm, and benefit, and review proce- dures for minimal-risk research. Because of our charge and expertise, we have addressed these issues for research in the social, behavioral, and economic sciences (SBES); however, we believe that many of our recommendations are also relevant to biomedical research and mul- tidisciplina~y research that uses such methods as laboratory experi- ments, surveys, unstructured interviews, participant observation, and secondary analysis of existing data. In this concluding chapter we consider system-level issues that we believe need continued attention. By system-level issues, we mean matters that involve the organization of components of the participant protection system and the relationships among actors in the system. The issues that we address fall into five broad categories: (1) guid- ance and support for IRBs; (2) qualifications and performance stan- dards for IRBs and researchers; (3) communication among IRBs and researchers; (4) organization of and among IRBs; and (5) the devel- opment of national policy for human research participant protection. With limited time and resources, our discussion of these topics is lim- 165

OCR for page 165
166 PROTECTING PARTICIPANTS AND FACILITATING SOCIAL AND BEHAVIORAL SCIENCES RESEARCH ited. We highlight and endorse relevant recommendations of other groups and offer additional recommendations in a few areas that are particularly important for human participant protection in SBES re- search. GUIDANCE AND SUPPORT FOR IRBS Developing OHRP Guidance for IRBs Throughout our report we stress the need for the Office for Hu- man Research Protections (OHRP) to develop authoritative guidance for IRBs in several areas: appropriate ways to obtain and document in- formed consent for different types of SBES research (Chapter 4), good practices for confidentiality protection (Chapter 5), and effective re- view of minimal-risk research (Chapter 61. The process for developing such guidance will be easier in some areas than in others. The most difficult area, we believe, relates to guidance on effec- tive review of minimal-risk research that is, guidance on applying the definitions of research and involvement of human subjects and when it is appropriate to exempt research or to conduct an expedited rather than full board review. To be helpful and compelling for IRBs, such guidance needs to include concrete examples for a variety of research topics and methods. It will not be easy to develop guidance for effective review of mini- mal-risk research in SBES fields for two reasons. The first impediment is the range of SBES research disciplines, topics, and methods, each of which presents somewhat different issues for human participant pro- tection and the determination of minimal risk. A second impediment is that, unlike the case of some minimal-risk biomedical procedures (e.g., drawing blood below a specified amount), there are no evidence-based classifications of the risk level for specific SBES procedures (e.g., spe- cific survey questions). Without such evidence, views may differ on, for example, whether and under what circumstances a survey question about alcohol use is minimal risk or more than minimal risk: When does such a question cause, at most, only temporary anxiety or embar- rassment, and when may it cause longer lasting psychological trauma? Both the Institute of Medicine (IOM) committee and the National Bioethics Advi- sory Commission (NBAC) support the concept of review commensurate with risk: see Institute of Medicine (2002:Executive Summary) and National Bioethics Advisory Com- mission (2001:Rec.2.5). The IOM committee also recommended (Rec.3.3) that "the Office for Human Research Protections, with input from a broad spectrum of research disci- plines and participant groups, should coordinate the development of guidance for risk classification."

OCR for page 165
SYSTEM ISSUES 167 Currently, IRBs are under pressure to be as risk-averse as possible. Without specific guidance, they are likely to assume the worst; how- ever, in so doing, they may needlessly add to their workload and im- pede useful, ethically responsible research. We suggest that OHRP, relevant professional associations, investi- gators, IRBs, and other interested groups regard the development of specific guidance as a long-term process that is carried out in a se- ries of incremental steps.2 Because the object is to develop guidance and not regulations, the process need not be as cumbersome or time- consuming as the regulatory process. One possible mode of operation is for OHRP to establish liaisons with relevant professional associa- tions. Each association would be charged, in turn, to work with its members to develop specific examples to include in guidance about what is and is not research involving human subjects and what should be exempted or receive expedited review.3 OHRP should issue such ex- amples for public comment as soon as they are available, not waiting for all possible examples from all relevant associations. It would also be very useful if SBES research funders were to spon- sor research on risks posed by various SBES procedures, such as re- search on the circumstances in which participants believe that labora- tory experiments or survey questions pose risks of temporary or longer- lasting harms of various types. Such research would be very helpful in enabling IRBs (and researchers) to make decisions about risk levels on the basis of science and not unsupported judgment (see also Recom- mendation 4.1 about the need for research on the effectiveness of alter- native informed consent procedures and Recommendation 6.3 about the need for researchers to debrief participants about perceived and actual risks and harms). Supporting IRBs The human research participant protection system has developed as an add-on to the scientific research enterprise in the United States, without explicit recognition of the need to provide adequate financial support for IRB operations and adequate rewards for IRB service. Both the IOM committee and NBAC make strong recommendations 2 participant groups should also be involved in the development of guidance on minimal-risk and other pertinent issues, although participant involvement may be dif- ficult to obtain because SBES research covers a wide range of populations as distinct from a group of patients with a specific disease as is characteristic of much clinical re- search. 3 Currently, SBES professional association guidelines on human research participant protection are addressed to researchers, not to IRBs.

OCR for page 165
168 PROTECTING PARTICIPANTS AND FACILITATING SOCIAL AND BEHAVIORAL SCIENCES RESEARCH for adequate funding for IRBs. The IOM report (Institute of Medicine, 2002:Rec. 2.3) states: Research sponsors and research organizations public and private should provide the necessary financial support to meet their joint obligation to ensure that Human Research Participant Protection Programs have adequate resources to provide robust protection to research participants. NBAC is even more explicit. Its report recommends (National Bio- ethics Advisory Commission, 2001:Rec. 7.1) adequate funding for fed- eral operations to oversee human research participant protection; in- clusion of separate allocations in federal research programs for over- sight activities; allowing research organizations to request funding for IRB and other protection activities; and additional funding from fed- eral agencies, other research sponsors, and research organizations for IRB and other protection activities. We agree with the intent of these recommendations, noting that there may be different mechanisms to achieve adequate funding for IRB and other protection activities (e.g., allowing direct charges to grants and contracts or allowing such charges to be included in in- direct costs). In addition, we encourage research organizations to provide adequate financial and nonfinancial recognition for service on IRBs so that such service is viewed in a positive way and not as an unmitigated burden. In addition, universities and other research or- ganizations should establish methods for evaluating the quality of IRB service and honor exemplary service by IRB members and chairs. QUALIFICATIONS AND PERFORMANCE STANDARDS Education and Training Because of the work of many groups, including the National Sci- ence Foundation, the Office of Behavioral and Social Science Research in the National Institutes of Health (NIH), professional associations, individual IRBs, and individual researchers, there are now some ma- terials available to help SBES researchers understand their responsi- bilities for human research participant protection and to learn how to navigate the maze of relevant regulations and procedures for project approval (see Appendix B; see also Oakes, 20021. There is much less guidance for IRBs or for researchers in other fields on how to han- dle different types of SEES research methods and populations stud- ied. The available guidance for IRBs from OHRP and other sources is

OCR for page 165
SYSTEM ISSUES 169 mostly general in nature, and training modules for researchers offered by NIH are oriented to clinical biomedical research. Education and training for both IRBs and researchers can be improved. Looking to the future, we see three main concerns that are specific to SBES research. First, training materials are needed for IRBs that review SBES research. IRBs need in-depth discussions and analysis of the varieties of SBES research, issues that different types of research raise for human research participant protection, and effective ways to address those issues. The development of appropriate training mate- rials for IRBs will require the joint efforts of OHRP, relevant SBES professional associations, and IRBs that are experienced in reviewing SBES research.4 Second, in the short term, training for SBES researchers of neces- sit~r must involve short courses at professional association meetings, on-line self-study modules, and the like. The training mechanisms should focus on specific issues of participant protection in different types of SBES research, such as appropriate consent processes for special populations, as much as possible. Such training should also reinforce the obligations of researchers to contribute to the participa- tion protection system in every way possible from submitting proto- cols for review that are ethically responsible and likely to contribute to useful knowledge to being willing to serve on IRBs. Third, for the longer term, research institutions, particularly in their graduate programs, will need to develop ways to instill principles, practices, and responsibilities for ethical research as part of the ba- sic professional education of SBES (and biomedical) researchers. The means to do so could include ethics modules as part of undergradu- ate courses, ethics courses as part of graduate training, and in-service courses for researchers to refresh understanding and address new de- velopments in human research participant protection. Many advisory bodies and commentators have highlighted the im- portance of education for IRBs and researchers in their recommen- dations (e.g., Advisory Committee on Human Radiation Experiments, 1996:Rec. 9; National Bioethics Advisory Commission, 2001 :Rec. 3.11. For example, the Committee on Assessing the System for Protecting 4The Social and Behavioral Sciences Working Group, which is continuing to operate even though its parent National Human Research Protections Advisory Committee lost its charter in October 2002 (see Appendix B), is planning an activity in July 2003 on good practices for IRBs for the review of SBES research protocols The activity will include a workshop followed by preparation of a document that is intended to help train IRB members (Felice Levine, American Educational Research Association, 2002, personal communication).

OCR for page 165
170 PROTECTING PARTICIPANTS AND FACILITATING SOCIAL AND BEHAVIORAL SCIENCES RESEARCH Human Research Participants of the Institute of Medicine (2002:Rec. 2.4) recommended:5 Research organizations should ensure that investigators, Institutional Review Board members, and other individu- als substantively involved in research with humans are ade- quately educated to perform their respective duties. The Of- fice for Human Research Protections, with input from a va- riety of scholars in science and ethics, should coordinate the development and dissemination of core education elements and practices for human research ethics among those con- ducting and overseeing research. Accreditation One response to heightened concern about the adequacy of the IRB system for protecting human research participants has been a call for greater oversight of IRBs by the federal government (see, e.g., Once of Inspector General, 1 998b; legislation introduced by Senator Kennedy). Another response has been for private groups to develop voluntary ac- creditation programs for IRBs. The IOM committee devoted its entire first report to accreditation issues (Institute of Medicine, 20011. The committee's final report (Insti- tute of Medicine, 2002) also contains several recommendations about accreditation, including Recommendation 6.4, which calls for contin- uation of efforts to develop voluntary accreditation programs: Voluntary accreditation should continue to be pilot tested as an approach to strengthening human research partici- pant protections. The Department of Health and Human Services should arrange for a substantive review and evalu- ation of the accreditation process after five years, to be con- ducted under the purview of an independent entity. The final report of the NBAC (2001) goes even further. It recom- mends not only that "Sponsors, institutions, and independent Institu- tional Review Boards should be accredited in order to conduct or re- view research involving human participants.... " (Rec. 3.4), but also that "All investigators, Institutional Review Board members, and Insti- tutional Review Board staff should be certified prior to conducting or sThe TOM report also recommends training for research participants so that they "understand their potential role in any study in which they enroll, the rationale underly- ing that study, and importantly, what is required of them to prevent unanticipated harm to themselves and to maintain the scientific integrity of the study" (Institute of Medicine, 2002:Rec. 4.2).

OCR for page 165
SYSTEM ISSUES 171 reviewing research involving human participants.... " (Rec. 3.31. The report asserts (p. xiv) that "Although accreditation and certification do not always guarantee the desired outcomes, these programs, which generally involve experts and peers developing a set of standards that represents a consensus of best practices, can be helpful in improving performance." While not taking a position at this time on the benefits and costs of accreditation or certification programs, we support the IOM recom- mendation for continued testing and review of IRB accreditation pro- grams. We offer two additional points. First, accreditation programs should involve researchers from the range of SBES (and biomedical) disciplines and take cognizance of appropriate review practices for dif- ferent types of SBES research methods and populations studied. In this regard, it is encouraging to note that the Association for the Ac- creditation of Human Research Protection Programs added seats to its board for SBES researchers and has pilot-tested accreditation proce- dures at research institutions that conduct SBES research. Second, accreditation programs need, so far as possible, to empha- size the spirit and not just the letter of the Common Rule regulations. For example, accreditation should focus on how IRBs review propos- als to assure an appropriate process for obtaining informed consent and not just the documentation of such consent (see Chapter 41. In ad- dition, accreditation programs should have standards for review com- mensurate with risk (see Chapter 61. Conducting reviews commen- surate with risk helps IRBs allocate their limited resources to ensure protection of human participants while enabling responsible research to proceed. For example, a standard might be that IRBs should exempt a high proportion (specified range) of research that is eligible for ex- emption. Another standard might set different targets for the length of time to complete reviews depending on risk and the type of review conducted. COMMUNICATION AMONG IRBS AND RESEARCHERS IRB-Researcher Interaction Recommendation 7.1: To improve IRB-researcher commu- nication and facilitate the review process, IRBs should: clearly distinguish and justify changes to research pro- tocols that are required for human participant protec- tion from suggested changes that are advisory; and develop ways to work cooperatively with investigators, such as providing opportunities for face-to-face meet-

OCR for page 165
172 PROTECTING PARTICIPLE ED FACILITATING SOCIAL ED BEHAVIOR SCIENCES SEARCH ings to discuss significant changes in research proto- cols that the IRB requires. We believe that miscommunication between IRBs and researchers may be one of the reasons that SBES researchers have often been frus- trated with the IRB system, while IRBs, in turn, may sometimes be- lieve that SBES researchers are not paying sufficient heed to human participation protection requirements. Clear, open communication be- tween IRBs and investigators is needed to facilitate the preparation of research protocols that adequately describe participant protection procedures and the timely review of research protocols by IRBs. To the extent that researchers better understand the functions of and con- straints on IRBs and IRBs better understand researchers' concerns for maintaining the integrity of their research design and reaching closure on a timely basis, the smoother the review process is likely to be. Scientific Review The Common Rule (45 CFR 46.1 1 1a) charges IRBs to determine for each protocol that "risks to subjects are minimized: (i) By using pro- cedures which are consistent with sound research design...." This language is based on language first proposed by the Office for Protec- tion from Research Risks (OPRR) in NISI in August 1979 (see Box A-6 in Appendix A). The OPRR proposal also included a provision for IRBs to determine that "the research methods are appropriate to the objec- tives fof] the research and the field of study." This provision would have considerably expanded the technical review function of IRBs, but it did not survive the battle over the OPRR proposals (see Chapter 31. Anecdotal evidence (including the experience of panel members) suggests that IRBs sometimes require technical changes to SBES re- search designs that are not necessary for human participant protection and that go beyond the expertise of the IRB members. For example, an IRB may require a change in wording of a survey question without fully understanding the purpose of the question or the research that went into testing the proposed wording; the IRB may also overestimate the risk that the question poses to human participants. Quantitative evidence on this point is hard to find. According to the 1995 Bell survey (Bell, Whiton, and Connelly, 1998:Figure 40), IRB chairs reported that they rarely criticized research designs; most criti- cisms instead were of consent forms. Similarly, investigators reported that they were required to modify their proposed scientific design only 6 percent of the time and how they recruited participants only 11 per- cent of time. Most commonly (78 percent of the time), they were re-

OCR for page 165
SYSTEM ISSUES 173 quired to modify the consent form. In response to a subjective assess- ment, 56 percent of IRB chairs and 55 percent of IRB members agreed or strongly agreed that "this IRB's reviews improve the scientific qual- ity of research done on human subjects," but only 37 percent of inves- tigators shared this perception. None of the survey results is reported separately for SBES research. From these data, it is possible that SBES researchers are overstat- ing the propensity of IRBs to require inappropriate design changes that are not needed for human participant protection. Nevertheless, to minimize both the reality and the perception, we urge IRBs to clearly identif y and justify changes that the IRB requires on behalf of human participant protection and to offer other suggestions on a purely advi- so~r basis.6 Involvement of Investigators The Common Rule does not specify how IRBs should communicate with investigators except to require an IRB to inform the investiga- torks) "in writing of its decision to approve or disapprove the proposed research activity, or of modifications required to secure IRB approval of the research activity" (45 CFR 46.109d). The 1995 Bell survey found that 42 percent of low-volume IRBs routinely encouraged investiga- tors to attend IRB meetings in person or to be reachable by telephone. In contrast, only 17 percent of high-volume IRBs followed this policy. Similarly, our review of IRB websites at 47 large universities found that only 15 percent invite attendance by investigators (another 9 per- cent have investigators sit outside the meeting to be available to answer questions). We believe that IRBs should consider various ways to develop more open, less adversarial communications with researchers. Greater open- ness has the potential to facilitate understanding, resolve misunder- standings, improve the efficiency and timeliness of review, and build trust in the IRB system. Individual IRBs should consider the best means for improved communication, taking account of workload vol- ume and other factors. Such means could include opening part of 6The IOM committee goes further, recommending that research organizations estab- lish three related bodies for review of research protocols. One body would consider sci- entific issues; another body would consider financial conflict of interest; the two bodies would each make recommendations to a Research Ethics Review Board, which would have final approval authority but concentrate its own efforts on ethical issues (Institute of Medicine, 2002:Rec. 3.1, 3.2). We believe that such a structure requires careful con- sideration of its merits and costs before being considered for adoption. It is possible that such a structure could separate scientific review too much from ethical review or add more steps to the approval process.

OCR for page 165
174 PROTECTING PARTICIPANTS AND FACILITATING SOCIAL AND BEHAVIORAL SCIENCES RESEARCH IRB meetings to investigators (and, possibly, potential research par- ticipants) or scheduling face-to-face meetings for the IRB or IRB chair with the investigator to discuss significant changes to protocols. Clear Guidance from IRBs to Researchers To facilitate better communication of IRB expectations, so that re- searchers submit research protocols for review that fully address hu- man participant protection issues, we believe that IRBs should pro- vide clear guidance about what constitutes an acceptable protocol. A useful practice in this regard could be to post on websites outstand- ing examples of approved research protocols that meet high standards for participant protection in such areas as confidentiality protection and informed consent. Examples of protocols that were exempted or received expedited review, covering a variety of disciplines and meth- ods used, could also be posted. Currently the IRB websites of major research universities often do not provide any more guidance on these matters than is contained in the Common Rule itself. For example, only 13 percent of these IRBs provide guidance for requesting exemption that does not simply repeat the Common Rule list of eligible categories of research, and only 11 percent provide guidance on confidentiality protection. However, over half of these IRBs provide guidance on in- formed consent, and 45 percent provide on-line training modules or guide books.7 Another potentially useful practice could be to publish, on an ongo- ing basis, the titles, names of principal investigators, and review clas- sification (exempt, expedited, full board review) of projects approved by an IRB. Such information would allow researchers preparing pro- tocols to identify work similar to theirs, contact the principal investiga- tors, and otherwise capture the policies of their IRBs from past board decisions. Finally, IRBs should be clear when their standards for re- view are more stringent than the Common Rule. Appeals Process There is currently no provision in the Common Rule for investiga- tors to appeal an adverse decision from an IRB. Although an inves- tigator may argue back and forth with an IRB about changes to the research protocol, ultimately, the IRB's decision about what changes 7 See Appendix D; we have not evaluated the relevance and usefulness of on-line training and guidance.

OCR for page 165
SYSTEM ISSUES 175 must be made to secure approval is final.8 Backing up that authority is the provision in the Common Rule whereby an IRB may "suspend or terminate approval of research that is not being conducted in ac- cordance with the IRB's requirements or that has been associated with unexpected serious harm to subjects" (45 CFR 46.1131. The purpose of lodging ultimate approval authority with an IRB was to insulate IRBs from pressures from institutional officials to ap- prove research that might be important to the institution but that the IRB determined would not adequately protect participants. Finding ways to enable IRBs to resist pressures that could potentially compro- mise ethical judgments was and is a legitimate goal. However, some researchers believe that the lack of an appeals process gives IRBs too much power over the conduct of research. Further study is needed to find better ways to maintain IRB integrity and yet also allow re- searchers greater access to the decision process. The panel discussed the desirability of recommending that a formal appeals process be added to the human participant protection process. Such a provision would provide a recourse to an investigator who be- lieved that IRB-required changes would harm the scientific integrity of the proposed research but not provide added participant protection. Some panel members wanted to recommend an appeals procedure, but other panel members feared that such a procedure could add another layer of bureaucratic oversight to the IRB system. These members hope that implementation of Recommendation 7.1 to improve IRB-research interaction and other steps such as clearer guidance from IRBs to re- searchers and more explicit guidance from OHRP to IRBscould go a long way to alleviate the possible need for a formal appeals process. ORGANIZATION OF AND AMONG IRBS Organization and Staffing of IRBs The Common Rule requires that IRBs have at least five members with varying backgrounds, including at least one scientist and one non- scientist, and at least one member not affiliated with the research insti- tution, but it does not otherwise specify IRB composition, size, staffing, or other organizational features. The several thousand IRBs in the United States differ in their size, disciplines and expertise of members, level of staffing, and other resources for their operations. As one would The Common Rule states that "Research covered by this policy that has been ap- proved by an IRB may be subject to further appropriate review and approval or dis- approval by officials of the institution. However, these officials may not approve the research if it has not been approved by an IRB" (45 CFR 46.1 12).

OCR for page 165
176 PROTECTING PARTICIPANTS ED FACILITATING SOCIAL ED BEHAVIOR SCIENCES SEARCH expect, high-volume IRBs have more members, are supported by more hours of administrative staff, and make greater use of consultants, on average, compared with low-volume IRBs (see data from Bell, Whiton, and Connelly, 1998, in Chapter 21. Some research institutions have more than one IRB for heavy work- loads and to better align expertise of members with type of research reviewed. From our review of IRB websites of 47 major research insti- tutions, 13 percent of these institutions have two IRBs, and 23 percent have three or more, including one institution that has half a dozen IRBs, each assigned to a specific research area, operating under the umbrella of an executive committee. The Department of Veterans Affairs (VA) recently sponsored research to develop "optimal" staffing costs for operating human research par- ticipant protection programs at VA medical centers. From interviews with participant protection experts, data from the 1995 Bell survey (Bell, Whiton, and Connelly, 1998), and data from a VA system on number of studies involving human participants at VA centers, Wagner and Barnett (2000) developed cost models for a hypothetical medium- volume center (averaging about 370 total IRB reviews per year) and a hypothetical high-volume center (averaging about 1,380 total IRB re- views per year). The cost models considered staff requirements in light of workloads and assumed greater efficiencies in handling reviews by high-volume compared with low-volume IRBs. The models projected that a high-volume center should have four IRBs supported by an ex- tensive staff; a medium-volume center should have two IRBs supported by somewhat fewer staff. The study also concluded that regional IRBs should be put in place to handle reviews for two or more low-volume centers to ensure that appropriate expertise and experience is available in the review process. The VA study represents an interesting effort to cost out staffing levels for IRBs and to propose an optimal organization for IRBs at the VA's medical centers. However, the study noted that "an optimally staffed and funded IRB does not guarantee high-quality reviews," and it called for research on factors related to quality and how quality re- lates to cost (Wagner and Barnett, 2000:31. We agree that research on the relationship of quality, funding and staffing levels, and IRB organization is important. Without such research it is not obvious, for example, whether it is preferable to have separate IRBs for differ- ent research fields or to have a system in which there is some type of cross-disciplinary review of research protocols. Too narrow a fo- cus for an IRB may lead it to be too uncritically accepting of the re- search protocols it reviews, but too wide a focus is likely to mean that few (if any) of the IRB members have appropriate subject expertise.

OCR for page 165
SYSTEM ISSUES 177 Similarly, research on the role that administrative staff should play in helping the IRB process (e.g., not only specific tasks, but also amount of discretionary judgment) would be useful. The need for these and similar research studies on IRB staffing and organization is a strong argument for the kinds of data that we recommend be gathered on IRB operations and on the risks and harms of different types of SEES (and biomedical) research and for the research that we recommend on developing appropriate performance or quality indicators for the IRB process (see Chapter 61. Research Involving Multiple Sites An increasing number of research protocols involve investigators and data collection at more than one site. This is true not only for biomedical research (e.g., clinical trials that may enroll participants at dozens of sites), but also for many kinds of SEES research. Examples include evaluation studies of the effects of the 1996 welfare reform act in several cities by consortia of researchers at different locations and ethnographic studies of school violence at multiple sites. There are no easy answers to how best to protect human partici- pants in multisite studies. The IRB system was established in part to permit local community input the Common Rule requires that IRB membership be diverse in terms "of race, gender, and cultural back- grounds and sensitivity to such issues as community attitudes" (45 CFR 46.107a). The Common Rule also states that "each institution fin a mul- tisite project] is responsible for safeguarding the rights and welfare of human subjects" (45 CFR 46.1 141. Obtaining the concurrence of all involved IRBs can be time con- suming and frustrating because of differences in IRB practices and standards for project approval. Even more important, the integrity of a multisite research design may be jeopardized if, because of differing IRB requirements, there is insufficient uniformity of procedures across all sites at which participants are recruited. The Common Rule permits organizations involved in cooperative research projects to "enter into a joint review arrangement, rely upon the review of another qualified IRB, or make similar arrangements for avoiding duplication of effort," although such arrangements require the concurrence of the relevant federal agency (45 CFR 46.1 141. The IOM report recommends stream- lining the review of multisite clinical trials, but its recommendation does not really solve the problem of how to keep IRBs from second- guessing each other and declining to cede their review authority to another IRB.9 9The TOM recommendation states: "The review of multisite trials should be stream-

OCR for page 165
178 PROTECTING PARTICIPANTS AND FACILITATING SOCIAL AND BEHAVIORAL SCIENCES RESEARCH We do not have a ready solution to this problem. We believe it should be a high-priority issue for OHRP and the Secretary's Advi- so~y Committee on Human Research Protections. Currently, there are efforts under way to establish regional IRBs that deal with multisite research in the region. For example, several universities and hospi- tals that frequently conduct multisite research projects may establish a single IRB to review those projects. Experience gained from these IRBs may, over time, be helpful in determining effective structures and procedures for review of multisite research. DEVELOPING hL\TIONAL POLICY FOR HUMAN RESEARCH PARTICIPANT PROTECTION Leadership for National Policy Development Leadership in developing national policy and providing adequate oversight of human research participant protection is a fundamental obligation of the federal government given its role as a major research sponsor and its obligation under the Constitution to promote the gen- eral welfare of the population. Because of ongoing changes in society, cultural values, research techniques, and other factors, there is a con- tinuing need for the government to review and modify, as appropriate, its policies, guidance, and oversight with respect to participant pro- tection. In turn, the government needs to receive advice and recom- mendations on human protection issues from a continuing body that represents a range of relevant expertise and backgrounds and is in fact and in perception independent of political concerns or manipulation. In this spirit, the IOM committee recommended (Institute of Medicine, 2002:Rec. 7.11: Congress should authorize and appropriate funding for a standing independent, multidisciplinary, nonpartisan expert Committee on EIuman Research Participant Protections whose membership would include the perspective of the re- search participant. We support the IOM recommendation and suggest that ways to pro- mote independence of such a committee is to give it a long-term charter and use staggered terms for members. Furthermore, it could be useful lined, 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" (Institute of Medicine, 2002:Rec. 3.7).

OCR for page 165
SYSTEM ISSUES 179 to have members nominated by professional associations in biomed- ical and SBES disciplines, in addition to members appointed by the president and Congress. NBAC did not speak to the issue of an appropriately constituted ad- visory committee, but it recommended that an independent agency be created to lead and coordinate federal oversight in this area (National Bioethics Advisory Commission, 2001:Rec. 2.21: To ensure the rights and welfare of all research participants, federal legislation should be enacted to create a single, in- dependent federal office, the National Office for Human Re- search Oversight (NOHRO), to lead and coordinate the over- sight system. This office should be responsible for policy de- velopment, regulatory reform, . . . research review and mon- itoring, research ethics education, and enforcement. We were not charged to consider and do not take a position on the proper form or location of a federal agency for human participant protection. However, we believe it likely that an independent agency, with responsibility for developing unified, comprehensive federal regu- lations and guidance (see NBAC, 2001:Rec. 2.3), and advised by a con- tinuing, independent expert committee, would bring desirable qualities to federal activities in this area. Properly funded and organized, such an agency should be well positioned to provide leadership, involve ex- perts from the full range of research disciplines, bring the views of research participants to bear, and protect federal policy in this area from partisan concerns. Involvement of SBES Researchers in National Policy Setting Recommendation 7.2: Any committee or commission that is established to provide advice to the federal government on human research participant protection policy should repre- sent the full spectrum of disciplines that conduct research involving human participants. In particular, such a body should include members who represent the range of the so- cial, behavioral, and economic sciences. The IOM recommendation for an independent continuing advisory committee to the federal government on human research participant protection calls for the committee membership to be multidisciplinary. As the history of human participant protection policy in the United States indicates (see Chapter 3), SBES researchers have had signif- icant input to the policy development process, but they have had to

OCR for page 165
180 PROTECTING PARTICIPANTS AND FACILITATING SOCIAL AND BEHAVIORAL SCIENCES RESEARCH struggle to be heard. We believe strongly that, as a matter of course, the SBES research community should be involved along with the biomedi- cal community in providing policy advice to the federal government. The benefits of such involvement would include not only increased support for and understanding of human participant protection poli- cies among SBES researchers, but also useful cross-fertilization of ideas and knowledge between SBES and biomedical researchers about such topics as confidentiality protection and effective informed con- sent. Such cross-fertilization is increasingly important given the grow- ing interdisciplinary nature of much research today. We note that the Secretary's Advisory Committee on Human Research Protections (for the U.S. Department of Health and Human Services) was chartered in October 2002, replacing the National Human Research Protections Advisory Committee. The members of the new committee were an- nounced in January 2003. The committee includes two psychologists but does not represent other SBES disciplines nor does it include par- ticipant representation. CONTINUING SYSTEM EVOLUTION The U.S. human research participant protection system is multi- layered, requires the cooperation of many different components (each with somewhat different perspectives), and is continually changing as various forces affect one or another part of the system. Such complex- ity and change can be unsettling for IRBs and others in the system who want clear guidance about how to meet their responsibilities for human participant protection. At present, IRBs are under pressure to take a legalistic approach by ignoring the flexibility in the Common Rule that permits informed consent procedures to be appropriately tailored for protection of specific populations or that permits review commensu- rate with risk. In the area of confidentiality protection, IRBs and re- searchers may not be fully aware of technology-driven changes that are increasing the risks of disclosure of information about participants. We believe that the adoption of the recommendations in our re- port on effective informed consent, enhanced confidentiality protec- tion, and review commensurate with risk will result in better guidance for IRBs and researchers on these topics and facilitate cooperative in- teraction and more consistent application of policies and procedures. We expect that implementation will be gradual as evidence is obtained and guidance is developed that gives IRBs assurance to move forward and that addresses other challenges to the system. For example, it will take time to develop a confidentiality protection certification program

OCR for page 165
SYSTEM ISSUES 181 such as we recommend in Chapter 5, but, as it comes on line, the result should be greater access to research data that embodies appropriate, state-of-the-art protections to minimize the risk of harmful disclosure. We also recognize that no report represents the last word on what should be done and that continued evaluation and modification of guid- ance (and, sometimes, regulations) will be needed in the future. To help guide the human research participant protection system as it moves forward and to enable IRBs and others to cope constructively with change, we see two critical needs. The first is for an ongoing data col- lection program on the operations of IRBs and the conduct of research that can inform policy and the public, as we recommend in Chapter 6. The second, as we recommend above, is for an ongoing independent advisory body for the federal government that can bring to bear key perspectives from researchers in all relevant fields, including SEES fields, as well as from participants. These two initiatives are important not only to help the development of appropriate policy and guidance, but also to increase public trust in the ability of the U.S. system to pro- tect the many volunteers who make it possible to conduct research.

OCR for page 165
182 PROTECTING PARTICIPANTS AND FACILITATING SOCIAL AND BEHAVIORAL SCIENCES RESEARCH performance standards for IRBs and researchers; (3) communica- tion among IRBs and researchers; (4) organization of and among IRBs; and (5) the clevelopment of national policy for human re- search participant protection. With limited time ant! resources, our discussion of these topics is limited. We highlight ant! endorse rele- vant recommendations of other groups ant! offer additional recom- mendations in a few areas that are particularly important for human participant protection in SBES research. GUIDANCE AND SUPPORT FOR IRBS Developing OHRP Guidance for IRBs .;; Throughout our report we stress the need for the Office for Hu- man Research Protections (OHRP) to clevelop authoritative guid- ance for IRBs in several areas: appropriate ways to obtain anti document informed consent for different types of SBES research (Chapter 41; good practices for confi(lentiality protection (Chapter 5~; and effective review of minimal-risk research (Chapter 61. The process for developing such guidance will be easier in some areas than in others. The most clifficult area, we believe, relates to guidance on effec- tive review of minimal-risk research that is, guidance on applying the definitions of research and involvement of human subjects and when it is appropriate to exempt research or to conduct an expe- dited rather than full-board review.) To be helpful and compelling for IRBs, such guidance needs to include concrete examples for a variety of research topics and methods. It will not be easy to develop guidance for effective review of minimal-risk research in SBES fields for two reasons. The first impediment is the range of SBES research disciplines, topics, and methods, each of which presents somewhat different issues for hu- man participant protection en c] the determination of minimal risk. A second impediment is that, unlike the case of some minimal- risk biomedical procedures (e.g., drawing blooc! below a specified amount), there are no evidence-based classifications of the risk level Moth the IOM committee and NBAC support the concept of review commen- surate with risk: see Institute of Medicine (2002:Executive Summary) and National Bioethics Advisory Commission (2001:Rec. 2.51. The IOM committee also recom- mended (Rec. 3.3) that "the Office for Human Research Protections, with input from a broad spectrum of research disciplines and participant groups, should co- ordinate the development of guidance for risk classification."