Click for next page ( 60


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 59
Regulatory History THERE ARE MANY ACCOUNTS that trace the evolution of U.S. poli- cies and regulations for human research participant protection in federally funded research generally and in experimental bio- medical research specifically. For example, the report of the Advisory Committee on EIuman Radiation Experiments (1996) provides a very informative history of regulation and human protection issues in ra- diation experiments and other biomedical research, covering not only the U.S. Department of Health and Human Services (DHHS), but also the U.S. Departments of Defense, Energy, and Veterans Affairs. But nowhere is there a comprehensive history of human participant pro- tection in social, behavioral, and economic sciences (SBES) research. Problems in such research have generally not received as much media attention as those in clinical trials and other biomedical experiments; also, the diversity of research questions and methods under the um- brella of SBES research is so broad as to make it difficult to construct a thorough-going history (see Chapter 21. fine are not able to make up for the lack of a comprehensive his- to~y here. Our summary below focuses on key events in federal policy making and regulation for SBES research from the end of World War II through the present.) Our discussion identifies five major periods, the end dates of which mark significant events in the development of regulations, major disputes, and media attention. The history of federal regulation for human research participant protection shows relatively less emphasis on issues of protecting the confidentiality of information from individual respondents, in contrast to the attention devoted to such issues as the definition of research in- volving human participants and the elements of informed consent. Im- ~Key sources that we consulted for this history include Advisory Committee on Human Radiation Experiments (1996:Ch. X); Beauchamp et al. (1982:Ch. 1); Gray (1982); McCarthy (1998); National Bioethics Advisory Commission (2001:App.C). Mc- Carthy (1998) covers attitudes toward and problems with human research participant protection in the United States prior to 1945. See Appendix A for changes in regulatory language from 1974 to the present on such topics as minimal risk, criteria for institu- tional review board (IRB) review, basic elements of informed consent, exempt research, and others. 59

OCR for page 59
60 PROTECTING PARTICIPANTS AND FACILITATING SOCIAL AND BEHAVIORAL SCIENCES RESEARCH portent developments in confidentiality protection, which we consider a major concern for much SBES research, have mainly occurred in federal statistical agencies and SEES data archives. More recently, ex- plicit federal confidentiality regulations have been enacted for research that uses medical records from health care providers and insurers. We review the history of confidentiality protection at the beginning of our discussion of confidentiality issues in Chapter 5. FROM 1945 TO 1966 The biomedical experiments performed by Nazi researchers during World War II focused attention on the ways in which human partici- pants could be seriously harmed by research. Many unwilling partici- pants suffered permanent injury, shortened life expectancy, psycholog- ical trauma, and even death in these experiments (e.g., measuring how long humans could survive in ice water). The second Nuremberg Mil- itary Tribunal condemned such research as a crime against humanity. In 1947 the judges proposed a list of ten principles, the "Nuremberg Code," which researchers should be obligated to respect in conducting medical experiments that involve human subjects. The code called for voluntary consent, minimization of harm, and a determination that the research benefits outweighed the risks to participants (see Annas and Grodin, 19921. Biomedical research exploded in scope and volume after the war. Professional organizations and government agencies began to develop codes of ethics for human research participant protection that reflected the Nuremberg Code. In 1953-1954 the National Institutes of Health (NIH) established an ethics review committee (the Medical Advisor Board) for intramural research at its clinical center and adopted a pol- icy that all human research participants at the center must provide informed consent, although written consent was not always required for sick patients. By the mid-1960s many biomedical research orga- nizations had voluntarily established ethics review mechanisms. But instances of unethical research kept coming to light, and public pres- sure built for explicit government regulation. The thalidomide tragedy spurred congressional action, as did reve- lations at congressional hearings about the common practice in which doctors gave patients samples of experimental drugs without the pa- tient knowing the experimental nature of the drugs or consenting to their use. The 1962 Kefauver-Harris amendments to the Food, Drug, and Cosmetic Act required informed consent in testing of experimen- tal drugs, although consent was not required if it was deemed by the

OCR for page 59
REGULATORY HISTORY 61 physician to be infeasible or not in the patient's best interest. Reports of other questionable research, including a 1963 study in which poor elderly patients at the Jewish Chronic Disease Hospital in New York City were injected with live cancer cells without their consent, led the U.S. Public Health Service (USPHS) to develop a policy on human re- search participant protection that was issued in 1966. The 1966 USPHS policy required that every research institution re- ceiving grant dollars from the agency establish a committee to review federally funded research projects for conformance with human partic- ipant protection. On December 12, 1966, in response to questions, the U.S. surgeon general issued a "clarification" of the USPHS policy, mak- ing explicit that review committees (the forerunners of IRBs) were to review "all investigations that involve human subjects, including inves- tigations in the behavioral and social sciences." With regard to SEES research, he said (Memorandum from the Surgeon General to Heads of Institutions Receiving Public Health Service Grants, December 12, 1966; quoted in Gray, 1982:3311: There is a large range of social and behavioral research in which no personal risk to the subject is involved. In these circumstances, regardless of whether the investigation is classified as behavioral, social, medical or other, the issues of concern are the fully voluntary nature of the participation of the subject, the maintenance of confidentiality of infor- mation obtained from the subject, and the protection of the subject from misuse of the findings.... ESBES research] may in some instances not require the fully informed consent of the subject or even his knowledge- able participation. The original USPHS policy, as Gray (1982:331) notes, "contained no systematic analysis of the ethical issues at stake in research involving human subjects." Nor did the surgeon general explain why or under what circumstances informed consent in SEES research might not be required. FROM 1966 TO 1974 The years subsequent to the first USPHS policy guidelines saw con- tinued interest in and attention to human research participant protec- tion issues among regulators, researchers, and members of Congress.

OCR for page 59
62 PROTECTING PARTICIPANTS AND FACILITATING SOCIAL AND BEHAVIORAL SCIENCES RESEARCH Federal Government Activity In 1971, in response to requests for clarification and evidence of highly variable implementation of the USPHS policy, the U.S. Depart- ment of Health, Education, and Welfare (DHEW)2 published "The In- stitutional Guide to DEIEW Policy on Protection of Human Subjects." This guide, known as the "Yellow Book," defined "risk" as exposure to the possibility of harm, including physical, psychological, sociolog- ical, or other harms, "beyond the application of those established and accepted methods necessary to meet [participants'] needs." (This state- ment distinguished research activities from activities to provide a ser- vice, such as medical treatment.) The Yellow Book explicitly identi- fied kinds of possible harm that could arise in research conducted with such typical SBES methods as interviews, observations, and secondary analysis of existing data, including "varying degrees of discomfort, ha- rassment, invasion of privacy, or... a threat to the subject's dignity through the imposition of demeaning or dehumanizing conditions." The Yellow Book discussed methods for protecting against disclo- sure of confidential data. It stated that IRBs should make sure that sec- ondary analysis was "within the scope of the original consent." Such consent could be oral or written, obtained "after the fact following de- briefing," or could be "implicit in voluntary participation in an ade- quately advertised activity." There was no repetition of the surgeon's general statement allowing less than fully informed consent under some circumstances. Building on the Yellow Book, DHEW issued comprehensive regu- lations in May 1974 for the protection of human research participants (45 Code of Federal Regulations [CFR] 46), which stated that the de- partment would not support any such research unless first reviewed and approved by an IRB.3 4 IRBs were to determine whether human subjects were at risk (defined as the possibility of "physical, psycho- logical, or social injury"; whether risks were outweighed by benefits to the subjects and the importance of the knowledge sought; whether the rights and welfare of subjects were adequately protected (although 2DHEW was the predecessor agency to DHHS. 3 The 1974 regulations referred to a "committee" and not to an IRB; the IRE ter- minology was adopted in a subsequent 1975 amendment to the Public Health Service Act. 4 Later in 1974 DREW published regulations for additional protections for pregnant women and fetuses and for prisoners; these regulations were revised in 1978 following the report of the National Commission for the Protection of Human Subjects of Biomed- ical and Behavioral Research (see below); they became subparts B and C of 45 CER 46, while the basic regulations became subpart A. Additional protections for children in subpart D were first codified in 1983.

OCR for page 59
REGULATORY HISTORY 63 there was no specific language about privacy or confidentiality pro- tection); and whether "legally effective informed consent" would be obtained by "adequate and appropriate methods." Informed consent was to be documented in writing. While the regulations included pro- visions for waiver of written consent, they were difficult to understand and appeared to be very narrowly drawn. The regulations specified that IRBs were to include at least five members of varying backgrounds, including individuals who were able to "ascertain the acceptability of proposals in terms of institutional commitments and regulations, applicable law, standards of professional conduct and practice, and community attitudes." Gray (1977), among others, argued that this language could give IRBs license to object to research on political rather than ethical grounds. A government staff member (Tropp, 1979, 1982) later asserted that the pressure of congressional action (see below) resulted in hasty publi- cation of the 1974 regulations, which were formulated primarily by the health components of DHEW, without participation from other agen- cies that supported SEES research and knew the SBES research com- munity. Further intradepartmental negotiation was planned to pro- duce regulations appropriate for all types of research, but it did not take place. National Research Act During this period, public concern about unethical biomedical ex- periments escalated. Two experiments in particular captured public attention. In the Willowbrook study, from 1956 to 1972, children who were residents of the Willowbrook (Staten Island, New York) State School for the Retarded were injected with a form of hepatitis. Par- ents, in order to have their children admitted to the only available area of the school, the research wing, had to consent to the study (see Beecher, 19701. In the Tuskegee, Alabama study, begun in 1932, USPHS physicians followed several hundred African-American men who had syphilis. No treatment was given to these men even after the discovery of penicillin. The study was not discontinued until 1973 (see Jones, 19811. The Willowbrook study had been reviewed by an ethics committee, and the Tuskegee study apparently had also had such a review, but neither study was stopped until the media reports and sub- sequent public reactions. In 1973 Senator Edward Kennedy (D-MA) held hearings and intro- duced legislation to establish an independent National Human Experi- mentation Board to regulate all federally funded research with human participants, not just that funded by DHEW. However, this approach

OCR for page 59
64 PROTECTING PARTICIPANTS AND FACILITATING SOCIAL AND BEHAVIORAL SCIENCES RESEARCH did not garner sufficient support, so Senator Kennedy introduced in- stead what became the National Research Act of 1974. The act en- dorsed the regulations about to be promulgated by DHEW (see above) and established a National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research. The commission was charged to review the IRB system and to advise DHEW and the Congress about ethical issues in research involving vulnerable popu- lations, such as pregnant women, fetuses, children, prisoners, and in- stitutionalized mentally ill or retarded people. The commission was to identify basic ethical principles for research with human participants and to recommend ways to ensure that research studies followed these principles. FROM 1974 TO 1981 National Commission The National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research conducted its work over a 4-year period. In addition to holding hearings and commissioning papers, it contracted with the University of Michigan for an in-depth study of the IRB system at a sample of 61 institutions, involving interviews with several thousand IRB members, researchers, and research participants (Cooke, Tannenbaum, and Gray, 1978; see also Appendix D). This survey offered three major findings: (1) by comparison with a study for 1969 (Barber et al., 1973),IRBs had become much more ac- tive in requiring modifications in proposed research, particularly with regard to informed consent;5 (2) the selection of participants seemed to distribute risks and benefits fairly evenly among various popula- tion groups; and (3) researchers, as well as IRB board members, sup- ported the IRB system overall. The study also revealed two prob- lems: First, not only did virtually all modifications to informed con- sent procedures deal with consent forms and not the process, but also IRB-modified consent forms were not more complete or easier to read than other forms. Second, the support for the IRB system among SBES researchers was somewhat less strong than among biomedical researchers (see below). The commission's report on IRBs was issued in 1978; it supported the basic IRB system and made numerous recommendations to streng- 5The Barber et al. (1973) study, the first to survey IRB operations, interviewed a single individual at each of 300 institutions with a biomedical IRB, 70 percent of which had existed prior to the 1966 USPHS requirement. It found that relatively few IRBs ever required modifications of protocols: 34 percent had never modified or rejected a project.

OCR for page 59
REGULATORY HISTORY 65 then and clarify its scope and procedures. Recommendations included: definitions of "research" and "human subject;" an expedited review procedure whereby IRBs could delegate the review of minimal-risk protocols to the chair or a subset of IRB members; an expanded list of types of information to disclose to obtain informed consent; and a clearer specification of the conditions under which written consent could be waived. The commission also recommended that IRBs should: determine that study methods are appropriate (i.e., review protocols for scientific soundness); determine that adequate procedures are in place to protect privacy and protect data confidentiality; not consider possible future harm from research results in making risk-benefit deter- minations;6 and provide an opportunity for investigators to respond in person or in writing to IRB decisions. Finally, the commission recom- mended that DHEW should issue regulations applicable to all research over which it had regulatory authority and that Congress should pass legislation to cover human participants in all research over which the federal government might have regulatory authority.7 In 1979 the commission issued the landmark Belmont report, which articulated three basic principles for research studies involving hu- mans: respect for persons, beneficence, and justice. This report quickly became the bedrock on which subsequent work on protecting human research participants was built. SBES Activity In both this period and earlier, the SBES research community de- voted considerable attention to participant protection issues. A liter- ature review conducted for a 1982 volume on Ethical Issues in So- cial Science Research (Beauchamp et al., 1982) turned up more than 3,500 relevant citations. First and most prominent in the literature was Herbert C. Kelman, who published a collection of early papers in 1968 as A Time to Spealc. Other leading publications in the years 1967- 1980 covered such topics as ethical dilemmas in SBES research, ethics of social experimentation, legal issues in SBES research, politics and SBES research, ethics of social intervention, ethical issues in behav- ior modification, studying deviance, and confidentiality protection (see Beauchamp et al., 1982:6-7,37-381. 6 Such harm could occur, for example, when results are unfavorable to a population group to which research participants belong. 7 Congress never acted on this recommendation; legislation is pending at present that would extend federal protections to all research involving human participants (see "Developments Since 1991," below).

OCR for page 59
66 PROTECTING PARTICIPANTS AND FACILITATING SOCIAL AND BEHAVIORAL SCIENCES RESEARCH BOX 3-1 Examples of Ethically Troubling SBES Research from the 1 970s Milgram's (1974) "Obedience to Authority" Summary Stanley Milgram, a social psychologist, carried out an experiment in the 1 960s in which volunteers were recruited to be "teachers" who were asked to administer an electric shock of increasing intensity to a "learner" for each mistake made during the experiment. The fictitious story given to these "teachers" was that the experiment was exploring effects of punishment (for incorrect responses) on learning behavior. The "teacher" was not aware that the Learner" in the study was an actor who merely indicated discomfort as the "teacher" increased the (fake) electric shocks. When the "teacher" asked whether increased shocks should be given, he or she was verbally encouraged to continue. Sixty percent of the "teachers" obeyed orders to punish the "learner" to the very end of the 450-volt scale; no subject stopped before reaching 300 volts. Many subjects were distressed by what they were asked to do. Subjects were debriefed at the end of the experiment and offered counseling if they wanted to examine their own behavior and feelings. Milgram later carried out variations of this same experiment. Commentary Milgram's experiments aroused intense controversy about the conditions under which it is appropriate to use deception, particularly when the consequence is intense (although apparently transitory) psychological stress and even when subjects are informed about the deception at the conclusion of the experiment. Milgram subsequently resurveyed his original subjects, most of whom said they supported the experimentation because it provided important information about how ordinary people could be induced to behave in reprehensible ways in response to a recognized authority, but this finding did not stop the controversy. Humphreys' (1975) "Tearoom Trade" Summary Laud Humphreys, a graduate student in sociology, conducted a study for his dissertation of homosexual behavior among men of high social standing in a large Midwestern city. Pretending to be gay and a lookout for the police, he observed men entering a public rest room in a city park, confirmed that they engaged in anonymous homosexual acts, recorded their license tag numbers, obtained their names from a contact in the Bureau of Motor Vehicles, and confirmed their identity and social status. A year later he disguised himself and interviewed them in their homes about their personal lives under the pretext of conducting a social health survey. When his dissertation was published, many of the men and their families were able to recognize themselves from the details that were reported about them. Humphreys' research had the perhaps beneficial result that police stopped raiding "tea rooms" when they learned that many of the men frequenting them were not people they wanted to arrest. Commentary Had the Common Rule been in effect at the time of "Tearoom Trade," his research would not likely have been approved. Given the stigma attached to homosexuality, the study would not be categorized as minimal risk, and, therefore, deception could not be used (see Box A-9 in Appendix A on the criteria for waiver of informed consent).

OCR for page 59
REGULATORY HISTORY 67 BOX 3-1 (continued) Whether norms have changed to the extent that such a study might be viewed as minimal risk today is not clear; however, informed consent would certainly be required before study results could be published, and other changes in study procedures would undoubtedly be required before it could be approved. Zimbardo's (1971) Stanford Prison Experiment Summary In summer 1971 psychologist Philip G. Zimbardo studied the reactions of male volunteers (Stanford University students) to a mock prison environment in which some of them were assigned roles as prisoners, others as guards. The experiment elicited such strong reactions from the participants, including abuse of some "prisoners" by some "guards," that Zimbardo halted the 2-week study after only 6 days (see Faden and Beauchamp, 1986:178-179; hitp:/www.prisonexp.org [4/10/03]). Commentary This study did not involve deception. Whether such a study would be approved today is unclear. Evidence from Milgram, Zimbardo, and others that strong negative reactions are not uncommon would likely make IRBs reluctant to approve further research of this type. However, an investigator could propose steps to minimize risk, such as planning prompt interventions of the type that were implemented by Zimbardo in the prison experiment, that could make the study acceptable, and it would have the potential to achieve important results for understanding social interactions in stressful situations. Social scientists and other researchers also engaged in vigorous de- bates about the ethics of particular research projects, including the Milgram "obedience to authority" experiments, Humphreys' study of the "tearoom trade," and Zimbardo's study of prison behavior (see Box 3-11. None of these projects caused long-lasting physical harm to any participant, but there was concern that they may have caused strong short-term psychological effects and perhaps even long-lasting psycho- logical trauma. However, for none of them, to our knowledge, has there been any documented lasting harm to participants. As a consequence of these debates and the growing concern with human participant protection in SBES research, both the American Anthropological Association and the American Sociological Associa- tion adopted new codes of ethics in 1971. The American Psychological Association, which had first adopted a code of ethics for clinical work and research in 1953, sponsored extensive research and discussion on ethical issues specific to psychological research and adopted a revised ethics code in 1973 (Faden and Beauchamp, 1986; see also Chapter 41. With regard to the 1974-1978 work of the national commission, there is no available evidence that concerns about SEES research

OCR for page 59
68 PROTECTING PARTICIPANTS AND FACILITATING SOCIAL AND BEHAVIORAL SCIENCES RESEARCH BOX 3-2 SBES Concerns in the 1 970s Problems of 1974 IRB Regulations for SBES Research IRBs' insistence on wriffen consent when it was not needed and would damage the research. Infringements of academic freedom resulting from IRBs' application of risk-benefit criteria. The general lack of fit between the regulations and the ethical issues that arise in some types of research (e.g., ethnography). The failure of the regulations to address explicitly certain issues (e.g., deception, privacy, confidentiality of data, and the use of research for teaching purposes) and certain types of research (e.g., fieldwork or evaluation research). The conflict that can develop between writlen consent requirements and researchers' responsibility to assure confidentiality of the identify of subjects. The lack of clear definition of the applicability of the regulations (e.g., just what is a human subject). The harm to research that can result from application of consent requirements to certain research, such as observational studies in public places and studies based on existing records. The use of IRBs to protect vested interests (e.g., the alleged tendency of IRBs in some medical schools and hospitals to discourage social research within their walls). Some IRBs' lack of reasonable flexibility in interpreting regulatory language. Some IRBs' lack of qualifications to review certain types of studies. The application of review requirements to research not funded by the federal government. . . The lack of appeal procedures. SOURCE: Verbatim quotes from Gray (1982:354) of testimony provided to the National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research. played a part in its formation (the "behavioral research" part of its title was not explicitly defined in its charter). As specified by Congress, the commission membership included six people who were not re- searchers and five who were: three physicians and two psychologists were chosen to fill the researcher slots. Although the commission did not represent the breadth of SBES expertise (or biomedical expertise, for that matter), it did seek input from SBES researchers. Most of their concerns, according to Gray (1982:335), "pertained to harm done to re- search interests while providing no benefit to the interests of subjects" (see Box 3-21. There were differing viewpoints: a paper commissioned

OCR for page 59
REGULATORY HISTORY 69 from a sociologist who served as chair of an IRB that primarily re- viewed SBES research argued that current regulations did not impede research, provided they were interpreted reasonably (Barber, 19791; a paper commissioned from another sociologist argued that existing regulations did not properly address issues of informed consent and confidentiality in SBES research (Reiss, 19791. Data from the University of Michigan survey conducted for the com- mission revealed that almost all researchers believed the IRB system had helped protect human participants (at least to some extent), al- though higher percentages of SBES researchers than of biomedical researchers were critical of the system in some respects. For exam- ple, excluding IRB board members, 54 percent of SBES researchers agreed that the IRB system had impeded the progress of research at their institution (at least to some extent), compared with 43 percent of biomedical researchers (Gray, 1 9 82 :Table 1 6 .2~. ~ 1979 Proposed Revision of Regulations In response to the national commission's recommendations, the NIH Office for Protection from Research Risks (OPRR) drafted a re- vision of 45 CFR 46, which was published for comment in the Federal Register in August 1979.9 The proposed revision raised a firestorm in the SBES research community. The major problem appears to have been that the regulations proposed to extend the IRB system to all re- search involving human participants at institutions that received DHEW funds, regardless of whether the particular protocol was funded by DHEW, and to studies in all fields that sought generalizable knowl- edge by using methods that collected information by which living or dead individuals could be identified. (See Boxes A-1 and A-3 in Ap- pendix A, which trace changes in language on applicability of the reg- ulations and the definition of human subject, respectively.) This broad scope could have subjected standard historical, document-based re- search, interviews with public officials, observation of behavior at pub- lic events, and other such studies to IRB review, including student and investigator studies funded by the research institution itself or by an- other federal agency and not DHHS. Furthermore, IRB review was explicitly to consider the appropriateness of the research methods for Researchers who were IRB members were much less likely to agree that the system impeded research (30 percent of SBES IRB board members did so, as did 26 percent of biomedical IRB board members). 9OPRR was established in the office of the NIH director in the early 1970s; it su- perceded the Institutional Relations Branch of the Division of Research Grants, which formerly had responsibility for oversight of human research participant protection (Mc- Carthy, 1998:313).

OCR for page 59
70 PROTECTING PARTICIPANTS AND FACILITATING SOCIAL AND BEHAVIORAL SCIENCES RESEARCH the objectives of the research and the field of study (see Box A-6 in Appendix A). The proposed revised regulations included provisions for IRBs to exempt certain types of research from review and to review other kinds of research by an expedited procedure (see Boxes A-4 and A-5 in Ap- pendix A). The intent of these provisions was to exempt or expedite much SBES research, such as that conducted by surveys, public ob- servation, and study of documents. The exemption provisions were restricted to research in which it was not possible to identify the hu- man subjects (Alternative A) or that presented no or only minimal po- tential for invasion of privacy (Alternative B). Similarly, the expedited review procedures were restricted to minimal-risk projects in speci- fied categories, and the proposed definition of minimal risk set a high standard namely, that risks of harm be no greater than those experi- enced in the daily lives of healthy individuals (see Box A- 11 in Appendix A). SBES researchers argued that all of these features of the proposed regulations subjected too many projects to full-scale review by an IRB when such review was not necessary or useful or even (in some argu- ments) constitutional. On the positive side, the proposed regulations responded to some SBES concerns that had been expressed in testimony before the na- tional commission. Specifically, provisions were added to require IRB review to consider the adequacy of proposed data confidentiality pro- tections, to enable IRBs to modify some or all of the elements of in- formed consent, and to enable IRBs to waive the requirement for writ- ten documentation of consent (see Boxes A-6, A-9, and A-10 in Ap- pendix A). These generally well-received changes (which remained in the revised regulations) were overshadowed, however, by the negative reactions to provisions that SBES researchers viewed as inimical to good research. Patullo (1982:373) characterizes the ensuing period until January 1981, when 45 CFR 46, subpart A, was revised, as follows: For eighteen months, hundreds of individuals, some prin- cipally concerned with maintaining a healthy research en- terprise and some alarmed at basic incursions upon accus- tomed freedoms, devoted thousands of hours to persuading those who write the regulations that single-minded concen- tration on precluding every possibility of harm to subjects had produced a proposal that threatened much greater so- cial harm than it might possibly prevent.~ ~Patullo was among those who argued against the 1979 proposed regulations on the basis that they threatened research and also among those who argued that the regula-

OCR for page 59
REGULATORY HISTORY 71 The head of OPRR, which drafted the regulations, gave his view of this tumultuous period as follows (McCarthy, 1984:8-91: Threats and epithets were hurled at us from many sides. These comments can be summarized as follows: 'Social and behavioral research is essentially harmless to individuals and benign to society. If you do not accept this view. . . we will organize ourselves to see that you and your misguided staff are drawn and quartered.' The charges were led by Ithiel de Sola Pool [1979, 1980], who insisted that... our four pages of fine print in the Federal Register were about to lay waste to the First Amendment of the Constitution.... Friendly champions of social and behavioral sciences showed us how to back away from our unpopular positions while continuing to offer what we felt were reasonable pro- tections for the dignity and rights of subjects involved in so- cial and behavioral research to say nothing of saving the face and the jobs of OPRR staff. What were the indignities to subjects that we felt needed attention? The Wichita jury bugging case, the tearoom trade research, the decision of a Georgia court concerning Medi- caid co-payment experiments and our own unpleasant memories of Psychology 101 and Sociology 102, when we felt we had better humor our professors... so we 'volun- teered' as research subjects rather than risk grade discrimi- nation! We discovered that we could write exemptions for broad categories of social and behavioral research categories in which subjects' behavior seemed to us little different than the commerce of daily life. It has been estimated that up to 80~o of social and behavioral research funded by our De- partment is now exempt. For the rest, we thought it not unreasonable to concern Institutional Review Boards with matters of privacy and confidentiality, and with efforts to protect unsuspecting and vulnerable subjects. The final regulations, issued in January 1981 by DHHS (which re- placed DHEW in May 1980) responded both to the recommendations of the national commission (see above) and the concerns raised by SEES researchers. With regard to the latter, the regulations somewhat narrowed the definition of a human subject and provided for certain tions were developed without evidence that SEES research caused physical harm (see Oakes, 2002:7).

OCR for page 59
72 PROTECTING PARTICIPANTS AND FACILITATING SOCIAL AND BEHAVIORAL SCIENCES RESEARCH kinds of human subjects research to be exempt from IRB review. A "human subject" was defined as "a living individual about whom an investigator (whether professional or student) conducting research ob- tains (1) data through intervention or interaction with the individual, or (2) identifiable private information" fitalics added]. This definition excluded some SEES research from IRB review, such as observational studies in which private information was not obtained and studies of information from past public records. In addition, the regulations provided that four categories of human subjects research could be ex- empted by an IRB from review (see Box A-4 in Appendix A): research in educational settings on new or established instruc- tional strategies and techniques, curricula, or classroom manage- ment methods; . research involving educational tests, if identifiers cannot be linked to the subjects; survey and interview research and observational research, unless identifiers are being collected, disclosure could be damaging to the participant, and the data pertain to sensitive subject behavior; and studies using existing data, documents, records, and the like, if these materials are publicly available or if the data will not be recorded in a manner that would allow linkage with individuals. Finally, backing off from the 1979 proposal, the regulations applied only to research sponsored by DHHS, although language was included indicating DHHS's concern that the interests of all human research participants, regardless of funding source, be protected at institutions that received DHHS funding. Also backing off from the 1979 pro- posal, the regulations excluded a requirement for IRB review of the appropriateness of the proposed research design and methods, requir- ing only that IRBs determine that risks to subjects are minimized "by using procedures which are consistent with sound research design." One observer concluded that "most social research is now exempt from the regulations either because it is specifically exempted or be- cause it is not federally supported, thoweverl, the extent to which IRBs thin backing off from this position, which had been a recommendation of the na- tional commission, DHHS responded to concerns of the SEES community, a growing opposition to federal regulation in general, and a specific conclusion of the 1980 Presi- dent's Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research that the National Research Act of 1974 could not be used to justify extension of DHHS regulations to non-DHHS-funded research.

OCR for page 59
REGULA TORY HISTORY 73 will actually continue to review such research remains to be seen" (Gray, 1982:3441. FROM 1981 TO 1991 With the adoption of the 1981 regulations, a period of relative calm ensued. A handful of studies of IRB operations were carried out. One study, which surveyed 341 IRB chairs soon after the passage of the regulations, found that almost all IRBs at that time had decided not to use the option to exempt protocols from review that fell under one of the four eligible categories of educational, social, and behavioral research (Grundner, 19831. A somewhat later study examined how IRBs were functioning from the perspective of political scientists; it queried 115 chairs of political science departments that offered Ph.D. programs and chairs of IRBs at those institutions (Cleary, 1987; see also Appendix D). Generally, most chairs were supportive of the IRB system, reporting that most political science protocols (88 percent) cleared without change. Problems iden- tified included uncertainty and lack of information regarding infold ailed consent and confidentiality protection; confusion at some institutions as to whether unfunded student research was covered; and variability in IRB procedures with regard to exemptions and infold consent. In 1981 the President's Commission for the Study of Ethical Prob- lems in Medicine and Biomedical and Behavioral Research (which met from 1980 to 1983) recommended that a Common Rule be developed that would apply to all federally supported research involving human participants. In response, in 1982 the President's Science Adviser, Of- fice of Science and Technology Policy, appointed an interagency com- mittee to develop a common framework on the basis of the DHHS regu- lations. The committee published a proposed common policy in 1986. Its work culminated on June 18, 1991, when 45 CFR 46, subpart A, was adopted in the regulations of 15 departments and agencies and by the Central Intelligence Agency by legislation (see Box 1-1 in Chapter 11. The Food and Drug Administration also modified its regulations to agree in large part with 45 CFR 46, subpart A. In addition to making the Common Rule regulations applicable to research conducted, supported, or regulated by any of the federal agency signatories, the 1991 regulations incorporated several other changes from the 1981 DHHS version (see Appendix A). First, the defi- nition of research was expanded to include research development, test- ing, and evaluation. Second, two categories of research that could be exempted were added: research and demonstration projects involving

OCR for page 59
74 PROTECTING PARTICIPANTS AND FACILITATING SOCIAL AND BEHAVIORAL SCIENCES RESEARCH public benefit or service programs and taste and food quality evalu- ation and consumer acceptance studies. Third, the requirements for exempting survey and observational research were modified so that IRBs could more readily decide not to exempt as much research as before (see Box A-4 in Appendix A). The list of types of research that were eligible for expedited review, provided they were deemed to be of minimal risk, was not changed at that time, but it was later amended in a Federal Register notice in 1998: many more categories of SBES research were added to the list than had been included in the original (1981) list (see Box A-5 in Appendix A). DEVELOPMENTS SINCE 1991 Raising the Alarm Even as the federal regulatory system for protection of human re- search participants was strengthened by the widespread adoption of the Common Rule, a spate of reports in the mid- to late 1990s con- cluded that IRBs were less and less able to meet their responsibili- ties. These reports dealt almost exclusively with concerns arising from biomedical, particularly clinical, research. The first such report emerged from the Advisory Committee on Hu- man Radiation Experiments, convened in 1994 to follow up media re- ports of federally sponsored radiation research with human partici- pants conducted between 1944 and 1974 that violated ethical no~s.~2 The committee also investigated the current state of human partici- pant protection by reviewing the regulatory system, examining a sam- ple of recent research proposals, and interviewing past, current, and prospective research participants. While concluding that "significant advances" had occurred in human participant protection since the 1940s and 1950s (Advisory Committee on Human Radiation Experi- ments,1996:510), the committee found "evidence of serious deficien- cies in some parts of the current system." These included substantial variability in the performance of IRBs and inattention to problems of informed consent for people with diminished decision-making capabil- ity, as well as confusion among participants as to whether they were involved in experimentation or treatment. The U.S. General Accounting Office (GAO) in 1996 found that the ability of IRBs to operate effectively was impaired by heavy IRB work- loads, lack of expertise on IRBs to review complex research, failure pouch experiments included injecting hospitalized patients with plutonium, likely without their knowledge or consent, and intentional release of radiation into the envi- ronment without public notice.

OCR for page 59
REGULATORY HISTORY 75 of IRBs to exercise continuing review over research projects, and lack of adequate facilities and support for IRB operations. The GAO also found that OPRR lacked time and funding to conduct site visits to re- view IRB operations, even though the value of such visits in identifying IRB compliance problems was known.~3 Beginning in 1998 the DHHS Office of Inspector General issued a series of reports warning that IRBs were being overwhelmed by the complexity and volume of research they had to review. The first report (Office of Inspector General, 1998b:ii-iii) concluded that "the effective- ness of IRBs is in jeopardy" because "they face major changes in the research environment," "they review too much, too quickly, with too little expertise," "they conduct minimal continuing review of approved research," "they face conflicts that threaten their independence," "they provide little training for investigators and board members," and "nei- ther IRBs nor HEIS devote much attention to evaluating IRB effec- tiveness." Furthermore, the report claimed that IRBs lacked adequate resources to do their job and were being pushed by regulations to focus on paperwork requirements more than on basic ethical issues. The National Bioethics Advisory Commission, established by pres- idential appointment in 1995, concluded in its final report (2001:4-8) that several factors underlay these gloomy assessments of the IRB sys- tem. They included: a large increase between 1985 and 1996 in the volume of research resulting from a two-fold increase in federally funded research and a three-fold increase in industry-funded research (particu- larly by pharmaceutical firms); a growing propensity for academic medical centers to develop ties with industry for research funding; an increase in industry-sponsored research conducted through nonacademic organizations; the emergence of stand-alone IRBs that operate on a fee-for-service basis; a growing number of clinical studies conducted at multiple, some- times hundreds, of sites; i3Between 1990 and 1995, OPRR conducted only 15 site visits during the course of over 200 compliance investigations triggered by charges of misconduct. No site visits were conducted randomly, nor were any visits conducted as part of assurance negotia- tions, when OPRR first reviewed an institution's written procedures for conformity with the Common Rule (U.S. General Accounting Office, 1996:19).

OCR for page 59
76 PROTECTING PARTICIPANTS AND FACILITATING SOCIAL AND BEHAVIORAL SCIENCES RESEARCH . and new technologies, such as research on genetic links to dis- eases and new gene therapies for medical treatment, and changes in public attitudesi4 that pose new ethical challenges. All of these factors overloaded IRBs with no offsetting increases in the resources available for their work. Federal Response In response to these reports and to media attention on tragic events in biomedical studies (see below), the federal government stepped up its regulator efforts. In June 2000, the Office for Human Research Protections (OHRP) was established in the Once of the Secretary of DHHS, taking over responsibilities from OPRR and acquiring new responsibilities (see http://ohrp.osophs.dhhs.gov t4/10/031; 67 Federal Register 10217, March 6, 20021. OHRP was given a broad mandate, not only to monitor the operations of IRBs that review DHlIS-funded research, but also to provide guidance on human research participant protection for the federal and nonfederal sectors, develop educational programs, and exercise leadership for human participant protection in cooperation with other federal agencies. A standing committee of outside experts, the National Human Research Protections Advisory Committee (NHRPAC), was established to provide continuing advice to OHRP. Prior to the establishment of OHRP, OPRR had become somewhat more active in enforcement: from one on-site inspection in 1997, OPRR conducted ten on-site inspections in 1998-2000 (all at medical research centers). Several of these reviews resulted in temporary suspension of all or some research at the visited institutions (Office of Inspector General, 2000:2~.~5 OHRP emphasized working cooperatively with re- search institutions and associations to develop training programs for IRBs and investigators and to improve IRB operations, but it, too, on occasion took drastic action. In July 2001 OHRP suspended nearly all research at Johns Hopkins University when a healthy young woman participating in a study about asthma medications died from an over- ~4Examples are research that is conducted with the assistance of community organi- zations and initiatives by some very ill patient groups to obtain greater access to clinical trials, seeing participation as a benefit, not a burden. Din May 1999, OPRR suspended all research at the Duke University Medical Cen- ter for 5 days and required several corrective actions, including re-review of DHHS- funded research approved by the IRB, establishment of a second IRB, and development of educational programs for IRB staff and members and researchers (National Bioethics Advisory Commission, 2 00 1: 5 6).

OCR for page 59
REGULATORY HISTORY 77 dose of an asthma-inducing substance (see Keiger and De Pasquale, 2002~.16 In these and other widely publicized incidents, such as the death of Jesse Gelsinger in 1999 in a University of Pennsylvania gene trans- fer experiment, it appeared that the researchers failed in one or more respects to follow the protocol approved by the IR:B. Nevertheless, the events increased calls for reform of the IRB system, such as recommen- dations by the National Bioethics Advisory Commission (2001:xi-xxi) for legislation to establish human participant protection for all pub- licly and privately funded research and a single independent federal office to lead and coordinate the oversight system; a single set of fed- eral regulations and guidance; accreditation of IRBs; and certification of IRB members. In fall 2001 DHHS commissioned the Institute of Medicine to un- dertake a thorough review of the IRB system, including proposed schemes for voluntary accreditation of IRBs by nonprofit organiza- tions (Institute of Medicine, 2001, 2002~. Congressional committees held hearings, and Representatives DeGette (D-CO) and Greenwood (R-PA) introduced a bill in May 2002 that would extend federal human participant protection to all federally funded research and require the harmonization of the Common Rule and Food and Drug Administra- tion regulations. Senator Kennedy (D-MA) introduced a bill in October 2002 that would expand protections to all research involving human participants. It would also strengthen OHRP, providing for a 6-year term for the director. SBES Involvement During this same period, researchers involved in SBES studies in- creasingly raised concerns that IRBs were overreacting to the increased scrutiny of their operations by putting minimal-risk studies through as time-consuming a review as clinical trials and other higher risk stud- ies and by rigidly adhering to the main provisions of the regulations rather than taking advantage of their options for flexibility. Some ar- gued (although less strongly than in earlier debates about IRBs and SBES research) that the IRB system infringed on academic freedom to conduct research (see, e.g., Shea, 20001. Many argued that IRBs were unnecessarily impeding research and sometimes increasing the risk to participants by their bureaucratic stance for example, by requiring written consent even when such consent could endanger participants thin all, work on 2,400 protocols was halted, some for 5 days and some for longer pe- riods. Research at Johns Hopkins' Homewood campus, which operated under a different federal assurance, was not affected.

OCR for page 59
78 PROTECTING PARTICIPANTS AND FACILITATING SOCIAL AND BEHAVIORAL SCIENCES RESEARCH by linking them to the study (see, e.g., American Association of Uni- versity Professors, 2001; Brainard, 2001; Moreno, 2001; Shea, 20001. In 1998, at the behest of the SBES community, additional categories of minimal-risk SBES research were added to the list of research cat- egories that IRBs could review under expedited procedures (see Box A-5, Appendix A). SBES researchers also fought to become involved in key groups that were part of the debate and that often initially excluded SBES disciplines. In January 2001, NHRPAC (the advisory body to OHRP) was expanded from 11 to 17 members in order to include SBES re- searchers. In June 2001, NHRPAC established an ad hoc Behavioral and Social Science Working Group, which is currently operating as a separate entity with private foundation funding. In September 2002, DHHS declined to renew NHRPAC's 2-year charter. A new Secretary's Advisor Committee on Human Research Protections was chartered in October 2002 with 11 members. The membership of the commit- tee was announced in January 2003; it includes only two SBES re- searchers, both psychologists.~7 The National Science Foundation (NSF), a principal funder of SBES research, established an ad hoc Social, Behavioral, and Economic Sub- committee for Human Subjects. The subcommittee developed a guide for SBES researchers, which was put up on the NSF Policy Office web- site as "Frequently Asked Questions and Vignettes: Interpreting the Common Rule for the Protection of Human Subjects for Behavioral and Social Science Research" (National Science Foundation, 20021. A new Association for the Accreditation of Human Research Protec- tion Programs (AAHRPP) added 3 seats for SBES researchers to its 21-member board and involved SBES researchers in pilot tests of ac- creditation procedures at several dozen research institutions. (See Appendix B for information about groups active in and resources for human research participant protection.) i7See http://www.hhs.gov/news/press/2003press/20030 ~ 1 3a.html [2/23/03]. In De- cember 2002 the presidents of the National Academy of Sciences, the National Academy of Engineering, and the Institute of Medicine wrote a letter to DHHS Secretary Thomp- son stating that "it is critical for this important body to represent all of the scientific disciplines engaged in research with human participants that is, the biological, clini- cal, social, behavioral, and economic sciences." i8AAHRPP was established in May 2001 as the national accrediting arm of Public Re- sponsibility in Medicine and Research. It is developing a voluntary, peer-driven human research accreditation program, using a site visit model; see Appendix B.

OCR for page 59
REGULATORY HISTORY 79 CONCLUSION This brief history makes clear that the concerns of the SBES re- search community about the IRB system are not new. In fact, the lists of concerns voiced in 1974 (see Box 3-2) are remarkably similar to those being voiced today. Moreover, both today and three decades ago, some researchers believe that the regulations need to be changed, while others maintain that the regulations have sufficient flexibility and the real concern is with their interpretation. fine conclude that the current focus for improvement of participant protection and facilitation of research in SBES should be on the devel- opment of guidance and other means to encourage IRBs to avail them- selves appropriately of the flexibility in the regulations rather than on further changes in the regulations themselves. A primary reason for our conclusion is the past successes of the SBES research community with respect to regulatory change. Thus, in the late 1970s, SBES re- searchers successfully participated in a process that resulted in the cur- rent provisions for exemption and expedited review. In the late 1990s, SBES researchers influenced the process that resulted in an expanded list of types of research that, if minimal risk, could be reviewed with an expedited procedure. The regulations also provide for flexibility in procedures for obtaining and documenting informed consent. As we discuss in subsequent chapters, OHRP and other actors in the protection system, such as accreditation organizations and pro- fessional associations, must assert leadership in providing guidance that can help IRBs use the flexibility in the Common Rule to carry out their responsibilities more effectively. To assure the development of ap- propriate guidance, SBES researchers must actively continue to seek formal and informal channels for input to OHRP and other relevant agencies and organizations. The SBES research community must also actively seek to develop a knowledge base that can inform OHRP, IRBs, and researchers about appropriate procedures for informed consent, balancing of risks and benefits, and other aspects of ethically responsi- ble research.

OCR for page 59
80 PROTECTING PARTICIPANTS AND FACILITATING SOCIAL AND BEHAVIORAL SCIENCES RESEARCH decided not to use the option to exempt protocols from review that fell uncler one of the four eligible categories of eclucational, social, and behavioral research (Grundner, 1983~. A somewhat later stucly examined how IRBs were functioning from the perspective of political scientists; it queried 115 chairs of political science departments that offered Ph.D. programs and chairs of IRBs at those institutions (Cleary, 1987; see also Appendix D). Generally, most chairs were supportive of the IRE system, re- porting that most political science protocols (who) cleared without change. Problems identified included uncertainty ant] lack of infor- mation regarding informed consent and confidentiality protection; confusion at some institutions as to whether unfuncled student re- search was covered; and variability in IRB procedures with regard . to exemptions en cl informed consent. In 1981 the Presiclent's Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research (which met from 1980 to 1983) recommended that a Common Rule be ~leveloped that wouic! apply to all federally supported research involving human participants. In response, in 1982 the Presiclent's Science Adviser, Office of Science and Technology Policy, appointed an interagency committee to develop a common framework on the basis of the DHHS regulations. The committee published a pro- posed common policy in 1986. Its work culminated on June 18, 1991, when 45 CFR 46, subpart A, was adopter! in the regulations of 15 departments and agencies and by the Central Intelligence Agency by legislation (see Box 1-1 in Chapter 11. The Food ant! Drug Acimin- istration also mortified its regulations to agree in large part with 45 CFR 46, subpart A. In addition to making the Common Rule regulations applicable to research conducted, supported, or regulated by any of the fecI- eral agency signatories, the 1991 regulations incorporated several other changes from the 1981 DHHS version (see Appendix A). First, the definition of research was expanded to include research clevel- opment, testing, anct evaluation. Second, two categories of research that could be exempted were addecI: research and demonstration projects involving public benefit or service programs and taste and food quality evaluation and consumer acceptance studies. Third, the requirements for exempting survey ant! observational research were mollified so that IRBs could more readily decide not to ex- empt as much research as before (see Box A-4 in Appendix A). The list of types of research that were eligible for expedited review, pro- 5