1 Although at present such committees are widespread and are legally required for research involving human subjects funded by the U.S. Department of Health, Education, and Welfare (DHEW), as recently as the early 1960s only a small minority of medical institutions had such committees. In 1966 the Public Health Service (PHS) initiated the requirement that a committee of the investigator's “institutional associates” review proposals for research involving human subjects as a condition for the receipt of a PHS grant. In 1971 this requirement was extended to include all experiments involving human subjects conducted under a DHEW grant or contract. The Food and Drug Administration (FDA) issued regulations in the same year requiring that proposals for the clinical investigation of new drugs be submitted to such a committee. 36 Fed. Reg. 5,037 (1971) (codified in 21 C.F.R. § 312.1 (1978)). Building on the then-existing DHEW policy Congress in 1974 mandated that the Secretary of Health, Education, and Welfare require that each entity applying for a grant or contract under the Public Health Service Act establish such a committee to review biomedical and behavioral research involving human subjects at or sponsored by the entity. 42 U.S.C.A. § 289/-3(a) (1976). The implementing regulations provide that no activity involving human subjects is to be supported by DHEW grants or contracts unless such a committee has reviewed and approved such activity. 45 C.F.R. § 46.102 (1978). For a discussion of the genesis and development of these committees see Bernard Barber et al., Research on Human Subjects 145–48 (New York: Russell Sage Foundation, 1973); Donald T. Chalkley, Developing Guidelines, 21 Clinical Research 777 (1973); William J. Curran, Governmental Regulation of the Use of Human Subjects in Medical Research: The Approach of Two Federal Agencies, in Paul A. Freund, ed., Experimentation with Human Subjects 402 (New York: George Braziller, 1970).Google Scholar

2 See Henry K. Beecher, Research and the Individual: Human Studies 12–15 (Boston: Little, Brown, & Co., 1970); U.S. Senate, Committee on Labor and Public Welfare, Subcommittee on Health, Quality of Health Care—Human Experimentation, Parts 1, 2, & 3, Hearings on S.974, S.878, S.J. Res. 71, 93d Cong., 1st Sess. (1973). Among the studies that have been given wide publicity because of their alleged ethical deficiencies are the Tuskegee Syphilis study, the live cancer cell study at Jewish Chronic Disease Hospital, and Milgram's study of obedience. See U.S. Department of Health, Education, and Welfare, Public Health Service, Final Report of the Tuskegee Syphilis Study Ad Hoc Advisory Panel (n.p., Apr. 28, 1973); Jay Katz, Experimentation With Human Beings: The Authority of the Investigator, Subject, Professions, and State in the Human Experimentation Process 9–65 (New York: Russell Sage Foundation, 1972); Stanley Milgram, Obedience to Authority: An Experimental View (New York: Harper & Row, 1974); Kenneth Ring, Kenneth Wallston, & Michael Corey, Mode of Debriefing as a Factor Affecting Subjective Reaction to a Milgram-Type Obedience Experiment–an Ethical Inquiry, in Katz, supra, at 395. Barber and his associates concluded from interviews with more than 300 researchers at two medical institutions that “while the large majority of our samples of biomedical researchers seems to hold and live up to high ethical standards, a significant minority may not.” Barber et al., supra note 1, at 52.Google Scholar

3 Dorothy Nelkin, Threats and Promises: Negotiating the Control of Research, 107 Daedalus 191–92 (Spring 1978): The scientific community has persistently resisted public control. Only when research has direct technological applications are scientists willing to concede the need for regulation. In a survey of 800 scientists, seventy-seven percent agreed that “The pursuit of science is best organized when as much freedom as possible is granted to all scientists.”Google Scholar

4 Detailed guidelines have been issued regulating research involving fetuses, pregnant women, and in vitro fertilization, 45 C.F.R. §§ 46.201-.211 (1978) and prisoners, 43 Fed. Reg. 53,655 (1978). Special regulations have been proposed but not as yet adopted for research involving the mentally disabled and children. See 43 Fed. Reg. 2,084 (1978) (report of National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research on research involving children); 39 Fed. Reg. 30, 648–57 (1974).Google Scholar

5 Existing regulations provide that the committee shall determine with respect to each proposal whether: (1) The risks to the subject are so outweighted by the sum of the benefit to the subject and the importance of the knowledge to be gained as to warrant a decision to allow the subject to accept these risks; (2) The rights and welfare of any such subjects will be adequately protected; and (3) Legally effective informed consent will be obtained by adequate and appropriate methods. 45 C.F.R. § 46.102(b) (1978).Google Scholar

6 Dr. Veatch of the Hastings Institute criticizes the jury analogy, rightly pointing out that committees are dominated by scientists from the institution. Robert M. Veatch, Human Experimentation Committees: Professional or Representative? Hastings Center Rep., Oct. 1975, at 31. Although a number of review boards contain an unrelated minister or attorney, and less often a business executive or “housewife,” and while the regulations mandate some outside participation, most members of committees are employees. While very reflective of the academic and scientific community, such boards are not drawn from the broader community affected by research.CrossRefGoogle Scholar

7 45 C.F.R. § 46.106(b) (1978).Google Scholar

8 Id.Google Scholar

9 Barber et al., supra note 1, at 145–67.Google Scholar

10 Id. at 159–60.Google Scholar

11 A survey taken for the National Commission found that investigators with more experience with human subjects committees tended to present better proposals. University of Michigan Survey Research Center, Institute for Social Research, A Survey of Institutional Review Boards and Research Involving Human Subjects, in National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research, Appendix to Report and Recommendations, Institutional Review Boards, DHEW Pub. No. (OS) 78–0009, at 1–67–1–69 (Washington, D.C.: Government Printing Office [1978]) (hereafter cited as National Commission Survey). At least eight members on four of the committees included in the present study stated that they had observed some change either generally or on specific aspects of proposals. One member stated that he had observed no improvement in the quality of the consents during the one year he had been on the committee. Several other members either indicated uncertainty on the point or made ambiguous comments.Google Scholar

12 See Benjamin S. DuVal, Jr., Educational Research and the Protection of Human Subjects, 1977 A.B.F. Res. J. 477.CrossRefGoogle Scholar

13 For the findings of the National Commission survey on this point see infra at 578. Data from the present study are limited and conflicting. Although one committee in the present study exhibited no substantial difference in the treatment of biomedical and behavioral proposals, some members of another committee felt that their committee had tended to impose more restrictions on biomedical than on behavioral proposals. The reliability of the performance data on which Barber and his associates relied is also open to some question. The findings were based on estimates by institutional officials, some of whom were not committee members. See Barber et al., supra note 1, at 14. Respondents were given no guidance whether modifications in a proposal that resulted from the action of a subcommittee of the human subjects committee were to be revisions required by the human subjects committee. Neither were respondents advised whether when a proposal was silent on a point, for instance, how confidentiality of data was to be preserved and the investigator supplied information at the request of the committee, this was to be counted as a revision of the proposal. My experience affords some basis for believing that estimates given by the chairman and members of a committee may often be reasonably accurate. Nevertheless, the great variation in the estimates given by members of one committee in the present study show that in some instances, at least, these estimates may not be reliable.Google Scholar

14 See National Commission Survey, supra note 11, at 1–299–1–300. In the introduction to the report the authors state that the data “concern 80 boards” (p. 1–11) while the tables reporting committee structure and procedures generally report an N of 75. The sampling design called for a sample of 67 institutions, 11 of which were selected because they “were thought to be unusual or unique in some way (a large volume of research, specialized research, or some other unusual characteristic).”Id. at 1–297. For selection of the remaining institutions the universe was broken down by type of institution and within each type by geographical region and by volume of research funded by the National Institutes of Health (NIH). The sampling rates were so constructed as to include in the sample a greater proportion of those types of institutions and, within each institutional type, those institutions which conducted a high volume of NIH-funded research. Id. at 1–294–1–299. Nine of the 67 institutions initially selected were unable or unwilling to participate. Substitutes were chosen for 3. Id. at 1–299.Google Scholar

15 Computed from table VI.45, National Commission Survey, supra note 11, at 1–208. The figures reported are not directly comparable to those in the Barber group study, since they are based upon committee action on a sample of the projects activated at the institutions studied. Since proposals that were flatly disapproved or on which the committee for some reason never took final action were never activated, the proportion of proposals submitted for review that were approved without modification or clarification should be somewhat lower than the figures actually reported. Six percent of the biomedical researchers and 1 percent of the behavioral researchers interviewed in the National Commission survey reported having one or more proposals disapproved. Bradford H. Gray, Human Subjects Review Committees and Social Research 9 (paper delivered at Annual Meeting of American Association for the Advancement of Science, Feb. 17, 1978).Google Scholar

16 National Commission Survey, supra note 11, at 1–41.Google Scholar

17 Id. at 1–37, 1–208.Google Scholar

18 Id. at 1–48. Since the proportion of biomedical research varies with the type of institution, these findings are difficult to reconcile.Google Scholar

19 Id. at 1–41–1–45.Google Scholar

20 Gray's figures are based on his own review of committee records. The 79 proposals represent only 60 percent of the proposals reviewed by the committee during the 1969–70 academic year. For reasons deriving from his research objectives Gray excluded psychiatric, pediatric, epidemiological, behavioral, and health care projects. Bradford H. Gray, Human Subjects in Medical Experimentation: A Sociological Study of the Conduct and Regulation of Clinical Research 25, 29 (New York: John Wiley & Sons, 1975).Google Scholar

21 Id. at 37.Google Scholar

22 Dale H. Cowan, Human Experimentation: The Review Process in Practice, 25 Case W. Res. L. Rev. 533 (1975).CrossRefGoogle Scholar

23 Eugene J. Millstein, The DHEW Requirements for the Protection of Human Subjects: Analysis and Impact at the University of California, Berkeley 7–8 (Berkeley: University of California Research Management Improvement Project, 1974). The report does not give a breakdown of the nature of the proposals but does indicate the percentage of proposals originating in different schools, colleges, and departments. Although none of the proposals originated in hospitals or medical schools and the great bulk originated in the School of Education and a variety of departments, colleges, schools, and institutes in the social sciences, proposals reviewed by the committee included some from a biomedical research laboratory, from the student health service, and from the departments of physical education, bacteriology and immunology, and biochemistry.Google Scholar

24 See, e.g., Powell v. Alabama, 287 U.S. 45 (1932) (the importance of counsel in arriving at a just decision); Goss v. Lopez, 419 U.S. 565, 583 (1975) (“requiring effective notice and informal hearing permitting the student to give his version of the events will provide a meaningful hedge against erroneous action”); 5 Wigmore, Evidence $1367 (Boston: Little, Brown & Co., 1974) (the value of cross-examination in “testing the value of human statements”). The important role that procedure plays is, of course, not confined to criminal proceedings or to procedures that are constitutionally guaranteed. The controversy surrounding the class action, for example, turns in part on the alleged impact of that procedural device on the results, and there is evidence that some suits brought as class actions would not have been feasible in the absence of the class action procedure. See Benjamin S. DuVal, Jr., The Class Action as an Antitrust Enforcement Device: The Chicago Experience (II), 1976 A.B.F. Res. J. 1273, 1287–1302.Google Scholar

25 See Hans Zeisel & Shari Seidman Diamond, The Jury Selection in the Mitchell-Scans Conspiracy Trial, 1976 A.B.F. Res. J. 151. Zeisel and Diamond conclude: It is undoubtedly true that to some extent the verdict in a jury trial depends on the composition of the jury, even though we do not know as yet the extent to which this is true. Nor do we know the conditions that make the composition of the jury more or less important. But there are good data supporting the lawyer's belief, however exaggerated, that the composition of the jury may affect the verdict. Id. at 173. For a discussion of earlier studies see James H. Davis, Robert M. Bray, & Robert W. Holt, The Empirical Study of Decision Processes in Juries, in June Louin Tapp & Felice J. Levine, eds., Law, Justice, and the Individual in Society: Psychological and Legal Issues 326 (New York: Holt, Rinehart & Winston, 1977).Google Scholar

26 A relatively small number of institutions entrust the principal responsibility for approving and disapproving proposals to ad hoc panels appointed to review a specific project. These panels may vary in size, in whether the members deliberate jointly or review the proposals individually, and in the vote required for approval. Review by such panels may be supplemented by review by other persons or groups before the proposal is submitted to the ad hoc panel and by provision for review of a panel's decision by some other person or committee under specified circumstances. For a description of one such committee see Melman, Kenneth L., Grossman, Michael & Curtis Morris, R. Jr, Emerging Assets and Liabilities of a Committee on Human Welfare and Experimentation, New Eng. J. Med. 28 2 427 (1970).CrossRefGoogle Scholar

27 When committee no. 2 reviews all the decisions of committee no. 1–both those approving and those disapproving proposals–it is possible either to view committee no. 1 as the human subjects committee and committee no. 2 as an appellate review committee, or committee no. 2 as the human subjects committee and committee no. 1 as an advisory prior review committee. Moreover, the institution's designation of one of these committees as its “institutional review board” may have little bearing on the actual interaction of the two groups.Google Scholar

28 See National Commission Survey, supra note 11, at 1–29, 1–31, 1–184, 1–250. In particular, members with a biomedical orientation (1) were more likely than behavioral scientists to feel that compensating subjects in proportion to the degree of risk to induce participation was improper, (2) were more likely to disapprove of withholding information from subjects about the design or purpose of the research when the disclosure might affect the validity of measurements, and (3) were less likely to express approval of research that involved both great risk and great importance but that was of no direct benefit to the subjects.Google Scholar

29 “[M]edical issues and risk ranked relatively high on the list for biomedical scientists; informed consent and scientific merit ranked high for behavioral and social scientists; while informed consent, consent forms, and the issues of privacy and confidentiality ranked relatively high for the other members.”Id. at 1–29. It should be noted, however, that no aspect of research proposals was so identified by more than a third of any type of member or by 20 percent of behavioral and social science or other members.Google Scholar

30 Id. at 1–60–1–62. No finding was specifically reported regarding the existence of a correlation between occupational and professional heterogeneity and percentage of proposals modified. Heterogeneity was defined in terms of whether the committee contained (1) biomedical researchers, (2) behavioral researchers, (3) those with full-time administrative positions in the institution, (4) those with legal backgrounds, and (5) community representatives. None of the committees included in the sample included fewer than three of these groups.Google Scholar

31 Id. at 1–61.Google Scholar

32 Id. at 1–33–1–34, 1–196.Google Scholar

33 Id. at 1–55, 1–196.Google Scholar

34 Id. at 1–56.Google Scholar

35 Id. at 1–33–1–34, 1–197–1–198.Google Scholar

36 Id. at 1–57. Prior review was not found to be related to the frequency of modifications concerning risk, scientific design, or confidentiality. The study did find a positive relationship to investigators' evaluations of the soundness of committee decisions.Google Scholar

37 Id. at 1–34–1–35, 1–203.Google Scholar

38 Only about 6 percent of medical school committees and 10 percent of university committees (where the university committee did not have jurisdiction over a medical school) but about one-third of hospital committees provided for such review. Id. at 1–200.Google Scholar

39 Id. at 1–59. No analysis appears to have been made of the effect of review of decisions disapproving or requiring modification of proposals. The Barber group study, however, found that review committees whose decision processes included a formal appeal procedure were more likely to have a relatively high rate of negative action. Barber et al., supra note 1, at 161–62.Google Scholar

40 Listed as examples of controls are the size of the committee, the type of research it reviews, and workload. National Commission Survey, supra note 11, at 1–11. For the most part in stating findings the report does not indicate whether or what controls were used.Google Scholar

41 In fairness it should be noted that the investigators acknowledge some of the limitations of the analysis. See National Commission Survey, supra note 11, at 1–6 n.1 and 1–11.Google Scholar

42 It should be noted that hospital and medical school committees did review some behavioral and survey research (e.g., survey research on problems related to medicine) and that research that could be broadly characterized as biomedical was conducted at institutions that had no medical school or hospital component. Biochemists on the nutrition faculties of universities studied the effect of experimental diets; professors of physical education studied the relation of exercise and health. The distinction in the kinds of research is, therefore, not a clean one.Google Scholar

43 The National Commission survey classified institutions with regard to NIH funding for research as follows: high volume, $4,000,000 or more; medium volume, $500,000-$3,999,999; low volume, less than $500,000. By this standard three committees were at institutions with a high volume of NIH-funded research, two committees were at institutions with a medium volume of NIH-funded research, and one committee was at an institution with a low volume of NIH-funded research. The institution with the lowest volume was a large state university which could be expected to and did have a large amount of research involving human subjects. None of the committees selected for study had reviewed fewer than 45 new proposals during the previous year. In comparison, the committees in the National Commission survey sample had reviewed on average 43 proposals during the year on which the study focused. National Commission Survey, supra note 11, at 1–50.Google Scholar

44 For a discussion of problems arising from the increase in judicial caseload, see, e.g., Henry J. Friendly, Federal Jurisdiction: A General View 15–54 (New York: Columbia University Press, 1973); Glenn R. Winters, ed., Court Congestion and Delay: Selected Readings (Chicago: American Judicature Society, 1971); Warren E. Burger, The State of the Federal Judiciary-1972, 58 A.B.A.J. 1049 (1972). The National Commission survey found that the volume of work that had to be accomplished by a committee was regarded as at least a somewhat serious problem by more than one-third of the committee members interviewed and was one of the most common problems listed by committee members. National Commission Survey, supra note 11, at 1–210.Google Scholar

45 At one institution the scope of jurisdiction had been expanded during the year preceding my visit. Prior to that time investigators had been required to submit only DHEW-funded research for review, although the submission of other research for which outside funding was sought was encouraged. The National Commission survey found that most committees reviewed unfunded research and research funded by sources other than DHEW, corroborating a similar finding by Barber and his associates. National Commission Survey, supra note 11, at 1–31–1–32, 1–187. See also Barber et al., supra note 1, at 149.Google Scholar

46 In one instance I made a return visit to attend a meeting of the committee; in another instance I was never informed of a committee meeting.Google Scholar

47 Information on the structure and operating procedures was obtained principally from the chairman and in some instances the executive secretary of the committee. Information on committee performance and the contributions of different kinds of members was obtained from interviews with the chairman and members, supplemented by my review of the minutes and observation of committee meetings.Google Scholar

48 For the first two committees I studied, I asked the chairman what objections had been raised during the previous year by committee members; in later interviews I asked specifically whether particular kinds of objections had been raised.Google Scholar

49 A shorter form could be used if the investigator also submitted a protocolGoogle Scholar

50 Including the chairman and executive secretary.Google Scholar

51 This was confirmed by an outside member of the committee who stated that the chairman of the human subjects committee held “the real appointing power.” The executive secretary stated that the chairman of the research committee consulted with the two administrators and possibly other members in making appointments.Google Scholar

52 Two members were not interviewed, but information on the listing of members suggests that neither one was a member of a racial or ethnic minority.Google Scholar

53 The psychology department had what was referred to as “our human subject pool”–student; taking the course in general psychology. Proposals that contemplated use of this pool had to be reviewed by a separate departmental committee to determine whether the proposal was a “legitimate” request for “this valuable resource,” to determine relative priority of investigators, to determine who was to read student papers submitted in lieu of participation as a subject, and similar matters.Google Scholar

54 The chairman of the human subjects committee appeared to equate “risks” with “significant issues of human rights and subject welfare.” It is far from clear that the two standards are equivalent. For instance, a proposal which involved some risk but a risk that was clearly outweighed by a great benefit might well not be viewed as presenting a significant issue of human rights and subject welfare.Google Scholar

55 Confirmed by the chairman and several members. Although the chairman stated that if the departmental committee disapproved the project it came before the human subjects committee unless the department found that the proposal was too far out of line, it would appear that the department chairman has the discretion to either refer the proposal or flatly disapprove it. One departmental committee member noted that the investigator could “of course appeal” to the human subjects committee “if he wanted to”–but he did not suggest that this had ever been done.Google Scholar

56 I was surprised by his statement and asked a follow-up question. “Only one that you remember sending on to the … [human subjects committee] even for consideration?” He responded: “In this department. In other words, we were able to handle it here…. We were able to do our own policing.” Statements of two other members confirmed that the committee reviewed only a very small fraction of research proposals involving human subjects.Google Scholar

57 Although the members tended to confirm the chairman's description of the operating procedures, there was some disagreement regarding the frequency of meetings.Google Scholar

58 Who had resigned from the committee because he was taking a position at another institution. He had not been replaced as of the time of my visit.Google Scholar

59 The committee had met about three months prior to the meeting I attended. A former member indicated, however, that the committee met only every six months, and at the meeting I attended the committee adopted a policy calling for quarterly meetings.Google Scholar

60 The six-month term apparently grew out of dissatisfaction with the performance of a nonmedical member who failed to turn in his evaluation sheets.Google Scholar

61 One nonmedical member stated that she had been selected for the committee because she was a neighbor of the chairman. A second knew two committee members personally, noting that one had been her doctor for some time and that “his wife is a personal friend.”Google Scholar

62 The other community woman had some training in medicine and was married to a physician. While she had no current association with a medical school or hospital, she said that her “background, working background, has always been on the edges of the medical field.”Google Scholar

63 Or at least all that I interviewed were. One member indicated that to the best of his knowledge there were no blacks on the committee.Google Scholar

64 Although the institution had no significant hospital component, it did have a small clinical research center in which biomedical research was conducted.Google Scholar

65 Two areas were not dealt with explicitly on the questionnaire (although some information on both points might be disclosed in response to other questions): (1) the existence of other means of obtaining the information sought; (2) the characteristics of the subject population, e.g., age, mental competency, or whether institutionalized.Google Scholar

66 Another member declared: “Sometimes the department head will not rubber stamp it and will actually go through it. But too often it does get rubber stamped.”Google Scholar

67 “If it is something which is clearly a no-risk problem or something that the committee has given me automatic authority to handle myself.” The statements of other members are vague regarding the existence of a blanket power to give provisional approval to no-risk proposals.Google Scholar

68 A student member stated that the chairman would generally appoint two M.D.s and one “other” member to review medical proposals. He stated further that students had been asked to review proposals (he had reviewed about 25) but that until recently students' areas of expertise had not been taken into account in making the appointments.Google Scholar

69 One member thought that only a two-thirds vote was required.Google Scholar

70 This committee had no general provision for provisional approval of proposals. There was a limited provision which had been used according to the chairman “three times or so” during the past year for emergency interim approval, which authorized interim approval if the investigator certified that a patient had a life-threatening disease that had not responded to any known standard therapeutic measure and there was reasonable evidence that a new or experimental procedure might benefit the patient, and there were no alternative experimental therapeutic procedures which would incur equal or less risk. The approval of the chairman, the executive secretary, and the subcommittee was required. This provisional approval was subject to ratification at the next meeting of the full committee.Google Scholar

71 Information acquired in an interview with committee chairmanGoogle Scholar

72 The executive secretary stated that the clinical research unit committee “will not appro protocol that has not been previously approved” by the human subjects committee.Google Scholar

73 I have no information on the affiliation of those members who were not interviewedGoogle Scholar

74 The points covered included the following: (1) a description and an assessment of the extent of the potential physical, psychological, social, legal, and other risks; (2) a description of alternative research procedures that the investigator had considered and an explanation of why such procedures had not been adopted; (3) procedures for minimizing risks including procedures for safeguarding the confidentiality of data; (4) the probable effectiveness of procedures adopted to minimize risks; (5) the potential benefits to the individual and to society; (6) an analysis of the risk/benefit ratio; (7) the characteristics of the subject population and the justification for using such special groups as prisoners, children, the mentally disabled, or others whose ability to give voluntary informed consent may be open to question; and (8) the procedures to be employed in obtaining informed consent. There was no inquiry regarding the type or amount of compensation to be offered for participation, nor was there any explicit inquiry into the source or nature of the data to be gathered or into the purpose of the research. Information on the source and nature of the data and the purpose of the research might be revealed, however, in response to some of the more general inquiries.Google Scholar

75 The procedure was described by the chairman and set forth in policy guidelines stating the review procedure. Two members of one subcommittee confirmed that subcommittees had met but indicated that such meetings were very rare. Several other committee members asserted that the subcommittees never met.Google Scholar

76 There was provision for emergency approval by the chairman or the department chairman of unexpected changes in a protocol. The procedure apparently was applicable only to changes in proposals that the committee had already approved–the chairman gave as examples cases where an investigator wished to use a different source or wished to use a different experimental drug that he felt was better. The chairman stated that in such situations after the proposal had been reviewed by the subcommittee, he sent the proposal to the full committee and polled the members by telephone. He said that no emergency cases had arisen during the previous year and that where such cases had arisen in the past, he had ultimately called an emergency meeting at the request of one of the members.Google Scholar

77 Typical was one member's statement that we “just sort of agree” to the proposals approved by the subcommittee and discuss only “the ones in which there have been objections by the review committee members.” One member asserted, however, that very often the committee discussed a proposal even where there was no dissenting vote on the subcommittee.Google Scholar

78 The chairman was not asked this question.Google Scholar

79 Of the 45 new proposals, 7 were approved as submitted. Of the 7, 2 did not involve human subjects and were approved as beyond the jurisdiction of the committee and 2 were umbrella proposals in which the committee approved a program encompassing a number of projects on the understanding that specific studies involving human subjects conducted under the program would be submitted for approval. Twenty-seven proposals were approved with some modification or clarification. No final action had been taken by the committee on the remaining 11 proposals during the year. Action on 8 of the 11 proposals had been deferred by the committee pending submission of additional information or a revised proposal by the investigator or some action by the research committee. In one instance the proposal was disapproved by the department head after objections had been raised by the committee and before a final decision had been rendered. One proposal was withdrawn by the investigator prior to consideration by the committee.Google Scholar

80 Of approximately 250 new proposals. The committee approved with some modification, condition, or clarification 61 percent of the proposals. Action was deferred on another 12 percent with no final action taken during the year, most often because the investigator did not respond to committee inquiries or submit an acceptable consent form. There were no substantial differences between clinical and behavioral/social science research in the proportion of proposals approved as submitted, approved with modifications or clarifications, and disapproved. Because the committee did not assign unique numbers to proposals, some double counting may have resulted, and the percentages are therefore reported as approximations. The extent of the possible error is, however, small, amounting to at most 7 out of approximately 250 proposals.Google Scholar

81 See text at notes 15–17 supra.Google Scholar

82 Of approximately 125 new applications that had been considered in the previous year. It appeared from the minutes that the committee had required modifications or clarifications in about 20 percent of the proposals and had deferred action on another 6 percent.Google Scholar

83 Another member estimated that the committee had required modifications or clarifications in two-thirds of the proposals it had considered. That the minutes may substantially understate the committee's impact is indicated when action at committee meetings on provisionally approved proposals is compared to action on provisions reviewed by assigned reviewers but not given provisional approval (either because immediate action had not been requested or because the reviewers had raised some issue that could not be resolved prior to the meeting). Of 35 proposals that had been given provisional approval on the recommendation of the assigned reviewers, the committee ratified the approval of 30 without requiring modification or clarification. In contrast, only 4 of the 29 proposals that had been reviewed by the assigned reviewers and had not received provisional approval were approved without modification or clarification. A number of factors may have contributed to the difference in the committee's behavior toward these two categories of proposals (see infra at 649–50), but clearly one contributing factor is that problems resolved in advance of the committee meeting when proposals have been given provisional approval are not dealt with until the committee meeting absent a request for provisional approval.Google Scholar

84 The committee was almost equally divided between those who saw no difference in the committee's treatment of behavioral and biomedical proposals and those who maintained that the committee was stricter toward biomedical proposals. It may be that some of those who saw no difference were saying only that they felt that the committee was equally conscientious and applied the same standards to both types of proposals. Other members, however, quite clearly did feel that the committee raised objections to behavioral proposals as often as it did to biomedical proposals.Google Scholar

85 Because of the operating procedures of this committee, the chairman was the only person interviewed who was in a position to be aware of all the objections. Unfortunately, the chairman's recollection on other matters was not always accurate. Two members gave estimates similar to the chairman's. Five other members thought that most proposals went through without modification or clarification. It is difficult to interpret the estimates given by committee members, however. The estimates with one exception purport to encompass both subcommittee action and full committee action and not to be limited to a particular member's subcommittee. Apart from the chairman (and possibly one other member), however, a member had no reliable way of knowing the objections and questions raised and cleared up at the subcommittee level other than his own objections and, perhaps, the objections of other members of the subcommittee on which he served. Since the five members who thought that most proposals were approved without modification included members of all three subcommittees, the variation in the estimates cannot be explained by the hypothesis that the three subcommittees differed in their behavior. Two other members responded to my question by stating that they could not answer it with respect to the subcommittee level, one explaining that “I think that there are many minor suggestions that are rapidly answered by the investigator and don't ever come to the attention of the full committee.”Google Scholar

86 Based on estimates by the chairman. Because of the nature of the review process at this institution, only the chairman was aware of the action taken by the full committee on proposals considered by mail ballot and thus could provide an accurate estimate. The reliability of the chairman's estimate is enhanced in this instance by the fact that it was based on a recent review of the minutes. The estimates given by other members varied. Several members confirmed the chairman's estimate. One member stated that he had raised objections to virtually all the proposals reviewed and another that he had raised objections to 15 to 20 percent of the proposals; these estimates of personal action are not necessarily inconsistent with the chairman's estimate of committee action.Google Scholar

87 Slightly more than half the proposals reviewed were proposals up for recertification. The procedure for reviewing proposals up for recertification was substantially the same as that used for new proposals. It should also be noted that the 10 percent figure reflects only “modifications” required by the committee. It does not cover cases in which the investigator responded to committee inquiries except to the extent that the chairman viewed the response as constituting a modification of the proposal. Because of the system of review by mail employed by this committee, such requests probably were confined largely to cases in which the full committee or an ad hoc committee of experts met on the proposal.Google Scholar

88 National Commission Survey, supra note 11, at 1–17. Changes were made in an additional 12 percent as the result of conversations between committee members and investigators prior to their review.Google Scholar

89 Fifty-four percent of the clinical and 50 percent of the behavioral/survey proposals.Google Scholar

90 Consent appears to have been the most frequent basis for objection on Committees C and F as well, but the frequency with which it was raised cannot be estimated. Most members of Committee F viewed some aspect of the consent procedure or consent form as their most frequent basis for objection. Similarly, on Committee C consent issues were identified by the chairman and by two members as probably the most frequent basis for objection. (Only 10 percent of the proposals reviewed by Committee C, however, were modified on any basis.)Google Scholar

91 On Committee A the most common modification, however, was the requirement, imposed in 14 instances, that one or more standard sentences be added to the consent form concerning the subject's opportunity to ask questions about procedures and the subject's right to withdraw from the study at any time without prejudice, and that a decision by the subject not to participate in the research would not prejudice his right to treatment. The inclusion of these points was required as a matter of committee policy, and the requirement that consent forms not containing these sentences be modified to include them appeared to occur as a matter of course.Google Scholar

92 National Commission Survey, supra note 11, at 1–18.Google Scholar

93 University of Michigan Survey Research Center, Institute for Social Research, Research Involving Human Subjects 30–31 (Oct. 2, 1976) (the National Commission Survey preliminary report).Google Scholar

94 The regulations provide that informed consent shall ordinarily be documented either by an instrument that embodies all the elements of an informed consent (a “long form”) or by an instrument indicating that the basic elements of informed consent have been presented orally to the subject (a “short form”). 45 C.F.R. § 46.110 (1978). A committee may permit modification of either primary procedure if it finds “(1) [t]hat the risk to any subject is minimal, (2) that the use of either of the primary procedures … would surely invalidate objectives of considerable immediate importance, and (3) that any reasonable alternative means for attaining these objectives would be less advantageous to the subjects.”Google Scholar

95 Although the policy memorandum lists both the long form and the short form as the “primary” procedures for documenting informed consent, the investigator questionnaire provides that if there is any element of risk the elements of an informed consent shall be clearly stated in a document to be signed by the subject or by a legally authorized representative. Both documents indicate that there are circumstances in which the committee may accept oral consent or dispense entirely with consent, although the investigator questionnaire appears to equate oral consent with the short-form procedure.Google Scholar

96 Indeed, on two occasions, both involving behavioral or survey research, the committee conditioned approval on the investigator modifying the proposal to eliminate written consent, replacing it by reliance on implied consent. At least one of these proposals involved a mail questionnaire. It seems quite probable that other proposals for behavioral or survey research that were approved without modification envisioned implied consent.Google Scholar

97 The nature of the subject population may also be a factor. The executive secretary explained that “obtaining informed consent from a population group that is working on the project is entirely different, we believe, than [securing consent from] a group of individuals who have never seen a hospital before.”Google Scholar

98 A minority of the members of Committees A and E had taken the position that consent should be obtained for such research. Committee E required that consent of some type be obtained where such materials were obtained for research.Google Scholar

99 The executive secretary of Committee E stated that such a procedure had been discussed on perhaps two occasions, but he could not recall any instance in which it had been done. The chairman of Committee D said that his committee had “rarely if ever” imposed such a requirement. The National Commission survey found that while a few committees always required a cooling-off period, most seldom or never required such a procedure. National Commission Survey, supra note 11, at 1–43. For a discussion of the need for a cooling-off period see Comment, Proposed Changes for Obtaining Consent from Experimental Subjects, 1 Law and Human Behavior 403, 415–16 (1977).Google Scholar

100 The chairman of Committee C stated that monetary inducements were infrequent and that the highest payment he could recall was $50 when the subject was required to spend the night. The executive secretary of Committee E stated that in the vast majority of cases payments were in the $5 to $25 range. The chairman of Committee F maintained that payments had been nominal. This was confirmed by two other members, one of whom said that students were paid at student labor rates, which were set according to university guidelines and amounted to perhaps $2 or $3 an hour. The chairman of Committee B when asked about the amount of compensation responded, “Oh, a dollar maybe, in some cases.” The National Commission survey found that in the vast majority of projects investigators did not pay subjects and that where subjects were paid, the amount was small, “frequently $l-$5.” National Commission Survey, supra note 11, at 1–20.Google Scholar

101 From $75-$150 to $25. Although some members may have viewed the problem as one of “unfair competition,” other members apparently did see the problem as one of undue inducement.Google Scholar

102 An analysis of the previous year's minutes shows no instance in which compensation had been reduced. The question to the chairman, however, was not limited to the previous year.Google Scholar

103 The interviews clearly indicated that the issue came up rarely, if at all, on Committee A. On Committee F most members commenting felt that the issue had not been raised.Google Scholar

104 No proposals had come before Committee A involving deception, and it is unclear whether any proposals had come before Committee E.Google Scholar

105 Half the clinical proposals in which the committee required some modification or clarification on grounds other than consent were eventually approved without modification except as to the consent. Of 34 instances in which the committee sought additional information concerning the risks, benefits, purpose, or research procedures, 22 were approved without modification other than to the consent. On the other hand, the committee did reject outright several clinical proposals and require modifications in a number of others, while in still other instances action was deferred and the proposal never approved.Google Scholar

106 Although an analysis of the minutes indicated that the committee had required modifications or clarifications on this basis in connection with only 6 of more than 125 proposals it had considered, apart from consent no other issue was a more frequent basis for requiring modification or clarification. Issues of privacy and confidentiality were among the issues most commonly listed by members as issues that they had raised.Google Scholar

107 The questioning on this point was generally open-ended, and no systematic effort was made to determine whether a committee had required an investigator to take specific steps. Thus, the lack of an affirmative indication that a committee required investigators to adopt a specific procedure for safeguarding confidentiality does not mean that no such action was required–although for Committees A, D, and E, it does mean that no such action was disclosed by the minutes.Google Scholar

108 The National Commission survey found that 60 percent of the committees surveyed had approved proposals which gathered data anonymously. National Commission Survey, supra note 11, at 1–74. No differentiation was attempted in the National Commission survey between steps that investigators voluntarily undertook and those that the human subjects committees had required investigators to take. The survey also found that 70 percent of the committees surveyed had approved proposals where identifying information was stored separately and that 37 percent had approved proposals where identifying information would be destroyed at a specified time. Id.Google Scholar

109 The chairman of the psychology department at Gondor State University felt, however, that Committee B had been “quite reasonable” in imposing restrictions by comparison with committees at other institutions. He noted particularly that the retention of data for use in longitudinal studies had been permitted.Google Scholar

110 The executive secretary of Committee E indicated that investigators often indicated that materials would be kept in a locked file cabinet.Google Scholar

111 The questionnaire asked for information on age; type of institution (e.g., penal, hospital, etc.); whether the subject is voluntarily or involuntarily committed; whether the subject is part of any special group (e.g., military, probation, school); and whether the mental or physical health of the subject is relevant to the research.Google Scholar

112 In another instance a proposal for a study involving a captive population was disapproved by the department head after questions were raised by the committee.Google Scholar

113 The chairman of the psychology department maintained that blacks and disadvantaged groups may have been underrepresented as research subjects. “Because our faculty is white and our students are white we may be shut out of parts of the community.”Google Scholar

114 A psychologist on the committee stated that department members were opposed to requiring participation but rather circulated sign-up lists. This member later acknowledged that there was one department in which there might be a little informal pressure on students in a given course to exchange with each other to do student projects. It was less clear whether participation had been required at the other two institutions. Committee F had no policy regarding the use of students. But while the chairman could not recall any instance in which the investigator had proposed requiring students to participate as a course requirement, several members of Committee F indicated that an objection had been raised to a proposal in which class credit was used as an inducement. The executive secretary of Committee A stated that “the committee doesn't particularly like the idea” of using students and that “we try to use primarily either patients” or “in-house people”–nurses, employees–and that volunteers were obtained by passing the word around among the interns and residents. However, one member stated that the committee had never turned down a proposal because it involved the use of students and that students were in fact used as subjects. He stressed that the committee required informed consent and no coercion.Google Scholar

115 The department chairman stated that about 10 percent chose to opt out of the subject pool.Google Scholar

116 The policy was the outgrowth of a situation in which the committee felt that the students were under a subtle pressure to participate. Under this policy the use of students was permissible in some situations, and it appeared that students were in fact often used for some types of low-risk procedures as, e.g., when blood was needed for research purposes.Google Scholar

117 National Commission Survey, supra note 11, at 1–35–1–36, 1–204, 1–207.Google Scholar

118 Id. at 1–35–1–36, 1–44.Google Scholar

119 Among the points covered were (1) the current status of the project (whether it was still under way, or if it had been abandoned, the reason for abandonment); (2) the number of subjects involved in the project during the preceding 12 months and how many of these subjects were minors, pregnant, prisoners, or mentally retarded; (3) details on how consent had been obtained together with a description of problems encountered in obtaining consent; (4) whether complications had occurred; (5) whether there had been or were contemplated alterations in the research procedures; and (6) whether the safeguards in the protocol previously approved by the committee had been adequate.Google Scholar

120 No proposals were flatly disapproved. Of the applications for recertification, 25 percent were approved with modifications or clarifications. Action on 2 percent of the proposals was deferred and no final action taken during the year. One proposal was approved in part with modifications, approved in part as an umbrella proposal subject to the requirement that the investigator submit particular applications for review, and in part deferred. Although apart from consent, no modification or clarification had been required on more than two occasions, the kinds of modifications and clarifications covered a broad range.Google Scholar

121 The chairman stated that when instances of noncompliance came to light, he generally tried to work the matter out informally by a telephone call to the investigator.Google Scholar

122 The proposal was assigned to a member of the human subjects committee for detailed review and was presented to the committee as if it were a new proposal. A proposal up for recertification did not have to go before the research committee, however, unless there had been some change in the research procedures.Google Scholar

123 The minutes disclosed one instance in which the committee had required that the investigator provide it with a full report of all the results of the study to date. On another occasion the committee approved the continuation of a project on the condition that it be provided with a full progress report after six months.Google Scholar

124 The executive secretary noted that at times the committee would question an informed consent that it had approved when the proposal was originally presented because “their thinking process has evolved over the last year and things which they took for granted are now being questioned.” By far the most common basis for modification of proposals up for recertification was the omission of standard sentences, and it can be inferred that when the proposal was originally reviewed, the committee's policy with reference to these sentences had not fully evolved.Google Scholar

125 This was confirmed by another member who stated that the committee had taken the position that “it would not certify or look at the new proposals of the investigator unless the changes were made” and that “the second response [of the committee] was anguish–at how they can enforce this on a more general basis.” The committee appeared to be seriously considering proposals to initiate monitoring procedures of some type. In addition, the executive secretary stated that he was conducting a chart review in certain cases where the committee felt that there might be a violation to see “that the informed consent is being signed and used the way we had agreed to.”Google Scholar

126 The investigator was required to disclose the number of subjects studied, how the subjects were induced to participate, how informed consent was obtained, any complications that may have been attributable to the experimental procedures, the scientific information obtained, and any additions or changes in the protocol since the last approval. He was required to supply a sample copy of the consent form in current use, to state the number of subjects to be studied during the next 12 months, to describe in detail any proposed changes or additions, and to summarize any new pertinent literature. The chairman stated that if the proposal had no problems only annual recertification was required. If the proposal did involve “some potential risk,” recertification might be required after 6 months, 3 months, or even shorter periods. In the majority of cases, however, the proposals were approved for 6 months or a year.Google Scholar

127 The chairman of the psychology department stressed that the members of the department's committee were “active participants in the department's graduate programs and in this capacity were kept up to date on the status of projects. He said that while investigators were not required to file reports, they were asked to submit any substantial changes “and the frequency of resubmittals” was sufficient to give us “some confidence that this is actually done.” Nonetheless in one instance a departure from committee requirements had come to light only because a member of the community had come in and questioned the department chairman about the project.Google Scholar

128 Committee D always and Committee F generally had required only annual recertification. A member of Committee F stated that in perhaps three or four instances during the past year the committee had required that the investigator report back after three months.Google Scholar

129 In most (perhaps all) instances the minutes show that no change was contemplated in the proposal. It is possible that cases in which some change was contemplated were grouped in the minutes with new proposals.Google Scholar

130 The members of Committee C expressed confidence that at least those proposals seeking outside funding were submitted.Google Scholar

131 On Committee F this may have resulted from the fact that the requirement that unfunded research be submitted had only recently been instituted and even some committee members were not aware of it. One member who was unaware of the requirement stated that proposals were not submitted for review both because investigators were unaware of the requirement–“that's probably the primary reason”–but also because faculty members questioned the need for review of some kinds of proposals. Echoing a concern expressed by several members, he declared: “I don't think that everything should go to the central committee, because it's just going to … flood the committee.” The chairman of Committee D stated that biomedical proposals generally were referred to the committee but that social science projects came to the committee primarily if they were funded or if the faculty person involved felt that the study should be reviewed. Both he and the grants administrator viewed the lack of a more effective mechanism for assuring the submission of proposals as a defect in the committee's procedures.Google Scholar

132 Studies of small group behavior confirm that the decisions of committee members acting individually and the decisions made by a committee after group discussion are likely to differ. See Dean G. Pruitt, Choice Shifts in Group Discussion: An Introductory Review, 20 J. Personality & Soc. Psych. 339 (1971). One theory advanced to explain the difference attributes the shift to the influence of the arguments advanced in group discussion. This theory would seem particularly persuasive in a context in which the members of the group cannot be assumed to be equally aware of the relevant considerations. For a suggestion that a number of factors were at work and a general comparison of individual and group decision making and problem solving, see A. Paul Hare, Handbook of Small Group Research 307–29 (2d ed. New York: Free Press, 1976).Google Scholar

133 Comments were mixed on the mail ballot at Committee B. A meeting was held if requested by any member, and it was suggested by one member that meetings had been held on most proposals.Google Scholar

134 Another medical member agreed that “to a certain extent” interchange might have been inhibited by the lack of meetings. Two other medical members and all of the nonmedical members interviewed thought more meetings would be useful. The law academic, while noting that there were five or six instances in which he had approved proposals with comments, stated that he would feel freer to raise questions at a meeting. The minister declared that “as soon as you have a discussion things begin to well up”–new considerations “come to light … questions … lead to questions….” While there was strong sentiment in favor of more meetings, most members felt that given the number of proposals that the committee had to review, meetings on all proposals simply were not feasible.Google Scholar

135 The former law member stated that the “lay” members, including himself, were less effective than they otherwise might be because they were unaware of the scientific objections that might exist. The lack of feedback from the committee had been a serious problem for him. He said that he really did not know what criticisms other members were raising and similarly they did not know what criticisms he was raising.Google Scholar

136 Their perception received some confirmation from the minutes. Of 35 proposals given provisional approval by the assigned reviewers, 30 were approved by the full committee without modification or clarification. In contrast, of 29 proposals that had been reviewed by the assigned reviewers and no provisional approval granted (either because immediate action had not been requested or because the reviewers had raised some issue that could not be resolved prior to the meeting), only 4 were approved without modification or clarification. Although a number of factors probably contribute to this difference (see infra at 649–50), proposals that had not been given provisional approval are likely to include a number to which the assigned reviewers had raised objections.Google Scholar

137 The volume of work was one of the more commonly listed problems for committees included in the National Commission survey as well, being listed as at least a somewhat serious problem by one-fourth of the committees and as a minor problem by an additional 38 percent. National Commission Survey, supra note 11, at 1–37–1–38, 1–209.Google Scholar

138 The chairman of Committee D stated that originally the members had received copies of all proposals but that as the volume of work increased the committee had been forced to use assigned reviewers even for nonurgent reviews. A member of Committee F, commenting on the pro forma review given most proposals at the full committee level, stated that “it would be good if we could review all of these proposals, but that isn't feasible.” Members of Committee C defended the failure of the committee to meet on all proposals on the ground that given the number of proposals that the committee had to review, meetings on all proposals simply were not feasible. There were, one member said, “just too damn many of them.” The National Commission survey found that the assignment of subcommittees and individual committee members to review proposals was more likely to be made by committees with heavy workloads. Nevertheless, the frequency with which these assignments were made did not seem to be associated with the number of member-hours spent on each proposal. National Commission Survey, supra note 11, at 1–56.Google Scholar

139 The assigned reviewers at Committee C could require review by an ad hoc committee of experts and could require that a proposal be reviewed at a committee meeting rather than by mail.Google Scholar

140 Interaction between the subcommittee and the full committee took a variety of forms. In some instances where the subcommittee had deferred action pending the response of the investigator to inquiries or had approved the proposal on condition that the investigator supply additional information, the investigator supplied the requested information in the interim between the subcommittee and full committee meetings and the full committee approved the proposal without restriction. In some instances the subcommittee referred the proposal without recommendation to be decided by the full committee. Apart from such cases, both the interviews and an analysis of the minutes indicated that in most instances the full committee ratified the recommendation of the subcommittee and that when the full committee had overturned the subcommittee recommendation, the change had generally been in the direction of imposing additional restrictions. Where the full committee imposed a restriction going beyond those recommended by the subcommittee, about half the time the additional restriction was to the consent form.Google Scholar

141 One community woman said that she went along with the subcommittee on the medical aspects but not the other aspects. The other community woman had been on the committee a very short time, but indicated that she tended to rely on the subcommittee. Her comments stressed her inability to evaluate the medical aspects of proposals.Google Scholar

142 The chairman indicated that “once in a while” a member would raise a question about a proposal recommended for approval. But while occasionally a member may have had some independent knowledge of the contents of a proposal, generally members were in no position to raise questions on proposals unanimously recommended for approval, and there was widespread agreement that the full committee generally gave such proposals only “perfunctory” or “pro forma” review.Google Scholar

143 See note 142 supra.Google Scholar

144 I have no data on this point for Committee D.Google Scholar

145 Particularly noteworthy was the difference in outlook among committee members regarding the use of deception in research, for it is clear that the psychology department's review committee had approved a number of such proposals without referring them to the human subjects committee.Google Scholar

146 The three types were (1) proposals that did not “really” involve research or “directly involve humans in any way,” (2) projects up for annual recertification when the project involved no risk to the subjects, and (3) proposals that after having been approved by the committee had not been funded for six months and were up for recertification prior to commencement of the study. All three categories are consistent in spirit with the recommendation of the National Commission regarding the circumstances in which such a provisional review procedure may be appropriate, although the first category is too broad and amorphous to satisfy the requirement that such categories be carefully defined. 43 Fed. Reg. 56,182 (1978). The chairman listed as an example of the first category of proposal studies using blood or tissues that had been obtained in routine management of patients. Although not one of the types of procedures specifically listed by the Commission, such studies are noninvasive and would appear to ordinarily involve minimal risk to subjects. Such studies clearly present less risk than the collection of blood samples by venipuncture—one of the types of research that is specifically listed. Both types of studies may present privacy problems in particular circumstances if data of a sensitive variety are collected.Google Scholar

147 He gave as examples of the kinds of situations in which the committee had authorized him to grant provisional approval the use of discarded blood samples or of pathological tissues collected at an autopsy and covered by an autopsy permission form.Google Scholar

148 In one other instance action was deferred and (as far as my notes show) no final action was taken during the year.Google Scholar

149 At the meeting of Committee C that I attended the committee ratified (en masse, without discussion) the provisional approval given by the chairman to 32 proposals. Presumably these 32 proposals had all been submitted since the committee's previous meeting three months before.Google Scholar

150 During the year preceding my visit approximately 35 proposals had been given provisional approval on the recommendation of the assigned reviewers. The committee had required modification of 4 of these proposals and in one instance had deferred action, requesting that the investigator supply additional information. The provisional approval of the remaining proposals was ratified without modification or clarification. Of 29 proposals reviewed by the assigned reviewers where no provisional approval had been granted (either because immediate action had not been requested or because the reviewers had raised some issue that could not be resolved prior to the meeting), only 4 were approved without modification or clarification.Google Scholar

151 Information was obtained on the performance of the clinical research unit committee at Gondor State. Although the proposals reviewed by the clinical research unit committee were only a small fraction of the total workload of the human subjects committee, they were probably a larger fraction of the proposals that presented a serious risk. According to its chairman, that committee had disapproved at least one and required modification in a majority of the proposals reviewed in the last year. The required modifications were related to the consent procedures and research design, the reduction of the risk to subjects, and, on rare occasions, to preservation of the confidentiality of data. Although there was disagreement regarding the extent to which the committee did more than evaluate scientific merit, it was clear that to some extent the clinical research unit committee did involve itself in subject protection issues. See discussion supra at 601–2.Google Scholar

152 See text at 608, 612, and 616 supra.Google Scholar

153 Both the medical ethicist and the head of the oncology department—while representing polar viewpoints in their positions on issues coming before the committee—agreed that proposals involving undue risk generally did not reach the human subjects committee. The head of the oncology department attributed this to the prior review by the department head and the research committee. The medical ethicist stated that there had been an occasion or two on which the committee had seen a proposal for an experiment which it would have objected to only to be told at the meeting that it did not get through the research committee.Google Scholar

154 Studies of the effect of size on group behavior in other contexts have found “that group size has significant effects on aspects of individual and group performance, on the nature of interaction and distribution of participation of group members, on group organization, on conformity and consensus, and on member satisfaction.” Edwin J. Thomas & Clinton F. Fink, Effects of Group Size, in A. Paul Hare, Edgar F. Borgatta, & Robert F. Bales, eds., Small Groups: Studies in Social Interaction 525, 534 (rev. ed. New York: Alfred A. Knopf, 1965). See Richard O. Lempert, Uncovering “Nondiscernible” Differences: Empirical Research and the Jury-Size Cases, 73 Mich. L. Rev. 644, 684–98 (1975); Hare, supra note 132, at 214–31. Although increasing the size of the group was found to have both negative and positive effects, larger groups had tended to produce more ideas and had performed better in solving problems that required a number of different inputs than did groups with fewer members. These results are important for the human subjects committee context, since the review of proposals requires the recognition of a broad range of issues and bringing to bear a number of different items of information.Google Scholar

155 For a discussion of the effect of composition on the performance of human subjects committees see pp. 659–62 infra.Google Scholar

156 Some studies of small groups and of juries have found that inequality of participation of members in group discussions increases as groups become larger. See Hare, supra note 132, at 219–21; James H. Davis, Group Performance 73 (Reading, Mass.: Addison-Wesley Pub. Co. 1969); But see Thomas & Fink, supra note 154, at 528 (questioning the generality of these findings). There are also indications that groups may tend to be influenced by their more talkative members. See Lempert, supra note 154, at 692–94. The significance of these findings for human subjects committees is doubtful, however. First, we cannot infer from the greater inequality of participation that the committee is deprived of relevant arguments that would be introduced if it were smaller. It may mean only that less vocal members speak only when they have a point that had not been raised by a more vocal member. Second, human subjects committees are composed of persons who are chosen because of their specialized knowledge or perspective. One would expect—and the interviews tended to confirm—that those members with specialized knowledge were expected to speak and that their views were given particular weight by other members on issues falling within their areas of expertise.Google Scholar

157 See Hare, supra note 132, at 219.Google Scholar

158 The finding of the National Commission survey that, after controlling for workload and type of research, the size of the committee had no independent relation to performance, National Commission Survey, supra note 11, at 1–49, is not inconsistent. The effects of committee size are varied and in part indirect. Although the size of Committee E made possible structural and compositional features that contributed to the effectiveness of review, it does not follow that larger committees are more effective on the whole than smaller committees. The most that can be said is that increasing the size of the committee makes possible a more varied membership and the creation of review structures not open to a smaller committee. A sidelight of interest is that Committee E, in addition to being the largest of the committees studied, had the heaviest workload. A heavy workload may tend to tead to an increased size in an effort to provide additional manpower to share the workload. Or it may be that both the size of the committee and the workload are a product of the size of the institution and of the number of persons at the institution who are actively engaged in research.Google Scholar

159 Maurice Rosenberg, Judicial Discretion of the Trial Court, Viewed from Above, 22 Syr. L. Rev. 635, 641–42 (1971) (footnote omitted).Google Scholar

160 Kenneth Culp Davis, Administrative Law Text 525 (3d ed. St. Paul, Minn.: West Pub. Co., 1972). Although the scope of judicial review of administrative action ranges from zero to one hundred per cent, that is, from complete unreviewability to complete substitution of judicial judgment on all questions, the dominant tendency in both state courts and federal courts is toward the middle position known as the substantial-evidence rule. Under this rule, the court decides questions of law but it limits itself to the test of reasonableness in reviewing findings of fact. Broadly, questions of law include not only common law, statutory interpretation and constitutional law, but also questions of administrative jurisdiction, of fair administrative procedure, and of protection against arbitrary or capricious action or abuse of discretionGoogle Scholar

161 9 Charles Alan Wright and Arthur R. Miller, Federal Practice and Procedure §§ 2522, 2524, 2531 (St. Paul, Minn.: West Pub. Co., 1971); 11 id. §§ 2806, 2810. A judgment notwithstanding a jury verdict may be entered only on a finding that there is not sufficient evidence for a jury reasonably to find a verdict for the party against whom the motion is directed. The trial judge may also overturn a jury verdict and order a new trial if the judge finds that the verdict is against the clear weight of the evidence.Google Scholar

162 “From the earliest times appellate courts have been empowered to reverse for errors of law, to announce the rules which are to be applied, and to ensure uniformity in the rules applied by various inferior tribunals.” Charles Alan Wright, The Doubtful Omniscience of Appellate Courts, 41 Minn. L. Rev. 751, 779 (1957).Google Scholar

163 See Paul D. Carrington, The Power of District Judges and the Responsibility of Courts of Appeals, 3 Ga. L. Rev. 507, 527 (1969).Google Scholar

164 “The trial judge doesn't have the appellate judge's advantage of many days of contemplation before deciding questions of law. He has to more or less shoot from the hip.” John J. Sirica, To Set the Record Straight 72 (New York: W. W. Norton & Co., 1979).Google Scholar

165 Carrington, supra note 163, at 527; see Rosenberg, supra note 159, at 642.Google Scholar

166 5 Roscoe Pound, Jurisprudence 606–7 (St. Paul, Minn.: West Pub. Co., 1959).Google Scholar

167 The National Commission Survey, supra note 11, at 1–34, 1–35, 1–200–1–202, found that only 20 percent of the committees surveyed provided routinely for review within the institution of decisions approving proposals and that only 13 percent provided automatically for review of decisions disapproving proposals. The survey reported, however, that nearly half of the committees allowed investigators to appeal a decision. Unfortunately, however, this total included committees which allowed the investigator to appeal an adverse decision to it—in effect to request a rehearing. No separate figures were given on committees which allowed an investigator to appeal to some other person or group within the institution.Google Scholar

168 The findings of the National Commission survey suggest that the experience of Committee E probably was typical, but are less clear as to Committee D. Less than 5 percent of the committees surveyed reported instances in which committee decisions approving proposals had been reversed. National Commission Survey 1–34. But see Mortimer J. Lipsett, John C. Fletcher, & Marian Secundy, Research Review at NIH, Hastings Center Rep. Feb. 1979, at 18, 19–20 (discussing the experience with review panels at NIH). It was reported that half of the committees which had a decision appealed had a rejection reversed by a subsequent review, but it is unclear what proportion of the committees with an appeals mechanism had had a decision reversed or to what extent the reversals were by the human subjects committee itself. National Commission Survey, supra note 11, at 1–35. Although members of committees whose project approvals were reviewed by a second committee did not think that the later review influenced their committee in any way, no findings were reported regarding whether members of committees whose decisions disapproving or modifying proposals were subject to further review had been influenced in any way by the later review. Id. at 1–34–1–35.Google Scholar

169 Research on the effect of an assigned decision rule on small group behavior has largely been confined to juries. See James H. Davis, Group Decision and Procedural Justice, in M. Fishbein ed., Progress in Social Psychology (Hillsdale, N.J.: Erlbaum Associates, 1979, forthcoming, reviewed in draft). The Supreme Court's decisions in Apodaca v. Oregon, 406 U.S. 404 (1972), and Johnson v. Louisiana, 406 U.S. 356 (1972), upholding the constitutionality of less than unanimous verdicts in noncapital state criminal trials, have generated increasing scholarly interest in the effect of permitting less than unanimous jury verdicts. As Zeisel has noted, abandonment of the unanimity requirement is an effective way of nullifying the potency of minority viewpoints, since it allows the majority to simply disregard minority arguments. Hans Zeisel, … And Then There Were None: The Diminution of the Federal Jury, 38 U. Chi. L. Rev. 710, 721–24 (1971). Studies of mock juries suggest that majority rule juries do not necessarily halt discussion when the requisite majority has been reached. See Michael J. Saks, Jury Verdicts: The Role of Group Size and Social Decision Rule 92–93 (Lexington, Mass.: D.C. Heath & Co., Lexington Books, 1977) (majority rule juries found to continue discussion for substantial periods); Norbert L. Kerr, Robert S. Atkin, Garold Stasser, David Meek, Robert W. Holt, & James H. Davis, Guilt Beyond a Reasonable Doubt: Effects of Concept Definition and Assigned Decision Rule on the Judgments of Mock Jurors, 34 J. of Personality & Soc. Psych. 282, 292 (1976). The total deliberation time of unanimous rule juries had tended to be longer, however (see Saks, supra, at 93; Kerr et al., supra, at 289; James H. Davis, Group Decision and Procedural Justice, supra), and one study found that more than half of the majority rule juries in the study did end their discussions when the requisite majority was obtained. See Kerr et al., supra, at 292. Moreover, Saks in his study of mock juries found that once a majority had been reached, further discussion had never resulted in dislodging that majority—suggesting that while majority rule juries may not terminate discussion immediately upon achieving a majority, members of the majority may thereafter simply disregard minority arguments. Saks, supra, at 94. The available evidence strongly suggests that the difference in end result between unanimous rule and majority rule juries is likely to be small: that juries in which a majority initially (at the commencement of jury deliberations) favors conviction usually convict under a unanimity requirement; that juries in which a majority initially favors acquittal usually acquit. See Harry Kalven, Jr., & Hans Zeisel, The American Jury 487–88 (Chicago: University of Chicago Press, Phoenix Books, 1971) (study of actual juries); Saks, supra, at 99–101; James H. Davis, Group Decision and Procedural Justice, supra; Kerr et al., supra, at 293. Some shifts do occur, however. Kalven and Zeisel found that 5 percent of the juries in which 7 to 11 members initially favored conviction rendered a verdict of acquittal. Kalven & Zeisel, supra. See, e.g., Kerr et al., supra, at 288 (a four-to-two initial preference was frequently insufficient to determine the result of a unanimous-rule six-person jury); William F. Godwin & Frank Restle, The Road to Agreement: Subgroup Pressure in Small Group Consensus Processes, 30 J. Personality & Soc. Psych. 500 (1974) (study of ad hoc groups). Although there are similarities in function between the jury and the human subjects protection committee, differences in composition, continuity of membership, and operating procedures preclude an assumption that the effect of a majority as opposed to a unanimous decision rule on jury behavior is directly applicable to human subjects committees. Because members of human subjects committees are likely as a result of their past association on the committee and, for many, in other activities at the institution to have developed feelings of collegiality, they may be less disposed to cut off discussion when a majority has been reached and more willing to shape committee action to accommodate the views of minority members. The costs of accommodating minority sentiment are less in the human subjects context. The decision for a human subjects committee is rarely one of voting a proposal up or down but rather whether to impose a restriction that the majority may feel to be unnecessary but are unlikely to find harmful.Google Scholar

170 Nearly two-thirds of the committees included in the National Commission survey required only a majority vote. Twenty-four percent required that proposals be unanimously approved. Only 2 percent reported that they required approval by other than a majority or unanimous vote. National Commission Survey, supra note 11, at 1–195.Google Scholar

171 See note 69, supra, and accompanying text.Google Scholar

172 This was clearly the case on Committees A, D, and E and appears to have been true of the other three committees as well. The evidence was least satisfactory as regards Committee C, which reviewed most proposals by mail. The chairman stated that while the committee operated officially by majority vote, he liked to see the committee as close to unanimous as possible and that if on mail review a member raised objections, it was his practice to send a note to the committee indicating that some members had raised objections and asking them to take another look at the proposal.Google Scholar

173 National Commission Survey, supra note 11, at 1–57, 1–195. Barber and associates reported that 91 percent of their respondents stated that when their committee approved a proposal, its action was almost always unanimous. Forty-two percent reported a unanimous vote requirement. Barber et al., supra note 1, at 154. See also Gray, supra note 20, at 33.Google Scholar

174 Six if the committee with student members is counted as a committee with “other” members.Google Scholar

175 Per capita. Unless otherwise indicated, comparisons of the comparative behavior of different kinds of members should be understood to refer to the behavior of an “average” member of that type.Google Scholar

176 Only one member suggested that nonmedical members had tended to object more often. Those members who stated that the biomedical members objected more often generally attributed this to their greater familiarity with the issues.Google Scholar

177 Several members thought that no proposals or virtually no proposals involving behavioral or social science research had been discussed at the full committee level during the preceding year. Some members thought that the behavioral scientists had objected most often, others that the biomedical members had objected most often, still others that the “lay” members objected most often, and some saw no difference in the frequency with which the three groups tended to object.Google Scholar

178 The chairman of Committee C estimated that the medical and nonmedical members objected with about equal frequency. Because most of the work was done by mail, members (except for the chairman) tended to be unaware of the responses of other members to most proposals. The members who responded to the question—some but not all of whom expressly noted that their responses were limited to behavior at committee meetings—gave varying answers. Several nonstudent members and one student member of Committee D did suggest that the students had tended to be somewhat more critical of proposals than had other members. One of the two psychologists, moreover, maintained that the psychologists were “probably somewhat more critical than the mean of the committee, but I would say a little less so than the students on the committee.” On the other hand, two student members denied that as a group students had been more critical. At the meeting I attended I could discern no difference in the frequency with which different types of members raised objections.Google Scholar

179 Given the larger number of biomedical members on Committee F, one would, of course, expect them as a group to raise a broader range of objections. But while averages are not very meaningful when the “groups” have only two or three members, the biomedical members do seem, member for member, to have raised more kinds of objections than the behavioral science members. The biomedical members averaged six kinds of objections, the behavioral scientists three kinds of objections.Google Scholar

180 While only two members stated that they had raised the first of these issues and only one that he had raised the second, 6 of the 10 biomedical members questioned stated that they had raised questions with reference to the risk/benefit ratio, one listing it as among his most frequent objections.Google Scholar

181 Questions relating to the soundness of the research design were infrequently raised on Committee A since the issue was regarded as the province of the research committee. There were substantial indications, however, that when the issue was raised the committee had tended to rely on members with expertise regarding the substance of the proposal.Google Scholar

182 The chairman, the nurse, and three physician-members all saw a need for the creation of a separate committee to review biomedical proposals in addition to or instead of the human subjects committee.Google Scholar

183 One member of Committee A suggested that the limited range of medical specialties represented on the committee had led it to scrutinize proposals solely from the standpoint of privacy and consent, ignoring the professional aspects. He said that the so-called clinical members either did not attend or were not really clinical members in the sense of being engaged in conducting research on human subjects. Although the member was clearly correct that only a limited range of specialties was represented on the committee, this limitation was counterbalanced by the requirement of review by the research committee and the overlapping memberships of the two committees. Moreover, while the primary focus of this committee, as on most of the committees studied, was indeed on consent, a substantial number of questions and objections went to the extent of the physical risk, the risk/benefit ratio, and the need for additional safeguards. See discussion at 54–55 supra.Google Scholar

184 Although several members of Committee E suggested that there were some differences between clinicians and other biomedical members in the issues they tended to raise, most questioned on the point saw no such division and none perceived any difference in the willingness of the two groups to approve proposals without modification. Most of the medical members, both clinicians and nonclinicians, indicated that they had initiated an objection relating to the extent of the physical risk. But while no nonclinician stated that this was one of his most frequent objections or that he had raised it more often than had other committee members, two clinician members named this as an objection they had raised more often than other members, one stating that it was one of the issues that he had raised most frequently. A third clinician, from pediatrics, stated that he was more likely than other members to raise a question if the proposal involved children or kidney disease.Google Scholar

185 Exclusive of psychiatrists, who while they have a foot in both camps have been classified for present purposes with the biomedical members.Google Scholar

186 None of the behavioral scientists indicated raising a question relating to (1) discomfort or in-convenience to the subjects, (2) the competence of the investigator, (3) the soundness of the scientific design, (4) the appropriateness of the use of disadvantaged or captive subjects, (5) the use of deception, (6) the inducements offered for participation in the research, or (7) whether the benefit was sufficient to justify the risk. One of the behavioral scientists stated that he had raised a question about the extent of the physical risk or the extent of the psychological risk, and he indicated that in each instance he had raised the issue on only one occasion. Only one of the behavioral scientists stated that he had raised a question as to the adequacy of the safeguards for protecting the physical well-being of the subjects (once) or the intelligibility of the language of the consent form.Google Scholar

187 Even this is doubtful. Although five of the eight members to express an opinion (including two of the behavioral science members) maintained that this issue had been raised most often by the behavioral scientists, only one behavioral scientist stated that he had raised the issue and then only once during the past year. It is possible, however, that a former behavioral science member may have been more active than any of the current members.Google Scholar

188 One biomedical member stated that he was not competent to judge and didn't care about surveys. “As far as I am concerned they can ask people whatever they want. If there is a risk in them [surveys], then somebody else will have to look into them.”Google Scholar

189 On the issue of deception, one student member stated that the medical people were not particularly interested in the issue and that the psychologists, while interested in the issue, had differing positions on the appropriateness of deception. The chairman stated that the social scientists, the humanist, and the students had tended to be “more concerned about informed consent or … the subtleties of ethics … than maybe the physicians might be.” Various other members, however, saw no difference in the frequency with which various kinds of members raised these issues.Google Scholar

190 Although research-oriented members appeared to be less critical of research proposals than law and religion members, there were differences among the research scientists in their reaction to particular issues. It is, for instance, clear that the two psychologist members were much more comfortable with the use of deception than was the political scientist. Both psychologists indicated that the departmental committee had approved proposals involving the use of deception. Consent was unnecessary, they felt, where informing the subjects in advance would vitiate the experiment and nothing “really disagreeable or adverse” would be done to the subjects. In contrast, the political scientist declared, “I guess the sort of thing that bothers me most is what some of our colleagues in psychology occasionally get into, … lying to subjects. I find that unacceptable myself whether it's psychologically damaging or not.”Google Scholar

191 A senior clinician agreed that the nonmedical members had played an important role in bringing in a different perspective and in sensitizing other members to issues of which they had previously been unaware—although he felt that over time the difference in the kinds of issues raised diminished. Yet another of the M.D.s concurred that while there was little difference in the kinds of objections raised, the nonmedical members raised particular objections that the medically oriented members had overlooked. He stressed, however, that he saw relatively little difference between the objections raised by the “middle-of-the-road professionals” and the nonmedical members.Google Scholar

192 The chairman and three other members indicated that they looked to (among others) one or both of the ethicists on the committee on issues of privacy and confidentiality.Google Scholar

193 All the members for whom I have data identified the philosophy professor as someone whose views carried particular weight on this issue. The chairman stated, however, that under particular circumstances there were other members whose views would carry equal or greater weight. If the question, for instance, was the reason for using a black population in a piece of biomedical research, the views of someone else might carry greater weight. By inference one senior clinician did not find the philosophy professor persuasive on this issue since he indicated that he regarded the objection as a silly one.Google Scholar

194 The assessment of the political scientist was less clear. He stated that he thought that both he and the sociologist were less likely to raise objections because of their greater familiarity with social research. When asked what objections he had raised, he responded by stating that he had raised very few and went on to list only one that he had raised. Later, however, he said that while an individual's discipline “does influence his perspective heavily … once we start talking about it I don't feel I'm reacting much differently than the lawyer or the person from religion.” He added that he doubted that the committee behaved differently with the law and religion members on it than it would have if they were not on it. Although these statements are difficult to reconcile, one interpretation is that while the law and religion members had a different perspective and raised objections that he would not have raised, all members shared a commitment to ethical research and any differences were resolved in committee discussions. Only the religion member saw no differences. He maintained that “everyone asks value questions, legal questions, confidentiality questions with equal frequency of distribution and with equal sensitivity.” While he felt that an attorney on the committee had shown greater sensitivity than he had, “that's a personal characteristic.” He acknowledged, however, that he brought to the committee academic expertise regarding religion that another member might not have.Google Scholar

195 This comment was echoed by the political scientist, who declared that because he had “engaged in so many survey operations” he had a better sense than could a religion professor of the psychological impact of various kinds of questions and in consequence was “less uptight about some of the kinds of questions than some of the other members of the committee.”Google Scholar

196 The interviews were replete with references to his contributions on a wide variety of issues. The chairman stated that he was as likely as anyone to raise issues of physical risk, psychological risk, undue inconvenience and discomfort, or that the benefits were insufficient to justify the risks. Another member stated that this member had “made good contributions” but “as often as not … he speaks to the same sorts of issues that I would or some other medical types would,” for instance, the adequacy of the consent or safety of a drug.Google Scholar

197 Because most of the work was done by mail, members other than the chairman tended to be unaware of the responses of other members with the exception of those cases that were discussed at committee meetings. The members who did comment (some, but not all, of whom expressly noted that their response was limited to behavior at committee meetings) gave varying responses. Although the chairman and several members felt that in some instances proposals may have received more support from “lay” members than from the medical members, this was uniformly attributed to a lack of knowledge of risks inherent in the proposal. No member recalled an instance in which the nonmedical members were willing to take greater risks than were the medical members because of the importance of the scientific objective.Google Scholar

198 Two of the nonmedical members of Committee C identified this as one of their most frequent bases for objection, while only one of the medical members named it as an issue he had raised, and none indicated that he regarded it as one of his most frequent objections. There was considerable evidence that the former law member in particular had raised this issue more often than had most other committee members. There was no kind of issue that the “lay” members were generally perceived as raising more often than other kinds of members on Committees E and F. When we consider only the issues that had been frequently raised on Committee E, the nonmedical members probably had the greatest impact on the intelligibility of the language of the consent form. In addition, a majority of the members of Committee E questioned on the point felt that the nonmedical members had objected as often as or more often than the medical members on the following issues: (1) whether the procedure involves excessive discomfort or inconvenience to the subjects, (2) whether the psychological risk is excessive, (3) whether the procedure unduly invades the subject's privacy or fails adequately to protect the confidentiality of data, (4) whether the consent form contains all of the necessary kinds of information, (5) the appropriateness of the use of disadvantaged or captive subjects, (6) whether the inducements for participation are excessive, and (7) the legal competence of the subjects to consent. As to no issue, however, did more than one-third of those questioned feel that nonmedical members had objected more frequently. It should be noted that despite a widespread perception of major “lay” input on the issue of psychological risk, only 1 “lay” member, but 9 of the 15 medical members, stated that they had raised this issue. sOn Committee F the most frequent basis for objection from all segments of the committee appears to have been to the adequacy of the consent form and procedures, and there was a broad measure of agreement that questions as to both the intelligibility and completeness of the consent form were equally likely to come from all three groups. The law member asserted that he had raised a broad range of issues and had raised issues of privacy and confidentiality more frequently than most other members. But while one member did feel that the law member was (along with two nonphysician biomedical members) among those who raised the issue of privacy most often, none felt that as a group the law member and the clergyman had tended to raise any kind of objection more frequently than the other kinds of members.Google Scholar

199 On Committee C the former law member was the only member to state that he had objected to the use of disadvantaged or captive subjects. As to most other members, however, I have no clear indication whether they had raised questions regarding the use of captive subjects.Google Scholar

200 On Committee F the attorney stated that other members had turned to him for advice in connection with legal niceties—discovery of documents, liability, and the like. Other members confirmed that the views of the attorney had been sought on the legal aspects of at least one proposal. Four members declared that his views carried special weight with them on the legal competency of subjects to consent and two other members that his views were influential on privacy questions. On Committee E the attorney stated that other members turned to him for advice on legal issues—offering as examples the issues of informed consent and privacy. Five members indicated that they did give particular weight to the views of the attorney on privacy issues, and six stated that they accorded particular weight to his views on the question of the legal competency of the subjects to consent. On Committee C one medical member, who expressed great doubt about the contribution of the nonmedical members to the work of the committee, stated that he could not recall an occasion where the law member had initiated an objection but that usually “we ask” the law member “about this or that thing.” Another member agreed that the law member had been “more than anything else … a resource person,” someone who “tends to answer questions that the rest of us raise about (the) legality of a particular position or how should this consent form be modified to be a better document, this type of thing.” Both the current and former law members of Committee C perceived their roles as law related: making sure that the legal rights of subjects were protected; advising the committee as to the legal implications of its actions.Google Scholar

201 On Committee F on a few occasions members had asked the clergyman for his views on the moral aspects of a proposal before the committee. On Committee E, although neither clergyman appears to have been regarded as an important resource person, on particular issues there were one or two members who stated that they gave weight to the views of one or both clergymen.Google Scholar

202 The “lay” members of Committee E themselves expressed serious doubt regarding the extent of their contribution. On Committee F most of those expressing a judgment felt that apart from their role as resource persons the attorney and the minister had exerted little influence on the behavior of the committee. On Committee C there was general agreement that, with the possible exception of the law members, the nonmedical members had not had much impact on the behavior of the committee.Google Scholar

203 The minister described himself as the kind of person who lies low until he is sure of himself. But while he had not been very outspoken at meetings, the minister had objected with some frequency in writing.Google Scholar

204 This view was expressed by a broad spectrum of committee members. Two of the students confirmed that they saw themselves as providing “the students' viewpoint, which is most often from the subject's side.” In addition, several members stressed that the students were younger and less sophisticated, and one stated that the students brought a perspective “of the radical left, to a greater extent than the average of the committee.” No member suggested that students were easier on student proposals than were other members, and one member flatly denied this.Google Scholar

205 On Committee A the absence of any member with legal training had led the committee to rely for legal advice on an administrator. The administrator stated that while he did not have legal training, other members turned to him on purely legal issues as a result of his association with hospital counsel. There was, as one member noted, no one on the committee who had technical skills in the area of privacy and confidentiality, and, while the chairman and another member indicated that they looked to the administrator on the question of the legal competence of the subjects to consent, no member was widely looked to on that issue-one member explicitly noting the absence of an attorney.Google Scholar

206 National Commission Survey, supra note 11, at 1–29–1–30. *While this article was in final galleys DREW and the FDA issued proposed regulations on the protection of human research subjects. 44 Fed. Reg. 47,688, 47,698 (Aug. 14, 1979). Although there are some differences, in most pertinent respects, the proposed regulations are in conformity with the National Commission's recommendations.Google Scholar

207 Under the proposed regulations to approve a research proposal the committee must find that: (A) The research methods are appropriate to the objectives of the research and the field of study; (B) Selection of subjects is equitable; (C) Risks to subjects are minimized by using the safest procedures consistent with sound research design and, whenever appropriate, by using procedures being performed for diagnostic or treatment purposes; (D) Risks to subjects are reasonable in relation to anticipated benefits to subjects and importance of the knowledge to be gained; (E) Informed consent will be sought under circumstances that provide sufficient opportunity for subjects to consider whether or not to participate and that minimize the possibility of coercion or undue influence; (F) Informed consent will be based upon communicating to subjects, in language they can understand, information that the subjects may reasonably be expected to desire in considering whether or not to participate, generally including: (I) That an Institutional Review Board has approved the solicitation of subjects to participate in the research, that such participation is voluntary, that refusal to participate will involve no penalties or loss of benefits to which subjects are otherwise entitled, that participation can be terminated at any time, and that the conditions of such termination are stated; (II) The aims and specific purposes of the research, whether it includes procedures designed to provide direct benefit to subjects, and available alternative ways to pursue any such benefit; (III) What will happen to subjects in the research, and what they will be expected to do; (IV) Any reasonably foreseeable risks to subjects, and whether treatment or compensation is available if harm occurs; (V) Who is conducting the study, and who is funding it, and who should be contacted if harm occurs or there are complaints; and (VI) Any additional costs to subjects or third parties that may result from participation; (G) Informed consent will be appropriately documented, unless the Board determines that written consent is not necessary or appropriate because (I) the existence of signed consent forms would place subjects at risk, or (II) the research presents no more than minimal risk and involves no procedures for which written consent is normally required; (H) Notwithstanding the requirements of paragraphs (E), (F) and (G) above, informed consent is unnecessary (I) where the subjects' interests are determined to be adequately protected in studies of documents, records or pathological specimens and the importance of the research justifies such invasion of the subjects' privacy, or (II) in studies of public behavior where the research presents no more than minimal risk, is unlikely to cause embarrassment, and has scientific merit; (I) There are adequate provisions to protect the privacy of subjects and to maintain the confidentiality of data; and (J) Applicable regulatory provisions for the protection of fetuses, pregnant women, prisoners, children and those institutionalized as mentally infirm will be fulfilled. Recommendation (4), 43 Fed. Reg. 56,179–80 (1978). For the current standards see note 5 supra.Google Scholar

208 43 Fed. Reg. 56,178 (1978).Google Scholar

209 Id. In some respects the Commission's requirements concerning composition are less detailed than the existing regulations. There is no counterpart in the Commission proposal to the existing requirement that to enable the committees “to ascertain the acceptability of applications and proposals in terms of institutional commitments and regulations, applicable law, standards of professional conduct and practice, and community attitudes,” they contain members whose “concerns” lie in these areas. 45 C.F.R. § 46.106 (b)(1) (1978).Google Scholar

210 43 Fed. Reg. 56,178 (1978).Google Scholar

211 Id.Google Scholar

212 Id. at 56,182.Google Scholar

213 The Commission states: “To require that IRB determinations be made by unanimous vote might result in a serious retardation of the review process, would place excessive power in the hands of single members, and would create an incentive for mitigating the diversity of viewpoints represented on the IRB.”Id.Google Scholar

214 Id.Google Scholar

215 Id. The Commission's recommendations regarding expedited review are permissive and hedged with a number of caveats, the most important of which are set forth in the text. The categories of research that are to be given expedited review are to be defined by the committee on the basis of its experience and must be approved by DREW before the expedited procedure can be used. The Commission lists a number of types of research procedures for which expedited review “may be appropriate”: (A) Collection (in a nondisfiguring manner) of hair, nail clippings and deciduous teeth; (B) Collection for analysis of excreta and external secretions including sweat, saliva, placenta expelled at delivery, umbilical cord blood after the cord is clamped at delivery, and amniotic fluid at the time of artificial rupture of the membranes prior to or during labor; (C) Recording of data from adults through the use of physical sensors that are applied either to the surface of the body or at a distance and do not involve input of matter or significant amount of energy into the subject or an invasion of the subject's privacy. Such procedures include weighing, electrocardiogram, electroencephalogram, thermography, detection of naturally occurring radioactivity, diagnostic echography, and electroretinography; (D) Collection of blood samples by venipuncture, in amounts not exceeding 450 milliliters in a six-week period, from subjects 18 years of age and over who are not anemic, pregnant or in a seriously weakened condition; (E) Collection of both supra- and subgingival plaque, Provided, The procedure is no more invasive than routine prophylactic scaling of the teeth and the process is accomplished in accordance with accepted prophylactic techniques; (F) Voice recordings made for research purposes such as investigations of speech deficits; (G) Moderate exercise by healthy volunteers; (H) The use of survey research instruments (interviews or questionnaires) and psychological tests, interviews and procedures that are part of the standard battery of assessments used by psychologists in diagnostic studies and in the evaluation of judgmental, perceptual, learning and psychomotor processes, Provided, That the subjects are normal volunteers and that the data will be gathered anonymously or that confidentiality will be protected by procedures appropriate to the sensitivity of the data; (I) Program evaluation projects that entail no deviation for subjects from the normal requirements of their involvement in the program being evaluated or benefits related to their participation in such program; and (J) Research using standard protocols or noninvasive procedures generally accepted as presenting no more than minimal risk, even when done by students. Id. at 56,178. The Commission stressed, however, that a study may entail more than minimal risk to subjects even though it involves procedures that ordinarily present no more than minimal risk.Google Scholar

216 Id. The Commission notes specifically that when there is any problem regarding “informed consent, reduction of risk, etc.,” the proposal should be referred to the full committee and that the investigator should always be able to request full committee review.Google Scholar

217 Id. at 56,182–83. The Commission states that it believes “that an IRE should have the final word at its institution regarding the ethical acceptability of proposed research involving human subjects.”Google Scholar

218 Id. at 56,178.Google Scholar

219 Id.Google Scholar

220 Id. at 56,179. The Commission also states that the consent process or the research itself may be observed on a routine or sample basis in the committee's discretion.Google Scholar

221 The Commission explains: “The involvement of a physician or therapist as an investigator may have significant advantages for patients and make available to them new forms of therapy. However, research interests may compromise the therapist's sound judgments regarding therapeutic goals. The involvement of a neutral third party may reduce the possibility of such a conflict of interest occurring.”Id. at 56,179.Google Scholar

222 “DHEW does not regulate substantive decision-making, which is the responsibility of the IRB'S alone.”Id. at 56,177.Google Scholar

223 Id. at 56,178.Google Scholar

224 Id.Google Scholar

225 Id. at 56,179.Google Scholar

226 See text at notes 162–66 supra.Google Scholar

227 Barber et at., supra note 1, at 161–62. The Barber group found that 37 percent of the institutions with a formal appeal procedure but only 22 percent of those without one had a relatively high rate of revision, rejection, and/or withdrawal. As they recognized, however, the presence of an appeals mechanism may have been the consequence of a high rate of negative actions, since the negative actions may have generated pressures for the establishment of an appeals procedure. The establishment of an appeals mechanism at Committee D was apparently the result of unhappiness of an investigator with an adverse committee decision. In the present study the two committees with the highest modification rates had no appeals procedures. When we recall that any committee that required revisions (however slight) in more than 10 percent of proposals was classified by the Barber group as having a “high” rate of modification, what is surprising is that as many as 22 percent of the committees which infrequently required modifications had gone to the trouble to create an appeals mechanism. Professor Robertson has also urged the creation of a mechanism for “rehearing or appeal,” not to improve the effectiveness of review but to assure investigators of fair treatment and a recourse for grievances. John A. Robertson, Ten Ways to Improve nun's, Hastings Center Rep., Feb. 1979, at 29, 30 (emphasis added). See also Jay Katz & Alexander Morgan Capron, Catastrophic Diseases: Who Decides What? 233–34 (New York: Russell Sage Foundation, 1975).Google Scholar

228 43 Fed. Reg. 56,178–79. The Commission rightly states that “observation of the consent process or conduct of research is both a difficult and delicate task.”Id. at 56,179. The same can be said of interviews with subjects. Such steps are likely to be resented by many investigators as an unwarranted intrusion in the conduct of the research and ought not to be undertaken in the absence of some reason to believe that the investigator may not be adequately protecting subjects.Google Scholar

229 See pp. 657–58 supra.Google Scholar

230 In Johnson v. Louisiana, 406 U.S. 356 (1972), and Apodaca v. Oregon, 406 U.S. 404 (1972), five justices expressed the view that in federal cases a unanimous verdict was constitutionally required by the sixth amendment, and at least one circuit has so ruled. See Sincox v. United States, 571 F.2d 876, 878 (5th Cir. 1978). For pre-Apodaca statements that a unanimous verdict was required by the sixth amendment, see, e.g., Andres v. United States, 333 U.S. 740, 748–49 (1948); Patton v. United States, 281 U.S. 276, 288–90 (1930) (dictum). Unanimity at the federal level is mandated by the Federal Rules of Criminal Procedure. F.R. Crim. P. 31(a). See also ABA Standards Relating to the Administration of Criminal Justice, Trial by Jury standard 15–1.1 (2d ed. tentative draft 1978). Although the Court has sustained (by a five-to-four vote) the constitutionality of conviction on a three-fourths vote in state criminal cases (see Apodaca v. Oregon, supra, and Johnson v. Louisiana, supra), the constitutionality of conviction on a simple majority vote is doubtful and the tradition of requiring greater than a simple majority strongly ingrained. Only a few states allow conviction by less than a unanimous verdict, and none permit conviction upon less than a two-thirds vote. See ABA Standards, supra, at 5. Other instances in which more than a simple majority has been required include the requirement of a vote of two-thirds of the senators voting to ratify a treaty and the requirement of a two-thirds vote of each house to propose constitutional amendments.Google Scholar

231 A requirement of a two-thirds vote may provide a partial answer to concerns that have been expressed (see Robert M. Veatch, The National Commission on Elias: An Evolutionary Approach, Hastings Center Rep., Feb. 1979, at 22, 26–27) regarding the dominance of scientists on such committees.Google Scholar

232 43 Fed. Reg. 56,182 (1978).Google Scholar

233 See discussions at pp. 145–46 supra.Google Scholar

234 The retention of the records of expedited review as part of the committee's records is not an adequate substitute, for it is doubtful that members would often go to the trouble to consult them. One member of Committee D, while defending the committee's practice of distributing only limited information on proposals on the ground that the full proposal was available to the members in the committee's files, conceded that he had never looked at the material on file with the chairman.Google Scholar

235 See discussion at 660–62 supra.Google Scholar

236 See text at notes 154–58 supra.Google Scholar

237 Similar issues are raised by the question of the impact of size on the behavior of juries. In a pair of much-criticized decisions, Williams v. Florida, 399 U.S. 78, 100 (1970), and Colgrove v. Bat-tin, 413 U.S. 149 (1973), the Supreme Court upheld the constitutionality of the 6-member jury. More recently a unanimous Court concluded that in the context of a criminal trial a jury of 5 did not pass constitutional muster. Ballew v. Georgia, 435 U.S. 223 (1978). There was no opinion for the Court, but Mr. Justice Blackmun, after surveying the commentary generated by the Court's prior decisions, concluded that the “purpose and functioning” of the jury were seriously impaired by a reduction in size to below 6 members since smaller juries (1) were less likely to be representative of the community, (2) were less likely to have members who remember each of the important pieces of evidence, and (3) were more likely to be influenced by the biases of its members. The argument that juries of 12 are more representative than juries of 6 is based on sampling theory. See Lempert, supra note 154, at 668–84. Zeisel, … And Then There Were None, supra note 169, at 710. But while juries are (more or less) chosen by lottery from the pool of available jurors, human subjects committees are not. A committee of 6 chosen by chance is less likely to contain the varied competencies needed for the review of proposals than a committee of 5 chosen in conformity with the Commission's recommendations. When committees are chosen by design rather than randomly selected, whether a larger committee is more diverse than a smaller committee depends upon the behavior of the appointing bodies. Thus, to the extent the argument that larger juries are more representative rests on sampling theory, its applicability to the performance of human subjects committees is doubtful. All that can be said is that increasing the size of the committee permits the inclusion of a broader range of members. The conclusion that jury size influences jury behavior was also based, however, on empirical research concerning the behavior of small groups and the effect of an increased membership on the decision-making process. See note 154 supra. A recent study of mock juries concluded: “Larger juries … produce longer deliberations, more communication, far better community representation, and, possibly, greater verdict reliability (consistency).” Saks, supra note 169, at 107.Google Scholar

238 One of the nonmedical members of Committee C stated that she had been selected because she was a neighbor of the chairman. Other noninstitutional members of both this and other committees indicated that prior to their membership on the committee they had known personally the chairman, a committee member, or the appointing officer.Google Scholar

239 Whether as a group noninstitutional members had been less effective than other members is difficult to say, as they were few and varied enormously in the extent of their contributions. See, however, the finding of the National Commission survey that “other” members were less likely than biomedical or behavioral members to initiate modifications or requests for information and somewhat more likely to describe themselves as being less influential than the average member. National Commission Survey, supra note 11, at 1–178–1–184. Although presumably most of the noninstitutional members fell within the “other” category, it should be noted that the “other” category also includes law professors, philosophy professors, and other academics from the humanities who were affiliated with the institution.Google Scholar

240 In a preliminary draft Campbell and Cecil proposed: “Where organizations exist appropriate to representing the interests of the participating groups, these should be represented on the … [committee] or at its hearings. Such groups would include the Welfare Rights League, unions of social workers and teachers, the Parent-Teacher Association, and community protest groups.” Donald T. Campbell & Joe Shelby Cecil, Protection of the Rights and Interests of Human Subjects in the Areas of Program Evaluation, Social Experimentation, Social Indicators, Survey Research, Secondary Analysis of Research Data, and Statistical Analysis of Data from Administrative Records, Recommendation 3 (Jan. 3, 1976, draft as quoted in 3d draft, Apr. 1977, at 3A). This recommendation was dropped from the third draft because “[w]hile we concur in the desirability of having such persons on Boards, along with substantial proportions of nonresearchers, we have been unable to develop a recommendation that would insure such representation and still be feasible.” Donald T. Campbell & Joe Shelby Cecil, Protection of the Rights and Interests of Human Subjects in the Areas of Program Evaluation, Social Experimentation, Social Indicators, Survey Research, Secondary Analysis of Research Data, and Statistical Analysis of Data from Administrative Records 7 (3d draft, Apr. 1977). See also Robertson's suggestion that in order to reduce researcher dominance “[i]nstitutional control over appointments … could be limited.” Robertson, supra note 227, at 30.Google Scholar

241 43 Fed. Reg. 56,178 (Nov. 30, 1978). The Commission recommends that committees include at least one person not otherwise affiliated with the institution, that they include at least one person familiar with the ethical issues in research involving human subjects, and that at least one-third but not more than two-thirds of the members be scientists.Google Scholar

242 National Commission Survey, supra note 11, at 1–61.Google Scholar

243 See Gray, supra note 20, at 41.Google Scholar

244 The National Commission survey found that a majority of the forms it reviewed required at least some college education to read. National Commission Survey, supra note 11, at 1–26.Google Scholar

245 Id. at 1–61.Google Scholar

246 The requirement that juries represent a cross section of the community is independent of whether the exclusion of a particular group can be expected to have an influence upon jury behavior. A jury selection system that systematically excludes women or blacks is unconstitutional without any showing that the exclusion is likely to have any impact. See Taylor v. Louisiana, 419 U.S. 522, 527 (1975). In Taylor the Court declared: ‘[i]t is part of the established tradition in the use of juries as instruments of public justice that the jury be a body truly representative of the community.’ To exclude racial groups from jury service was said to be ‘at war with our basic concepts of a democratic society and a representative government.’” 419 U.S. at 527 (quoting from Smith v. Texas, 311 U.S. 128, 130 (1940)).Google Scholar

247 National Commission Survey, supra note 11, at 1–61.Google Scholar

248 That the inclusion of women does not affect performance should not be surprising. There is, a priori, no reason to suppose that women scientists would view research proposals differently from their male counterparts. One might, however, have expected minority group members to be more sensitive to some kinds of risks—most particularly the undue use of disadvantaged subjects—than other committee members. Most of the minority group members in the National Commission survey were employed by the institution and were involved in the biomedical area. Only a small minority reported their discipline or field of training as law or religion. Id. at 1–171–1–172. This is of some importance since black scientists may be much like white scientists and yet black attorneys quite different from white law academics in the objections they raise.Google Scholar

249 Committee A had reviewed approximately the same number of proposals as did the average committee in the Commission's survey. See National Commission Survey, supra note 11, at 1–51.Google Scholar

250 Id. at 1–56. Of the committees included in the present study only Committee B did not assign proposals for detailed review to either one or more committee members or to a subcommittee.Google Scholar

251 Various members stressed the “tremendous workload” imposed particularly on members of the subcommittee on new applications. One of the clergymen stated “the workload of the subcommittee has become almost impossible.” One member suggested that the volume of proposals resulted from the requirement that the committee review low-risk proposals. Although the subcommittee was freed of the need to review proposals up for recertification, it considered each new application individually at a meeting of the subcommittee. But while the volume of proposals clearly was perceived to be a serious problem, more than half the members interviewed indicated that their committee work took less than 4 hours a week. Apart from the executive secretary, only one member claimed to spend as much as 7 hours a week on committee business. Compare the finding of the National Commission survey that the average committee member devoted 55 hours a year (or just over an hour a week) to the work of the committee. National Commission Survey, supra note 11, at 1–214.Google Scholar