Click for next page ( 226


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



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

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

OCR for page 225
D Selected Studies of IRB Operations: Summary Descriptions Studies of institutional review boards (IRBs) are listed in chronologi- cal order; see References for full citations. Information is not available with which to evaluate the quality of individual studies. Barber, B., J.J. Lally, J. Makarushka, D. Sullivan, 1973, "Research on Human Subjects: Problems of Social Control in Medical Experimen- tation" This is the first known survey of IRBs, conducted in 1969; it in- cluded 300 biomedical IRBs, of which 70 percent had existed prior to the 1966 U.S. Public Health Service policy requirements. A single individual was interviewed at each institution. The survey found that relatively few IRBs required modifications of protocols: 34 percent had never modified or rejected a project. Gray, Bradford H., Robert A. Cooke, Arnold S. Tannenbaum, 1978, "Research Involving Human Subjects" This article in Science is a condensed version of the report of a study that the authors conducted for the National Commission for the Protec- tion of Human Subjects of Biomedical and Behavioral Research under a contract to the University of Michigan (see Cooke, Gray, and Tannen- baum, 19781. The study covered a wide variety of issues and topics; highlights are excerpted below. A probability sample of 61 institutions, stratified by type of institu- tion and weighted by research volume, was drawn from a list of more than 420 institutions with IRBs approved by the U.S. Department of Health and Human Services; the survey covered research reviewed by IRBs at these institutions between July 1, 1974, and June 30, 1975. In- terviews were conducted between December 1975 and July 1976 with 225

OCR for page 225
226 PROTECTING PARTICIPANTS AND FACILITATING SOCIAL AND BEHAVIORAL SCIENCES RESEARCH more than 2,000 research investigators whose proposals had been re- viewed, over 800 IRB members, and almost 1,000 participants or third parties (e.g., parents) who consented on participants' behalf. The sample comprised IRBs at medical schools (including those that share an IRB with their university; 59 percent); separate univer- sity IRBs (18 percent); hospital IRBs (15 percent); and other. About 60 percent of studies reviewed were biomedical; about one-third were behavioral research; and the rest involved secondary analysis of data or tissue samples. IRBs varied widely in size, number of proposals re- viewed per year, how often convened, and member-hours of IRB work per year. Fifty percent of IRB members were biomedical scientists; 21 percent were behavioral scientists. Great diversity was evident in IRB procedures. For example, half had provisions for investigators to appeal IRB decisions, half did not; half assigned proposals to individual members for intensive review; half took formal votes; two-thirds accepted majority approval, one- fourth required unanimity; one-fourth said that investigators always attended meetings at which their projects were reviewed. IRBs were active in modifying protocols: 50 percent of projects were modified or the IRB required more information; 44 percent were approved as is (6 percent had no data). These figures indicate greater activity by IRBs in modifying research than in the Barber et al. study (19731. In the Gray et al. study, 14 percent of the IRBs modified ev- ery study; 22 percent modified no more than one-third. The largest percentage of modifications involved consent forms and procedures. One-fourth of investigators judged their projects to be without risk; one-fourth judged their projects to have no more than a "very low" probability of "minor" medical or psychological complications; the re- mainder judged their projects to involve a "low" probability of minor complications or a "very low" probability of "serious" complications. IRBs often modified consent forms, but there were no significant differences in the average readability or completeness scores between the original and modified consent forms. Those agreeing that "the review procedure has improved the quality of scientific research done at this institutionat least to some extent" were 78 percent of IRB biomedical members, 69 percent of biomedical researchers, 62 percent of IRB behavioral and social science members, and 55 percent of behavioral and social science researchers. Those agreeing that "the review committee makes judgments that it is not qualified to makeat least to some extent" were 21 percent of IRB behavioral and social science members, 28 percent of IRB biomedical members, 43 percent of biomedical researchers, and 49 percent of behavioral and social science researchers.

OCR for page 225
SELECTED STUDIES OF IRB OPERATIONS: SUMMARY DESCRIPTIONS 227 High percentages (96-99 percent) of all four groups agreed that "the human subjects review procedure has protected the rights and welfare of human subjects at least to some extent." IRBs that made more requests for information or frequent modifi- cations of protocols were likely to be viewed less favorably than less active IRBs. Goldman, Jerry, and Martin D. Katz, 1982, "Inconsistency and Insti- tutional Review Boards" The investigators submitted three identical research protocols to 32 IRBs at major universities with medical colleges. Twenty-two IRBs participated by reviewing the three protocols and documenting their judgments. Each protocol contained serious ethical issues, had scien- tific design flaws, and provided an incomplete consent form. The par- ticipating IRBs varied in their judgments, and a substantial number approved the flawed designs. Although there was considerable consis- tency in objecting to the consent forms, IRBs did not identify specific deficiencies consistently. Subsequently, the chair of one of the participating IRBs challenged Goldman and Katz, who responded in kind. This chair objected to the conclusion that IRBs needed more regulation so as to be more consis- tent. This chair also disagreed with the position of Goldman and Katz that it is part of the responsibility of IRBs to consider scientific design. Chlebowski, Rowan T., 1984, "How Many Protocols Are Deferred? One IRB's Experience" This study examined one IRB with jurisdiction over the clinical re- search program of a university-affiliated, major teaching hospital. The study reviewed the actions of this IRB on new protocols (n = 168) and continuing reviews (n = 138) over a 12-month period. Of new protocols, 27.9 percent were approved without comment, 64.4 percent were ap- proved with comment, and 7.7 percent were deferred (none were dis- approved). Reasons for deferral included: (1) study design was judged not likely to answer the research question, which made it difficult to determine the risk/benefit ratio; (2) concerns of lay members of IRB regarding adequacy of protection for participants; and (3) major prob- lems with the consent form. Of 13 deferred proposals, 9 were resub- mitted and approved, 4 were dropped. The IRB reached a unanimous decision in 97 percent of cases.

OCR for page 225
228 PROTECTING PARTICIPANTS AND FACILITATING SOCIAL AND BEHAVIORAL SCIENCES RESEARCH Cleary, Robert E., 1987, "The Impact of IRBs on Political Science Re- search" This study surveyed 1 15 chairs of political science (PS) departments that offered Ph.D. degree programs and chairs of IRBs at these insti- tutions. The response rate for political science chairs was 47 percent (541; the response rate for IRB chairs was 68 percent (781. For 30 institutions, responses were received from both PS and IRB chairs. PS chairs reporting an IRB at their institution: 51 of 53. PS chairs reporting experience with an IRB: 27 of 53. Characterization of experience: 16 positive, 4 negative. Social scientists on IRB: 25 yes, 4 no; PS members on IRB: 10 yes, 19 no. IRB covered federally funded student research: 29 yes, O no; un- funded student research: 26 yes, 4 no. IRB requires written advance informed consent: 26 yes, 4 no. PS protocols cleared without change: 159; cleared after changes: 22; denied: O. Departments reporting significant problems with IRB: 3 yes, 26 no. Chairs reporting IRB increased protection for human subjects: 19 yes, 6 no. IRB chair responses largely agreed with department chair responses; O said that PS had problems with IRB, 54 said IRB had increased pro- tection, 6 said it had not. Problems identified included uncertainty and lack of information regarding informed consent and confidentiality; also, confusion at some institutions as to what research is covered, particularly unfunded stu- dent research. Rules also differed: one IRLB required advance written consent for surveys; another specified verbal consent. Some institu- tions deleted the exemption for research with public officials or spec- ified that such research had to be limited to official responsibilities of public officials. Advisory Committee on Human Radiation Experiments, 1996, "Final Report" Research Proposal Review Project: This project sampled 125 research protocols involving human par- ticipants (84 involving ionizing radiation, and 41 not involving radia- tion) that were approved and funded by the Departments of Health and EIuman Services, Defense, Energy, and Veterans Affairs or by the Na- tional Aeronautics and Space Administration between fiscal years 1990 and 1993. Of the 84 radiation protocols, 31 were extramural (primar- ily from universities) and 53 were intramural. A committee member

OCR for page 225
SELECTED STUDIES OF IRB OPERATIONS: SUMMARY DESCRIPTIONS 229 reviewed an additional 93 protocols with regard to informed consent issues. The documents obtained about each of the 125 protocols were re- viewed by two individuals, including at least one Advisory Committee member. The reviewers rated each protocol on ethical concerns, from 1 (no concerns) to 5 (serious concerns), and on level of risk (minimal, more than minimal). The reviewers also identified factors that resulted in poor ratings. Overall ratings were distributed as follows: 1 34 percent; 234 percent; 318 percent, 4 or 5 14 percent. All studies receiving a 4 or 5 were also considered greater than minimal risk. Factors contributing to poor ratings (3, 4, or 5) fell into three cat- egories: (1) factors likely to affect how well potential participants un- derstood the research and how they could benefit or be harmed (e.g., consent forms suggesting that participants might benefit from being treated by experimental drugs when such an outcome was highly un- likely); (2) factors likely to affect the voluntary nature of decisions about participation; and (3) approaches to the inclusion of people with limited or questionable decision-making capacity. Subject Interview Study (SIS): The SIS sample included almost 1,900 patients in oncology or car- diology clinics at medical institutions in Eve areas of the country; all sampled patients received a brief interview, and 103 of them who were research participants received a longer interview. The brief interview covered general attitudes toward medical re- search; understanding of such terms as clinical trial and medical exper- iment; beliefs about research participation; reasons for participating or not participating in research (when applicable); and demographic and other background information. The overall response rate was 95 percent. Nearly 40 percent of patients had been research participants or in- vited to be participants. The attitudes of these patients were generally favorable to research; most felt free to decline or to leave the project. Bell, James, John Whiton, and Sharon Connelly, 1998, "Evaluation of NIH Implementation of Section 491 of the Public Health Service Act, Mandating a Program of Protection for Research Subjects" This is the most recent major study of IRBs. The study universe was defined as 491 IRBs that in 1995 operated with multiple project assurances under 45 CFR 46 and that had conducted more than 10 initial reviews of human participant research protocols in the previous

OCR for page 225
230 PROTECTING PARTICIPANTS AND FACILITATING SOCIAL AND BEHAVIORAL SCIENCES RESEARCH year. Five groups received questionnaires: IRB chairs and institution officials at all 491 institutions; IRB administrators at 300 institutions; 4 investigators at each of the 300 institutions (1,200 investigators); 4 IRB members at each of 160 IRBs (640 members). Response rates were 80 percent or higher for IRB chairs (394), administrators (245), and institutional officials (4001; rates were 68 percent for IRB members (435) and 53 percent for investigators (632~. Topics covered included: . Person-time effort (total person-time of all IRB personnel, chair effort, member effort, administrator effort, institutional official effort, investigator effort on initial review); Effort per review (per initial review, per continuing review); Other information on effort (meeting time per review, duration of initial review, unimplemented protocols, multiple IRB reviews); Opinions about burden (overall efficiency, getting into inappro- priate areas); General opinions and ratings relative to adequacy (rating of over- all adequacy, effect of initial review on protocols, effect of IRB ac- tion versus other influences at the institution, relative effect of dif- ferent IRB activities, effect on scientific quality, influence of work- load on protection adequacy, bias/lack of expertise, investigators' ability to participate in review decisions, relative impact/burden of federal requirements); . Concerns, modifications, and other review outcomes (approved as submitted, concerns raised in initial review, protocol modifica- tion, conditions on approval, actions on multicenter protocols); Other IRB actions (suspension or termination of approved re- search, overruling exemption determinations); Reports of potential problems (serious investigator noncompli- ance, withinjurisdiction harms, legal actions by subjects, sub- jects' complaints, informed consent process, problems with in- vestigators); Suggested changes at the local level (enhancing IRB procedures and structure, education and training, additional resources); Suggested changes at the federal level (enhancing regulations and practices).

OCR for page 225
SELECTED STUDIES OF IRB OPERATIONS: SUMMARY DESCRIPTIONS 231 A few reported measures are distinguished by type of research (bio- medical, SBES). A problem in evaluating the findings is that the spon- soring agency, the National Institutes of Health Office for Protection from Research Risks, never funded completion of the second volume of the study, which includes the questionnaires and other technical infor- mation. (Questionnaires were kindly provided to panel staff by James Bell Associates.) Panel on Institutional Review Boards, Surveys, and Social Science Research, January 2003, Staff Review of University IRB Websites (un- published) Panel staff drew a sample of 48 of the 151 universities classified as Doctoral/Research Universities-Extensive" in 2000 by the Carnegie Foundation for the Advancement of Teaching. These institutions typi- cally offer a wide range of baccalaureate programs, and they are com- mitted to graduate education through the doctorate. During 2000, they awarded 50 or more doctoral degrees per year across at least 15 disci- plines. The final sample size was 47 institutions because one institution provided no information about IRBs on its website. The staff examined IRB websites for the 47 institutions in December 2002-January 2003. These sites varied in the amount of information provided; also, some sites may have been more up to date than others. Based on the website materials only, the staff determined answers to the following questions: 1. How many IRBs does the university have? 1 IRB for the university 2 IRBs 3 or more (range is from 3 to 7) 64%* 13 23 * One university also has a College of Education Human Subjects Committee that can review unfunded projects involving no more than minimal risk. 2. Does the website list names of IRB members? Yes 45% No 55

OCR for page 225
232 PROTECTING PARTICIPANTS AND FACILITATING SOCIAL AND BEHAVIORAL SCIENCES RESEARCH 3. Does the IRB require that undergraduate research be reviewed? Yes Only if will be made public or is a thesis 6 Entire class, not individual projects 6 No 9 Could not determine 75% 4 4. Does the IRB allow research to be exempted? Yes 91%* No 9 * At one university, department chairs make exemption decisions. 5. Is there guidance (beyond the Common Rule) for requesting exemption? Yes No Not applicable (does not exempt any research) 6. Does the IRB allow research to be expedited? Yes 87% No 13 13% 79 9 7. Is there guidance (beyond the Common Rule) for requesting expedited review? Yes No Not applicable (does not expedite any research) 4% 83 13 8. Is there guidance on informed consent, such as a template, that goes beyond the Common Rule? Yes, guidance is appropriate for SEES research 64% Yes, but guidance specifies inappropriate elements 9 Guidance simply repeats Common Rule No guidance is provided 11 17

OCR for page 225
SELECTED STUDIES OF IRB OPERATIONS: SUMMARY DESCRIPTIONS 233 9. Is human participants protection training required for researchers? Yes No No, but there is a link to NIH on-line training No, but there is a link to university on-line training 10. Are there on-line training modules and/or guide books? 57% 23 11 9 Yes (developed by university), no SEES module 40% Yes (developed by university), SEES module 4 Reference to NIH on-line training only No 23 32 11. Is there guidance on confidentiality protection (beyond just stating a requirement)? Yes 11% No 89 12. What information is provided on how long the IRB review process will take? Notice of frequency of IRB meetings Estimate of time to allow for review No guidance provided Minimum time for review for those providing time estimates (n = 20) At least 1 week At least 2 weeks At least 3 weeks At least 4 weeks At least 6 weeks Longer than 6 weeks 30 43 26 100 95 85 75 30 20 13. Is there guidance on minimal-risk research beyond the Common Rule definition? Yes 0% No 100

OCR for page 225
234 PROTECTING PARTICIPANTS AND FACILITATING SOCIAL AND BEHAVIORAL SCIENCES RESEARCH 14. Is there any provision for investigators to meet with IRB face-to-face? Yes, may attend meeting May wait outside meeting to answer questions Meeting time and place listed, but no invitation 9 No 72 15%