Research Status of Case Reports for Medical School Institutional Review Boards FREE

To the Editor: The role of case reports in generating medical knowledge is debated.¹ The US Government Code of Federal Regulations (CFR) requires all human subjects' research be reviewed by institutional review boards (IRBs). It defines research as “a systematic investigation, including research development, testing and evaluation, designed to develop or contribute to generalizable knowledge” but does not specifically address case reports.² Therefore, each IRB implicitly defines the research status of case reports by whether such studies require its approval. In addition, some IRBs function as privacy boards to meet requirements of the Health Insurance Portability and Accountability Act (HIPAA); they may also review case reports in this capacity.³ We therefore assessed whether medical school IRBs require review of case reports; if so, whether they do so because they consider a single case report as research under the federal definition or to meet privacy requirements.

We surveyed the 124 continental US medical schools (survey available from the author).⁴ Participation was voluntary. A research assistant initially attempted to obtain information from the main campus Web sites; if this was unsuccessful, the survey was administered to IRB personnel by telephone or email. The initial question for each institution was, “Does a single case report require IRB approval prior to presentation or publication?” A response of “No” was considered an indication that a case report was not research under the federal definition. A response of “Yes” was followed up by questions asking whether the institution considers a case report research as defined by the CFR, whether the IRB also functions as a privacy board, and whether the review is required because of research status or privacy requirements. The study was approved by the IRB of the University of Tennessee College of Medicine in Chattanooga.

We were able to obtain information from 116 (94%) of the 124 schools (Table 1). Of these, 74 (64%) were public. The most frequent responders were IRB directors (n = 47; 41%). Ninety-one (78%) of the responding schools did not require IRB approval for a case report.

Of the 25 (22%) of the schools requiring IRB approval, 16 (64%) were public; 10 (40%) functioned only as IRBs; while 15 (60%) functioned both as an IRB and privacy board (Table 2). None of the 25 required full IRB board review; 17 (68%) required IRB notification; and 8 (32%) did not have a fixed policy but considered whether the report should have full or expedited review on an individual basis. Six (24%) conducted reviews because they considered case reports to be research, 8 (32%) for privacy requirement, and 11 (44%) for both purposes. Six of 9 (67%) private schools considered a case report research under the CFR definition compared with 12 of 16 (75%) public schools. Of the 15 committees that functioned as both an IRB and privacy board, 7 (46%) required review to meet privacy requirements and 8 (53%) to meet research and privacy requirements.

This study was limited in that it included only main campuses of continental US medical schools. It may not reflect procedures of nonuniversity IRBs that comprise the remainder of the 3504 US IRBs.⁵ Nevertheless, it indicates that, de facto, most medical school IRBs do not consider a single case report to represent research under the federal definition; only 17 (15%) of the responding schools review case reports for this reason. The variability reflects the lack of guidance to IRBs from the CFR, combined with lack of consensus in the research community. Consistency would improve if the federal government or Association of American Medical Colleges would state explicitly whether they consider a case report to be research under the federal definition.

Further research should consider what it is that needs protection in a case report, and who is responsible for overseeing that protection: the authors, the IRB of the authors’ institution, or the publishing journal.