PURPOSE: To compare how different institutional review boards (IRBs) process and evaluate the same multiinstitutional educational research proposal of medical students' quality of life. METHOD: Prospective collection in 2005 of key variables regarding the IRB submission and review process of the same educational research proposal involving medical students, which was submitted to six IRBs, each associated with a different medical school. RESULTS: Four IRBs determined the protocol was appropriate for expedited review, and the remaining two required full review. Substantial variation existed in the time to review the protocol by an IRB administrator/IRB member (range 1-101 days) and by the IRB committee (range 6-115 days). One IRB committee approved the study as written. The remaining five IRB committees had a median of 13 requests for additional information/changes to the protocol. Sixty-eight percent of requests (36 of 53) pertained to the informed consent letter; one third (12 of 36) of these requests were unique modifications requested by one IRB but not the others. Although five IRB committees approved the survey after a median of 47 days (range 6-73), one committee had not responded six months after submission (164 days), preventing that school from participating. CONCLUSIONS: The findings suggest variability in the timeliness and consistency of IRB review of medical education research across institutions that may hinder multi-institutional research and slow evidence-based medical education reform. The findings demonstrate the difficulties of having medical education research reviewed by IRBs, which are typically designed to review clinical trials, and suggest that the review process for medical education research needs reform. © 2007 Association of American Medical Colleges.