Sunday, April 3, 2005
Hall of Mirrors (Hilton Cincinnati Netherland Plaza)
Session: 1223, Graduate Student Poster Session, 3:00 PM

Structure, Personnel, Activities and Fees of Domestic Institutional Review Boards

Nancy Novotny, Clinical Research Coordinator, Nursing Department, University of Illinois at Chicago, 10711 Sleepy Hollow Road, Peoria, IL 61615 and Nina Collins, Professor, Family and Consumer Sciences, Bradley University, Bradley, Peoria, IL 61604.

Purpose: Responsibilities of Institutional Review Boards (IRBs) have dramatically increased. Lacking recent research on the structure or workload of IRBs, informed recommendations for the future of IRBs and optimal researcher interaction with IRBs are impossible. The purpose of this study was to describe current structure, personnel, activities, and fees of domestic IRBs. Subjects and Methods: A cross-sectional survey of domestic IRBs was conducted March-July 2004. Postal mailing to a random selection of registered IRBs included stamped return envelopes. Reminders were emailed at 1-month. As this method yielded an unacceptably low response rate of 48/276 (17%), 1163 questionnaires were emailed to additional IRBs with identifiable addresses. Results: The 136 respondent IRBs consisted of 12 ± 7 regular members, and 1.7 ± 2.8 full time equivalents (FTEs). Meetings occurred 12 ± 9 times annually and lasted 107 ± 54 minutes. IRBs reviewed 0-1148 new studies (median 37) and 0-11,107 total actions (median 158). IRBs provided education for board members (78%), personnel (65%), and investigators/community (52%). Of 113 IRBs that responded to fee-related items, 28% charged $570 ± $110 for full review of new studies. Total actions were dichotomized at the median. Compared to low volume IRBs, higher volume IRBs: (a) provided more education (Chi Sq 17.67, df=2, p < .001), (b) spent 24 fewer minutes/action (95% C.I. –34 to -13; t=-4.43, df=101, p < .001), and (c) charged higher fees (t=5.81, df=111, p < .001). A positive correlation between FTEs and total actions existed for high volume IRBs (r=.527, p < 0.001), but not low (r=-.153, p=0.235). Conclusions. When you have seen 1 IRB, you have seen 1 IRB. Wide variation in the IRB structure and workload is partly explained by annual workload. Among lower volume IRBs there was no indication that staffing was based on workload; a threshold effect may exist. Researchers may adjust their IRB interactions upon consideration of an IRB's unique operations.

Session #1223 - Graduate Student Poster Session

The 29th Annual MNRS Research Conference (April 1-4, 2005)