Background: Multicenter studies are becoming more common, and variability in local institutional review board (IRB) assessments can be problematic.
Objective: To investigate the variability of IRB responses to a multicenter observational study of children presenting to emergency departments.
Methods: The authors collected the original IRB applications, subsequent correspondence, and a survey assessing submission timing and response and the nature of IRB queries. The study was conducted as part of the Emergency Medicine Network http://www.emnet-usa.org
Results: Of 37 sites initiating the IRB process, 34 (92%) participated in this IRB-approved study. Institutional review boards returned initial applications in a median of 19 days (IQR, 11–34 d), and 91% considered the protocol to be minimal risk. Of 34 submissions, 13 required no changes, 18 received conditional approvals, and 3 were deferred. The median time from initial submission to final approval was 42 days (IQR, 27–61 d). Seven sites did not participate in patient recruitment: two had institutional issues, one obtained IRB approval too late for participation, and four sites (12%) reported that IRB hurdles contributed to their lack of participation. Nonetheless, 68% of sites that recruited patients reported that the overall experience made them more likely to participate in future multicenter research.
Conclusions: There was substantial variation in IRB assessment of a standard protocol in this study. The burden of the application process contributed to some investigators not participating, but the majority of investigators remain enthusiastic about multicenter research. A national IRB may streamline the review process and facilitate multicenter clinical research.