2

Abstract

Introduction: Many potential barriers exist in the Institutional Review Board (IRB) approval process. The Critical Illness Outcomes Study (CIOS) is a multicenter observational study of critical care patients at 37 U.S. sites. In order to avoid selection bias, participating sites applied for a waiver of informed consent through the IRB. IRB concern regarding waiver of consent was identified as a potential barrier to IRB approval. The purpose of this study was to identify both site-specific as well as IRB barriers to approval of an observational study. Hypothesis: We hypothesize that site-specific characteristics and IRB interpretation of the criteria for waiver of informed consent are barriers to conducting large observational trials. Methods: CIOS investigators who applied for IRB approval were surveyed regarding site-specific characteristics and IRB responses. Survey data were obtained via an electronic survey. Results: 35 out of 37 (95%) participating CIOS sites responded to the survey, in addition to two sites that did not receive approval. 75% of IRB proposals were submitted by an experiened investigator who had previously submitted 5 or more proposals, while 25% were submitted by an investigator who had previously submitted less than 5 proposals. The median time to approval of the IRB protocol was 26 days at sites where the submitting investigator had submitted 5 or more proposals, while the median time to approval was 97 days at sites where the investigator had submitted less than 5 proposals (p=0.049). Four sites independently cited IRB concern regarding waiver of consent as a major barrier to approval and were required to perform revisions or meet with the Board regarding this concern. Conclusions: Our study suggests that the experience of the submitting investigator is inversely related to the time to approval of an IRB protocol. In addition, IRB concern regarding waiver of informed consent was identified as a major barrier to approval. These barriers should be addressed in an effort to expedite approval of minimal risk, multicenter, observational studies. Introducing a centralized IRB should be considered as an option to minimize these barriers and avoid duplication of effort.