Abstract No. 248 - IR peer, a newly developed peer review scoring system designed for interventional radiology practice

Abstract

Purpose Peer review systems have been used for years in medical, paramedical and diagnostic radiology fields to promote patient safety and healthcare outcomes, but no dedicated peer review system has been developed specifically for Interventional Radiology (IR) practice. We introduce here ‘IR Peer’, a novel peer review system designed to address the specificities of IR activity, usable as a primary or complementary quality assurance tool. Materials and Methods IRB-waived retrospective review of a prospectively encoded pilot database aimed at demonstrating the feasibility of IR Peer in a multi-physicians practice. This scoring system uses peer review during morning rounds of selected IR cases from prior days, a 5-items questionnaire and an ordinal answer scale that grades reviewers’ agreement with imaging findings, procedural/technical management, early outcomes and follow-up plan. Patient lists from IR Peer and from MandM Conferences were compared to evaluate the amount of overlap and capability of IR Peer to help detecting adverse events. Results 417 consecutive reviews of IR Attendings’ cases by their peers were performed in 163 consecutive patients over 18 months. 94% of cases were reviewed by 2 or 3 IR Attendings. Each question was answered 99-100% of the time. Participation rate improved significantly over time, currently 3.2% over the last semester. Answers showed disagreement in 10% of cases (2% by a single reviewer, 8% by several reviewers). Most disagreements related to procedural technique (24% disagreements vs. 3-5% in other four questions). Although IR Peer alone is an imperfect tool to detect complications (it failed to find 89% of cases discussed at MandM), it could still capture many more (11%) adverse events than its alternative (i.e., not doing IR Peer: 1.7% adverse events found; p Conclusion IR Peer is feasible, relevant, and easy to implement in a multi-physician IR practice. When used along with other QA processes, it might help detecting adverse events for MandM Conferences; the latter will require further confirmatory research.