Random Versus Nonrandom Peer Review: A Case for More Meaningful Peer Review.

OBJECTIVE: Random peer review programs are not optimized to discover cases with diagnostic error and thus have inherent limitations with respect to educational and quality improvement value. Nonrandom peer review offers an alternative approach in which diagnostic error cases are targeted for collection during routine clinical practice. The objective of this study was to compare error cases identified through random and nonrandom peer review approaches at an academic center.

METHODS: During the 1-year study period, the number of discrepancy cases and score of discrepancy were determined from each approach.

RESULTS: The nonrandom peer review process collected 190 cases, of which 60 were scored as 2 (minor discrepancy), 94 as 3 (significant discrepancy), and 36 as 4 (major discrepancy). In the random peer review process, 1,690 cases were reviewed, of which 1,646 were scored as 1 (no discrepancy), 44 were scored as 2 (minor discrepancy), and none were scored as 3 or 4. Several teaching lessons and quality improvement measures were developed as a result of analysis of error cases collected through the nonrandom peer review process.

CONCLUSIONS: Our experience supports the implementation of nonrandom peer review as a replacement to random peer review, with nonrandom peer review serving as a more effective method for collecting diagnostic error cases with educational and quality improvement value.