Application of the RADPEER™ Scoring Language to Interpretation Discrepancies Between Diagnostic Radiology Residents and Faculty Radiologists

Abstract

The objectives of this study were to assess the rate of discrepancies between radiology residents and faculty radiologists at an academic hospital using the RADPEER(™) scoring language and to determine if a correlation existed between assigned RADPEER score (2, 3, or 4) and the clinical significance of the error. Over 19 months, preliminary resident interpretations were graded using the RADPEER scoring system. A retrospective review of discrepant cases was performed to assess the percentage of discrepancy, change in clinical management, and identification of error patterns. Of 2,255 preliminary interpretations, 29 discrepancies (1.29%) were judged to be potentially clinically significant. Of these, 14 (0.62%) resulted in immediate changes in clinical management. Discrepancies assigned RADPEER scores of 3 or 4 were significantly more likely to be judged clinically significant than those assigned scores of 2 (54.5% of 33 studies graded 3 or 4 and 7.7% of 142 studies graded 2, P < .0001). CT imaging generated a higher percentage of discrepancies that were predicted to be clinically significant than plain-film radiography, as well as a higher percentage of discrepancies that resulted in immediate changes in management, but the incidence of each remained low overall (≤2.1%). Discrepancy rates in this study are comparable with previously reported data for discrepancies between attending radiologists and those between attending radiologists and residents data. A significant correlation was observed between increasing RADPEER scores and the clinical significance of discrepancies. This study supports the use of the RADPEER scoring language as both a resident quality assurance measure and an educational tool for quality improvement.