1252. What Do Residents Miss on Night-Float? A Comprehensive Retrospective Analysis of Preliminary Report Discrepancies
Authors * Denotes Presenting Author
  1. Mindy Wang *; University of Texas at Houston
  2. Nathan Doyle; University of Texas at Houston
  3. Jennifer Hanak; University of Texas at Houston
  4. Kevin Sweet; University of Texas at Houston
  5. Eduardo Matta; University of Texas at Houston
  6. Susanna Spence; University of Texas at Houston
Multiple studies have explored the rate of discrepancies among preliminary radiology resident interpretations and attending final interpretations across multiple environments including the overnight emergency setting. However, few studies have reported the specific findings missed or overcalled. The purpose of this retrospective analysis is to report and categorize the most common discrepancies found in preliminary radiology resident reports.

Materials and Methods:
There were 41,289 preliminary resident reports generated in the 2019-2020 academic year (July 2019-June 2020) at our institution, one of the busiest Level 1 trauma centers in the United States. Attending radiologists provide feedback after overreading resident preliminary reports by categorizing them into “Agree”, “Minor Discrepancy”, and “Major Discrepancy” groups. A major discrepancy is defined as a time-critical change to the preliminary report that impacts immediate patient management. A minor discrepancy is defined as a change that has/may have an impact on management, but is not immediately time-critical. Our discrepancy review committee consisting of residents and attending radiologists continually monitor discrepancy data collected through our Picture Archiving and Communication System (PACS)-integrated feedback software. For this retrospective analysis, all discrepant preliminary reports overread by radiology attendings for the 2019-2020 academic year were evaluated. The discrepant findings were identified and categorized by topics. Exclusion criteria included studies without a recorded discrepant finding and duplicate reports.

854 resident preliminary reports were categorized as discrepant over the review period, with 185 major (24.2%) and 578 minor (75.8%) discrepancies. The overall discrepancy rate of all studies was 2.07%, and an overall major discrepancy rate of 0.45%. 950 total individual discrepancy topics were evaluated from these discrepant studies, with 232 major (24.4%) and 718 minor (75.6%) discrepant topics. The most common traumatic discrepancies were fractures/ligamentous injuries, namely compression fractures and rib fractures. Among extremity fractures, the most common discrepancies were femur fractures and shoulder joint injuries. Pulmonary opacities were the most common discrepancy in pediatric imaging. In abdominal imaging, the most common discrepancy was appendicitis. In chest imaging, the most common discrepancies were subsegmental pulmonary emboli and pulmonary nodules. In neuroimaging, the most common discrepancy was extra-axial intracranial hemorrhage.

Identifying the most common types of findings that residents misread is highly informative when working to improve resident report accuracy. By separating minor versus major discrepancies, we can focus our greatest educational efforts on common discrepancies that impact time-critical patient management, while subcategorization of the most common minor discrepancy types allows for broader educational targets across all subspecialty areas.