Auditing Report Order in Emergency Teleradiology Using Routine Event Logs

This study aims to describe recorded report-order inversions in emergency teleradiology and assess their interpretation when assignment and active interpretation times are unavailable. We reconstructed computed tomography (CT) transmissions from July 2025 through June 2026, requiring a unique exact linkage and stable priorities. An inversion occurred when a later examination received its first recorded report before an earlier examination still unresolved at the same provider. Eligible head-abdomen pairs were analyzed using dependence-preserving intervals, an empirical conditional reference and secondary models incorporating retrieved imaging workload proxies. Among 8170 examinations, 6,051 had the standard ED urgent category. Later head CT inverted 1060/1761 earlier-abdomen pairs (60.2%; 95% confidence interval, 57.5-62.7%); later abdominal CT inverted 154/1038 earlier-head pairs (14.8%; 12.3-17.5%). Observed-minus-reference differences were - 4.5 percentage points (- 7.3 to - 1.6) and - 11.4 (- 14.2 to - 8.8). Counts were usable for 3474/3565 target examinations and 2665 pairs. Report-order asymmetry did not demonstrate preferential dispatch. Observed inversion proportions were below the empirical conditional reference probabilities. This does not establish protective scheduling. Routine exports support a descriptive audit; additional workflow events are necessary to identify mechanisms or compare dispatch policies.