Diagnostic Discrepancies Between Emergency Department Assessment and Internal Medicine Discharge Diagnosis: A...

Background/Objective: Diagnostic safety is a major challenge in emergency departments (ED), where clinicians frequently make decisions under time pressure and with incomplete information. Prospective data on diagnostic discrepancies in Swiss EDs remain limited. To determine the frequency of diagnostic discrepancies between the initial ED diagnosis and the final Internal Medicine (IM) discharge diagnosis, identify associated patient-, physician-, and context-related factors, and evaluate clinical outcomes associated with diagnostic discrepancy. Methods: We conducted a prospective observational study including 515 patients admitted from the ED to an IM ward and managed by 44 physicians at a Swiss tertiary non-university hospital. The initial ED diagnosis was compared with the IM discharge diagnosis (or diagnosis on day 28 if the patient remained hospitalized). Diagnostic discrepancies were classified using a predefined algorithm with independent expert opinion where required. Generalized linear mixed-effects models were used to assess associations between diagnostic discrepancy and mortality, in-hospital transfers and length of stay; a multivariable logistic regression model was used to identify factors associated with diagnostic discrepancy. Results: Diagnostic discrepancies were identified in 10.1% of patients (n = 52). These patients had longer hospital stays (9.3 ± 9.2 vs. 6.9 ± 5.8 days; p = 0.069) and were more frequently transferred within the hospital (17% vs. 5.8%; p = 0.006). After adjustment, diagnostic discrepancies were associated with higher odds of in-hospital transfer (OR 3.35; 95% CI 1.47-7.66; p = 0.004) and a 20% longer stay (exp β = 1.20; 95% CI 1.00-1.45; p = 0.049), although the latter was attenuated after adjustment for comorbidity. Specialist involvement in the ED was independently associated with lower odds of diagnostic discrepancy (OR 0.30; 95% CI 0.10-0.72; p = 0.015), whereas each one-point increase in physician-perceived diagnostic difficulty was associated with higher odds (OR 1.42; 95% CI 1.07-1.90; p = 0.016). Conclusions: Diagnostic discrepancies occurred in approximately one in ten patients admitted from the ED to IM and were associated with increased in-hospital transfer and, less robustly, with prolonged hospitalization. Prospective multicenter studies should evaluate strategies to reduce diagnostic discrepancies and improve diagnostic safety.