A method for assessing emergency department performance using patient outcomes

Objective: To determine the rates of correct patient disposition after an ED evaluation. Methods: In a university pediatric hospital, a 25% random sample of ED patients for 4 consecutive months was reviewed, after exclusion of minor injuries and patients triaged to the nonurgent clinic. Patients were categorized into one of 4 outcomes on the basis of inpatient resource use: appropriate admission, inappropriate admission, appropriate release, or inappropriate release. A 10% random sample of released patients was contacted by telephone to detect patients who sought care elsewhere after ED release. Results: 642 of 2,682 ED patients (23.9%) were admitted; 159 (24.7%) were inappropriately admitted, and 26 (1.3%) were inappropriately released. The correct identification of the need for hospitalization (sensitivity) was 94.9%, and for release (specificity) 92.7%. Overall, the correct classification rate was 93.1%. Inappropriate admissions were associated with diagnoses of trauma, seizures, and burns. Conclusion: Inappropriate admissions occur at a substantial rate and occur more commonly than inappropriate releases. The correct disposition of patients is a practical and meaningful outcome-based measure of the quality of ED care. This methodology is suitable for use in other EDs.