810 - Provider-in-Triage Prediction of Hospital Admission After Brief Patient Interaction
Friday, May 15, 2020
Location: Majestic Ballroom: Majestic
Background and Objectives: We sought to determine if emergency physician providers working in the triage area (PIT) of the ED could accurately predict the likelihood of admission for patients at the time of triage. Such predictions, if accurate, could decrease the time spent in the ED for patients who ultimately are admitted to the hospital by hastening downstream workflow.
Methods: This is a prospective cohort study of PIT providers at a large urban hospital. Physicians were asked to predict the likelihood of admission and confidence of prediction for patients after evaluating them in triage. This included at the discretion of the provider: possible chart review, brief interview and brief physical exam. Physicians also predicted the service to which the patients would be admitted. Measures of predictive accuracy were calculated, including sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV).
Results: 33 physicians (16 attendings, 17 residents) evaluated 300 patients and made predictions. The average patient age was 48 (range 18-94) and 55% were female. 70 (23.33%) patients were admitted (4.48% observation, 85.07% general care/telemetry, 7.46% progressive care, 2.99% ICU). Of these, 82.09% were admitted to general medicine and 14.93% were admitted to general surgery. Overall, physicians predicted 91 patients would be admitted of which 52 actually were. The sensitivity of determining admission for the entire cohort was 74.29%. The specificity was 83.11%. The PPV was 57.78% and the NPV was 92.57%. When correctly predicting admission, physician accuracy of predicting the admitting service was 86.54% (Kappa=0.56; 95%CI 0.29-0.83). When physicians were at least 80% confident in their predictions, the predictions improved: Sensitivity 96.2%, Specificity 93.6%, PPV 75.8%, NPV 99.2%.
Conclusion: The accuracy of physician providers-in-triage of predicting discharge was better than the accuracy of predicting admission. Those predictions made with confidence >80% were dramatically more accurate. Physicians predicted with reasonable accuracy the service to which patients were ultimately admitted. These results indicate that while general predictions of admission are likely inadequate to guide downstream workflow, predictions in which the physician is confident could provide utility.