Challenges in Predicting Discharge Disposition for Trauma and Emergency General Surgery Patients

Changes in discharge disposition and delays in discharge negatively impact the patient and hospital system. Our objectives were¹ to determine the accuracy with which trauma and emergency general surgery (TEGS) providers could predict the discharge disposition for patients and² determine the factors associated with incorrect predictions.

Discharge dispositions and barriers to discharge for 200 TEGS patients were predicted individually by members of the multidisciplinary TEGS team within 24 h of patient admission. Univariate analyses and multivariable logistic least absolute shrinkage and selection operator regressions determined the associations between patient characteristics and correct predictions.

A total of 1,498 predictions of discharge disposition were made by the multidisciplinary TEGS team for 200 TEGS patients. Providers correctly predicted 74% of discharge dispositions. Prediction accuracy was not associated with clinical experience or job title. Incorrect predictions were independently associated with older age (OR 0.98; P < 0.001), trauma admission as compared to emergency general surgery (OR 0.33; P < 0.001), higher Injury Severity Scores (OR 0.96; P < 0.001), longer lengths of stay (OR 0.90; P < 0.001), frailty (OR 0.43; P = 0.001), ICU admission (OR 0.54; P < 0.001), and higher Acute Physiology and Chronic Health Evaluation II scores (OR 0.94; P = 0.006).

The TEGS team can accurately predict the majority of discharge dispositions. Patients with risk factors for unpredictable dispositions should be flagged to better allocate appropriate resources and more intensively plan their discharges.