Using RFID Data to Improve the Identification of Abandonment Behavior in an Emergency Department: Clinical Policy Implications

Abstract

Identifying which Emergency Department (ED) patients are likely to leave without being seen (LWBS) could enable interventions that reduce LWBS rates. Machine Learning (ML) models that updated these predictions as patients wait were developed and validated, to correctly identify more patients who LWBS. Using a dataset of 150,959 patient visits to the ED of an academic medical campus, two types of classification models were developed: (1) a static model that uses patient and ED census information at the time of arrival to predict the risk to LWBS; and (2) a time-dependent model that updates the predictions based on new information after 30 minutes for patients who are still waiting in the ED. Preliminary results show that the time-dependent model reduces the number of missed LWBS cases by approximately 50% as compared to the static model, without incurring any additional false-positives.