Predicting the Unpredictable: A Data-Driven Machine Learning Model for Emergency Department Waiting Room Surge Status.

Objective: To develop a machine learning algorithm to predict emergency department (ED) waiting room surge status (green, yellow, and red categories) 4, 8, and 12 hours in advance. Patients and Methods: We designed a retrospective cohort with prospective validation conducted between July 1, 2023, and April 30, 2025. A total of 154,956 encounters were analyzed; features included ED operational metrics and timestamps with 72-hour lagging data. Deep neural network and gradient boosted decision tree (XGBoost) models were trained to predict 3 predefined surge categories: green (<15 waiting patients; 48.8% of hours), yellow (15-30 waiting patients; 38.8% of the hours), and red (≥31 waiting patients; 12.3% of the hours). Results: The XGBoost model found strong predictive performance across all forecast horizons. Area under the curve (AUC) curves showed excellent discrimination between green and yellow levels, ranging from 0.87 to 0.91 AUC across all time horizons, demonstrating reliable differentiation between normal and moderate surge conditions. The model showed acceptable discrimination for red levels, with AUCs of 0.76 and 0.77, meeting commonly accepted thresholds for clinical forecasting tools, particularly given the difficulty of predicting rare, high-volume situations. The operational accuracy of 68% to 70% showed strong real-world multiclass operational forecasting over prolonged windows up to 12 hours in advance. Conclusion: Using operational metrics with timestamps, XGBoost models can differentiate between different levels of ED surge states with meaningful accuracy. This ability to forecast risk of high volumes provides a window of opportunity to proactively modify operations.