On Quantifying and Forecasting Emergency Department Overcrowding at Sunnybrook Hospital Using Statistical Analyses and Artificial Neural Networks

Emergency department (ED) overcrowding is a challenge faced by many hospitals. One approach tomitigate overcrowding is to anticipate high levels of overcrowding. The purpose of this study was toforecast a measure of ED overcrowding four hours in advance to allow clinicians to prepare for highlevels of overcrowding. The chosen measure of ED overcrowding was ED length of stay compliancemeasures set by the Ontario government. A feed-forward artificial neural network (ANN) was designedto perform a time series forecast on the number of patients that were non-compliant. Using the ANNcompared to historical averages, a 70% reduction in the root mean squared error was observed as wellas good discriminatory ability of the ANN model with an area under the receiver operating characteristiccurve of 0.804. Therefore, using ANNs to forecast ED overcrowding gives clinicians an opportunity to beproactive, rather than reactive, in ED overcrowding crises.