A Hybrid Approach for Forecasting Patient Visits in Emergency Department

An accurate forecast of patient visits in emergency departments (EDs) is one of the key challenges for health care policy makers to better allocate medical resources and service providers. In this paper, a hybrid autoregressive integrated moving average–linear regression (ARIMA–LR) approach, which combines ARIMA and LR in a sequential manner, is developed because of its ability to capture seasonal trend and effects of predictors. The forecasting performance of the hybrid approach is compared with several widely used models, generalized linear model (GLM), ARIMA, ARIMA with explanatory variables (ARIMAX), and ARIMA–artificial neural network (ANN) hybrid model, using two real‐world data sets collected from hospitals in DaLian, LiaoNing Province, China. The hybrid ARIMA–LR model is shown to outperform existing models in terms of forecasting accuracy. Moreover, involving a smoothing process is found helpful in reducing the interference by holiday outliers. The proposed approach can be a competitive alternative to forecast short‐term daily ED volume.