Long-term forecasting of regional demand for hospital services

Abstract

Many western countries undergo substantial demographic changes at present. This is particularly challenging for the health care industry since resources have to be set up and arranged well in advance to be able to cover future patient demand. The objective of this article is to present a method for forecasting regional demand for hospital services. The problem of forecasting regional patient volumes is based on three components. First, population forecasts provided by local authorities serve as a basis for the projections. Second, future per-capita demand is forecasted to account for sociological and medical trends. Forecasting methods in this step include autoregressive integrated moving average models, exponential smoothing models, neural nets, and regression models. Third, patient volumes are anticipated merging the projections of the population and per-capita demand for the respective age and sex groups. The proposed method is applied to publicly available data concerning discharges from German hospitals over 18 years. Results indicate that considering the age structure of the population in the catchment area of the hospital and taking into account trends of significantly changing per-capita demand are crucial for accurate forecasts.