Forecasting the demand for inpatient services for specific chronic conditions.

While the proposed forecasting methodology has a well-established record in evaluating economic time-series, there is minimal, if any, use of this technique in projecting hospitalizations for specific chronic conditions. Using an established taxonomy of disease codes for alcoholism and alcohol abuse in a national inpatient database, a monthly time-series of hospitalizations was modeled. The model derived in both statistically adequate and accurate in forecasting future monthly demand for inpatient hospitalizations. This type of model specifications could be used by hospital planners and policy makers in evaluating monthly resources for specific chronic conditions.