A Model for Allocating Budgets in a Closed System Which Simultaneously Computes DRG Allocation Weights

This paper presents a quadratic programming model for allocating a global budget to individual hospitals. In the model, allocation weights are estimated for Diagnosis Related Groups (DRGs) which reflect the average historical costs of treating an additional case in each of nine DRG systems. The model is illustrated using data from the Department of Defense (DoD). Allocated budgets and DoD specific DRG marginal cost weights are estimated using information on three factors; the facility designated functions in the military health care system; hospital output; and sets of price structures used in other systems. The model's weights are used to allocate the DoD's global budget to its treatment facilities under constraints. The model minimizes the difference (in a least squares sense) between the facilities' current budgets and their computed budgets, while constraints are designed to produce budgets representing differences in the type and volume of patients and to preserve certain specialized institutional capabilities.