MEASURING HOSPITAL EFFICIENCY WITH FRONTIER COST-FUNCTIONS

This paper uses a stochastic frontier multiproduct cost function to de rive hospital-specific measures of inefficiency. The cost function inc ludes direct measures of illness severity, output quality, and patient outcomes to reduce the likelihood that the inefficiency estimates are capturing unmeasured differences in hospital outputs. Models are esti mated using data from the AHA Annual Survey, Medicare Hospital Cost Re ports, and MEDPAR. We explicitly test the assumption of output endogen eity and reject it in this application. We conclude that inefficiency accounts for 13.6 percent of total hospital costs. This estimate is ro bust with respect to model specification and approaches to pooling dat a across distinct groups of hospitals.