Measuring hospital performance using multi-product stochastic cost frontier analysis in Florida hospitals.

摘要

After a wave of hospital consolidation activities occurred in the mid-1990s, there has been an increasing concern about whether hospitals use resources efficiently. In this study I used multi-product stochastic cost frontier functions comprising three outputs and six factor prices to estimate the total costs, static efficiency (inefficiency score), and dynamic efficiency (technical change and total factor productivity growth rate) of Florida hospitals during the period 1996--2001. Assuming that cost efficiency is affected by hospital characteristic variables, the models include ownership status, the size of beds, competition, payment methods, location, and market demand conditions. Following Coelli's technique, inefficiency scores were estimated. The results show a slight improvement in inefficiency score compared with the previous studies. It should be noted that MEDICARE and MEDICAID help increase inefficiency, but competition measured by the HHI (Herfindahl-Hirschman Index) fails to improve efficiency. As expected, for-profit (i.e., investor-owned) hospitals are more efficient than not-for-profit hospitals. Hospitals with larger numbers of beds are more efficient than smaller hospitals, whose results may support mergers and acquisitions of hospitals occurred in the mid-1990s. Also, the results show positive technical change and total factor productivity growth rate during the period of this study. My study recommends the conservative use of multi-product cost frontier methods in panel data to devise better government reimbursement policies for Medicare and Medicaid programs.