Performance of US teaching hospitals: a panel analysis of cost inefficiency.

This research summarizes an analysis of the impact of environment pressures on hospital inefficiency during the period 1990-1999. The panel design included 616 hospitals. Of these, 211 were academic medical centers and 415 were hospitals with smaller teaching programs. The primary sources of data were the American Hospital Association's Annual Survey of Hospitals and Medicare Cost Reports. Hospital inefficiency was estimated by a regression technique called stochastic frontier analysis. This technique estimates a "best practice cost frontier" for each hospital that is based on the hospital's outputs and input prices. The cost efficiency of each hospital was defined as the ratio of the stochastic frontier total costs to observed total costs. Average inefficiency declined from 14.35% in 1990 to 11.42% in 1998. It increased to 11.78% in 1999. Decreases in inefficiency were associated with the HMO penetration rate and time. Increases in inefficiency were associated with for-profit ownership status and Medicare share of admissions. The implementation of the provisions of the Balanced Budget Act of 1997 was followed by a small decrease in average hospital inefficiency. Analysis found that the SFA results were moderately sensitive to the specification of the teaching output variable. Thus, although the SFA technique can be useful for detecting differences in inefficiency between groups of hospitals (i.e., those with high versus those with low Medicare shares or for-profit versus not-for-profit hospitals), its relatively low precision indicates it should not be used for exact estimates of the magnitude of differences associated with inefficiency-effects variables.