4142.0: Tuesday, November 14, 2000 - 12:50 PM

Abstract #7680

Why does hospital inefficiency vary?

Michael D. Rosko, PhD, Health Administration Department, Widener University, One University Place, Chester, PA 19013, 610-499-4322, michael.d.rosko@widener.edu

The research summarized in this paper examines the determinants of cost inefficiency in urban, short-term, general hospitals (n=1,966) in fiscal year 1997. Cost inefficiency is defined as the percentage difference between actual costs and the costs predicted to occur at the best practice frontier for a given mix of outputs and input prices. Cost inefficiency was estimated by stochastic frontier regression. Primary sources of data are Medicare Hospital Cost Reports, AHA Annual Survey, and Area Resource File.

The stochastic cost frontier was estimated with a generalized translog cost function. The cost function used inpatient discharges, other patient days (e.g., those in hospital-based SNFs or long-term care units) and outpatient visits as outputs, adjusted by measures of inpatient and outpatient case-mix. Input prices for capital and labor also were included in the model.

Preliminary analysis suggests most of the observed variation from the best practice frontier may be attributed to inefficiency rather than to random error. Further, mean cost inefficiency was about 18% in 1997 and it was positively associated with not-for-profit ownership and location in highly competitive areas. Cost inefficiency was inversely related with HMO market penetration, hospital dependence on Medicare and Medicaid revenue, and volume of uncompensated care.

Since the application of stochastic frontier analysis to the health care industry is in its infancy, the paper concludes with an extensive discussion of the potential problems with this technique and suggestions for future research.

Learning Objectives: 1. Discuss detrminants of hospital inefficiency. 2. Explain the concept of stochastic frontier analysis. 3. Discuss implications for payment policy. 4. Discuss implications for further reearch

Keywords: Cost Issues, Hospitals

Presenting author's disclosure statement:
Organization/institution whose products or services will be discussed: None
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.

The 128th Annual Meeting of APHA