Application of the stochastic frontier regression estimates to the assurance region DEA model: A case in a Catholic health care system

In this study, Stochastic Frontier Regression will be the main investigative tool used to analyze the data. The study includes a comparison of Data Envelopment Analysis with Stochastic Frontier Regression in the analysis of technical efficiency. The study will also use the regression estimators from the Stochastic Frontier Regression model as weights in the Assurance Region Data Envelopment Analysis model. A Midwestern Catholic Healthcare System with 22 hospitals has agreed to provide data for this research.; The output variables in the order of importance to hospital efficiency are; outpatient visits, inpatient days, and emergency room visits. The input variables in the order of importance to hospital efficiency are; payroll expenses, total hospital full-time equivalent, other expenses, and hospital beds. The average cost inefficiency of the hospitals in the system was 9% indicating that the system as a whole is operating 9% above the best-cost frontier.; The weight constrained Data Envelopment Analysis indicated a different set of results. The number of efficient hospitals evaluated using the different scaling methodologies decreases significantly.; In comparing the variables that affect efficiency of hospitals the following conclusions are relevant. Majority of the hospitals that are efficient are located in urban areas. The quantity of output produced by efficient hospitals compared to proportional use of inputs is better for efficient hospitals in contrast to the inefficient hospitals. The average inpatient days and average other admissions are higher in the case of efficient hospitals compared to inefficient hospitals. The average numbers of hospital beds are very close for both types of hospitals. Efficient and inefficient hospitals on an average spend the same amounts of inputs in terms of labor expenses, other expenses, and total full-time equivalent to produce outputs. The average labor intensity is the same for both types of hospitals.