Technical and scale efficiencies of province and district level hospitals in Mongolia

Abstract

To provide the empirical evidence of inefficiencies to health policy by measuring efficiencies of province and district level hospitals in Mongolia. To reach the goal first we determined the technical and scale efficiencies of the hospitals and then determine the factors of inefficiencies. There are 21 provincial and 10 district hospitals included in this study with available data for 2003 to 2005. Data envelopment analysis (DEA) technique was used to assess technical and scale efficiencies. The DEA model was used three inputs and two outputs. Following three physical inputs were used which are number of medical doctor, number of nurse and number of hospital bed. Two outputs we consider were number of inpatient days and number of outpatient visits. To identify and evaluate factors on technical efficiency, the efficiency score for each hospital, calculated using DEA, was used as the dependent variable in regression model, we include four independent variables representing the factors impact on efficiency performance of the hospitals which are average length of stay, Per capita Health budget, Number of elderly, Urban/Rural. To identify the factors of scale efficiency, the efficiency score was used as the dependent variable in regression model; independent variables are number of medical doctor, ratio of doctor, nurse and number of hospitals bed which are can represent the size of the hospitals. The DEA models estimated for the period from 2003 to 2005 indicate average technical and scale efficiency scores were ranking from 0.559 to 1 at the provincial hospitals and 0.768 to 1 at the district hospitals. The coefficient of number of elderly has negative and significant, confirming the expectation that a high number of elderly who are living that area is negatively associated with efficiency. The coefficient of urban also has positive and significant impact on the efficiency, expecting that the efficiency performance has high in urban areas. The number of medical doctor has significant and positive association with scale efficiency other two variables which are ratio of medical doctor/nurse and hospital bed was not significant on hospital scale efficiency.