Annals of Management Science


The task of measuring and benchmarking hospital performance, focusing mainly on measures of productivity and efficiency, is one of attempting to collect and homogenize data that allow researchers applying models that avoid comparisons in a basis of ‘apples-to-oranges”. Based upon a refinement of data envelopment analysis (DEA) – the cross-efficiency method (X-DEA), the current paper ranks 756 hospitals in Spain. A detailed analysis is done according to the purpose of hospital, the Autonomous Region where hospitals are located, the type of property, the specialty and the type of agreement with the National Health System. A method is also proposed to determine the relative weight of each dimension for all the hospitals and for each category based on the specialty which can be used as a tool by hospital managers in order to establish a program of improving the ranking position of the hospital identifying the critical performance factors. Multivariate analyses of the X-DEA score are performed using recursive partitioning referred to as classification and regression tree (CART) analysis. We conclude big hospitals (>1022 beds) underperform small hospitals (beds). The performance of intermediate size hospitals depend more on features like specialty and type of property. This issue is quite relevant these days in Spain as economic situation is obliging policy makers to cut costs in most of the activities of the public sector, being health system one of the main controversial sectors under scrutiny.