Efficiency Determination of Hospitals of Shiraz University of Medical Sciences using Simple and Super Efficiency DEA Models

Background and objectives: Managers will be able to allocate resources optimally using performance evaluation data, specifically, efficiency evaluation of organizations. An efficiency evaluation of hospitals can be provided by determining the optimal scales for each hospital, to which optimal allocation of resources will be possible. Therefore, the present study aimed to evaluate the efficiency of hospitals using super efficiency model in comparison with the simple DEA model. Methods: This is a descriptive-analytic and a cross-sectional study performed from 21 March 2015 to 20 January 2016. The study population consisted of 11 hospitals affiliated to Shiraz University of Medical Sciences. The data were collected into 3 sections: input indicators, output indicators, and environmental-social indicators. In this study, super efficiency and simple DEA models were used for evaluating the efficiency of the hospitals Results: Based on the super efficiency model, hospitals 9, 5, and 4 had the highest efficiency. According to the input-oriented CCR model of DEA, among 11 studied hospitals, 4(36%) were efficient and 7(64%) were inefficient. Hospitals 3, 5, 7, and 9 were considered to be the most efficient ones with an efficiency score of 1. The mean efficiency score of the hospitals was 0.845 and their maximum and minimum efficiency scores were 1 and 0.462, respectively. In terms of surplus production factors, the input of specialists had the highest surplus value. Conclusion: The study indicated that the super efficiency model led more accurate efficiency scores and provided a complete ranking of hospitals compared to the CCR model. Furthermore, managers should reconsider the number of hospitals and their allocation, improve their efficiency, and reduce repetitions by reducing their size or the scale of those kinds of hospitals which their efficiency was below the optimal line. Large hospitals should be divided into small-scaled and small-sized ones which could be managed more easily.