Using a two-stage data envelopment analysis to estimate the efficiency of county hospitals in China: a panel data study

Background Chinese county hospitals have substantially enlarged their scale and service volume since the new health-care reform in 2009. In the context of scale expansion, continuously efficient operations are important considerations. In this study, we aimed to estimate the efficiency of Chinese county hospitals after such expansion and to establish whether and how efficiency is affected by various factors. Methods We studied three kinds of efficiency: pure technical efficiency, scale efficiency, and technical efficiency (=pure technical efficiency × scale efficiency). We included 140 county hospitals from Henan province and 100 county hospitals from Jiangsu province during 2010–12. A two-stage data envelopment analysis was used. In the first stage, the efficiencies of county hospitals were estimated using data envelopment analysis. In the second stage, overall technical inefficiencies were regressed against two environmental factors and five institutional factors using Tobit regression. The Health and Family Planning Commission of Jiangsu and Henan province, China, provided us the data for hospital efficiency estimation. No patient information was involved in the study. Findings 20 (2·8%) of 720 sample hospital units were technically efficient overall, with a mean efficiency of 0·573 (SD 0·188). 39 (5·4%) were only technically efficient, and 94 (13·1%) were efficient in scale, with mean efficiencies of 0·615 (0·184) and 0·932 (0·132), respectively. Scatter plot of the relation between scale efficiency and hospital beds showed that Chinese county hospitals were likely to be efficient at roughly 200–600 beds. Tobit regression analysis showed that hospital size and average length of stay had negative associations with technical efficiency, whereas occupancy rate, bed-to-nurse ratio, and nurse-to-physician ratio are positively associated with technical efficiency. Moreover, hospitals in Jiangsu province are more efficient than those in Henan province. Interpretation The Chinese county hospitals sampled were technical inefficient, and pure technical inefficiency was the driving force. Useful ways to improve efficiency in these hospitals could include taking measures to improve pure technical efficiency, controlling hospital bed scale to balance efficiency with quality, coordinating the ratio of physicians, nurses, and beds, and improving the hospital management ability to reduce average length of stay and enhance bed usage. Funding This work was supported by the National Natural Science Foundation of China grant number 71173081. The funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript.