Abstract
Benchmarking is increasingly used in performance monitoring of the health care sector, for self-assessment as well as external accountability - in order to promote efficiency. However, a prerequisite for achieving this objective is that the benchmarking process is based on scientifically sound methods. Using DRG-data from 19 Swedish hospitals in 1994-1995, we apply data envelopment analysis (DEA) to compare technical efficiency (productivity) of inpatient obstetric care. The results indicate an average potential for enhanced productivity of at least six-seven percent each year. Most hospitals show decreasing productivity between 1994 and 1995, partly due to their inability to adjust production capacity and costs to decreasing demand, From a policy perspective, DEA could be very useful in performance monitoring. An important advantage ofthe DEA method is that it helps identify the strengths and weaknesses of the various units (hospital departments) studied. We conclude that there are good reasons for adopting the DEA method for benchmarking in other types of hospital departments as well, and preferably on a larger scale, Le., with more observations, and also for including health outcome and quality indicators among the output variables.