OBJECTIVES: Hospital efficiency has been the subject of numerous health economics studies, but there is little evidence on how the chosen output and casemix measures affect the efficiency results. The aim of this study is to examine the robustness of efficiency results due to these factors. Comparison is made between activities and episode output measures, and two different output grouping systems (Classic and FullDRG). METHODS: Non-parametric data envelopment analysis is used as an analysis technique. The data consist of all public acute care hospitals in Finland in 2005 (n=40). RESULTS: Efficiency estimates were not found to be highly sensitive to the choice between episode and activity descriptions of output, but more so to the choice of DRG grouping system. Estimates are most sensitive to scale assumptions, with evidence of decreasing returns to scale in larger hospitals. CONCLUSIONS: Episode measures are generally to be preferred to activity measures because these better capture the patient pathway, while FullDRGs are preferred to Classic DRGs particularly because of the better description of outpatient output in the former grouping system. Attention should be paid to reducing the extent of scale inefficiency in Finland.
OBJECTIVES: Hospital efficiency has been the subject of numerous health economics studies, but there is little evidence on how the chosen output and casemix measures affect the efficiency results. The aim of this study is to examine the robustness of efficiency results due to these factors. Comparison is made between activities and episode output measures, and two different output grouping systems (Classic and FullDRG). METHODS: Non-parametric data envelopment analysis is used as an analysis technique. The data consist of all public acute care hospitals in Finland in 2005 (n=40). RESULTS: Efficiency estimates were not found to be highly sensitive to the choice between episode and activity descriptions of output, but more so to the choice of DRG grouping system. Estimates are most sensitive to scale assumptions, with evidence of decreasing returns to scale in larger hospitals. CONCLUSIONS: Episode measures are generally to be preferred to activity measures because these better capture the patient pathway, while FullDRGs are preferred to Classic DRGs particularly because of the better description of outpatient output in the former grouping system. Attention should be paid to reducing the extent of scale inefficiency in Finland.
Authors: Emma Medin; Kjartan S Anthun; Unto Häkkinen; Sverre A C Kittelsen; Miika Linna; Jon Magnussen; Kim Olsen; Clas Rehnberg Journal: Eur J Health Econ Date: 2010-07-29
Authors: José Luis Franco Miguel; Carmen Fullana Belda; José Manuel Cordero Ferrera; Cristina Polo; Roberto Nuño-Solinís Journal: BMC Health Serv Res Date: 2020-11-16 Impact factor: 2.655