Hospital specialisation within a DRG-framework : the Austrian case

Evaluation of the true relationship between costs and specialisation in hospitals is hindered by the lack of a standard measure. Specialised hospitals might produce at lower costs because their staff builds expertise and care is better organised. On the other hand specialised hospitals might be more costly because they systematically attract sicker patients within each diagnosis-related group (DRG) or have special equipment available. We compare three common measures of specialisation and introduce an alternative, which builds on the widely used Gini coefficient, and investigate the influence of the Austrian provincial health-policy making on specialisation. Although the four measures differ in definition, they show high concordance and prove to assess hospital specialisation in a robust way. With the exception of university hospitals, measured specialisation complies with the different hospital types as defined by legislation in Austria. We find no significant time trend towards more specialisation and legislation on provincial level seems to have a small impact on hospital specialisation. However, caution should be paid to skewness, so that outliers do not inappropriately influence the results when evaluating the true relationship between costs and the specialisation of hospitals. Overall, the Austrian DRG framework introduced in 1997 and regional regulation by the Provinces have not led to more specialised hospitals. This finding challenges the expected impact of activity based funding on specialisation, but it may reflect the lack of incentives set by the Austrian DRG framework and the Provinces. -- hospital specialisation ; hospital financing ; Herfindahl-Hirschman index ; information theory index ; Gini coefficient ; decomposition of inequality ; Austria