Study of cost by type of diagnosis in Switzerland

The development of cost information by DRG--experience in a Barcelona hospital

The determination of hospital patient costing adjusted by case-mix is one of the possible applications arising from the emergence of patient classification systems, such as Diagnosis Related Groups (DRGs). Most of the experiences up until now have given priority to the determination of the total costs by DRG, making it difficult to establish comparisons between hospitals. In this article, results of an experience in Barcelona with a direct patient costing model are presented, with these later being grouped into DRGs. Descriptive information is given together with statistical analysis of the intra-DRG variability and the relationship between costs and the length of stay. Patient costing systems represent a useful instrument for hospital management as they allow analyzing variation and its reasons at DRG level. It is in fact the primary level of information required in order to develop management control in hospitals.

[Main results of the Swiss study on DRGs (Casemix Study)]

Sponsored by the Health Administrations of nine cantons, this study was conducted by the University Institute of Social and Preventive Medicine in Lausanne in order to assess how DRGs could be used within the Swiss context. A data base mainly provided by the Swiss VESKA statistics was used. The first step provided the transformation of Swiss diagnostic and intervention codes into US codes, allowing direct use of the Yale Grouper for DRG. The second step showed that the overall performance of DRG in terms of variability reduction of the length of stay was similar to the one observed in US; there are, however, problems when the homogeneity of medicotechnical procedures for DRG is considered. The third steps showed how DRG could be used as an account unit in hospital, and how costs per DRG could be estimated. Other examples of applications of DRG were examined, for example comparison of Casemix or length of stay between hospitals.