Effects of German DRG 2004 on grouping of stroke care

Background: The introduction of the new funding-system based on DRG is in progress. The calculation of the cost weights for 2004 based on hospital data of the year 2002 is completed. Extensive changes of ICD-10-GM 2004, OPS-301 2004 and definitions of G-DRG were made. Compared to the G-DRG-System Version 1.0 the grouping of the data with the new G-DRG-System 2004 shows different results. The effects of these results may not be estimated exactly. They will be inspected with this study. Method: The sample includes 100 inpatient cases, who were treated in the stroke unit (admission and discharge) in the year 2003. The clinical data was grouped with the certificated grouper version 2003, the interim-grouper 2003/2004, the certified grouper version 2004 and, after recoding in accordance to the new classification-systems ICD-10-GM 2004 and OPS-301 2004, the data were again grouped with the grouper version 2004. Results: For the adjacent DRG B69 (TIA) the sample showed the same PCCL. Within the adjacent DRG B70 (cerebral apoplexy) we had considerable movement (revaluation and devaluation in cost weights). The DRG B70A with the highest resource consumption was less frequently grouped than the other DRG of this adjacent DRG with lower resource consumption. This effect is due to changes in the definitions manual. The table for additional diagnoses, referring complicating diagnoses, is cancelled in the G-DRG-system 2004. The PCCL-assignment there is now based on CC-values of all additional diagnoses. For DRG B42A (rehabilitation for diseases and disorders of the nervous system with catastrophic CC and complex diagnosis) cerebral ischemia is not included in the table "complex diagnosis". Patients with this diagnosis are grouped in the DRG B42B or B42C. The procedure-code for rehabilitation short-term treatment (8 - 551.0) does not trigger DRG B42 (rehabilitation for diseases and disorders of the nervous system). Conclusion: On the basis of this sample, suggestions on development of the G-DRG-System may be given. Estimation of the budget for the year 2004 with the grouper versions 2004 based on clinical data 2003 is not recommended without recoding of clinical data. Also the interim-grouper 2003/2004 brings different results. This study shows trends and problems with the G-DRG-System 2004. Further analyses is urgently recommended.