Two-stage DRG Grouping of Cerebral Infarction based on Comorbidity and Complications Reclassification

Background： Cerebral infarction（CI）is one of the leading cause of death in China since 2017, and the growing health care cost of CI is a huge challenge for the government. Chinese government is hopeful that the Diagnosis-related groups DRG system will reverse the trend of rising healthcare costs. However, the use of DRG grouping is challenging due to China’s large geographical area, unbalanced economic levels, and differences in disease spectrum. Aim： In this study, we propose a grouping optimization strategy adapted to different regions based on a two stage model. The first stage analyzed the comorbidity and complications (CCs) of CI and constructed a CCs scheme. The second stage optimize the DRG grouping based on the results of stage 1. Methods: We used data from a Healthcare Security Administration of a city in western China. We analyzed healthcare burden trends for 297,757 CI patients from 2011 to 2018, and used the data after DRG implementation to begin building a two-stage optimization model. We used regression analysis and decision trees to establish rules for the DRG classification model for different regions. Results: The average hospitalization cost was 1,565 US dollars, and the total cost in City C was 1.71 million US dollars. The classification of CCs in different regions is the key to optimizing DRG grouping. The two-stage grouping model can adapt to the needs of different regions. According to our research, only one city C, 8.15 million dollars would be saved if we used the CHS-DRG model and 8.71 million dollars would be saved if we used the two-stage model if we take full control to reduce the medical cost. according to our research.