Developing case-mix standards with the Diagnosis-related Groups for payment reforms and hospital management in China: A case study in Xiamen city

Abstract

Objective: To inform health insurance payment reforms and hospital cost management by investigating the case-mix of viral hepatitis, a common disease, in a Chinese city. Methods: All inpatient medical records of viral hepatitis of selected hospitals in Xiamen from 2008 to 2013 were collected. Both univariate and multivariate analyses were performed to test the association between hospitalization costs and other key variables. The CHAID model was employed to generate the Diagnosis-related Groups (DRGS). Evaluation indexes were adopted to evaluate the grouped results. Standard expenditures were estimated for each group. Results: Gender, age, marital status, mode of payment, admission condition, discharged condition and length of stay were found significantly associated with inpatient expenditures, with p values smaller than .001. Eleven diagnosis-related groups were obtained with the CHAID methods and three classification nodes were obtained from the CHAID decision tree, including age, admission status and mode of payment. The analysis revealed a considerable discrepancy in expenditures across groups with similar characteristics. This variance suggests the existence of inflated inpatient costs as well as inefficient use of medicalre sources. Conclusions: The analysis provided more accurate estimation on case-mix costs for a major disease in China. The standard expenditures are of reference values for the social insurance administration in calibrating reimbursement rates in its move to case-mix payment system and helpful for hospital cost management.