Diagnosis-related groups study of uterine leiomyoma patients based on E-CHAID

Diagnosis-related groups (DRG) is emerging as an effective tool for measuring the quality and efficiency of medical services. This study aims to explore a DRG combination scheme for control the hospitalization costs in patients with uterine leiomyoma (UL). This study enrolled a total of 3087 UL patients from a tertiary hospital in Chongqing between year 2017 and 2021. The data of front sheet of the medical record of each UL patient were retrospectively extracted. Mann-Whitney U test and Kruskal-Wallis H test, and subsequent multivariate logistic regression analysis were used to identify the significant factor influencing the hospitalization cost. Then, a decision tree model was performed to establish the DRG case mix model, and to calculate the payment standards and cost caps for each DRG group. The average age of all patients was 43.03 years old, the average length of stay (LOS) was 6.66 days, and the average total hospitalization cost was 22,436.11 yuan. The highest proportion of the total hospitalization cost was treatment cost (29.46%), followed by diagnostic cost (28.98%), consumable cost (26.01%), etc. Factors including surgery type, LOS, and complications or comorbidities had significant impact on hospitalization costs, among which surgery type were the main influencing factor. Using the three factors as classification nodes, a total of 10 DRG groups were established by a decision tree model. The value of reduction in variance was 0.68, indicating significant heterogeneity between groups. The coefficient of variation for hospitalization costs ranged from 0.08 to 0.44, indicating minimal variation and reasonable grouping result. The established hospitalization cost scheme for patients with UL is reasonable in this study. It provides a reference for the advancement and implementation of DRG payments in local area.