Optimizing case mix for per-diem payment of mental disorders based on E-CHAID decision tree analysis.

BACKGROUND

Mental disorders impose a substantial economic burden on patients, their families, and society. Although per-diem payment is commonly used for inpatient mental health services, the current per-diem payment standards for mental disorders remain relatively crude and fail to reflect the differences in actual resource consumption among patients with different characteristics. Therefore, this study analyzed the factors associated with per-diem inpatient costs among patients with mental disorders to provide evidence for the development of a refined case mix-based per-diem payment system suitable for mental health services provided in China.

METHODS

A total of 25,510 inpatients with mental disorders admitted to two hospitals in Chongqing between 2020 and 2024 were included in this study, and their demographic, clinical, and cost information collected. A univariate analysis and generalized linear model were used to identify the main factors associated with per-diem inpatient costs, and the Exhaustive CHi-squared Automatic Interaction Detection (E-CHAID) decision tree model was applied to construct a case-mix classification model for per-diem payments in mental health services.

RESULTS

The generalized linear model showed that sex, age, insurance type, discharge mode, hospital level, readmission plan, inclusion in clinical pathways, surgical status, number of comorbidities, number of hospitalizations, length of stay (LOS), and principal diagnosis were significantly associated with per-diem inpatient costs. Upon the use of LOS, hospital level, insurance type, and number of comorbidities as splitting variables, 10 case-mix groups were generated. The grouping scheme demonstrated high reliability and robustness, with a reduction in variance of 0.43 and coefficients of variation of < 1 across all groups (0.20-0.59). The per-diem payment standards for the 10 groups were 120.06-658.16 yuan.

CONCLUSIONS

LOS, hospital level, insurance type, and number of comorbidities were key factors for grouping inpatients with mental disorders, and the construction of a refined case mix-based per-diem payment system using the E-CHAID decision tree is a reasonable and appropriate approach.