Abstract: Medical insurance reform and medical expenses are people’s important livelihood issues. People pay attention to whether there are differences in the proportion of medical expenses paid by different types of medical insurance and the influencing factors of medical expenses. In this paper, the discharge settlement data of type 2 diabetes patients in the secretion department of a tertiary class-Ⅲ-A hospital in Chengdu, Sichuan Province are analyzed from January 2020 to December 2021 by using the high-dimensional permutation test and the interaction effect model to explore whether there is a significant difference in the medical expense out-of-pocket ratio between urban and rural resident medical insurance patients and urban employee medical insurance patients and the influencing factors of the medical expense out-of-pocket ratio, so as to provide a basis for the subsequent reform of medical insurance. The results of the study show that there are significant differences in the medical expense out-of-pocket ratio of patients with two types of medical insurance, and the increase of various expense ratios such as examination expense ratio would increase the medical expense out-of-pocket ratio. Moreover, under low dimensions, the results of the high-dimensional permutation test method used in this paper are consistent with those of the Hotelling T² permutation test.

Key words: medical insurance, medical expenses, high dimensional permutation test, interaction effect, multivariate regression