Influencing factors of hospitalization expenses of serious illness based on quantile regression
Liu Pei1, Zhong Shiyang2, Guo Wei3, Lü Yipeng4, Tang Yuanjie5
1 Department of Mathematic & Physics, Naval Medical University, Shanghai 200433, China;
2 Outpatient Department, Jiangsu Corps Hospital of Jiangsu of Armed Police Forces, Yangzhou 225003, China;
3 Department of Health Statistics, Naval Medical University, Shanghai 200433, China;
4 School of Public Health, School of Medicine, Shanghai Jiao Tong University, Shanghai 200025, China;
5 Jiangsu Corps Hospital of Jiangsu of Armed Police Forces, Yangzhou 225003, China.
Abstract：Objective To analyze the composition of hospitalization expenses, explore the related influencing factors and provide policy suggestions. Methods Inpatient data of medical expenses in a tertiary hospital in Yangzhou City, Jiangsu Province from 2016 to 2019 were used, and Stata 16 software was used to analyze the composition of inpatient medical expenses. On the basis of single factor analysis, a quantile regression model of inpatient medical expenses was constructed. Results Mean hospitalization cost of serious illness was 34 860 yuan, and the proportion of drug cost decreased by 8.94% over four years. Univariate analysis showed that the influencing factors of hospitalization expenses included age (χ2=63.467, P<0.001), marital status (Z=-4.305, P<0.001), source of patients (Z=-10.555, P<0.001), identity (Z=-4.207, P<0.001), way of payment (Z=-14.287, P<0.001), hospitalization days (χ2=1 930.540, P<0.001), operation (Z=-17.149，P<0.001), curative effects (χ2=49.104，P<0.001) and hospitalization times (χ2=111.937, P<0.001). The results of quantile regression model showed that way of payment, hospitalization expenses, operation, and hospitalized times had an impact on different quantile points of hospitalization expenses. Conclusion Compared with the traditional regression model, quantile regression model is more robust. Increasing the coverage of medical insurance, and carrying out health education can not only promote the utilization rate of medical insurance, but also reduce the medical expenses.
