An Empirical Study on Ambulatory Patient Groups of Diabetic Patients

Chen Meng'en; Wang Yan; Hou Haojia; Yang Jingyu; Wang Zhiwei; Zhang Xiaoxi

Abstract：Objective:It aims to analyzes the influencing factors of diabetes outpatient costs and conduct case grouping,providing references for chronic disease outpatient cost management and deepening of the medical insurance system.Methods:Outpatient medical insurance claims data of diabetic patients in Lanzhou during 2022-2023 were retrospectively analyzed.Univariate analysis and multivariate linear model were applied to explore the factors affecting outpatient medical cost.A decision tree model was applied for outpatient case grouping,followed by evaluation and cost estimation of the case combinations.Results:A total of 14 038 valid data were included.The outpatient medical cost for diabetes is associated with patient gender,ethnicity,age,type of diabete(P＜0.05).The E-CHAID decision tree model identified 9 case combinations,with Combination 9 having the highest number of cases and Combination 3 the lowest;Combination 4 had the highest standard cost,and Combination 3 the lowest.The coefficient of variation values for all case combinations were close to 1,reduction in variance values of 0.978,and related weight values ranging from 0.30 to 1.33.Conclusion:The core factors influencing the outpatient medical cost for diabetes patients are medication,diagnostic tests,and patient age.Grouping outpatient cases based on a decision tree model yields good results,with strong homogeneity within groups and heterogeneity between groups,and significant differences in cost standards across different combinations.It is recommended to strengthen the control of outpatient medical costs for diabetes,promote the development of a diversified outpatient payment system,collaboratively optimize chronic disease diagnosis and treatment protocols,and standardize electronic medical record documentation,so as to alleviate the economic burden on patients and advance the reform of outpatient medical insurance systems.