| Objective: With the aging of the population and the aging of the elderly population,the elderly’s demand for medical and health services is increasing.How to formulate a reasonable medical insurance payment policy to solve the contradiction between the limited health resources and the diversified health service needs of the elderly has become an important issue in the government’s health system reform.In this context,Dalian will comprehensively carry out DRG payment reform in 2021.In this study,middle-aged and elderly non-surgical inpatients with femoral fracture in the DRG group were taken as the research objects,and the population characteristics in the DRG group were analyzed to find the relationship between the disease characteristics of patients and medical costs,so as to provide references for establishing reasonable payment standards,guiding discharged patients with different service needs to be transferred downward,and improving the efficiency of medical insurance fund use.Methods: To understand the theories of medical insurance payment,combination of medical and nursing care,hierarchical diagnosis and treatment,social hospitalization and other aspects through literature.The data of the disease group from Dalian Medical Insurance information system was cleaned,and the information of 1689 patients was finally included in the study.Descriptive statistical analysis was made on the basic information of senile inpatients with femoral fracture without operation.Chi-square test was used to analyze the distribution of patients with different characteristics.Nonparametric test was used to analyze the basic situation of hospitalization expenses of patients with different sociological and clinical characteristics.Multiple linear regression model and support vector machine model were used to analyze the influencing factors of hospitalization costs.The decision tree model was used to group patients,and the population characteristics and hospitalization costs of patients in different groups were studied.Results: The mean age of patients was 78.80 years.Female patients accounted for67.61%;The patients in tertiary hospitals accounted for 81.05%;Outpatient admission accounted for 59.98%;Doctor’s order to leave hospital accounted for 91.42%;Medical insurance for employees accounted for 51.81%;Patients without complications accounted for 60.39%;72.94% of the patients had a 31-day re-hospitalization plan.The average length of hospitalization was 10.45 days,the average hospitalization cost was22480.41 yuan,and the average daily cost was 2125.07 yuan.The results of multiple linear regression model showed that the model had statistical significance.According to the effect size of various factors on hospitalization costs,the order was as follows:length of stay,31-day re-hospitalization plan,superior care,age,admission route,type of treatment,type of insurance insurance,complications,gender,way of discharge,fracture site.The results of the support vector machine model show that the factors influencing hospitalization costs and their importance are ranked as follows: Plan of re-hospitalization for 31 days after discharge: 0.2461,length of stay: 0.0962,age:0.0906,complications: 0.0775,route of admission: 0.0586,pulmonary infection: 0.0553,fracture site: 0.0519,cerebral infarction: 0.0503,way of discharge: 0.0486,diabetes:0.0436.The prediction accuracy of support vector machine model reached 87.86%.The results of decision tree model showed that the number of days in hospital,the 31-day re-hospitalization plan,age and fracture site were finally selected as classification nodes,and 7 combinations were constructed.The first layer screened by the model is the grouping of hospitalization days,which is the most important grouping variable.The group that consumed the least medical resources was the case combination with hospitalization days of 3-11 days and no re-hospitalization plan for 31 days,a total of330 cases,accounting for 19.5%,with an average hospitalization cost of 8668.74 yuan.The group that consumes the most medical resources is residential.Conclusion: The non-operative inpatients with fracture of femur were mainly female and elderly.The majority of patients in tertiary hospitals are employees’ medical insurance patients,which are characterized by a high emergency visit rate,acute illness,high incidence of diabetes,and a large number of patients with femoral shaft fracture and femoral trochanteric segment fracture.Most of the patients have the need for readmission plans.The majority of secondary hospitals are residents with medical insurance and outpatients,and most patients leave the hospital without medical advice.Most patients have a 31-day readmission plan.Taking patients with readmitted plans without complications as an example,the simulation according to the weight of DRG group found that the actual cost of tertiary hospitals was lower than the medical insurance limit,while that of secondary hospitals was higher than the medical insurance payment standard.In terms of hospitalization frequency,only a small number of patients had multiple visits,and social hospitalization did not occur.But at the same time,it also shows the importance of strengthening the capacity building of grassroots medical and elderly care services.If the combined capacity of grassroots institutions is not enough to meet the discharged patients’ demand for combined services,social hospitalization may occur,resulting in the waste of medical resources. |