Empirical Research on DRGs Path of Ovarian Cancer Hospitalization Expenses Based on Decision Tree Model
Objective:To explore the diagnosis-related groups(DRGs)path and hospitalization expenses standards for ovarian cancer,providing evidence for optimizing the disease grouping scheme and improving the level of hospital fine management.Methods:The study collected the first page information of 1585 patients diag-nosed with ovarian cancer who were discharged form Sichuan University West China Second University Hospital between January 1,2019 and September 30,2023.Univariate analysis and Logistic multiple linear regression a-nalysis were used to analyze the factors affecting hospitalization expenses.The decision tree model was em-ployed to establish and verify a case mix plan for ovarian cancer and analyze the over-standard costs.Results:①The results of Logistic multiple linear regression analysis showed that age,length of stay,number of comorbidi-ties,level of surgical procedure,whether rescue was performed,whether blood transfusion was performed,and whether there was metastasis were related factors that affected hospitalization costs(P<0.05).②1585 patients were divided into DRG1 to DRG11 using the decision tree model,where the first layer classification node was length of stay,the second layer classification node was whether blood transfusion was performed,followed by the number of comorbidities and whether metastasis was performed(P<0.001).Kruskal-Wallis H test showed that the11 groups had good inter-group heterogeneity(P<0.001)and good intra-group homogeneity[coefficient of variation(CV)0.15-0.30].③A total of 174 cases(10.98%)exceeded the upper limit of hospitalization expen-ses.The highest proportion of excessive expenses in the DRG7 group(hospital stay of 6-9 days,no blood transfusion,and≥5 comorbidities)was 15.11%.The cost of exceeding the standard cases is higher than that of below the standard cases,The proportion of medical service income is 4.82 percentage points lower.Conclu-sions:It is reasonable and scientific to use the decision tree model to group hospitalization expenses for ovarian cancer.medical institutions can focus on the analysis of the cost of patients with over-limit disease group by tak-ing the number of days in hospital as the breakthrough point,and control the cost for fine management.The ad-ministrative department can group DRGs according to the influencing factors in this plan.
Decision tree modelDiagnosis-related groupsCase mixOvarian cancerHospitalization expenses
国家科技期刊平台
NSTL
万方数据
