基于决策树模型的卵巢恶性肿瘤患者住院费用疾病诊断相关分组实证研究  

Empirical Research on DRGs Path of Ovarian Cancer Hospitalization Expenses Based on Decision Tree Model

作　　者：徐偲瑜 郎肖玲 李松阳 孙率 杨兰 曹雅楠[1] 周雪莹 安景欢 XU Siyu;LANG Xiaoling;LI Songyang(Department of Operation Management and Assessment,Sichuan University West China Second University Hospital,Key Laboratory of Birth Defects and Related Diseases of Women and Children,Chengdu Sichuan 610041,China)

机构地区：[1]四川大学华西第二医院运营管理与评价部出生缺陷与相关妇儿疾病教育部重点实验室,四川成都610041 [2]四川省儿童医院四川省儿童医学中心运营管理部,四川眉山620010

出　　处：《实用妇产科杂志》2024年第4期322-326,共5页Journal of Practical Obstetrics and Gynecology

基　　金：四川省医院协会医务管理分会(编号:SCYW24-27)。

摘　　要：目的:探索卵巢恶性肿瘤患者疾病诊断相关分组(DRGs)方案及费用标准,为优化该病种分组方案以及提高医院精细化管理水平提供依据。方法:收集四川大学华西第二医院2019年1月1日至2023年9月30日主要诊断为卵巢恶性肿瘤的1585例出院患者的病案首页信息,利用单因素分析和Logistic多元线性回归分析影响住院费用的相关因素,采用决策树模型建立卵巢恶性肿瘤病例组合方案并验证,并对超标费用进行分析。结果:①Logistic多元线性回归分析结果显示,年龄、住院天数、合并症及并发症个数、手术级别、是否抢救、是否输血以及是否有淋巴结转移是影响住院费用的相关因素(P<0.05)。②1585例患者经过决策树模型分为DRG1组~DRG11组,其中决策树第一层分类节点为住院天数,第二层分类节点为是否输血,其次为合并症及并发症个数和是否淋巴结转移(P<0.001)。Kruskal-Wallis H检验示,11个分组组间异质性较好(P<0.001),组内同质性较好[变异系数(CV)0.15~0.30]。③住院费用超过费用上限的共174例(占10.98%)。DRG7组(住院天数6~9天、无输血、合并症及并发症个数≥5个)的超标费用占比最高为15.11%。174例住院费用超标病例的各收费项目的平均费用均高于未超标病例组,超标病例组医疗服务的收入占比比未超标病例组少4.82%。结论:采用决策树模型对卵巢恶性肿瘤病例进行住院费用分组具有一定合理性与科学性。医疗机构可以通过住院天数为切入点,重点分析超标病例组患者费用情况,进行费用控制以进行精细化管理,行政部门可根据此方案中的影响因素进行DRGs。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 diagnosed 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 analysis were used to analyze the factors affecting hospitalization expenses.The decision tree model was employed 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 comorbidities,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 the 11 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 expenses.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.Conclusions:It is reasonable and scientific to use the decision tree model to group hospitalization expenses for ovarian cancer.medical in

关 键 词：决策树 疾病诊断相关分组 病例组合 卵巢恶性肿瘤 住院费用 

分 类 号：R737.31[医药卫生—肿瘤]