DIP支付方式下四川省某三甲医院病种人组影响因素分析——基于多水平Logistic回归模型

Analysis of the Factors Influencing Disease Enrollment in One Tertiary Hospital in Sichuan Province under DIP Pay-ment Methods Based on Multilevel Logistic Regression Models

目的:研究医保DIP支付背景下病例诊断入组的影响因素,为医疗机构提高入组率提供参考依据。方法:利用成都市某公立医院2021年1~11月17689例住院患者病历资料,进行多水平Logistic回归分析。结果:单因素分析结果显示入院途径、科室、参保类型、费用偏离度、年龄是住院病人能否入组的影响因素。住院费用(对数)、住院天数(对数)、性别对是否入组无统计学影响;多水平Logistic回归模型显示,住院费用、住院天数、性别对入组与否无统计学影响(P>0.05),入院方式、参保类型、年龄、费用偏离度是病人能否入组的正向影响因素(P<0.05)。结论:多水平模型结果显示,科室水平的入组结果存在一定的聚集性,病人入组与否不仅与个体特征有关,还受到所属科室的影响。实行按病种分值付费时,医院管理者应当考虑不同科室、不同病人特征对病种入组的影响,从而提升医院、科室的精细化管理水平。

Objective To investigate the influencing factors of case diagnosis enrollment under the background of medical in-surance DIP payment,and provide reference for medical institutions to improve enrollment rates.Methods A multi-level logistic regression analysis was conducted on the medical records of 17689 hospitalized patients in a public hospital in Chengdu from Janu-ary to November 2021.Results The results of univariate analysis showed that admission routes,departments,types of insurance coverage,cost deviation,and age were the influencing factors on whether hospitalized patients could be enrolled.The hospitaliza-tion cost,length of stay and gender had no statistically significant impact on enrollment.The multi-level logistic regression model showed that hospitalization costs,length of stay and gender had no statistically significant impact on enrollment(P>0.05),while admission methods,type of insurance,age,and cost deviation were positive influencing factors for patient enrollment(P<O.05).Conclusion The results of multi-level model showed that there was a certain degree of clustering in the enrollment results at the department level.Whether patients were enrolled was not only related to individual characteristics,but also influenced by the department they belong to.When implementing payment based on disease classification,hospital managers should consider the im-pact of different departments and patient characteristics on disease classification,which can improve the level of refined manage-ment in hospitals and departments.