超长住院日患者的影响因素分析与预测模型构建  

Analysis of influencing factors and construction of a prediction model for extended hospital stay patients

作　　者：沈可一 郑心月 肖晓月 吴丹 林昕皓 裴彤 孟雪晖 SHEN Keyi;ZHENG Xinyue;XIAO Xiaoyue;WU Dan;LIN Xinhao;PEI Tong;MENG Xuehui(School of Humanities and Management,Zhejiang Chinese Medical University,Hangzhou,Zhejiang Province,311402,PRC)

机构地区：[1]浙江中医药大学人文与管理学院,浙江省杭州市311402

出　　处：《中国医院》2024年第11期59-63,共5页Chinese Hospitals

基　　金：国家自然科学基金资助项目(72074064);浙江省医药卫生科技计划项目(2024KY1190)。

摘　　要：目的:探讨超长住院日发生的影响因素,并构建预测模型,为缩短平均住院日提供科学参考。方法:选取某三甲医院44901例出院患者作为研究对象,分为训练集35921例和验证集8980例,根据住院天数划分为超长住院组与常规住院组,采用多因素分析筛选超长住院的影响因素,构建列线图模型,采用C-index指数与校准曲线对模型进行评估。结果:发生超长住院日的患者共计1262例,占2.80%。回归分析结果表明,年龄、职业、手术/操作情况、疾病严重程度、高倍率病例、中药制剂和中医诊疗设备是影响超长住院日的因素(P<0.05);列线图模型的内部与外部验证C-index指数分别为0.883和0.887,校准曲线结果良好。结论:年龄、疾病严重程度、高倍率病例和治疗方式是超长住院日的影响因素,基于这些因素构建的列线图模型对超长住院日发生风险的预测效果较好。Objective:To explore the influencing factors of extended hospital stays and construct a prediction model,providing scientific reference for shortening the average length of hospital stays.Methods:A total of 44901 discharged patients from a Grade III Level A hospital were selected as the study subjects,divided into a training set(35,921 cases)and a validation set(8,980 cases).Patients were categorized into the extended stay group and the regular stay group based on the number of hospitalization days.Multivariate analysis was used to identify factors influencing extended hospital stays,and a nomogram model was constructed.The C-index and calibration curve were used to evaluate the model.Results:A total of 1,262 patients(2.80%)experienced an extended hospital stay.Regression analysis showed that factors such as age,occupation,surgery/procedures,disease severity,high-multiplier cases,traditional Chinese medicine(TCM)preparations,and TCM diagnostic equipment were significant influencers of extended hospital stays(P<0.05).The internal and external validation C-index of the nomogram model were 0.883 and 0.887,respectively,with good calibration curve results.Conclusion:Age,disease severity,high-multiplier cases,and treatment methods are the main factors influencing extended hospital stays.The nomogram model based on these factors provides good predictive performance for the risk of extended hospital stays.

关 键 词：超长住院日 住院日 影响因素 预测模型 

分 类 号：R197[医药卫生—卫生事业管理]