基于贝叶斯网络的老年缺血性脑卒中患者衰弱的影响因素分析  

Analysis of the factors influencing frailty in elderly ischemic stroke patients based on a Bayesian network

作　　者：乔丽敏 王明慧 赵雅宁 刘瑶 顾小颖 史雪菲 QIAO Limin;WANG Minghui;ZHAO Yaning;LIU Yao;GU Xiaoying;SHI Xuefei(College of Nursing and Rehabilitation,North China University of Science and Technology,Tangshan 063200,China;Affiliated Hospital of North China University of Science and Technology,Tangshan 063200,China)

机构地区：[1]华北理工大学护理与康复学院,河北唐山063200 [2]华北理工大学附属医院,河北唐山063200

出　　处：《现代医学》2025年第4期515-522,共8页Modern Medical Journal

基　　金：唐山市科技创新团队培养计划(18130218A);2021年度华北理工大学省属高校基本科研业务费项目(JQN2021038)。

摘　　要：目的:探讨老年缺血性脑卒中患者衰弱的影响因素并分析影响因素间网络关系,为制定个性化措施提供参考。方法:选取2022年8月至2023年7月在华北理工大学附属医院收治的904例老年缺血性脑卒中患者为研究对象并收集相关资料。采用倾向性评分匹配的方法对混杂因素进行匹配,单因素及多因素Logistic回归分析模型筛选患者发生衰弱的独立影响因素,使用R及Netica软件进行贝叶斯网络模型构建及风险推理。结果:成功匹配282对研究对象,多因素Logistic回归分析筛选后,脑血管病发生次数、体育锻炼、抑郁、自理能力、预后营养指数(PNI)、高血压被纳入贝叶斯网络模型。结果显示,自理能力、体育锻炼、PNI、高血压与衰弱直接相关,且均是衰弱发生的父节点。结论:老年缺血性脑卒中患者衰弱影响因素较多,贝叶斯网络能较好地揭示老年缺血性脑卒中患者衰弱与影响因素间的网络关系,找出其直接或间接的影响因素,为早期预防老年缺血性脑卒中患者发生衰弱提供科学依据。Objective:To explore the influencing factors of frailty in elderly ischemic stroke patients and analyze the network relationship among influencing factors to provide reference for developing personalized interventions.Methods:A total of 904 elderly ischemic stroke patients admitted to the Affiliated Hospital of North China University of Science and Technology from August 2022 to July 2023 were selected as the study subjects and relevant data were collected.The propensity score matching method was used to match confounders,and univariate and multivariate Logistic regression analysis models were used to screen independent influencing factors of patient frailty.Bayesian network modeling and risk reasoning were performed using R and Netica software.Results:282 pairs were successfully matched.Multivariate Logistic regression analysis indicated that the frequency of cerebrovascular disease episodes,physical exercise,depression,self-care ability,prognostic nutritional index(PNI),and hypertension were integrated into the Bayesian network model.The results showed that self-care ability,physical exercise,PNI and hypertension were directly related to frailty,and all serve as parent nodes in the occurrence of frailty.Conclusion:There are many influencing factors in elderly ischemic stroke patients.Bayesian network can effectively reveal the network relationship between frailty and influencing factors in elderly ischemic stroke patients,identifying both direct and indirect influencing factors.This approach provides a scientific basis for early prevention of frailty in elderly ischemic stroke patients.

关 键 词：缺血性脑卒中 衰弱 倾向性评分匹配 贝叶斯网络 影响因素 

分 类 号：R743.3[医药卫生—神经病学与精神病学]