首页|期刊导航|沈阳医学院学报|老年脑卒中患者衰弱风险预测模型的Meta分析

老年脑卒中患者衰弱风险预测模型的Meta分析OA

A meta-analysis of frailty risk prediction models in elderly stroke patients

中文摘要英文摘要

目的:系统评价老年脑卒中患者的衰弱风险预测模型与预测因子,为护理人员筛选合适的老年脑卒中衰弱风险评估工具提供参考,并为制定老年脑卒中衰弱患者的个性化干预方案提供依据.方法:检索PubMed、Web of Science、SCOPUS、The Cochrane Library、CINAHL、EMbase、知网、万方、维普和中国生物医学文献数据库,检索时限为建库至2025年9月.由2名研究人员独立筛选文献及提取数据,采用PROBAST工具评价文献质量,采用MedCalc和RevMan5.4软件对脑卒中衰弱预测模型的受试者工作特征曲线下面积(area under the curve,AUC)和预测因子进行Meta分析.结果:共纳入14篇文献,涵盖19个老年脑卒中衰弱风险预测模型,AUC为0.629~0.957,有17个预测模型的预测性能较好(AUC>0.7),Meta分析后AUC的合并效应量为0.846(95%CI:0.812~0.880),展示出良好的预测性能.文献质量评价结果为所有模型均存在高偏倚风险,但多数模型临床适用性较好.预测因子的Meta分析显示,年龄、吞咽障碍、多重用药、睡眠障碍、独居为老年脑卒中衰弱的预测因子(P<0.05).结论:当前老年脑卒中衰弱风险预测模型仍处于探索阶段,模型区分度好,但存在一定偏倚风险.未来还需关注模型开发过程中的数据来源与研究设计、纳入和排除标准设置、合理计算样本量与结果事件数与协变量个数比、连续变量处理方式、预测因子筛选方法、缺失值处理及模型性能和验证,以增强预测工具的科学性和实用价值.

Objective:To systematically evaluate the frailty risk prediction models and predictors in elderly stroke patients,so as to provide a reference for nursing staff in selecting appropriate frailty risk assessment tools for elderly stroke patients and to provide a basis for formulating personalized intervention plans for frail elderly stroke patients.Methods:PubMed,Web of Science,SCOPUS,The Cochrane Library,CINAHL,EMBASE,CNKI,Wanfang,VIP and the Chinese Biomedical Literature Database were searched from inception to Sep 2025.Two researchers independently screened the literature and extracted data.The PROBAST tool was used to evaluate the quality of the literature.MedCalc and RevMan 5.4 software were used to conduct meta-analysis of the area under the curve(AUC)of receiver operating characteristic curves and predictors of stroke frailty prediction models.Results:A total of 14 articles were included,encompassing 19 frailty risk prediction models for elderly stroke patients,with AUC values ranging from 0.629 to 0.957.Among these,17 prediction models had good prediction performance(AUC>0.7),and the pooled AUC from the meta-analysis was 0.846(95%CI:0.812-0.880),indicating good predictive performance.Quality assessment results showed that all models had a high risk of bias,but most had good clinical applicability.Meta-analysis of predictors showed that age,dysphagia,polypharmacy,sleep disorders,and living alone were predictors of frailty in elderly stroke patients(P<0.05).Conclusions:Current frailty risk prediction models for elderly stroke patients are still in the exploratory stage.The models show good discriminative ability but carry a certain risk of bias.Future research should pay attention to data sources and research design,inclusion and exclusion criteria settings,reasonable calculation of sample size and the number of events per variable(EPV),continuous variable processing method,predictor screening method,missing data management,as well as model performance and validation,so as to enhance the scientific and practical value of prediction tools.

侯佳怡;李壮苗;陈梅;张洁

福建中医药大学护理学院,福建 福州 350122福建中医药大学护理学院,福建 福州 350122福建中医药大学护理学院,福建 福州 350122福建中医药大学护理学院,福建 福州 350122

医药卫生

脑卒中衰弱风险预测Meta分析

strokefrailtyrisk predictionmeta-analysis

《沈阳医学院学报》 2026 (5)

465-475,11

福建省自然科学基金项目(No.2023J01874)

10.16753/j.cnki.1008-2344.2026.05.003

评论