摘要
目的探讨脑卒中半失能老年患者坠床跌倒相关因素,构建列线图预测模型,制定针对性防治措施,以降低坠床跌倒风险。方法回顾性收集2020年1月至2022年12月374例脑卒中半失能老年患者临床资料,按7∶3比例随机分为建模组、验证组。采用单因素、多因素Logistic回归方程筛选脑卒中半失能老年患者坠床跌倒风险。建立逐步Logistic回归、Lasso-Logistic回归模型进行参数估计。基于Lasso-Logistic回归方程筛选高危因素构建列线图预测模型并进行内部验证,评价该预测模型预测效能及临床效用。结果374例调查问卷,有效回收369份,369例脑卒中半失能患者按73比例随机分配样本量,最终建模组258例,验证组111例;跌倒史、睡眠障碍、抑郁状态、衰弱、夜尿≥3次/晚、Morse跌倒风险评分是坠床跌倒的危险因素,且经逐步Logistic回归证实其拟合、预测效果相对较好(P<0.05);列线图预测模型在建模组、验证组的C-index分别为0.813、0.842,AUC分别为0.813、0.842,且校准曲线、DCA曲线证实该模型校准能力、净获益值较高。结论跌倒史、睡眠障碍、抑郁状态、衰弱、夜尿≥3次/晚、Morse跌倒风险评分升高为脑卒中半失能老年患者坠床跌倒的危险因素,基于上述危险因素建立列线图预测模型对预测坠床跌倒风险具有临床应用价值,可为临床工作者提供可预见性的干预措施,减少坠床跌倒发生。
Objective To investigate the influencing factors for falling down from bed in semi-disabled elderly stroke patients,and to create a nomogram and formulate targeted prevention and treatment measures to reduce the risk of falling down from bed.Methods Clinical data of 374 semi-disabled elderly stroke patients from January 2020 to December 2022 were retrospectively collected and randomly divided into modeling group and verification group at a ratio of 7∶3.Univariate and multivariate Logistic regression analyses were performed to screen the risk factors for falling down from bed in semi-disabled elderly stroke patients.Stepwise Logistic regression and Lasso-Logistic regression models were established for parameter estimation.Based on the Lasso-Logistic regression equation,the risk factors falling down from bed in semi-disabled elderly stroke patients were screened to construct a nomogram,and internal verification was carried out to evaluate its prediction efficiency and clinical effectiveness.Results Among 374 cases,totally 369 questionnaires were effectively collected.A total of 369 semi-disabled elderly stroke patients were randomly divided into modeling group(n=258)and verification group(n=111)at a ratio of 7∶3.History of falling,sleep disorder,depression,frailty,nocturia≥3 times/night and Morse Fall Scale(MFS)score were the risk factors for falling down from bed in semi-disabled elderly stroke patients,which were verified with good goodness-of-fit and prediction effect by stepwise Logistic regression(P<0.05).The C-index of the nomogram in the modeling group and the verification group was 0.813 and 0.842,respectively,and the area under the curve(AUC)was 0.813 and 0.842,respectively.The calibration curve and decision curve analysis(DCA)confirmed that the model had high calibration ability and net benefit value.Conclusion History of falling,sleep disorder,depression,frailty,nocturia≥3 times/night,and MFS score are risk factors for falling down from bed in semi-disabled elderly stroke patients.A nomogram base
作者
任斯诗
叶菲
郑涛
詹凡
杨莉
REN Sishi;YE Fei;ZHENG Tao(Wuhan Red Cross Hospital,Hubei,Wuhan 430015,China)
出处
《河北医药》
CAS
2024年第6期924-929,934,共7页
Hebei Medical Journal
