摘要
目的探讨影响脑卒中发生风险的因素,构建并验证脑卒中发生风险的预测模型。方法将2021年1月-12月乌鲁木齐市某三甲医院住院患脑卒中患者作为研究对象,以同期未患脑卒中患者作为对照组,利用Lasso回归筛选出与脑卒中相关的变量,用于构建多因素Logistic回归分析模型;采用列线图实现模型的可视化,Bootstrap法对列线图预测模型进行内部验证,应用ROC曲线分析列线图预测模型预测发生脑卒中风险的能力。结果本研究纳入595名研究对象,有344(57.82%)人患脑卒中。Lasso回归筛选出7个与脑卒中发生风险相关的变量,分别是性别、年龄、小腿围、握力、舒张压、中性粒细胞百分比、钠离子;Logistic回归分析发现性别、小腿围、握力、舒张压、中性粒细胞百分比、钠离子是脑卒中发生风险的影响因素(均P<0.05)。应用ROC曲线分析列线图预测模型预测脑卒中发生风险的能力,0.80为曲线下面积,说明该模型的预测价值高。结论本研究建立的风险预测模型有较好的预测能力,可用于指导临床医务人员尽早识别出脑卒中的高风险人群,为早诊断早治疗提供依据。
Objective To explore the factors influencing the risk of stroke and establish and verify the prediction model of stroke.Methods The inpatients with stroke in a first-class hospital in Urumqi City from January to December 2021 were selected as the research objects,and patients without stroke during the same period were selected as the control group.Variables related to stroke were screened by Lasso regression for the construction of multivariate Logistic regression analysis model.The visualization of the model was realized by using the line graph.The Bootstrap method was used to internally verify the line graph prediction model,and the ROC curve was used to analyze the ability of the line graph prediction model to predict the risk of stroke.Results In this study,595 subjects were included,and 344(57.82%)of them suffered from stroke.Seven variables related to stroke risk were identified by Lasso regression,which were gender,age,calf circumference,grip strength,diastolic blood pressure,neutrophil percentage and sodium ion.Logistic regression analysis showed that gender,calf circumference,grip strength,diastolic blood pressure,neutrophil percentage and sodium ion were the influencing factors of stroke risk(all P<0.05).ROC curve was used to analyze the ability of the line graph prediction model to predict the risk of stroke.0.80 was the area under the curve,indicating the high predictive value of the model.Conclusion The risk prediction model established in this study has good predictive ability,which can be used to guide clinical staff to identify the high-risk population of stroke in advance,and provide basis for early diagnosis and early treatment.
作者
李辉
张亮
韩晓丽
LI Hui;ZHANG Liang;HAN Xiao-li(1474 hospital of Xinjiang,Urumqi,830000,China;The friendship hospital of Urumqi,Urumqi,830046,China)
出处
《新疆医学》
2023年第12期1435-1438,1457,共5页
Xinjiang Medical Journal
基金
新疆维吾尔自治区自然科学基金(项目编号:2016D01A022)
乌鲁木齐市卫生健康委科技计划(项目编号:202115)。
作者简介
李辉,女,副主任护师,研究方向:脑卒中预防,营养与疾病;通信作者:韩晓丽,女,硕士,主任医师,研究方向:营养与疾病,E-mail:hxl20030331@sina.com。
