中青年脑卒中病人社交回避风险预测模型的构建与验证

Construction and verification of social avoidance risk prediction model in young and middle-aged stroke patients

目的:构建和验证中青年脑卒中病人社交回避风险预测模型,为减少病人社交回避提供参考。方法:选取锦州市2所三级甲等医院共392例脑卒中病人为建模组,沈阳市某三级甲等医院127例脑卒中病人为验证组,评估病人的神经功能损伤、睡眠、日常生活活动能力、心理韧性、领悟社会支持及社交回避情况,通过单因素分析和多因素LASSO回归探讨病人社交回避的危险因素,并建立和验证列线图风险预测模型。采用校正曲线、受试者工作特征(ROC)曲线下面积(AUC)评价模型的预测效能。结果:建模组中青年脑卒中病人社交回避检出率为40.56%;LASSO回归结果显示,卒中部位、神经功能损伤、睡眠、日常生活活动能力、心理韧性、领悟社会支持是社交回避的影响因素。基于上述6个独立影响因素建立的列线图预测模型具有较好的区分度和可靠性,内部验证AUC为0.839,外部验证AUC为0.829。结论:中青年脑卒中病人社交回避发生率较高,构建的列线图模型有助于有效预测病人早期社交回避风险。

Objective:To construct and verify the social avoidance risk prediction model of young and middle-aged stroke patients,so as to provide reference for reducing the social avoidance of patients.Methods:A total of 392 young and middle-aged stroke patients from 2 tertiary grade A hospitals in Jinzhou were selected.Neurological function,sleep,ability of activities of daily living,mental toughness,perceived social support,perceived social support and social avoidance were measured.Through univariate analysis and multivariate LASSO regression analysis,the risk factors of patients′social avoidance were discussed,and the risk prediction model was established.The calibration plot and the area under the receiver operating characteristic(ROC)curve were used to evaluate the prediction efficiency of the model.Results:The detection rate of social avoidance in young and middle-aged stroke patients in modeling group was 40.56%.LASSO regression results showed that stroke location,neurological impairment,sleep,activities of daily living,psychological resilience and perceived social support were the influencing factors of social avoidance.The Nomogram prediction model based on the above 6 independent influencing factors had good discrimination and reliability.The area under the receiver operating characteristic(ROC)curve was 0.839 in internal verification and 0.829 in external verification.Conclusion:The incidence of social avoidance in young and middle-aged stroke patients is high,and the constructed prediction model is helpful to effectively predict the risk of early social avoidance in patients.