老年认知障碍患者伴发抑郁风险的列线图预测模型构建和验证OA北大核心CSTPCD

Objective To construct a predictive model of depression risk in older people with cognitive impairment.Methods Using cross-sectional data from 2018 wave of the Chinese longitudinal healthy longevity survey(CLHLS),2070 samples in the training set and 888 samples in the test set were selected,and the older people with cognitive impairment were categorized into the depressed group[Center for Epidemiologic Studies Depression Scale(CES-D)≥10,317 in the training set,155 in the test set]and the non-depressed group(CES-D＜10,1 753 in the training set,733 in the test set)according to whether they were depressed or not.Demographic characteristics,living habits,and medical history were investigated.A predictive model for the risk of depression in older people with cognitive impairment was constructed based on the results of Lasso regression and binary logistic regression analysis.Results Age,cohabitation,smoking history,sleep,outdoor activities,reading books and newspapers,television and radio,and history of cataracts were independent risk factors for depression in elderly patients with cognitive impairment.The area under the curve(AUC)of the nomogram model based on these 8 factors was 0.807.Conclusions The predictive model constructed in this study to construct the risk of depression in elderly patients with cognitive impairment has a high degree of accuracy and specificity,which can be used to identify the risk of depression in older people with cognitive impairment.