社区慢性病老年人社会衰弱风险预测模型的构建及验证OA

Objective:To construct the social frailty risk prediction model for elderly people with chronic diseases in communities,and to explore the predictive value of intrinsic capacity for social frailty among them.Methods:A total of 785 elderly patients with chronic diseases in the communities,who were surveyed in Lu'nan district of Tangshan city from July 2024 to May 2025,were selected as research subjects.LASSO regression was used to screen the variables.Logistic regression analysis was conducted to analyze the influencing factors.A column chart was drawn,and a baseline model without intrinsic ability variables and a prediction model including intrinsic capacity variables were constructed.Results:The detection rate of social frailty among community-dwelling elderly with chronic diseases was 23.69%.The results of Logistic regression analysis showed that age,educational level,duration of chronic disease,level of e-health literacy,satisfaction with community public services,family function,and intrinsic capacity were the influencing factors of social frailty in community-dwelling elderly with chronic diseases(all P<0.05).Two models were established,one without the intrinsic ability variable and the other with the intrinsic ability variable.The results showed that the prediction model had significantly improved reclassification ability and discriminative ability compared to the baseline model.The area under the ROC curve of the prediction model training set and validation set was 0.828 and 0.856 respectively.The DCA curve indicated that the model had good clinical practicability.Conclusions:The social frailty risk prediction model for elderly people with chronic diseases in the communities had certain predictive efficacy.It could be used to identify the risk of social frailty in elderly people with chronic diseases in the communities at an early stage.It could provide basis for taking timely preventive and intervention measures.