Development and Validation of a Risk Prediction Model for ADL Dysfunction in the Elderly

Background : Aging populations have led to increased chronic diseases, economic challenges, and labor shortages, significantly impacting China's public health. The elderly often seek to maintain independence, making research crucial to understand factors that lead to daily life dysfunction. This study aimed to develop and validate a predictive tool to identify risks of daily life dysfunction in the elderly. Methods: Data from 5081 participants aged 60 to 80 in the China Health and Retirement Longitudinal Survey(CHARLS) wave 3 was analyzed. Participants were divided into training and testing groups. Predictive factors were identified using Least Absolute Shrinkage and Selection Operator(LASSO) and multivariable logistic regression, resulting in a nomogram model. The model's performance was evaluated using Receiver Operating Characteristic(ROC) curves, the calibration plots, and Decision Curve Analysis(DCA). Results: It identified 12 factors as risk factors for daily life dysfunction in the elderly, which were integrated into the final model. The nomogram's predictive performance was deemed acceptable, with ROC curve values of 0.788 (95% CI: 0.771-0.804) for the training set and 0.794 (95% CI: 0.773-0.814) for the testing set. The calibration curve confirmed the model's accuracy, and the DCA showed it had good clinical utility. Conclusions: Twelve key factors were chosen to create a nomogram predicting daily life dysfunction in the elderly. This nomogram demonstrates strong evaluation performance and serves as a dependable tool for forecasting daily life dysfunction in the elderly.