Dynamic nomogram for predicting depression risk in middle-aged and older adults based on CHARLS

This study aims to develop and validate a nomogram-based risk assessment model for predicting high-risk individuals with depression in middle-aged and elderly populations. Data from the China health and retirement longitudinal study (CHARLS) in 2013 and 2018 were used to construct training cohort and validation cohorts. Spearman correlation analysis, LASSO regression, univariate and multivariate regression analyses were performed to select features, and the significant features were included in the nomogram model construction. The model was evaluated using calibration curves, ROC curves, and clinical decision curve analysis in the training, internal validation, and external validation cohorts to assess its predictive performance and clinical utility. The final selected features included gender, alcohol consumption, health status (self-rated health), life satisfaction, sleep disorders, activities of daily living score, cognition, hearing, and pain, which were used to construct the nomogram model. The AUC values of the model in the training, internal validation, and external validation cohorts were 0.823, 0.823, and 0.819, respectively. The calibration curves showed good consistency between the predicted and actual values, and the clinical decision curves indicated good clinical utility of the model. An interactive dynamic nomogram web application was also developed, allowing users to manually input variable values and instantly obtain predicted probabilities with confidence intervals, thereby enhancing the model's clinical usability and interpretability. The nomogram model developed in this study effectively predicts individuals at risk of depression in middle-aged and elderly populations, with good predictive performance and clinical utility. This model could assist in early screening and intervention for high-risk individuals, thereby reducing the burden of depression.