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

Area of Science:

• Gerontology
• Psychiatry
• Biostatistics

Background:

• Depression is a significant health concern in middle-aged and elderly populations.
• Accurate risk assessment is crucial for timely intervention and management.
• Existing prediction models may lack specific applicability to diverse demographic groups.

Purpose of the Study:

• To develop and validate a nomogram-based risk assessment model for predicting high-risk individuals with depression.
• To identify key predictors of depression in middle-aged and elderly populations.
• To enhance clinical utility through an interactive dynamic web application.

Main Methods:

• Utilized data from the China Health and Retirement Longitudinal Study (CHARLS) for training and validation cohorts.
• Employed Spearman correlation, LASSO, and regression analyses for feature selection.
• Evaluated model performance using calibration curves, ROC curves, and decision curve analysis.

Main Results:

• The nomogram incorporated gender, alcohol consumption, self-rated health, life satisfaction, sleep disorders, ADL score, cognition, hearing, and pain.
• Achieved AUC values of 0.823 (training), 0.823 (internal validation), and 0.819 (external validation).
• Demonstrated good calibration and clinical utility, supported by an interactive web tool.

Conclusions:

• The developed nomogram is an effective tool for predicting depression risk in older adults.
• The model offers good predictive performance and clinical utility for early screening.
• Facilitates early intervention, potentially reducing the burden of depression in this demographic.