Risk Prediction Models for Depression in Community-Dwelling Older Adults

Area of Science:

• Gerontology
• Psychiatry
• Epidemiology

Background:

• Late-life depression poses a significant public health challenge.
• Accurate risk prediction is crucial for timely intervention.
• Existing models may lack efficiency or broad applicability.

Purpose of the Study:

• To develop streamlined Risk Prediction Models (Manto RPMs) for identifying individuals at risk of developing late-life depression.
• To create a user-friendly tool for risk assessment and prevention planning.

Main Methods:

• A prospective study utilizing data from the Survey of Health, Ageing and Retirement in Europe (SHARE).
• Development of Manto RPMs using regression, LASSO penalty, and Artificial Neural Networks on a large cohort of community-dwelling adults aged 55+.
• Identification of 129 predictors from a comprehensive literature review of 227 longitudinal studies.

Main Results:

• Manto RPMs demonstrated satisfactory accuracy in predicting depression risk (AUC up to 0.81).
• Models derived from combined depressed and non-depressed participants showed higher predictive performance.
• A web-based risk calculator was developed based on the streamlined Manto model.

Conclusions:

• The Manto RPMs effectively identify community-dwelling older adults at risk for developing depression within a 2-year timeframe.
• The freely accessible web calculator (https://manto.unife.it/) supports personalized and population-level prevention efforts.
• These models facilitate targeted interventions to mitigate the impact of late-life depression.