Development of Prediction Models for Healthy Ageing in Community-Dwelling Middle-Aged and Older Adults: A...

Area of Science:

• Gerontology
• Public Health
• Computational Biology

Background:

• Healthy ageing is a key public health concern with undefined determinants.
• Predicting healthy ageing is crucial for developing targeted interventions.

Purpose of the Study:

• To develop and validate machine learning (ML) models for predicting healthy ageing in community-dwelling middle-aged and older adults.
• To identify key characteristics influencing healthy ageing outcomes.

Main Methods:

• A retrospective cohort study using data from the English Longitudinal Study of Ageing (participants aged 50+).
• Utilized sociodemographic, health, lifestyle, and psychosocial data.
• Developed ML models (decision tree, logistic regression, neural network, random forest) and employed SHapley Additive exPlanations for predictor analysis.

Main Results:

• The random forest ML model demonstrated the best predictive performance (AUC = 0.78).
• Positive predictors of healthy ageing included normal physical performance, higher household wealth, and younger self-perceived age (50-59).
• Negative predictors were physical inactivity, abdominal obesity, and lack of internet/email use.

Conclusions:

• ML models effectively predict healthy ageing in older adults based on baseline characteristics.
• Findings can inform public health strategies to promote active and healthy ageing.