Joint modeling of self-rated health and changes in physical functioning

Area of Science:

• Gerontology
• Biostatistics
• Public Health

Background:

• Self-rated health is a significant predictor of morbidity and mortality.
• Previous research links self-rated health to physical functioning levels and changes.
• No prior studies have jointly modeled longitudinal self-rated health and physical functioning.

Purpose of the Study:

• To develop and validate a joint statistical model for self-rated health and physical functioning trajectories.
• To investigate the relationship between perceived health status and the progression of physical disability over time.
• To improve the understanding of how self-perceptions of health influence functional decline in older adults.

Main Methods:

• A joint model combining a non-homogeneous Markov process for physical functioning states (e.g., activities of daily living [ADL], instrumental activities of daily living [IADL]) with a logistic regression model for self-rated health.
• Simulation studies to assess the model's properties and robustness to misspecification.
• Application to longitudinal data from the Cardiovascular Health Study (CHS) of older adults.

Main Results:

• The joint model effectively links self-rated health to both the current level of physical functioning and the risk of future functional decline.
• Self-rated health is significantly associated with difficulties in ADLs and IADLs.
• The proposed joint model demonstrates superior performance compared to empirical models based solely on observed functional changes.

Conclusions:

• Self-rated health is a crucial indicator associated with both existing functional limitations and the future risk of increasing disability in older adults.
• The developed joint model provides a robust framework for analyzing the interplay between perceived health and physical functioning over time.
• This approach enhances our ability to predict health trajectories and identify individuals at higher risk for functional decline.