Multi-state analysis of cognitive ability data: a piecewise-constant model and a Weibull model

Area of Science:

• Gerontology
• Biostatistics
• Epidemiology

Background:

• Cognitive decline is a significant concern in aging populations, impacting health trajectories and care needs.
• Longitudinal studies are vital for understanding the progression of health states over time.
• Accurate measurement of cognitive ability is essential for predicting future healthcare requirements.

Purpose of the Study:

• To develop and apply an illness-death model for estimating transition intensities between healthy and unhealthy states in older adults.
• To model cognitive decline trajectories, accounting for potential misclassification of cognitive improvement.
• To extend the methodology for estimating life expectancy with and without cognitive impairment.

Main Methods:

• Utilized longitudinal data from the Medical Research Council Cognitive Function and Ageing Study (UK, 1991-2005).
• Employed an illness-death model with time-varying transition intensities, using piecewise-constant and Weibull models.
• Modeled observed cognitive improvement as a misclassification process.

Main Results:

• The study successfully estimated transition intensities between cognitive health states.
• The methodology allowed for the estimation of life expectancy stratified by cognitive impairment status.
• The models provide insights into the dynamics of cognitive decline and its influencing factors.

Conclusions:

• The developed illness-death models offer a robust framework for analyzing cognitive aging and its impact on life expectancy.
• Understanding cognitive decline dynamics is critical for healthcare planning and resource allocation.
• The methodology can be applied to other longitudinal health studies to assess disease progression and survival outcomes.