Decline in learning ability best predicts future dementia type: the Freedom House Study

Area of Science:

• Neuroscience
• Gerontology
• Cognitive Psychology

Background:

• Early detection of dementia is crucial for timely intervention.
• Longitudinal changes in cognitive function may offer predictive insights into dementia type.
• Understanding predictors of dementia progression is vital for at-risk elderly populations.

Purpose of the Study:

• To investigate longitudinal changes in learning efficiency as a predictor of dementia type in healthy elderly retirees.
• To determine if the rate of change in memory performance predicts future Alzheimer's type versus non-Alzheimer's type dementia.
• To assess the predictive value of dynamic memory changes compared to baseline memory function.

Main Methods:

• Utilized latent growth (LG) models to analyze serial California Verbal Learning Test (CVLT) performance over 36 months.
• Developed a higher-order LG model to represent the dynamic change in learning efficiency (DeltaCVLT).
• Used DeltaCVLT to predict dementia type (none, Alzheimer's, non-Alzheimer's) adjusting for baseline factors.

Main Results:

• Nonlinear LG models best fit the data, indicating significant variability in baseline memory and learning efficiency decline.
• Age, baseline dementia type, and DeltaCVLT significantly predicted final dementia type.
• Baseline memory performance (CVLT1) did not independently predict dementia type when accounting for the rate of change.

Conclusions:

• The dynamic rate of change in learning efficiency is a significant predictor of future dementia type in elderly individuals.
• Serial memory assessments are valuable for identifying individuals at varying risk for developing dementia.
• Baseline memory alone is insufficient for predicting dementia type; tracking cognitive change over time is essential.