Investigating episodic-memory predictors of early-stage Alzheimer's disease

Abstract

A rapid increase in Alzheimer's disease (AD) patients is expected over the next 30 years. Accordingly, there is a critical need for early-stage AD detection methods that can enable professionals to treat the disease adequately. The present study considers the ability of episodic-memory measures to predict mild cognitive impairment (MCI) to AD conversion and thus, detect early-stage AD. Using a binary logistic regression, episodic-memory tests were compared to each other and to prominent neuroimaging methods in MCI converter (MCI participants who developed AD) and MCI non-converter groups (MCI participants who did not develop AD). Standard tests for AD (e.g., MMSE) were also compared to specific episodic-memory tests-using a principal component analysis-to test if standard tests can measure episodic memory. Our study acquired all data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) and tested participants over four years. Our results indicate that individual episodic-memory measures could predict MCI to AD conversion better than episodic memory, neuroimaging, and mixed models. We theorised that mixed models were worse than individual tests, as mixed episodic-memory models increase multicollinearity and neuroimaging measures had poor accuracy. Specifically, the most accurate predictors were the ADNI memory score in year one (56.4%), the RAVLT percent forgetting measure in year two (71.7%), and the logical memory test in years three (76.9%) and four (77.2%). Our results also indicated that standard tests could be used to measure episodic memory. In conclusion, our study highlighted the ability of episodic-memory tests to predict disease conversion and thus detect early-stage AD.

Date of Award	2019
Original language	English