[P2–434]: INTELLECTUAL FUNCTION VERSUS YEARS OF EDUCATION AS COGNITIVE RESERVE PROXIES: INTERACTION WITH ALZHEIMER's DISEASE NEUROPATHOLOGY AND IMPACT ON EPISODIC MEMORY AND EXECUTIVE FUNCTION

Michael Malek‐Ahmadi,Sophie Lu,Yan Yan Chan,Sylvia E. Perez,Kewei Chen,Elliott J. Mufson

DOI: https://doi.org/10.1016/j.jalz.2017.06.1090

2017-01-01

Alzheimer s & Dementia

Abstract:Years of education are the most common proxy for measuring cognitive reserve (CR) when assessing the relationship between Alzheimer's disease (AD) pathology and cognition. Alternatively, intellectual function may be a better measure of CR. The present study determined whether tests of vocabulary and reading ability were better proxies of CR than years of education in terms of their impact on episodic memory and executive function. Study population consisted of 249 older deceased and autopsied persons who died with a premortem clinical diagnosis of No Cognitive Impairment (NCI; n = 123), Mild Cognitive Impairment (MCI; n = 79), and AD (n = 47). CERAD neuropathological criteria determined pathology severity. Regression analyses using a CR by CERAD interaction term determined whether vocabulary scores, reading test scores, or years of education were associated with cognition. All models adjusted for clinical diagnosis, age at death, gender, APOE e4 carrier status, and Braak stage. For episodic memory, the CR by CERAD interaction term was statistically significant for the reading test and vocabulary, but not for education: Education (Beta = -0.009, SE = 0.01, p = 0.38); Reading Test (Beta = -0.02, SE = 0.01, p = 0.04); Vocabulary (Beta = -0.05, SE = 0.01, p<0.001). For executive function, only the Vocabulary test by CERAD interaction showed a significant effect: Education (Beta = -0.007, SE = 0.01, p = 0.55); Reading Test (Beta = -0.02, SE = 0.01, p = 0.15); Vocabulary (Beta = -0.04, SE = 0.01, p = 0.008). For Braak stage, only the interaction for Braak by Vocabulary within episodic memory was significant (Beta = 0.04, SE = 0.01, p <0.001). Among the four CERAD classifications, there was a significant association between vocabulary performance and cognition. However, respective regression slopes for each classification declined with increased plaque pathology. CR by clinical diagnosis interactions did not reveal a significant association with the cognitive outcomes.