TELOMERE ATTRITION AND AGE-RELATED COGNITIVE DECLINE: DISPARITIES BY POVERTY STATUS AND RACE

Abstract

Telomere length (TL) is a marker of biological aging. However, evidence for relations between TL attrition and age-related cognitive decline is equivocal. Further, it is unknown whether historically disadvantaged sociodemographic groups (e.g., African Americans (AAs) and/or those living in poverty), are most vulnerable to cognitive decline associated with TL attrition. Therefore, the present study examined prospective interactive relations of TL change (?TL), poverty status, and race with age-related cognitive decline across multiple domains over five years. Participants were 323 urban-dwelling adults (baseline mean age = 48; 50.8% female; 52.0% AA; 50.2% below poverty) from the Healthy Aging in Neighborhoods of Diversity across the Life Span study. Cognitive tests included Digit Span Forward and Backward; California Verbal Learning Test-II (CVLT) total learning, short-delay free recall, and long-delay free recall subtests; Benton Visual Retention Test; Trail Making Test Parts A and B; and Animal Naming. TL was assayed from peripheral blood mononuclear cells using quantitative polymerase chain reaction. Linear mixed-effects models examined prospective interactions of ?TL, poverty status, race, and age (indexing time) with cognitive decline, adjusted for sex, high school-or-greater education, and baseline TL. Results revealed a significant four-way interaction of ?TL�Poverty Status�Race�Age (b=-0.16, p=.023) such that greater TL attrition was associated with steeper decline in Digit Span Backward performance among Whites living in poverty and AAs living above poverty. Second, a significant three-way interaction of ?TL�Race�Age (b=-0.15, p=.005) showed that greater TL attrition was associated with steeper decline in CVLT long-delay free recall performance among Whites. Third, a significant two-way interaction of ?TL�Age (b = 0.09, p = .022) showed that, in the overall sample, greater TL attrition was associated with steeper decline in Animal Naming. Finally, sensitivity analyses showed that adjustment for hsCRP and the presence of cardiometabolic diseases (hypertension, diabetes mellitus, coronary artery disease, heart failure, myocardial infarction) did not cause attenuation of these effects. Findings suggest that associations between TL attrition and cognitive decline may be specific to working memory, delayed verbal memory, and verbal fluency, and vary by race and/or poverty status. Clarification of underlying mechanisms may ultimately support prevention and treatment of premature cognitive aging.