Abstract
Objectives: To examine associations between adiposity and adiposity change (loss, stable, gain) and subsequent longitudinal cognitive performance in African Americans in mid and late life. Design: Cohort study using linear mixed models. Setting: Genetic Epidemiology Network of Arteriopathy. Participants: African-American sibships with hypertension in Jackson, Mississippi (N = 1,108). Measurements: Waist circumference and body mass index (BMI) were measured at two examinations 5 years apart. Stable adiposity was defined as values within 5% of the first measure. A composite cognitive Z-score was derived from individual cognitive test Z-scores at two study visits 6 years apart. Results: Larger waist circumference was associated with greater rate of cognitive decline during follow-up (beta = −0.0009 per year, P =.001); BMI, change in waist circumference, and change in BMI were not associated with rate of decline. Loss of adiposity in midlife was associated with higher cognitive Z-scores in middle-aged individuals, and loss of adiposity in late life was associated with lower Z-scores in older adults (P =.01 for interaction between waist circumference and age; P =.04 for interaction between BMI and age). Simultaneous inclusion of waist circumference and BMI in the cross-sectional model suggested an association between larger waist circumference and poorer cognitive performance (beta = −0.009, P =.006) and between higher BMI and better cognitive performance (beta = 0.014, P =.06). Conclusion: The results suggested a differential pattern of the relationship between adiposity and cognition according to age (mid- or late life) and regional distribution of adiposity.
Original language | English (US) |
---|---|
Pages (from-to) | 1282-1288 |
Number of pages | 7 |
Journal | Journal of the American Geriatrics Society |
Volume | 65 |
Issue number | 6 |
DOIs | |
State | Published - Jun 2017 |
Keywords
- body mass index
- cognitive decline
- ethnicity
- longitudinal
- waist circumference
ASJC Scopus subject areas
- Geriatrics and Gerontology