Abstract P217: Directionality of Associations Between Daily Physical Activity and Cognition Among Older Adults in the Baltimore Longitudinal Study of Aging

Abstract

Physical activity is a modifiable risk factor for dementia. Associations between physical activity and cognition are likely bidirectional, as cognitive functions are required to initiate and perform activities. Understanding the directionality of these associations has potential implications for both dementia prevention and physical activity engagement. Yet, the directionality between cognitive function and free-living physical activity patterns is unclear. This study included 237 cognitively unimpaired older adults in the BLSA with cognitive test performance and ≥ 3 valid days of wrist accelerometry measured at 2 visits over an average of 1.8 (0.9-4.0) years. Cognitive domain factor scores were derived using confirmatory factor analysis. Accelerometry data were used to calculate daily physical activity patterns: 1) total activity counts (TAC), 2) time spent active, 3) time spent non-active, 4) active-to-sedentary-transition probability (ASTP), 5) mean active bout length, 6) sedentary-to-active transition probability (SATP), 7) mean non-active bout length. Bivariate latent change score models estimated directionality of associations between changes in cognitive factor scores and changes in daily physical activity patterns. Models were adjusted for age, sex, race, education, and number of comorbidities. Participants were on average 77 (60-96) years, 48% women, and 22% Black. After covariate adjustment, one SD higher in baseline TAC was associated with less memory decline (estimate = 0.089, SE = 0.045). One SD higher in baseline time spent non-active was associated with greater declines in language (estimate = -0.068, SE = 0.035), visuospatial processing (estimate = -0.079 SE = 0.034), and global cognition (estimate = -0.058, SE = 0.024). One SD higher in baseline ASTP was associated with greater declines in memory (estimate = -0.114, SE = 0.047) and visuospatial processing (estimate = -0.074, SE = 0.034), whereas mean active bout length was associated with less decline in these domains (memory: estimate = 0.121, SE = 0.046; visuospatial processing: estimate = 0.081, SE = 0.033). There were no associations with time spent active, SATP, or mean non-active bout length and changes in any cognitive factor score. Cognitive factor scores were not associated with changes in any activity pattern. Patterns relating to amount of daily activity, time spent sedentary, and probability of transitioning from active-to-sedentary states were primarily associated with changes in memory and visuospatial processing. This suggests increasing free-living movement or decreasing sedentary behavior might be related to short-term changes in cognition among older adults. Future studies are needed to determine causality of these associations and to evaluate directionality over longer periods of time.