0858 Impact Of Actigraphic Sleep Measures On Ambulatory Cognitive Performance In A Community-Based Sample Of Older Adults

Abstract

Abstract Introduction Few longitudinal studies link objectively assessed sleep and cognition, especially day to day differences in sleep as they relate to daily cognitive performance in ecologically-valid, natural environments. We examine the associations of sleep (actigraphy) with ambulatory ecological momentary assessments (EMA) of cognitive performance. Methods Analyses involved 225 participants enrolled in The Einstein Aging Study, a community based longitudinal cohort of older adults free of dementia at enrollment (Mage=77.27 years; 33% males; 47% Caucasian, 39% African American, 13% Hispanic). We examined between-person associations between actigraphic sleep duration and wake after sleep onset (WASO) with mean and variability across the day in cognitive performance. Cognitive performance was assessed via validated, smartphone-based EMA over a mean of 18 days. Six assessments/day included Symbol Match (processing speed), Color Dot (working memory), and Color Shape (memory binding). Models controlled for age, gender, ethnicity, education (years), clinically assessed mild cognitive impairment, and learning effects. Actigraphy data was collected throughout the study period. Results Sleep duration had a significant effect on within-person variability on ambulatory cognition: Color Dot, Symbol Match, and Color Shape (all p’s <0.001). Older adults with longer mean nightly sleep duration exhibited more stable cognitive performance over time versus those with shorter sleep duration; sleep duration did not predict mean levels of daily cognitive performance. Person-level means of WASO (0.99±0.45 hours/night) predicted mean levels on cognitive tests, independent of sleep duration. Older adults with less WASO/night exhibited better cognitive performance. One half hour less nightly WASO predicted 175ms shorter Symbol Match response time (p=0.004), 1.5% lower Color Dot error proportion (p=0.048), 0.07 points higher Color Shape accuracy (p<0.001). Older adults with less WASO/night also had less within-person variations in Color Dot (p<0.001), Symbol Match (p<0.001) and Color Shape performance (p=0.01). Conclusion Ambulatory cognitive performance assessed using EMA was related to actigraphic sleep. Poor sleep health may be a target for prevention of early cognitive changes that may precede onset of cognitive impairment and AD. Support Research was supported by the National Institute on Aging (NIA) award numbers P01AG003949 and R01AG056538.