Interdaily variability of rest-activity patterns and risk of dementia in older adults: The ARIC Neurocognitive Study.

introductionDay-to-day variability of rest-activity patterns may have a multifaceted relationship with dementia risk, but empirical evidence is sparse.

methodsWe included 2332 dementia-free participants (79 ± 4.6 years, 58% female, 24% Black) who had ≥ 3 days of ZioXT accelerometer data from the Atherosclerosis Risk in Communities Neurocognitive Study in 2016 and 2017. Variability of physical activity (PA) volume was operationalized as median absolute deviation (MAD) of (1) daily average PA, (2) average PA during the most active 10 hours (M10), and (3) average PA during the least active 5 hours (L5). Variability of PA timing was defined as MAD of (1) the start time of the M10 interval and (2) the start time of the L5 interval. Participants were followed until incident dementia, death, last contact, or December 31, 2022. We used cause-specific Cox regression to estimate variability measures with incident dementia.

resultsAfter a median 5.2-year follow-up, 319 (13.7%) of individuals developed dementia. Higher MAD of daily average PA and M10 were associated with lower hazard of dementia (dose-response trend: DISCUSSION: In this large older adult population, greater variability of daily PA volume, particularly during wake periods, and lower variability of PA timing, a marker of circadian phase, are associated with lower dementia risk. Research on the stability of interdaily rest-activity rhythms should consider potentially opposing relationships of types of PA variability with dementia risk.