To examine the seasonal difference in disability in older adults.

We analyzed data from 1121 participants from the Rush Memory and Aging Project, a community-based cohort study of older persons. Self-reported disability was assessed via three ordinal scales- the Katz scale for activities of daily living (ADL), the Lawton and Brody scale for instrumental activities of daily living (IADL), and the Rosow-Breslau scale for mobility disability, each of which quantifies the number of standard tasks performed independently. Physical activity and circadian rhythmicity were measured using multi-day actigraphy recordings, grip strength was quantified by dynamometry, and cognition was measured with a battery of 19 neuropsychological tests. We modeled these outcomes as a function of the cosine of the date of assessment using ordinal logistic (disability) or linear (activity, grip, cognition, circadian rhythms) regression models.

There was a robust association between date of assessment and ADL (p=0.036), IADL (p=0.0058), and mobility (p=0.0063) disability. Those measured in the spring peak had a higher odds of having more severe disability compared to participants assessed in fall (Odds Ratio (OR) 1.89 [95%CI 1.20-3.29] for ADL, OR 1.65 [95%CI 1.29-2.36] for IADL, and OR 1.61 [95%CI 1.25-2.23] for mobility disability. Moreover, seasonal rhythms of grip strength (p=0.0072), cognition (p=0.032), physical activity (p=0.00081), and circadian (p=0.016) rhythmicity were also seen. When the latter phenotypes were modelled together with disability, the association of season with disability was attenuated.

Season has a clinically significant association with disability in older adults, with greater severity in spring than fall. This may in part be due to seasonal rhythmicity in motor function, cognition and circadian rhythms.