Joint associations of physical activity and sleep duration with cognitive ageing: longitudinal analysis of an English cohort study.

BACKGROUND

Physical activity and sleep duration are key factors associated with cognitive function and dementia risk. How physical activity and sleep interact to influence cognitive ageing is not well explored. We aimed to examine the associations of combinations of physical activity and sleep duration with 10-year cognitive trajectories.

METHODS

In this longitudinal study, we analysed data from the English Longitudinal Study of Ageing collected between Jan 1, 2008, and July 31, 2019, with follow-up interviews every 2 years. Participants were cognitively healthy adults aged at least 50 years at baseline. Participants were asked about physical activity and nightly sleep duration at baseline. At each interview, episodic memory was assessed using immediate and delayed recall tasks and verbal fluency using an animal naming task; scores were standardised and averaged to produce a composite cognitive score. We used linear mixed models to examine independent and joint associations of physical activity (lower physical activity or higher physical activity, based on a score taking into account frequency and intensity of physical activity) and sleep duration (short [<6 h], optimal [6-8 h], or long [>8 h]) with cognitive performance at baseline, after 10 years of follow-up, and the rate of cognitive decline.

FINDINGS

We included 8958 respondents aged 50-95 years at baseline (median follow-up 10 years [IQR 2-10]). Lower physical activity and suboptimal sleep were independently associated with worse cognitive performance; short sleep was also associated with faster cognitive decline. At baseline, participants with higher physical activity and optimal sleep had higher cognitive scores than all combinations of lower physical activity and sleep categories (eg, difference between those with higher physical activity and optimal sleep vs those with lower physical activity and short sleep at baseline age 50 years was 0·14 SDs [95% CI 0·05-0·24]). We found no difference in baseline cognitive performance between sleep categories within the higher physical activity category. Those with higher physical activity and short sleep had faster rates of cognitive decline than those with higher physical activity and optimal sleep, such that their scores at 10 years were commensurate with those who reported low physical activity, regardless of sleep duration (eg, difference in cognitive performance after 10 years of follow-up between those with higher physical and optimal sleep and those with lower physical activity and short sleep was 0·20 SDs [0·08-0·33]; difference between those with higher physical activity and optimal sleep and those with lower physical activity and short sleep was 0·22 SDs [0·11-0·34]).

INTERPRETATION

The baseline cognitive benefit associated with more frequent, higher intensity physical activity was insufficient to ameliorate the more rapid cognitive decline associated with short sleep. Physical activity interventions should also consider sleep habits to maximise benefis of physical activity for long-term cognitive health.

FUNDING

UK Economic and Social Research Council.