The association between 24-hour movement behaviours and cognitive function in adults aged 55 and above: a cross-sectional compositional data analysis of the PASOCA study

Global life expectancy has consistently increased since 1950, resulting in more people living to an older age. However, maintaining optimal cognitive health is a challenge as ageing is accompanied by natural cognitive decline, which can affect daily functioning and quality of life. Importantly, modifiable lifestyle factors can play a role in promoting healthy ageing. Among these, physical activity (PA), sedentary behaviour (SB) and sleep have gained increasing attention for their potential contributions to cognitive health. This study investigates in greater detail how these 24-hour movement behaviours relate to cognitive function in older adults. Participants were 233 healthy adults aged 55 years and older (51.1% women; mean age 68.3 ± 7.7 years). Daily time spent in light PA (LPA), moderate-to-vigorous PA (MVPA), SB and sleep was derived from 7-day wrist-worn ActiGraphy (wGT3X-BT). Cognitive function, including short-term and long-term memory (STM, LTM), executive function (EF) and processing speed, was assessed using the Cambridge Neuropsychological Test Automated Battery (CANTAB) and expressed in z-scores. Compositional multiple linear regression was used to assess the association between time use and cognitive function. Compositional isotemporal substitution examined how hypothetical time reallocations between the different movement behaviours were related to cognitive function. Even after adjusting for age, sex, educational level, social isolation and multiple testing, time use was significantly associated with short-term memory (p = 0.01) and executive function (p = 0.001). Hypothetical time reallocations of 30-min from LPA to MVPA were associated with the largest significant mean differences of 0.19 [95% confidence interval 0.05–0.32] in STM z-scores and 0.21 [0.10–0.33] in EF z-scores. Notably, reallocating time from LPA or sleep to SB was also related to better EF z-scores. Importantly, reallocating even 5 minutes away from MVPA to any other behaviour was significantly associated with poorer z-scores in STM and EF. No significant associations were observed for long-term memory and processing speed. This study underscores the importance of considering 24-hour movement behaviours in cognitive health at older age. Dedicating time to moderate-to-vigorous PA seems to be important for specific cognitive domains. Longitudinal studies are needed to further explore these relationships, with a focus on detailed assessments of the various contexts in which PA and SB occur.