The Relationships between physical activity, sedentary behaviour, sleep, and dementia: A systematic review and meta-analysis of cohort studies

Objective: This study aimed to summarize the observational evidence from prospective cohort studies examining the associations of regular physical activity, sedentary behaviour, and sleep duration with incident dementia among community-dwelling adults aged 35 years and older. Methods: Systematic literature searches (1946 to August 2025) of CINAHL, EMBASE, MEDLINE, PSYCINFO, and SPORTDISCUS were performed. Eligible studies included community-dwelling adults aged 35 + years with at least one year of follow-up and valid measures of movement behaviours and dementia outcomes. Studies were excluded if they included participants with baseline dementia, lacked risk estimates for all-cause dementia. Grey literature was excluded. Random effects meta-analysis generated pooled risk ratio (RR) and 95% confidence intervals (CI). Primary exposures were defined using national thresholds for physical activity, sedentary time, and sleep duration. Subgroup analyses were performed by age and follow-up duration. Results: Forty-nine studies with physical activity (n = 2,855,529), 17 studies on sleep duration (n = 1,344,170), and three studies on sedentary duration (n = 295,809) were included. Regular physical activity significantly reduced the risk of incident dementia (pooled RR = 0.75, 95% CI = 0.68 to 0.82), though heterogeneity was substantial and partially explained by subgroup analyses. Prolonged sedentary behaviour (8 + hours/day sitting) increased dementia risk (RR = 1.27, 95% CI = 1.17 to 1.39) with low heterogeneity. Moreover, both short ( 8 hours; RR = 1.28, 95% CI = 1.15 to 1.43) sleep were linked to higher dementia risk compared with 7–8 hours. Heterogeneity was moderate to substantial. Conclusion: Regular physical activity, less sedentary time and appropriate nightly sleep (7–8 h) may be associated with reduced risk of dementia and are potentially modifiable factors in the prevention or delay of dementia. Future studies with middle-aged adults and longer-term follow-up including changes in movement behaviours over time are needed to better understand the relationship between physical activity, sedentary behaviour, and sleep for dementia risk. PROSPERO Registration: CRD42021272054.