Objective
Physical activity (PA) is beneficial in reductions of all-cause mortality and dementia. However, whether Alzheimer's disease (AD) risk is modified by PA remains disputable. This meta-analysis aims to disclose the underlying relationship between PA and incident AD.
Methods
Pubmed, Embase, Cochrane Library, and Web of Science were retrieved from inception to June 2023. Random-effects models were employed to derive the effect size, represented by hazard ratio (HR) and 95% confidence interval (CI).
Results
Twenty-nine prospective cohort studies involving 2068,519 participants were included. The pooled estimate showed a favorable effect of PA on AD risk decline (HR 0.72, 95% CI 0.65-0.80). This association remained robust after adjusting for maximum confounders (HR 0.85, 95% CI 0.79-0.91). Subgroup analysis of PA intensity demonstrated an inverse dose-response relationship between PA and AD, effect sizes of which were significant in moderate (HR 0.85, 95% CI 0.80-0.93) and high PA (HR 0.56, 95% CI 0.45-0.68), but not in low PA (HR 0.94, 95% CI 0.77-1.15). Regardless of all participants or the mid-life cohort, the protection of PA against AD appeared to be valid in shorter follow-up (<15 years) rather than longer follow-up (≥15 years). In addition to follow-up, the robustness of the estimates persisted in supplementary meta-analyses, meta-regression analyses, and sensitivity analyses.
Conclusion
PA intervention reduces the incidence of AD, but merely in moderate to vigorous PA with follow-up of less than 15 years, thus conditionally recommending the popularization of PA as a modifiable lifestyle factor to prevent AD.