Associations between community built environment and leisure-time physical activity among older adults: a systematic review and meta-analysis.

Objective: To systematically evaluate associations between the community built environment and leisure-time physical activity (LTPA) among older adults using three-level meta-analysis to account for dependent effect estimates within studies. Methods: PubMed, Embase, Web of Science, Scopus, the Cochrane Library, and SPORTDiscus were searched for cross-sectional studies. Built-environment exposures were grouped into nine domains, and LTPA outcomes into leisure-time walking (LTW), neighbourhood leisure-time walking (NLTW), and General LTPA. Odds ratios (ORs) with 95% confidence intervals (CIs) were classified as categorical contrast (CAT) or continuous unit (UNIT) effects and pooled separately using three-level random-effects models. Cluster-robust variance estimation with CR2 correction, heterogeneity and prediction intervals, subgroup analyses, meta-regression, publication-bias assessments, and sensitivity analyses were conducted. Results: Thirty-four studies contributed 344 CAT and UNIT effect estimates. In primary CAT analyses, more favourable built environments were associated with higher LTW (OR = 1.103, 95% CI 1.022-1.191) and General LTPA (OR = 1.180, 95% CI 1.089-1.278), with both associations remaining significant after CR2 correction. NLTW was not significantly associated with the overall built environment. In UNIT analyses, the positive association with NLTW lost significance after CR2 correction, while no consistent associations were observed for LTW or General LTPA. Prediction intervals for all three CAT models crossed OR = 1, indicating substantial contextual variation. Domain-specific analyses showed relatively consistent positive associations between aesthetics and LTW, parks and green spaces and General LTPA, and destination accessibility and LTW in UNIT analyses. Sensitivity analyses generally supported the LTW and General LTPA CAT findings. Conclusion: More favourable community built environments were associated with higher LTW and General LTPA among older adults, but associations varied by environmental attribute, activity type, and study context. Because the NLTW analyses were based on very few independent study samples, the NLTW findings should be regarded as hypothesis-generating rather than as established effects. Because all included studies were cross-sectional and some analyses showed substantial heterogeneity or limited study numbers, the findings should be interpreted as associations rather than causal effects. Longitudinal studies and natural experiments with refined exposure assessment are needed. Systematic review registration: https://www.crd.york.ac.uk/PROSPERO/view/CRD420261435582, Unique Identifier: CRD420261435582.