Objective: Apply more robust and additional analyses to a previous meta-analysis that reported statistically significant associations between leisure time physical activity (LTPA) and stroke.
Methods: A reanalysis of a previous meta-analysis that included nine prospective cohort studies representing 269,594 men and women 25-84 years of age and in which the association between LTPA and incident stroke was examined. Follow-up periods ranged from 7.7 to 32.0 years. Relative risks (RR) from each study were pooled using the inverse-heterogeneity model. Heterogeneity was examined using the Q statistic, inconsistency using I 2, and small-study effects using Doi plots and the LFK index. Influence and cumulative meta-analysis were also conducted.
Results: Using low LTPA as the reference, moderate LTPA was associated with a statistically significant reduction in the risk for stroke in men (RR = 0.79, 95% CI = 0.65 to 0.95) and a trend in women (RR = 0.88, 95% CI = 0.78 to 1.0). High LTPA was associated with a statistically significant reduction in the risk for stroke in both men (RR = 0.72, 95% CI = 0.60 to 0.86) and women (RR = 0.78, 95% CI = 0.66 to 0.92). No statistically significant heterogeneity was observed and inconsistency was low. However, potential small-study effects were observed. With each study deleted once, results remained statistically significant. Cumulative meta-analysis demonstrated stability in results since at least 2005.
Conclusions: Leisure time physical activity is associated with a reduced risk of stroke in both men and women. However, the small-study effects observed suggest the possibility that results may be exaggerated.