Physical Activity and the Risk of Hemorrhagic Stroke: A Population-Based Longitudinal Follow-Up Study in Taiwan

Background: Data on the relationship between physical activity (PA) and hemorrhagic stroke (HS) are limited in Asian populations. This population-based longitudinal follow-up study therefore investigates whether PA is associated with a reduced risk of HS in Taiwan. Methods: A total of 58,857 subjects who had participated in the Keelung Community-based Integrated Screening Program between 2005 and 2012 were enrolled. Information about their PA, obtained using questionnaires, was used to categorize them into three groups according to their average weekly time engaged in it: (1) no PA, (2) low PA (<90 min weekly), and (3) high PA (90 min per week or more). Cox proportional hazard regression was used to evaluate the effect of PA on HS. Stratified analysis by sex and comorbidities (diabetes mellitus, hypertension, and hyperlipidemia) were conducted to evaluate their impact on the relationship between PA and HS. Results: Compared to the no-PA group, the adjusted hazard ratio of HS for the low-PA group was 0.74 (95% CI, 0.57–0.96, p = 0.0219), and for the high-PA group, 0.72 (95% CI, 0.58–0.90, p = 0.004). The stratified analyses showed that, for the non-comorbidity strata, the beneficial effect of PA on reducing HS risk became stronger as PA increased. However, in the diabetes and hypertension strata, high PA did not appear to have any greater protective effect than low PA. Conclusions: Our findings suggested that even <90 min of PA per week might be beneficial to reduce HS risk. Such a low level of PA is likely to be more achievable and easier to maintain for the general population. Additionally, personalized recommendations based on pre-existing comorbidities may help optimize the beneficial effects of PA on HS prevention.