Independent and Joint Associations of Physical Activity and Fitness on Stroke in Men

Abstract

Introduction: Recent studies have demonstrated that physical activity (PA) and cardiorespiratory fitness (CRF) are independent predictors of stroke in men. The combined associations of these 2 factors are not well established. Objective: To investigate the independent and joint associations of PA and CRF with fatal, nonfatal, and total stroke in a group of men. Methods: The current analyses included 45 689 men aged 18 to 100 years who completed baseline sessions between 1970 and 2001. All participants had no known myocardial infarction, stroke, or cancer. Physical activity was measured by questionnaire, and CRF was assessed from a maximal treadmill exercise test. The National Death Index for fatal stroke and mail-back surveys for nonfatal stroke were used to ascertain cases. Cox regression analyses were used to estimate the risk of stroke outcomes. Results: There were 619 cases over 800 582 person-years of observation. Significant inverse associations were observed between CRF and fatal, nonfatal, and total strokes after adjustment for age and examination year (P for trend < 0.05 for each). No associations were found between PA and any of the 3 outcomes after adjusting for other covariates and CRF. Joint associations of 9 PA fitness groups showed less risk for total stroke in the moderate and high fitness categories. Conclusion: These findings suggest that CRF is an independent predictor of incident stroke in asymptomatic men.