Background: Associations of physical activity with all-cause mortality seem to be quite strong, but little is known about potential effect modifiers as sex, race/ethnicity, age, and obesity.
Methods: Data of the Third National Health and Nutrition Examination Survey (NHANES III), conducted 1988-1994 with mortality follow-up until 2006, were used to compare mortality risk between different levels of leisure-time physical activity (LTPA) and occupational physical activity (OPA). Cox proportional hazards models were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs).
Results: LTPA (n = 15,307) was inversely associated with all-cause mortality (HR 0.75, 95% CI 0.64-0.88 for regular vs. no LTPA). There was a statistically significant interaction with age (P = .03), with participants over 60 years of age benefitting more from regular or irregular LTPA. OPA was positively associated with all-cause mortality (HR 1.25, 95% CI 0.85-1.84 for high vs. low OPA), particularly among Mexican-Americans (HR 2.28, 95% CI 1.23-4.22); statistically significant interactions were observed for obesity and gender.
Conclusions: LTPA clearly predicts all-cause mortality. However, associations between OPA and all-cause mortality are unclear and need further research with special regard to ethnic differences.