Leisure-time aerobic physical activity and the risk of diabetes-related mortality: An analysis of effect modification by race-ethnicity

Abstract

To examine the relationship between aerobic moderate-to-vigorous intensity leisure-time physical activity (LTPA) and the risk of diabetes-related mortality in the NHANES III (1988-1994) while considering potential effect modification by race-ethnicity. The study sample (n = 14,006) included adults, 20-79 years of age, with Mobile Examination Center (MEC) data. An age-standardized physical activity score (PAS) was calculated from the self-reported frequency and intensity of 12 leisure-time aerobic activities. Three categories of PA were examined: inactive (PAS = 0), insufficiently active (PAS >0-<10), and active (PAS ≥10). Diabetes-related mortality was defined as death from diabetes mellitus. Cox Proportional Hazard models were used all analyses. A statistically significant reduction in risk was found for insufficiently active (HR 0.59, 95% CI 0.40-0.90) and active non-Hispanic black (NHB) (HR 0.54, 95% CI 0.34-0.88). Among active non-Hispanic white (NHW), a similar pattern of risk reduction was found, however, this relationship was borderline significance (HR 0.59, 95% CI 0.35-1.02, p = 0.06). Any volume of aerobic LTPA is beneficial in terms of reducing the risk of diabetes-related mortality. However, these benefits may differ by racial-ethnic group, with further research on health disparities in the area of PA being warranted.