Association of Physical Activity, Fitness, and Race: NHANES 1999-2004

CEASER, T. G., E. C. FITZHUGH, D. L. THOMPSON, and D. R. BASSETT JR. Association of Physical Activity, Fitness, and Race: NHANES 1999-2004. Med. Sci. Sports Exerc., Vol. 45, No. 2, pp. 286-293, 2013. Purpose: Regular physical activity (PA) can be used to improve cardiorespiratory fitness (CRF). Previous research has shown differences in CRF ((V) over dotO(2max)) among racial groups, but it is unclear how much of these differences can be explained by PA. Thus, we sought to examine the association between PA and CRF in different racial groups. Methods: As a part of the National Health and Nutrition Examination Survey (1999-2004), 3115 adults (18-49 yr) completed a submaximal graded treadmill exercise test to estimate (V) over dotO(2max). Independent variables were demographics (race, education, sex, partner status, and waist circumference), behavioral measures (smoking and alcohol consumption), self-reported PA from three domains (leisure-time, domestic, and transportational PA (MET.min.wk(-1))), and the proportion of PA at a vigorous intensity (VMET). CRF was the dependent variable. Multiple linear regression was performed using SUDAAN statistical software. Results: Results indicated that (V) over dotO(2max) was significantly higher for Mexican Americans (40.9 +/- 0.5 mL.kg(-1).min(-1)) and non-Hispanic Whites (40.2 +/- 0.3 mL.kg(-1).min(-1)) compared with non-Hispanic Blacks (37.9 +/- 0.6 mL.kg(-1).min(-1)) (P = 0.01). Demographics including race explained 18.5% of the variance in (V) over dotO(2max), with race being significant (P G 0.01) in the model. When PA was added to the model, the explained variance in (V) over dotO(2max) increased to 19.3% (P = 0.001). VMET was more predictive of (V) over dotO(2max) than total PA, and the model including VMET explained 20.4% of the variance in (V) over dotO(2max). Race remained a significant, independent predictor of (V) over dotO(2max) after VMET and overall PA were added to the model. Conclusion: Race, PA, and exercise intensity are important factors in explaining differences in CRF. After accounting for demographics, PA, and VMET, a large proportion of the variance in CRF remains unexplained. Thus, other factors should also be considered when examining racial/ethnic differences in CRF.