Background: The purpose of this study was to examine the extent to which physical activity (PA) is related to obesity in older adults when accounting for race/ethnicity.
Methods: Cross-sectional data were collected on 214 older adults (72.3 +/- 8.9 y; body mass index [BMI] 28.9 +/- 6.0; 151 females; 96 non-White). Measures of body height and mass were collected; BMI was calculated. PA was assessed via an electronic pedometer worn for seven consecutive days.
Results: "White" subjects accumulated 5036 +/- 286 steps/d. "Non-White" subjects accumulated significantly fewer steps/d (3671 +/- 253 steps/d; z = -3.45, P = 0.001). Race/ethnicity, income, age, gender, and steps/d accounted for 27.4% (P < 0.001) of the variance in BMI, with steps/d accounting for 21.2% (P < 0.001). The most influential factor in this model was PA level (beta = -0.510), followed by age (beta = -0.220), and finally gender being the least influential, but still a significant factor (beta = 0.168).
Conclusion: Although race/ethnicity and income have been associated with obesity levels, this study shows that older adults who accumulate more ambulatory activity tend to have healthier levels of BMI irrespective of race/ethnicity or income.