Physical activity and body weight: associations over ten years in the CARDIA study

BACKGROUND: Hypothesis 1 - sustained changes in physical activity are inver sely related to changes in body weight. Hypothesis 2-those who attenuate we ight gain because of a temporary increase in physical activity (PA) may mai ntain a lower body weight over time. METHODS: Data were from the Coronary Artery Risk Development in Young Adult s (CARDIA) Study, a cohort of black and white men and women, aged 18-30 y a t baseline, who attended up to five examinations over a 10 y period (n = 51 15 at baseline). Longitudinal associations between physical activity and bo dy weight changes were assessed, adjusting for secular trend, age, clinic s ite, education, smoking, alcohol intake, parity, percentage energy intake f rom fat, and changes in these variables over time. For hypothesis 1, concur rent associations of physical activity and body weight changes were examine d. For hypothesis 2, we explored whether weight gain attenuation associated with increased PA during the initial 2-3 y of follow-up was sustained over 5 y. The study 2 analyses were conducted with three separate 5 y intervals : baseline to year 5 (n = 3641), years 2-7 (n = 3160), and years 5 - 10 (n = 2617). RESULTS: Hypothesis 1 - change in physical activity was inversely associate d with change in body weight within all four race and sex sub-groups (P < 0 .005). The predicted weight change associated with change in physical activ ity was four to five times larger in participants who were overweight compa red with those who were not were overweight at baseline. Hypothesis 2 - an increase in physical activity during 2-3 y of follow-up was associated with an attenuation of weight gain that was sustained through 5 y of follow-up whether or not the physical activity increase was maintained during the lat er years. This finding persisted whether the starting point for the 5 y fol low-up was year 2, year 5 or baseline (women only). Comparing participants who increased physical activity with those who decreased physical activity in the first 2-3 y of follow-up (eg by at least 2 h per week of stationary cycling for at least 6 months per year), the mean 5 y weight gain attenuati on ranged from 0.8 to 2.8 kg. CONCLUSIONS: The results of these analyses support the need for public heal th messages for promoting increased physical activity for weight maintenanc e and attenuation of age-related weight gain, especially for higher weight sub-populations.