Associations Between Sedentary Time, Physical Activity, and Dual-Energy X-ray Absorptiometry Measures of Total Body, Android, and Gynoid Fat Mass in Children.

Document Type


Publication Date



Absorptiometry, Photon, Accelerometry, Adipose Tissue, Body Composition, Body Fat Distribution, Child, Exercise, Female, Humans, Male, Sedentary Lifestyle, Sex Factors, Time Factors


Negative health outcomes are associated with excess body fat, low levels of physical activity (PA), and high sedentary time (ST). Relationships between PA, ST, and body fat distribution, including android and gynoid fat, assessed using dual-energy X-ray absorptiometry (DXA) have not been measured in children. The purpose of this study was to test associations between levels of activity and body composition in children and to evaluate if levels of activity predict body composition by DXA and by body mass index percentile in a similar manner. PA, ST, and body composition from 87 children (8.8-11.8 yr, grades 3-5, 44 boys) were used to test the association among study variables. Accelerometers measured PA and ST. Body composition measured by DXA included bone mineral content (BMC) and fat and lean mass of the total body (TB, less head), android, and gynoid regions. ST (range: 409-685 min/wk) was positively associated with TB percent fat (0.03, 95% confidence interval [CI]: 0.00-0.05) and android fat mass (1.5 g, 95% CI: 0.4-3.0), and inversely associated with the lean mass of the TB (-10.7 g, 95% CI: -20.8 to -0.63) and gynoid regions (-2.2 g, 95% CI: -4.3 to -0.2), and with BMC (-0.43 g, 95% CI: 0.77-0.09). Moderate-to-vigorous PA was associated with lower TB (-53 g, 95% CI: -87 to -18), android (-5 g, 95% CI: -8 to -2]), and gynoid fat (-6 g, 95% CI: -11 to -0.5). Vigorous activity results were similar. Light PA was associated with increased TB (17.1 g, 95% CI: 3.0-31.3) and gynoid lean mass (3.9 g, 95% CI: 1.0-6.8) and BMC (0.59 g, 95% CI: 0.10-1.07). In boys, there were significant associations between activity and DXA percent body fat measures that were not found with the body mass index percentile. Objective measures of PA were inversely associated with TB, android, and gynoid fat, whereas ST was directly associated with TB percent fat and, in particular, android fat. Activity levels predict body composition measures by DXA and, in particular, android fat distribution.

Publication Title

Journal of Clinical Densitometry





First Page


Last Page




DOI of Published Version