|| Checking for direct PDF access through Ovid
We investigated whether type of physical activity (PA) (sports, running, and fitness/dance) engaged in during adolescence is associated with body composition in late adolescence or early adulthood.Data were drawn from 631 participants in the Nicotine Dependence in Teens study, a prospective investigation of students ages 12–13 yr at inception. Self-report PA data were collected at baseline, in grade 7, and every 3–4 months thereafter during the 5 yr of high school (1999–2005). Anthropometric indicators (height, weight, waist circumference, triceps, and subscapular skinfold thickness) were measured at ages 12, 16, and 24 yr. On the basis of prior exploratory factor analysis, PA was categorized into one of three types (sports, running, and fitness/dance). Regression models estimated the association between number of years participating in each PA type (0–5 yr) during high school and body composition measures in later adolescence or early adulthood.In multivariable models accounting for age, sex, and parent education, more number of years participating in running during adolescence was associated with lower body mass index, waist circumference, and skinfold thickness in later adolescence and early adulthood (all P < 0.01). This association was no longer apparent in models that accounted for body composition at age 12 yr. The number of years participating in sports was positively associated with body mass index in early adulthood (P = 0.02), but fitness/dance was not statistically significantly associated with any outcome.Sustaining participation in running, but not in other PA types, during adolescence was related to lower body composition in later adolescence and adulthood. However, more research is needed to determine whether this association is attributable to a relationship between PA and body composition in early adolescence.