•  
  •  
 

Abstract

The objective of this study was to examine the association of self-reported physical activity (PA) with body composition in 290 college students (49% male, 60% White) 18-25 years of age. Outcome measures included: self-reported PA levels calculated in MET-hrs·wk-1 from the International Physical Activity Questionnaire (IPAQ); body mass index (BMI; in kg·m-2); and body composition variables estimated by dual-energy X-ray absorptiometry (DXA). Mean activity levels of the sample were 39.8 ± 23.8 MET-hrs·wk-1. Participants were divided into quartiles of PA levels: ≥0 to <24.0, ≥24.0 to <34.0, ≥34.0 to <51.25, and ≥51.25 MET-hrs·wk-1 and body composition variables were compared by group. Chi-square analyses revealed a significant difference for gender by PA quartile [χ2 (3, N=290) = 32.42, p < 0.0001], and for gender by race by PA quartile [χ2 (9, N=290) = 37.82, p < 0.0001]. MET-hrs·wk-1 was inversely correlated with %BF (r = -0.40, p < 0.0001) but not BMI (r = 0.05, p = 0.43). When comparing body composition variables across PA quartiles, no significant differences were observed for BMI; however, subjects in the highest quartile of PA had a lower percent body fat (%BF) and fat mass (FM), and a higher lean-tissue mass (LTM) compared to subjects in the other three groups. In this cohort of young adults, participants in the highest activity group had a more fit body composition profile (e.g., lower %BF, lower FM, and higher LTM) which was not reflected in BMI and was independent of gender and race.