ObjectivesThe objective of this study was to examine if, in the general population, physically active adults have less body fat after taking body mass index (BMI) into account.DesignA cross-sectional analysis of participants recruited into UK Biobank in 2006–2010.SettingUK Biobank assessment centres throughout the UK.Participants119 230 men and 140 578 women aged 40–69 years, with complete physical activity information, and without a self-reported long-term illness, disability or infirmity.ExposuresPhysical activity measured as excess metabolic equivalent (MET)-hours per week, estimated from a combination of walking, and moderate and vigorous physical activity. BMI from measured height and weight.Main outcome measureBody fat percentage estimated from bioimpedance.ResultsBMI and body fat percentage were highly correlated (r=0.85 in women; r=0.79 in men), and both were inversely associated with physical activity. Compared with <5 excess MET-hours/week at baseline, ≥100 excess MET-hours/week were associated with a 1.1 kg/m2 lower BMI (27.1 vs 28.2 kg/m2) and 2.8 percentage points lower body fat (23.4% vs 26.3%) in men, and 2.2 kg/m2 lower BMI (25.6 vs 27.7 kg/m2) and 4.0 percentage points lower body fat (33.9% vs 37.9%) in women. For a given BMI, greater physical activity was associated with lower average body fat percentage (for a BMI of 22.5–24.99 kg/m2: 2.0 (95% CI 1.8 to 2.2), percentage points lower body fat in men and 1.8 (95% CI 1.6 to 2.0) percentage points lower body fat in women, comparing ≥100 excess MET-hours per week with <5 excess MET-hours/week).ConclusionsIn this sample of middle-aged adults, drawn from the general population, physical activity was inversely associated with BMI and body fat percentage. For people with the same BMI, those who were more active had a lower body fat percentage.