OBJECTIVESTo determine whether a large-scale physical activity intervention could affect body composition in primary school students in Beijing, China.

METHODSThe study design was one-year cluster randomized controlled trial of physical activity intervention (20 min of daily exercise in the classroom) with an additional year of follow-up among 4 700 students aged 8-11 years at baseline.

RESULTSAfter the one-year intervention, BMI increased by 0.56 kg/m(2) (SD 1.15) in the intervention group and by 0.72 kg/m(2) (SD 1.20) in the control group, with a mean difference of -0.15 kg/m(2) (95% CI: -0.28 to -0.02). BMI z score decreased by -0.05 (SD 0.44) in the intervention group, but increased by 0.01 (SD 0.46) in the control group, with a mean difference of -0.07 (-0.13 to -0.01). After another year of follow up, compared to the control group, children in the intervention group had significantly lower BMI (-0.13, -0.25 to -0.01), BMI z score (-0.05, -0.10 to -0.01), fat mass (-0.27 kg, -0.53 to -0.02) and percent body fat (-0.53, -1.00 to -0.05). The intervention had a more pronounced effect on weight, height, BMI, BMI z score, and body composition among obese children than among normal weight or overweight children. Compared to the control group, the intervention group had a significantly higher percentage of children who maintained or reduced their BMI z score at year 1 (P=0.008) and year 2 (P=0.04).

CONCLUSIONSThese findings suggest that 20 min of daily moderate to vigorous physical activity during the school year is a feasible and effective way to prevent excessive gain of body weight, BMI, and body fatness in primary school students.