Physical education and leisure-time sport reduce overweight and obesity: a number needed to treat analysis.

Background: School-based physical education (PE) and organised leisure-time sports participation (LTSP) represent important physical activity opportunities for children. We examined the preventive effect of increased PE as well as LTSP on overweight and obesity (OW/OB) in school children.

Methods: Longitudinal data from children attending 10 primary schools in the Danish municipality of Svendborg, comprising 6 intensive PE (270 min/week) and 4 control (90 min/week) schools were assessed. Age- and sex-specific cut-offs for body mass index (BMI) determined OW/OB status. Associations between OW/OB status and school type (intensive PE or control) or LTSP were investigated using mixed, multilevel logistic regression models. Significant parameter estimates were converted into number needed to treat statistics (NNT).

Results: In total, 1009 children (53.3% female; mean age 8.4 ± 1.4 years) were included in the analysis, with 892 children (52% female) being normal weight (NW) at baseline. Eighteen (NNT = 17.1; 95% CI [11.0, 226.1]) children attending an intensive PE school for 2 years, resulted in one fewer case of OW/OB compared with attendance at a normal PE school. For NW children, prevention of one case of OW/OB requires 36 (NNT = 35.8; 95% CI [25.1, 596.3]) children to participate in intensive PE for 2 years in comparison with normal PE. LTSP over 2 years may prevent OW/OB if 15 children participate in one LTSP session/week, 9 in two LTSP sessions/week and 8 in three LTSP sessions/week; for normal weight children, 25 children had to participate in one LTSP session/week, 16 in two LTSP sessions/week and 14 in three LTSP sessions/week.

Conclusion: We provide the first NNT estimates of school-based PE and LTSP to prevent the onset of OW/OB. PE, and separately, LTSP seem to have both a protective and a treatment effect against OW/OB in children.