In the Bunkeflo project, one elementary school increased duration of school physical activity (PA) to 200 minutes/week while 3 control schools continued with 60 minutes/week throughout the nine elementary school years. We then registered fractures in 3534 children, and evaluated the duration of PA, bone mass and muscle strength in a subsample (n=140) during the intervention and 3 years after. The PA intervention was associated with higher duration of PA both during and 3 years after the intervention. With each year of intervention, the fracture incidence rate ratio (IRR) declined in the intervention children so that it was 0.48 (95% CI 0.25, 0.91) the eighth year of intervention. These findings were accompanied by beneficial gain in musculoskeletal traits in both intervention girls and boys (p<0.05). Nine years of daily school-based PA is associated with a progressive reduction in fracture risk, accompanied by beneficial musculoskeletal gains and a more physically active lifestyle.