A high-impact exercise and a lifestyle intervention were implemented over a 9-week period; changes in bone and body composition were compared to controls. Sixty-one children volunteered from three randomly selected schools. Each school was randomly assigned to either a structured exercise (STEX) intervention, a life-style intervention (PASS) or control (CONT). Bone mineral content (BMC) and density (BMD) of total body, femoral neck and lumbar spine were measured as well as fat and lean mass at baseline and post-intervention by dual-energy X-ray absorptiometry. The STEX intervention resulted in an additional mean increase in total body BMC of 63.3 g (p = 0.019) and an additional increase of 0.011 g·cm-2 (p = 0.018) for BMD over increases observed by controls. Bone mineral increases observed for the PASS intervention were not significant compared to the control group (p > 0.05). Neither intervention produced significant increases in bone mineral at femoral neck or lumbar spine sites (p > 0.05) compared with the controls. No significant changes were found in fat mass index (p > 0.05), lean mass index (p > 0.05) or percent body fat (p = 0.09) in any groups. Structured impact exercise promoted significant and clinically relevant increases in bone measures, without significant changes to body composition. A larger, definitive randomised trial is needed to confirm the present results.