Exercise builds men's hip bones but leaves the spine behind

Bone density research has historically concentrated on women, because of the sharp bone loss that follows menopause, yet men also lose bone with age and suffer fractures with serious consequences. Exercise is the most accessible candidate for protecting bone, since bones adapt to the forces placed on them. The question this review set out to answer was whether randomised trials in men actually show exercise improving bone mineral density, and at which sites: the femoral neck of the hip, the lumbar spine and the lower limb, the regions that matter most for fracture.

The authors conducted a systematic review and meta-analysis of randomised controlled trials of exercise in men, deliberately using more robust pooling methods than earlier analyses. Eligible trials lasted more than 24 weeks, were published in English up to May 2020, and were found through three electronic databases, cross-referencing and expert review. The primary outcomes were changes in bone mineral density at the femoral neck, lumbar spine and lower limb. Standardised effect sizes were calculated from each trial and combined using the inverse heterogeneity model.

Exercise produced a statistically significant benefit on femoral neck bone mineral density, with a pooled effect of 0.21 standard units. At the lumbar spine, by contrast, the pooled effect of 0.10 was not statistically significant, so no benefit could be claimed there. On the strength of the hip finding, the authors recommend exercises that generate ground reaction or joint reaction forces, meaning impact work or movements where muscles pull hard on bone, for improving or maintaining femoral neck density. They judge that additional well-designed trials are unlikely to overturn the hip result.

This updates earlier analyses of exercise and bone in men with stricter methods, and firms up the hip conclusion to the point the authors consider it close to settled. The spine finding complicates the simple idea that exercise strengthens the skeleton uniformly: bone adapts locally to the loads it receives, and the interventions trialled so far may never have loaded the lumbar spine adequately. The open question the authors flag is whether exercise that directly loads the spine could succeed where the tested programmes did not, which would need new trials to answer.

For men who train, this is solid evidence that exercise involving impact or forceful loading supports bone density at the hip, one of the most fracture-prone sites in later life. It is also a useful corrective to the assumption that any training protects every bone equally: the spine showed no significant benefit from the exercise tested here. That is not proof exercise cannot help the spine, only that the right kind of trial has not been run. Bone, like muscle, appears to adapt specifically to where the load goes.