Effect of exercise on bone health in middle aged and older adults: hierarchical network meta-analysis of randomised trials

Abstract Objective To determine the type and dose of exercise needed to maintain bone mineral density (BMD) and reduce fracture risk in adults aged ≥40 years. Design Hierarchical bayesian network meta-analysis and multivariable dose-response network meta-analysis. Data sources PubMed, Embase, the Cochrane Library, and Web of Science from inception to January 2026. Eligibility criteria for selecting studies Randomised controlled trials comparing structured exercise interventions with non-exercise control or alternative exercise modalities in middle aged (40-60 years) and older (>60 years) adults. Main outcome measures The primary outcomes were changes in BMD (g/cm 2 ) at the lumbar spine, femoral neck, and total hip. The secondary outcome was incidence of fractures. Evidence certainty was rated using the Confidence in Network Meta-Analysis tool across six predefined domains. Results A total of 124 trials (18 429 participants) was included. For lumbar spine BMD, low certainty evidence showed that brisk walking or jogging (mean difference 0.013 g/cm 2 , 95% credible interval (CrI) 0.006 to 0.021) and low to moderate certainty evidence showed that combined aerobic-resistance exercises (0.013 g/cm 2 , 0.010 to 0.016) probably result in improvements compared with controls. For femoral neck BMD, brisk walking or jogging (0.009 g/cm 2 , 0.001 to 0.018; low to moderate certainty) and mind-body exercise (0.007 g/cm 2 , 0.002 to 0.013; moderate certainty) may provide the greatest benefit. For total hip BMD, only brisk walking or jogging (0.021 g/cm 2 , 0.005 to 0.038; low certainty) may provide the greatest benefit. Dose-response relations appeared non-linear and were generally an inverted U-shape. Minimal clinically important difference might be reached at about 400 metabolic equivalents of task in minutes per week (METs-min/week) for lumbar spine and around 600 METs-min/week for femoral neck and total hip BMD. . In fracture analyses (26 trials, 11 132 participants), mixed aerobic exercise (odds ratio 0.29, 95% CrI 0.11 to 0.78; low certainty) and mind-body exercise (0.58, 0.36 to 0.93; low certainty) may reduce fracture risk, whereas other modalities had neutral effects. Conclusions Exercise is effective in attenuating age related decline in BMD in adults aged ≥40 years. Based on low to moderate certainty evidence, brisk walking or jogging and combined aerobic-resistance exercises probably offer favourable benefits for BMD. Clinically meaningful benefits might generally be achieved at doses around 600 METs-min/week—equivalent to about 2-3 hours of brisk walking or jogging, while mixed aerobic and mind-body exercise may additionally reduce fracture risk. Given the overall low to moderate confidence in several estimates, these findings provide cautious support for the integration of dose specific, modality tailored exercise prescriptions into bone health guidelines. Systematic review registration PROSPERO NCT42024553562 .