Pooling 20 randomized trials, traditional Chinese mind-body training (tai chi, qigong and related practices) raised bone mineral density in middle-aged and older women, with a standardized mean difference of 0.55 (95% CI 0.24-0.87) and very high heterogeneity (I2 = 84.7%). Removing three highly influential trials shrank the estimate to 0.32 (0.22-0.42) across 17 trials. Dose-response modeling hinted at nonlinear effects, which the authors label exploratory.
The more trustworthy number is the smaller one. A drop from 0.55 to 0.32 after trimming outliers signals that a few small, probably optimistic trials inflated the headline result, a common pattern in exercise research. Even 0.32 is a small-to-moderate effect, and standardized differences obscure what clinicians care about: fracture risk and absolute BMD change at the hip and spine. The tai chi literature has long suggested real value for balance and fall prevention, so a modest skeletal benefit plausibly stacks with a fall-reduction benefit. Still, tai chi is low-impact, and resistance training and impact loading remain better supported for bone. The XGBoost-SHAP analysis is in-sample only and should be set aside, as the authors concede. Dose guidance is not yet defensible. Overall this is a confirmatory, incremental finding: tai chi is a safe, accessible add-on for women worried about bone loss, not a replacement for strength training, calcium and vitamin D adequacy, or medication where indicated.