Tai chi nudges bone density up in older women

Bone mineral density falls in women through midlife and beyond, and the standard advice leans on weight-bearing and resistance exercise, which not everyone will take up or sustain. Traditional Chinese mind-body practices such as slow, weight-shifting forms are popular, low-impact and widely accessible, so whether they do anything measurable for bone is a practical question. Individual trials have been small and have disagreed. This review set out to pool them and, separately, to ask whether more training produced more benefit.

The authors searched five databases, including Chinese-language sources, from their inception to April 2026 for randomised controlled trials testing Traditional Chinese mind-body training on bone mineral density in middle-aged and older women. Two reviewers independently selected studies, extracted data and rated risk of bias. Results were combined using a three-level random-effects model. They then stress-tested the result with influence diagnostics, leave-one-out analysis, Egger's test, trim-and-fill, subgroup analysis, meta-regression and Bayesian dose-response modelling, plus an exploratory machine-learning analysis of which study features predicted outcomes.

Twenty randomised trials went into the first pooled analysis, which showed a clear increase in bone mineral density with a standardised mean difference of 0.55. Diagnostics flagged three trials as heavily influential; with those removed, 17 trials remained and the pooled effect fell to 0.32, smaller but still statistically significant. Disagreement between studies was substantial in the initial analysis. Dose-response modelling hinted at non-linear relationships between training amount and bone outcomes, and the machine-learning step suggested training frequency, the site measured and participant age mattered to predictions. The authors label both of these exploratory rather than confirmed.

The finding supports the broader idea that regular, low-impact loading and balance work is not inert for the skeleton, while sizing the benefit modestly rather than dramatically. The drop from 0.55 to 0.32 once three outlying trials were excluded is a useful illustration of how fragile pooled estimates can be. The lumbar spine showing the most apparent improvement fits the pattern seen with other loading interventions, though the review cannot explain why. Open questions include the right frequency and duration, how long gains last, and whether any of this translates into fewer fractures, which these trials did not measure.

If you already practise or are drawn to slow Chinese mind-body forms, this is a reason to think they contribute something real to bone density rather than nothing at all. The honest framing is a small average gain, not a replacement for anything else you might do, and the studies disagreed enough that individual results will vary. Nobody can yet tell you how many sessions a week would matter, because the dose analysis was exploratory by the authors' own description. It is also worth remembering that bone density is a marker, and these trials did not track fractures.