Across 45 randomized trials and 16,240 older adults, mind-body exercise (tai chi, qigong, yoga-type practice) lowered the odds of being a faller by 38% (OR 0.62, 95% CI 0.46–0.85) and of recurrent falls by 43% (OR 0.57, 0.38–0.85) versus usual care. Multicomponent exercise cut fallers by 22% (OR 0.78, 0.67–0.91). Balance training and mind-body exercise modestly reduced fall-related injuries (ORs 0.76 and 0.68). No exercise type significantly reduced fractures.
The ranking of mind-body exercise fits earlier pairwise meta-analyses and the Cochrane review on tai chi, but this network approach is the first to compare modalities within a single framework. Tai chi plausibly works by combining balance, lower-limb strength, weight-shifting, and attention in one low-risk format that people sustain. The practical message for adults and clinicians: structured, balance-challenging programs beat generic activity, and adherence matters more than novelty. Caveats are substantial. Certainty of evidence ranged from moderate to very low, heterogeneity was moderate for fallers (I2 55%), and indirect comparisons between modalities rest on assumptions that trial populations are similar. The fracture null result likely reflects too few events rather than true absence of benefit, since fractures are rare and trials were underpowered for them. Trials also cannot blind participants to exercise assignment, which can inflate self-reported fall outcomes. Overall this is confirmatory rather than paradigm-shifting: it reinforces existing guidelines recommending balance and multicomponent training, while nudging tai chi-style practice higher on the list of options worth discussing with a doctor or physiotherapist.