Yoga and pilates reduce pain in chronic low back pain more effectively than conventional rehabilitation or no intervention, with yoga performing somewhat better than pilates. For older adults, however, an explicit caution applies: an intensive yoga programme increased fall risk in a large RCT, even though participants felt more confident on their feet.
Yoga and pilates have both been studied as treatments for chronic low back pain. A large network meta-analysis of 75 randomised controlled trials (RCTs) involving more than 5,000 participants shows that yoga has a moderate to large effect on pain reduction (SMD -1.76) and on physical functioning (SMD -1.72), both compared with conventional rehabilitation. Pilates also reduces pain significantly (SMD -1.52), but the effect is somewhat smaller than for yoga or tai chi. In the same analysis, pilates showed no statistically significant benefit on physical functioning. That sounds as though yoga clearly wins, but the difference is small and direct comparisons between yoga and pilates are largely absent.
For older adults with joint pain (osteoarthritis, neck pain), a systematic review in people with an average age above 50 points to moderate evidence that yoga relieves pain. Of the eight studies that measured pain, six showed a significant effect. The effects are comparable to those of other forms of exercise; yoga is therefore no miracle cure, but a useful option alongside other types of movement. Quality of life did not clearly improve in the short term.
An important safety signal comes from a large RCT (700 participants, aged 60+, published in Lancet Healthy Longevity 2025): a 12-month Iyengar yoga programme increased the number of fall incidents among community-dwelling older adults compared with a seated relaxation programme (0.87 versus 0.64 falls per person per year). This is a causally established effect. The researchers explicitly advise against this programme for fall prevention. Six participants experienced musculoskeletal adverse events. Notably, that same programme did improve balance confidence and the number of hours of physical activity per week, illustrating that how someone feels and how safe something is do not always go hand in hand.
For other applications the evidence is thinner. A broad review on ageing describes positive effects of yoga on balance, osteoporosis and Parkinson's disease, but yoga did not consistently outperform other forms of exercise. For menopausal symptoms (hot flushes, anxiety, sleep problems, pain), a 2024 review overview mentions yoga as an option with some empirical support, but without hard figures. For migraine there is limited positive evidence; for cancer pain, a joint guideline from two major oncology organisations concludes that there is insufficient evidence to make a recommendation.
Finally: no single form of exercise is demonstrably better than all others for chronic low back pain. Both a narrative review and the network meta-analysis confirm that tai chi, yoga and pilates perform comparably or better than conventional rehabilitation, but differ little from one another. The advice from the literature is to choose a form of movement that suits the individual's own preferences and circumstances.
The strongest evidence comes from one network meta-analysis (PMID 38035307, 75 RCTs, ~5,254 participants) for low back pain, and one large RCT for fall risk in older adults (PMID 41005344, n=700). The remaining claims are based on systematic reviews, narrative reviews and guidelines of varying quality. Direct head-to-head comparisons between yoga and pilates are scarce.