Guidelines include yoga, and put exercise firstStrong evidence
The American College of Rheumatology conditionally recommends yoga for knee osteoarthritis (tai chi is strongly recommended) and mind-body exercise for rheumatoid arthritis. EULAR (the European Alliance of Associations for Rheumatology) counts yoga as one kind of exercise and finds it "difficult to recommend one type of exercise over another". OARSI (the Osteoarthritis Research Society International) makes structured exercise a core treatment.
Sources 6, 3, 19, 20
Knee osteoarthritisEmerging evidence
Meta-analyses of 9, 8 and 14 studies found that yoga eased knee pain and stiffness and improved function, from very low to low certainty evidence with trials that varied a lot. A 2025 analysis of 92 trials (6,079 people with knee or hip osteoarthritis) ranked yoga third of five exercise types for pain, with no significant differences between them. Two small pilot trials in older adults found yoga eased knee symptoms more than waiting (36 women) and, at 8 weeks, more than aerobic and strengthening exercise (83 adults).
Sources 21, 4, 22, 23, 24, 25
As good as strengthening, in one trialEmerging evidence
In 117 adults in Tasmania, 12 weeks of yoga was as good as a strengthening programme for knee pain (a difference of about 1 point on a 100-point scale), and 7 of 27 other measures favoured yoga. Side effects possibly linked to the programmes were more common with yoga (15.5% against 6.8%), though none were serious and flare-ups of knee pain were about as common in both groups. The authors note their programme was designed for arthritis and doesn't reflect general community classes.
Source 5
Online videos alone: small, short-lived gainsEmerging evidence
In 212 adults with knee osteoarthritis, 12 weeks of unsupervised online yoga videos improved function a little but not pain when walking, and the benefits had gone by 24 weeks. Side effects were minor.
Source 26
Chair yogaEmerging evidence
In 131 older adults with hip or knee osteoarthritis who couldn't do standing exercise, chair yoga twice a week for 8 weeks reduced how much pain got in the way of life, and that lasted 3 months; pain, walking speed and tiredness improved only during the programme.
Source 17
Hip and hand osteoarthritisMixed evidence
We found no yoga trial in hip osteoarthritis on its own, and the Cochrane review of yoga for the hip or knee is still a protocol. For the hands, one small 1994 trial of weekly yoga for 8 weeks eased pain during activity and tenderness, but a 2019 meta-analysis found no clear effect.
Sources 6, 27, 28, 21
Rheumatoid arthritisEmerging evidence
Across 10 trials (840 people), yoga improved function, disease activity and grip strength but not pain, swollen or tender joints, or blood markers of inflammation. A 2025 review of 5 trials found moderate certainty for small improvements in disease activity and function, and very low certainty for pain. A 2026 review of 8 trials (651 people) found disease activity improved, but the change in function wasn't significant. The Cochrane review is still a protocol.
Sources 7, 8, 29, 30
One small new trialEmerging evidence
In 74 women with rheumatoid arthritis in remission, 8 weeks of yoga and of combined exercise both did better than usual care, and yoga eased pain more than combined exercise at 20 weeks, with a very large effect. It's one small trial, registered after it began, and the result needs confirming.
Source 31
Mood, sleep and fatigueEmerging evidence
Across rheumatic conditions, yoga eased depressive symptoms, anxiety and poor sleep, but not more than other exercise for depression, and most trials had a high or unclear risk of bias. In 26 young women with rheumatoid arthritis, 6 weeks of Iyengar yoga improved quality of life, mood and tiredness, and a review of 23 studies found better quality of life overall. The American College of Rheumatology's integrative recommendations cite low to moderate certainty evidence that mind-body approaches such as breathing and relaxation help depression, anxiety, fatigue and sleep in rheumatoid arthritis.
Sources 9, 10, 32, 3, 33
SafetyEmerging evidence
Across 94 trials, side effects were no more common with yoga than with usual care or other exercise. But many arthritis trials didn't report harms, and in one knee trial side effects possibly linked to the programme were more common with yoga than with strengthening (15.5% against 6.8%), though none were serious.
Sources 34, 5, 35