Most yoga trials are hatha-styleEmerging evidence
In 306 randomised trials, hatha was the most common style label (36 trials), and the share of positive conclusions didn't differ between styles; 91% of the trials concluded positively, which suggests some optimism in the research. The Cochrane back pain review says most of its trials used Iyengar, hatha or viniyoga. When you read that yoga helped something, it was usually a steady, mixed class like a hatha class.
Sources 8, 10
Back painStrong evidence
In 320 low-income adults in Boston, 12 weekly 75-minute hatha classes worked as well as 15 physiotherapy visits for function and pain, and both groups were about 21 percentage points less likely than an education group to use pain medicine, with gains holding at a year. In 313 UK adults, 12 classes improved back function at 3, 6 and 12 months, but not pain.
Across 21 trials and 2,223 people, yoga did about 1.7 points better on a 24-point function scale and 4.5 points on a 100-point pain scale than no exercise, gains the Cochrane review judged too small to matter clinically, and it was about as good as other back exercise.
Sources 9, 23, 10
Blood pressureMixed evidence
Across 49 trials, yoga lowered blood pressure by about 5/4 millimetres of mercury (mmHg), and by about 11/6 mmHg when classes included breathing and relaxation three times a week in people with high blood pressure. But an earlier review rated all 17 of its studies at unclear or high risk of bias, the newest review rates the evidence as very low quality, and against other exercise the difference shrinks or disappears.
Sources 24, 25, 26, 27
Low mood and anxietyEmerging evidence
Across 218 exercise trials, yoga was among the better types of exercise for depression, on very low-certainty evidence, and a review of 19 studies found yoga eased depressive symptoms in people with mental disorders. In 122 people still depressed on medication, weekly hatha classes didn't beat health education at 10 weeks but did over 6 months (51% against 31% responded).
In diagnosed mood and anxiety disorders, hatha didn't beat usual care or active comparisons in 18 low-quality trials. Yoga is an add-on, not a replacement for treatment.
Sources 28, 29, 30, 31
StressEmerging evidence
In 98 adults, 8 weeks of hour-long hatha classes three times a week reduced stress in the moment, but didn't change the daily pattern of the stress hormone cortisol.
Source 32
Older adultsEmerging evidence
In 118 inactive adults with an average age of 62, three hour-long hatha classes a week for 8 weeks improved balance, strength, flexibility and mobility as much as a stretching and strengthening class.
Source 11
Bladder leaks and sleepEmerging evidence
In 240 women aged 45 to 90, 12 weeks of hatha-based pelvic yoga cut leaks by about 2.3 a day, against 1.9 with a general conditioning programme, a difference that wasn't significant. Sleep didn't differ between the groups either.
Sources 12, 33
CancerMixed evidence
In 410 cancer survivors, 4 weeks of gentle hatha with restorative poses improved sleep quality and cut sleeping-pill use against usual care. But across 11 studies with 777 people, hatha or Iyengar yoga didn't improve cancer-related tiredness, depression or quality of life, on low to very low certainty evidence.
Sources 15, 34
Minor harmsEmerging evidence
In back pain trials, more people had minor problems with yoga than with no exercise (43 against 9 per 1,000), mostly more back pain.
Source 10