Meditation calms you down. Right? Let's check!

Meta-analysis of randomised trials. 3,515 people. trials lasted 4 weeks to 9 years, mostly short; typical mindfulness programme = MBSR, 20-27.5 hours of training over 8 weeks; outcomes at ~8 weeks and 3-6 months.

Many trials, added up. In each one, chance decided who did what.

Who: adults (18+) with a clinical condition or under stress: 15 trials in psychiatric populations (anxiety, depression, stress, chronic worry, insomnia), 5 in smokers/alcohol users, 5 in chronic pain, 16 in medical populations (heart disease, lung disease, breast cancer, diabetes, hypertension, HIV). Otherwise healthy people, children, yoga/tai chi and app/audio-only programmes were excluded

• WorryLessa little
• Feeling downLessa little

No sign it made people happier.

More to know, from our read of the paper:

• Effects are small (effect sizes ~0.2-0.4) and none of the conclusions reached HIGH strength of evidence; most grades were low or insufficient
• Only trials with active controls (education, attention control, or another therapy) were included; this is stricter than wait-list trials, which tend to show bigger effects. Letter by Loucks (JAMA Intern Med 2014;174:1194) argues usual-care-controlled trials were wrongly left out, so the review answers 'beyond attention and expectation', not 'vs doing nothing'
• 'No better than exercise, drugs or CBT' does NOT mean 'as good as': comparative trials averaged ~37 people per group and none was powered to show equivalence or non-inferiority
• Trials had high attrition, poor blinding of outcome assessment, unclear allocation concealment and missing intention-to-treat analyses; many old trials, mostly unregistered
• Participants had a clinical condition or were stressed; results do not automatically apply to healthy people, apps, or yoga/tai chi (excluded)

Scientists added up 47 tests where chance decided who got a meditation class and who got a different class of the same length (a meta-analysis of 47 randomised controlled trials, 3,515 adults who were stressed or had a health condition). Mindfulness classes, often about 8 weeks long, helped worry, low mood and pain a little; mantra meditation, including TM, did not improve anything they checked. There was no sign it boosted happy feelings, and the evidence on sleep and focus was too thin to say. Meditation was no better than exercise, medicine or other therapies (which doesn't mean it is as good). Healthy people and apps were not studied.

We read the abstract and the full text of this paper and checked every number on this page against it (20 facts, checked on 2026-10-03). The study design is named above. Cohort and other observational studies are described as linked, not proven. Found an error? Email theis@onestudyaday.com.