Yoga guide
Yoga is an ancient practice rooted in Indian philosophy. As most people do it today, it combines postures, breathing and relaxation or meditation, and it is one of the most studied kinds of exercise. The clearest benefits are for low mood and stress, with smaller gains for back and neck pain, balance, flexibility, leg strength, blood pressure and blood sugar.
Most trials are small and compare yoga with doing nothing, and where yoga has been compared with other exercise it usually does about as well, not better. It counts as strength exercise, but most styles are light activity, and it doesn't build bone, detox your body or, as far as trials show, prevent falls.
This guide goes benefit by benefit, with the numbers and how sure we can be, then covers how yoga fits the activity guidelines, the conditions it has been tested in, the claims that don't hold up and how to stay safe.
In the largest analysis of exercise for depression (218 trials), yoga eased symptoms about as much as walking or strength training, and people stuck with it well. It also lowers everyday stress. As with most exercise research, the certainty is low.
For long-lasting low back pain, yoga helps a little more than no exercise and about as much as other back exercise. Neck pain improves about as much as with other exercise.
Balance test scores improve in people over 60. But in the largest falls trial, 700 people aged 60 and over had about a third more falls with yoga classes than with seated relaxation.
Most styles are light, about a slow walk; brisk flows and power yoga reach moderate intensity. The NHS lists yoga as a muscle-strengthening activity, and in older adults it improves flexibility and leg strength.
Blood pressure fell by about 5 points against no exercise, and in type 2 diabetes average blood sugar fell a little. Where yoga has been compared with other exercise, it did about as well, not better.
Yoga doesn't build bone, no study shows it detoxes the body or works on particular organs, and the North American Menopause Society doesn't recommend it for hot flushes.
Yoga began as a spiritual practice rooted in Indian philosophy. As practised in the UK and US today, it usually combines postures (asanas), breathing techniques (pranayama) and meditation or relaxation, and styles range from gentle to physically demanding. Trials use different styles, which can give different results; our types of yoga guide explains them.
Source 16
A 2025 umbrella review, which pulls other reviews together, found 51 systematic reviews covering 579 trials and 28,403 people. The clearest effects were for depression, blood pressure, blood sugar and tiredness, with weaker evidence for pain and arthritis.
Read it as a map of where the evidence sits, not as a measure of how big the effects are: many of the reviews share the same small trials, and its ratings come from the authors' own statistical rules rather than a formal grading of certainty. An earlier overview of 26 reviews judged the evidence behind yoga generally low in quality, and a synthesis of 72 meta-analyses of mind-body exercise for spinal pain found many of them of critically low quality.
Most trials compare yoga with doing nothing or with usual care, and people know which group they are in, which can make effects look bigger. Where yoga has been compared with other exercise, it usually does about as well. The notes below say which comparison each result comes from.
Blood pressure trials averaged nearly 5 sessions a week of about an hour for 13 weeks. Trials in older adults ran 1 to 3 classes a week of 50 to 90 minutes for 4 to 28 weeks, often with practice at home, and one 12-week trial used a weekly class plus about 23 minutes a day at home. The falls trial ran two 1-hour classes a week for a year. Most of the benefits below come from regular practice over months.
Across 218 trials of exercise for depression (14,170 people), yoga eased symptoms by a moderate amount compared with usual care or a placebo tablet (33 yoga groups, 1,047 people), about as much as walking or jogging and strength training. People were among the most likely to keep going with it, and light activity such as hatha yoga still helped.
Only one study in that analysis was at low risk of bias, so confidence in yoga's effect is very low, as it is for most types of exercise. Smaller reviews point the same way, including in people with a diagnosed mental health condition. Use yoga alongside, not instead of, treatment for moderate or severe depression.
In stressed adults (13 trials, 1,026 people), perceived stress fell in the short term compared with no treatment, on low-quality evidence, and quality of life improved; a second review of 36 studies agreed. Anxiety symptoms eased a little compared with no treatment in 8 trials, but not clearly in people with a diagnosed anxiety disorder.
In a trial of 226 adults with generalised anxiety disorder, 54% responded to Kundalini yoga, 33% to stress education and 71% to CBT (cognitive behavioural therapy), so CBT stays the first treatment. The NHS suggests activities such as yoga to help you relax.
In 19 trials of women with sleep problems (1,832 women), sleep quality scores improved, though insomnia severity didn't. A network meta-analysis (which compares many types of exercise at once) of 86 trials found yoga improved sleep quality against no exercise, with no clear differences between six types of exercise. Our yoga for anxiety and sleep guide has the detail.
In a Cochrane review of 21 trials (2,223 adults), yoga gave small improvements over no exercise, too small to matter much clinically, and about the same as other back exercise. It caused more side effects than no exercise, mostly more back pain. In one trial of 228 adults, yoga classes did no better than stretching classes over 12 weeks.
NICE (the National Institute for Health and Care Excellence) suggests considering a group exercise programme, which can include mind-body exercise, for low back pain. Our yoga for back pain guide has the detail.
In 10 trials (686 people), yoga eased neck pain and disability more than usual care or other treatments, though the authors draw only a cautious conclusion, and 3 smaller trials against usual care agreed. A network meta-analysis of 40 trials found yoga, Pilates and tai chi about as good as strengthening and motor control exercise, at very low certainty.
In people aged 60 and over, yoga improved balance and mobility test scores (6 trials, 307 people), and a review of 22 trials agreed. A 2026 review of 21 studies found the benefit for balance while moving was no longer clear after allowing for small studies that may be missing.
In the SAGE trial, 700 people aged 60 and over who were new to yoga were offered 80 Iyengar-based yoga classes over a year, or two workshops of seated breathing and stretching (the seated relaxation group). The yoga group had about a third more falls (0.87 against 0.64 falls per person a year), and the authors say the programme shouldn't be recommended for preventing falls in its current form.
It is a single trial, which we've read in its published summary (the abstract), and most classes ran online during the COVID-19 pandemic. In surveys, people who practise yoga fall about as often as people who don't. Better balance scores don't mean fewer falls: if falls are a worry, choose a falls-prevention programme.
In older adults, yoga improved lower-body flexibility more than no exercise (7 trials, 431 people) and a little more than other exercise (3 trials, 225 people); upper-body flexibility didn't clearly change. Our yoga for flexibility guide covers the poses and hold times.
In older adults, leg strength improved against no exercise (7 trials, 485 people) and against other exercise (3 trials), though a 2026 review found the benefit borderline after allowing for small studies that may be missing. In a 12-week trial of 173 adults with an average age of 52, which wasn't randomised, push-ups rose by about 3 and curl-ups by 2 to 3.
In 49 trials (3,517 adults), yoga lowered blood pressure by about 5 mmHg (millimetres of mercury) systolic and 4 diastolic compared with no exercise. In people with high blood pressure practising 3 times a week, programmes that included breathing and relaxation lowered it by about 11 and 6, against 6 and 3 for programmes that didn't.
Yoga did no better than other exercise, and the quality of the evidence ranged from very low to moderate. Our exercise for high blood pressure guide covers what else helps.
In type 2 diabetes (12 trials, 864 people), HbA1c (a blood test of average blood sugar over 2 to 3 months) fell by about half a percentage point and fasting blood sugar by about 1.3 mmol/L (24 mg/dL), with small falls in cholesterol, though the trials varied a lot. A second review of 23 studies agreed.
In broader groups of adults the results for blood sugar were mixed, and against other exercise only HDL ("good") cholesterol differed, slightly in yoga's favour.
A beginner hatha session didn't raise the heart rate enough to meet fitness recommendations, and another study concluded hatha may have little, if any, benefit for heart and lung fitness. Training studies disagree: aerobic fitness rose in a 12-week trial with daily home practice, but didn't change after 8 weeks of hot yoga or in trials of older adults.
Quality of life improved in older adults, in stressed adults (moderate-quality evidence) and in women with breast cancer.
The NHS lists yoga among muscle-strengthening activities, which adults should do on at least 2 days a week, and says these are "not always an aerobic activity", so you still need 150 minutes a week of moderate activity. Adults aged 65 and over are advised to do activities that improve strength, balance and flexibility on at least 2 days a week.
Intensity is measured in METs (metabolic equivalents, where sitting still is 1): light activity is under 3 METs and moderate activity is 3 to 6. The 2024 Compendium of Physical Activities lists hatha yoga at 2.3 METs, vinyasa 2.7, hot yoga 3.0, sun salutations 3.5 and power yoga 4.0 (an estimate rather than a measured value), against 4.8 for brisk walking. Measured yoga sessions average about 3.3 METs.
An hour of vinyasa measured 3.6 METs in a lab, against 4.4 for brisk walking at a self-chosen pace. In community hatha and vinyasa classes, people spent about a third of class time at moderate intensity by heart rate, and a beginner hatha session was like walking at 3.2 km/h (2 mph).
Gentle yoga is light activity, like a slow walk, so it doesn't count toward your 150 minutes. Sun salutations, power yoga and brisk vinyasa reach moderate intensity for at least part of a class, and can count if you keep a pace where you can talk but not sing. For your heart and lungs, add walking, cycling or swimming.
It counts toward your 2 strength days, and in trials it built leg strength in older and untrained people. The World Health Organization (WHO) asks for muscle strengthening at moderate or greater intensity: chair, warrior and boat poses load your muscles most, but body-weight poses add little load as you get stronger, so add weights or bands if you want to keep building strength (our suggestion).
2 or 3 sessions a week of 30 to 60 minutes, for at least 8 to 12 weeks before you judge the results, alongside brisk activity for your 150 minutes. Trials ranged from a weekly class plus home practice to nearly 5 sessions a week. Light activity still counts for something: the UK Chief Medical Officers' 2026 update says it can match the benefits of moderate activity if you do it for longer.
Each condition has its own exercise guide on the site; these notes cover what the yoga trials found.
In 8 trials (756 people), yoga eased pain and stiffness against control. In a 12-week trial of 117 adults with an average age of 62, yoga (2 supervised sessions and 1 home session a week) was as good as strengthening exercise for knee pain. The American College of Rheumatology conditionally recommends yoga for knee osteoarthritis, as reported by the US National Center for Complementary and Integrative Health (NCCIH).
In 8 small trials (651 people), yoga lowered disease activity a little, but everyday physical function didn't change.
Source 53
In women with breast cancer (24 studies, 2,166 women), yoga improved quality of life and eased fatigue and sleep problems in the short term compared with no therapy, on moderate-quality evidence, but did no better than other exercise. In people with any cancer, yoga after treatment probably eases fatigue in the short term; during treatment its effect is very uncertain.
After a heart attack, yoga-based cardiac rehabilitation in 3,959 patients in India gave better self-rated health, and more people returned to their usual activities, than enhanced usual care, but there was no clear reduction in heart attacks, strokes or deaths (6.7% against 7.4%, too few events to be sure). In coronary heart disease, yoga didn't change death rates. In heart failure, 11 small trials found better fitness and quality of life.
In COPD (chronic obstructive pulmonary disease), breathing-focused yoga improved walking distance, by about 25 metres in 6 minutes, and lung function against no treatment in 11 trials (586 people); programmes built on yoga postures didn't. Compared with or added to pulmonary rehabilitation, mind-body exercise (mostly tai chi) gave inconclusive results, so yoga doesn't replace rehab.
In multiple sclerosis (MS), yoga eased fatigue more than usual care in two reviews (7 and 10 trials), but no more than other exercise. NICE advises that aerobic, resistive and balance exercises, including yoga and Pilates, may help MS fatigue. Hot classes need care with heat sensitivity: see our hot yoga guide.
In adults aged 55 and over, mind-body exercise including yoga improved thinking skills, more than aerobic or resistance exercise did, in an umbrella review of reviews that varied a lot.
Source 63
In 94 trials (8,430 people), side effects were no more common with yoga than with usual care or other exercise. In surveys of 9,129 practitioners, most problems were sprains and strains, 1.9% reported a serious one, and knee cartilage (meniscus) injuries were more common than in people who don't practise. US emergency department data show the highest injury rate in people aged 65 and over.
In case series, 9 people had spinal compression fractures linked to yoga poses, and 13 of 89 people with yoga injuries had compression fractures, mostly linked to extreme bending forward or backward. The UK osteoporosis consensus advises avoiding postures with a high degree of spinal flexion (rounding) if you have osteoporosis.
Hold each for 20 to 60 seconds with steady breathing, with a wall or chair nearby for the balance poses; sun salutations are the flow that reaches moderate intensity (doses from each move's page).
3D demoMountain Pose (Tadasana)Beginner · No equipment
3D demoChair pose (Utkatasana)Beginner · No equipment
3D demoWarrior I (Virabhadrasana I)Intermediate · No equipment
3D demoWarrior II (Virabhadrasana II)Intermediate · No equipment
3D demoTree pose (Vrksasana)Beginner · No equipment
3D demoWarrior III (Virabhadrasana III)Advanced · No equipment
3D demoSun salutation A (Surya Namaskar A)Intermediate · No equipmentLift your chest with your back muscles in the face-down poses, then finish with at least 5 minutes of slow breathing or relaxation: blood pressure programmes that included breathing and relaxation did better (doses from each move's page).
3D demoSphinx pose (Salamba Bhujangasana)Beginner · No equipment
3D demoLow cobra (Bhujangasana)Beginner · No equipment
3D demoLocust pose (Salabhasana)Beginner · No equipment
3D demoSlow breathing (6 breaths a minute)Beginner · No equipment
3D demoCorpse Pose (Savasana)Beginner · No equipment
3D demoLegs up the wall (Viparita Karani)Beginner · A wallStart with the beginner flow; sun salutations are the yoga sequence that reaches moderate intensity, and if falls are the worry, add the balance routine (our suggestions).
A 20-minute beginner yoga sequence: mountain, chair, warrior II, triangle, tree, down dog, cobra and child's pose.
20 minBeginner11 movesA stepA 20-minute flow built on sun salutations A and B, with warrior poses and a short cool-down.
20 minIntermediate8 movesNo equipmentA hip-opening yoga sequence: low lunge, lizard, goddess, pigeon, bound angle and happy baby.
20 minBeginner7 movesNo equipmentA 25-minute beginner session pairing each Pilates move with its yoga cousin: Pilates moves through repetitions, yoga holds the pose with the breath.
25 minBeginner16 movesNo equipmentBalance practice in 10 minutes: tandem stance, single-leg stands, marches, heel walks, tree pose and floor transfers.
10 minBeginner7 movesA wall, A bench or boxA slow, 10-minute floor routine to unwind before sleep: child's pose, twists, happy baby and slow breathing.
10 minBeginner7 movesA wallFor the brisk activity and strength work that sit alongside yoga.
For chest pain that feels tight or like squeezing, or spreads to your arms, neck or jaw; severe difficulty breathing, such as gasping, choking or not being able to get words out; lips or skin that turn pale, blue or grey; or if someone has passed out and isn't responding normally.
Source 69
If you have pain, tingling, weakness or numbness in both legs, loss of feeling around your genitals or anus, changes in your bladder or bowels or in how sex feels, or back pain that started after a serious accident.
Source 70
If you're suddenly more short of breath than usual, have chest pain with feeling sick or a racing, slow or fluttering heartbeat, or have back pain and feel hot, shivery or generally unwell. In the US, contact your doctor or urgent care.
If low mood or anxiety doesn't lift, or you have thoughts of harming yourself, see a GP or get urgent help. Our exercise for depression guide has the NHS advice.
US health advice is that pregnant women, older adults and people with health conditions should talk to their doctor and the yoga teacher about their needs, and that beginners should avoid extreme practices such as headstands, shoulder stands, the lotus position and forceful breathing.
With osteoporosis, fold from the hips with a long back and avoid deep or loaded rounding. With glaucoma or high blood pressure that isn't under control, keep head-down poses brief.
Stop for sharp or increasing pain, tingling or numbness travelling down an arm or leg, dizziness, or a pinch at the front of a joint, and come up slowly from forward folds and head-down poses.
No study of yoga has measured detoxification. The idea that heat removes impurities comes from the founder of Bikram yoga, not from research; sweat in hot classes is mostly water and salt; and a review of detox claims found very little clinical evidence for any of them. Your liver and kidneys do that job.
No trial has tested twists on digestion or the organs. Twists are a comfortable way to move your spine, which is reason enough to do them.
Source 74
Studies of yoga for thyroid disease tested whole programmes of postures, breathing and meditation, and only 2 of 13 were randomised. No study has isolated one pose's effect on an organ.
Source 75
In 11 studies of women aged 45 to 78, yoga changed bone density no more than control. The popular study that said a 12-minute daily routine reverses bone loss had no comparison group and analysed only 227 of 741 volunteers. For bone, add strength and impact exercise, as the UK osteoporosis consensus advises.
It improves balance scores, but in the largest falls trial, 700 people aged 60 and over had about a third more falls with yoga classes than with seated relaxation.
The NHS counts yoga as a muscle-strengthening activity, it built leg strength in trials of older adults, and power yoga and sun salutations are moderate intensity.
Most styles are light (2.3 to 2.7 METs), about a slow walk, and a beginner hatha session didn't meet fitness recommendations. Brisk flows are the exception.
In 26 trials, most at high risk of bias, markers of inflammation moved in yoga's favour, but not significantly, and no trial counted infections.
Source 77
Side effects are no more common than with other exercise, but sprains and strains happen, knee cartilage injuries are more common, and emergency department injury rates are highest in people aged 65 and over.
The North American Menopause Society's 2023 statement doesn't recommend yoga to treat menopausal hot flushes. Our exercise for menopause guide covers what helps.
Source 15
No study has tested it.
Source 78
The best evidence is for low mood, where yoga helps about as much as walking or strength training, and for everyday stress. It also modestly lowers blood pressure, helps blood sugar in type 2 diabetes, improves balance, flexibility and leg strength, and eases neck and long-lasting back pain about as much as other exercise. Most trials are small, so treat these as likely rather than certain.
Yes. The NHS counts yoga as a muscle-strengthening activity, and brisk styles such as power yoga and sun salutations are moderate-intensity exercise. Gentle yoga is light activity, about like a slow walk, so you still need 150 minutes a week of moderate activity such as brisk walking.
It counts toward the NHS's two strength days a week, and in trials it improved leg strength in older and untrained people. It won't build strength the way progressively heavier weights do, so add weights or bands if strength is your goal.
Mostly not: hatha yoga is light activity, at about 2.3 METs (where sitting still is 1). Vinyasa, power yoga and sun salutations reach moderate intensity for at least part of a class. For your heart and lungs, add walking, cycling or swimming.
For most people, no. It covers strength, balance and flexibility, but gentle yoga doesn't count toward the 150 minutes of moderate activity a week. Brisk flows can count, or you can add brisk walking or similar.
In the largest analysis of exercise for depression, yoga eased symptoms about as much as walking or strength training, and people stuck with it well. It also eases stress and anxious feelings. For a diagnosed anxiety disorder, talking therapy (CBT) worked better in a trial, so use yoga alongside treatment rather than instead of it.
A little. In a Cochrane review of 21 trials, yoga helped long-lasting low back pain slightly more than no exercise and about as much as other back exercise, but it caused more side effects than no exercise, mostly more back pain. Our yoga for back pain guide has the detail.
It improves balance test scores in people over 60, but it hasn't been shown to prevent falls. In the largest trial, 700 people aged 60 and over had about a third more falls with a year of yoga classes than with seated relaxation. If falls worry you, do a falls-prevention programme and practise balance poses near a wall.
Modestly: by about 5 points systolic and 4 diastolic on average, and more in people with high blood pressure, similar to other exercise. Programmes that included breathing and relaxation did better.
It doesn't build bone: in studies of women aged 45 to 78, bone density changed no more than in control groups. It can help balance and strength, but avoid deep or repeated forward rounding, which has been linked to spinal fractures in people with fragile bones.
No study shows that yoga, twists or hot rooms remove toxins. Sweat is mostly water and salt, and your liver and kidneys do the clearing.
Trials ran from one class a week plus home practice to nearly 5 sessions a week, usually for 8 to 12 weeks or more. Two or three sessions a week is a sensible start, and daily practice is fine if you vary the effort.
Exercise education, not a personal prescription, and clinical review of these guides is pending. If you are pregnant, have a health condition such as glaucoma, high blood pressure or osteoporosis, or have had recent surgery, check with a doctor, midwife or physiotherapist before starting, and stop for sharp or increasing pain. Health disclaimer.
