Pilates for back pain: what the research shows, and how to start safely

It helps long-lasting back pain more than doing little

In a Cochrane review of 10 trials (510 adults whose back pain had lasted 3 months or more), Pilates eased pain by about 14 points on a 0 to 100 scale in the short term and about 10 points at 3 to 12 months, and eased disability too, compared with minimal care such as advice. Exercise in general does about as well.

Sources 1, 2

It isn't clearly better than other exercise

Compared head-to-head with other exercise, Pilates gave similar results. NICE (the National Institute for Health and Care Excellence) and the World Health Organization (WHO) found no type of exercise clearly better than another, so choose the one you'll keep doing.

Sources 1, 3, 4, 5

Twice a week is a good start

In a 6-week trial of 296 adults, Pilates twice a week helped more than once a week, and three times a week wasn't clearly better than twice. Give it at least 6 weeks: programmes that add up to 20 hours or more tend to do better.

Sources 6, 7, 5

Back pain that has just started: keep moving first

The Cochrane review found no Pilates trials in back pain that had just started, and exercise of any kind made little difference in the first weeks. Stay active and carry on with daily life; a class can wait until you move more freely.

Sources 1, 8, 9

Sciatica, slipped discs and stenosis: little direct evidence

Two small trials in people with a slipped disc found Pilates helped. We found no trials of Pilates for spinal stenosis (a search of Europe PMC, October 2026). Let your symptoms guide which moves you use.

Sources 10, 11, 12, 13

Some back pain needs a doctor first

Numbness around your genitals or bottom, new problems peeing or pooing, or weakness or numbness in both legs need urgent help, not a class. The full list is further down this page.

Source 9

What the trials tested

The trials used supervised sessions of about an hour, taught by physiotherapists or instructors trained in Pilates, on a mat or on equipment such as the reformer, the Cadillac and the chair. Programmes averaged about 15 sessions, from daily to 3 times a week, over 10 days to 3 months.

In the largest trial of how often to go, physiotherapists certified in Pilates ran the sessions, with an average of 7.5 years' experience.

Sources 1, 14

Clinical Pilates and studio classes

Clinical Pilates is Pilates-based exercise chosen and supervised by a physiotherapist after an assessment, often built around the movements that ease your pain. A review describes it as a mix of Pilates and the McKenzie method, which is a theory rather than a tested claim.

A studio or gym class is usually a group mat or reformer class led by a Pilates instructor. Thirty Australian physiotherapists suggested supervised sessions of 30 to 60 minutes, twice a week for 3 to 6 months, for long-lasting back pain: an expert view, not a trial result.

Sources 15, 16, 17

Which back pain the trials studied

Almost all trials studied non-specific low back pain: pain with no specific cause, such as a fracture, an infection or a trapped nerve, found. Long-lasting (chronic) means 12 weeks or more. Most new back pain "often improves on its own within a few weeks", the NHS says.

Sources 1, 2, 9

Why it might help

Nobody knows exactly why it helps. Breathing is the only traditional Pilates principle that back pain trials mention, and some of the benefit may come from the attention and expectation that go with any supervised programme, as the authors of the largest trial and NICE's committee both noted. That it works by fixing a weak core or correcting posture hasn't been shown.

Sources 18, 6, 4

Better than doing littleStrong evidence

In the Cochrane review (10 trials, 510 adults, 7 trials at low risk of bias), Pilates reduced pain by about 14 points out of 100 in the short term (low-certainty evidence) and about 10.5 points at 3 to 12 months (moderate certainty, 2 trials), and disability by about 8 and 11 points, compared with minimal care. An earlier review of 14 trials found an advantage over usual care at 4 to 15 weeks but not at 24 weeks, and a 2024 review of 36 studies agreed on the direction.

For scale, across 249 trials, exercise of any kind reduced long-lasting back pain by about 15 points out of 100, the size its authors count as clinically important. The grade reflects how consistently the reviews agree; each rated its certainty low to moderate.

Sources 1, 19, 20, 2

Better than other exercise? Not clearlyMixed evidence

Head-to-head, Pilates and other exercise gave similar results for disability (moderate certainty) and mixed results for pain. A 2023 review of 11 trials found a small edge for Pilates over general exercise for pain, at low certainty, and no difference from stabilising exercise.

Network meta-analyses, which rank many types of exercise at once, put Pilates at or near the top: first for pain in a review of 89 studies, and among the most effective in reviews of 82, 118 and 217 trials. Another ranked tai chi first. All rated their certainty low or very low, and a ranking is the chance of coming first, not the size of the benefit.

The 89-study review (Owen 2020) received an expression of concern from its journal after comments from other researchers. The editors withdrew it on 27 July 2020, saying the paper needed no changes and that they had "full confidence in the findings".

Sources 1, 3, 21, 22, 23, 7, 24, 25

Guidelines recommend exercise, not Pilates in particularStrong evidence

NICE says to "consider a group exercise programme (biomechanical, aerobic, mind-body or a combination of approaches)" for a flare-up of low back pain, with or without sciatica, taking your needs and preferences into account. Its evidence review counts Pilates as biomechanical exercise, and NICE left this advice unchanged when it updated the guideline in July 2026.

The WHO says a structured exercise programme may be offered to adults, including older people, with long-lasting low back pain (a conditional recommendation, on low-certainty evidence), and found too little evidence to say one type beats another. A US physiotherapy guideline names Pilates among the active treatments to use, and the American College of Physicians recommends exercise for long-lasting back pain. The NHS says "Activities like walking, swimming, yoga and pilates may also help ease back pain".

Sources 26, 4, 5, 27, 28, 9, 29

A fuller Cochrane comparison is on the way

Cochrane published the plan for a network meta-analysis of exercise types, including Pilates, in June 2023. It has no results yet, so the rankings above may change.

Source 30

Side effects are uncommon and minorEmerging evidence

Only 2 of the 10 Pilates trials in the Cochrane review checked for side effects: one found none and the other only minor ones. Across exercise trials for back pain, side effects were mostly minor, such as muscle soreness, and NICE found no evidence of harm.

Sources 1, 2, 4

Twice a week beat once a weekEmerging evidence

In a trial of 296 adults with long-lasting back pain, everyone got an advice booklet, and three groups also did Pilates once, twice or 3 times a week for 6 weeks. All three Pilates groups improved more than the booklet group. Twice a week helped pain and disability more than once a week. Three times a week wasn't clearly better than twice in the reported 6-week results, though it was the option most likely to be cost-effective.

The authors noted that some of the effect may have come from extra attention or expectation.

Source 6

More sessions didn't speed things upEmerging evidence

In the same trial, daily pain ratings showed people improved at the same speed whether they went once, twice or 3 times a week.

Source 14

Session and programme length

In a network meta-analysis of 118 trials, the most helpful programmes had at least 1 to 2 sessions a week of Pilates or strength exercise, sessions under an hour (for core, strength or mind-body exercise) and 3 to 9 weeks of Pilates or core exercise. The WHO notes that supervised, tailored programmes adding up to at least 20 hours tend to do better, and an expert panel suggested 30 to 60 minutes twice a week for 3 to 6 months.

Sources 7, 5, 17

Our editorial starting point

Two sessions a week of 30 to 60 minutes, for at least 6 weeks. Think in months: 20 hours is, for example, two 1-hour sessions a week for 10 weeks. Add a short home routine on other days if you like, and keep going once you feel better: in older trials, the advantage over usual care had faded by 24 weeks.

Sources 6, 5, 19

Clinical Pilates did no better than general exerciseEmerging evidence

In a trial of 87 adults with back pain for 3 months or more, both groups had 60-minute sessions with a physiotherapist twice a week for 6 weeks, plus home exercise. The clinical Pilates group got a programme tailored after an examination. Both groups improved, with no difference between them at 6, 12 or 24 weeks.

Source 15

Mat and equipment eased pain equallyMixed evidence

In a trial of 86 adults, 12 sessions over 6 weeks of mat or equipment Pilates eased pain equally. At 6 months, equipment Pilates was slightly better for disability and fear of movement. It is one small trial; our reformer guide compares the two in more detail.

Source 31

Supervised Pilates beat home exercise, but not by muchEmerging evidence

In 145 adults, twice-weekly supervised mat Pilates for 6 weeks eased pain and disability more than home exercise straight afterwards, but by amounts the authors judged too small to matter, and by 6 months only quality of life still differed. In 66 people whose pain had lasted more than a few weeks but less than 3 months, supervised Pilates beat a home programme.

Sources 32, 33

We found no trial of physiotherapists against instructors

We found no trial comparing physiotherapist-led Pilates with instructor-led classes (a search of Europe PMC, October 2026). The trials used physiotherapists or instructors trained in Pilates.

Sources 34, 1

Who should see a physiotherapist first

Our suggestion: see a physiotherapist before a general class if you have leg pain, numbness or weakness, a diagnosed spondylolisthesis or an unstable spine, a recent spinal fracture or surgery, or pain that isn't improving after several weeks. Australian physiotherapists agreed that people with an unstable spondylolisthesis or a fracture may not benefit from Pilates. In many areas of the UK you can refer yourself to NHS physiotherapy without seeing a GP.

Sources 35, 9

SciaticaMixed evidence

NICE's group exercise advice covers back pain with or without sciatica, but its evidence review found no evidence on mind-body exercise in sciatica and mixed results for other exercise. The NHS says to "start gentle exercise as soon as you can" and not to sit or lie down for long periods, because moving isn't harmful even if it hurts. If a move sends pain further down your leg, leave it out.

Sources 26, 4, 36

Slipped (herniated) discMixed evidence

Two small trials found benefits. In 54 people with a slipped disc in the lower back, clinical Pilates 3 times a week for 6 weeks eased pain and disability more than no exercise, though side effects weren't recorded and nobody was followed up beyond 6 weeks. In 45 women, adding Pilates to water exercise improved mobility, disability and quality of life more than water exercise alone.

The NHS says "The type of exercise is not important as long as it does not cause pain or make your back pain worse".

Sources 10, 11, 13

Spinal stenosisMixed evidence

We found no trials of Pilates for spinal stenosis, a narrowing of the spaces the nerves run through in the lower back (a search of Europe PMC, October 2026). Our spinal stenosis guide covers the exercise that has been tested. If arching your back brings on leg symptoms, skip the arching moves.

Source 12

SpondylolisthesisMixed evidence

Spondylolisthesis is where a bone in the spine slips forward. The only Pilates study we found was a report on one patient. Australian physiotherapists agreed that people with an unstable slip may not benefit from Pilates. The NHS advises avoiding activities that make symptoms worse, such as bending, lifting, athletics and gymnastics, and physiotherapy to strengthen and stretch the muscles of the lower back, tummy and legs.

Its pain is often eased by sitting or bending forward, so test backbends such as the swan carefully, or skip them (our suggestion).

Sources 37, 12, 35, 38

Let your symptoms choose the direction

Moves that round your back (the chest lift, spine stretch and roll-downs) and moves that arch it (the swan, swimming and sphinx) suit different backs. If one direction brings on leg pain, or makes it spread further down the leg, leave those moves out and keep the neutral ones. No Pilates move has been shown to harm these conditions, and none has been shown to fix them.

Sources 13, 38

Better than doing nothingEmerging evidence

In a 2025 review of 8 trials (439 people with long-lasting neck pain), Pilates eased pain and disability more than no exercise, at very low to moderate certainty. A network meta-analysis of 40 trials found yoga, Pilates, tai chi and qigong together helped about as much as motor control and strengthening exercise, on very low quality evidence. In one trial of 64 people, Pilates twice a week for 12 weeks, with usual medication, eased pain and cut painkiller use more than medication alone.

Sources 39, 40, 41

Not better than other exerciseMixed evidence

Compared with other exercise, Pilates eased neck pain slightly less straight after treatment, at low certainty, with no difference in disability. Another review of 5 trials found it no better than other treatments. Choose what you'll keep doing.

Sources 39, 42, 40

Spare your neck

Leave out moves that put weight on your neck, such as the roll-over, jack knife and control balance, while you have neck pain or tingling in an arm. In curl-ups, let your hands carry your head, and rest it between repetitions. Most new neck problems start to settle within 8 weeks, according to NHS inform.

Sources 43, 44

Call 999 (911 in the US) or go to A&E

If you have back pain and any of these. The first four can be signs of cauda equina syndrome, pressure on the nerves at the base of the spine, which needs treating in hospital urgently. Don't drive yourself: ask someone to drive you or call an ambulance.

• pain, tingling, weakness or numbness in both legs
• a loss of feeling around your genitals or anus
• changes in your bladder or bowels, such as difficulty peeing, or peeing or pooing yourself
• changes in how your penis or vagina feels during sex, not being able to get or keep an erection, or not being able to orgasm
• chest pain
• pain that started after a serious accident, such as a car accident

Source 9

With sciatica, also go to A&E or call 999

If you have:

• sciatica on both sides
• weakness or numbness in both legs that's severe or getting worse
• numbness around or under your genitals, or around your bottom (anus)
• difficulty starting to pee, not being able to pee, or not being able to control when you pee, if this isn't normal for you
• not noticing when you need to poo, or not being able to control when you poo, if this isn't normal for you

Source 36

Get an urgent GP appointment or call NHS 111

If you have back pain and you feel hot, cold, shivery or generally unwell, or the pain is severe and started suddenly, or is getting worse quickly. In the US, contact your doctor or urgent care.

Source 9

See a GP

If you have back pain and:

• it hasn't improved after a few weeks of looking after it at home
• it's stopping you doing your day-to-day activities
• you're worried about it or struggling to cope
• you've lost weight without trying to
• there's a lump or swelling in your back, or your back has changed shape
• it doesn't improve with rest, or is worse at night
• it's worse when you sneeze, cough or poo
• it's coming from the top of your back, between your shoulders, rather than your lower back

Source 9

Neck pain: contact your GP practice urgently

Or NHS 111 if your practice is closed, if you have any of these. Our neck pain guide lists the signs that need 999.

• clumsiness in your hands, or numbness, pins and needles or weakness in one or both arms, that is getting worse
• problems with your balance or walking since your neck pain started
• neck pain with a severe headache

Source 44

During a session

The NHS says to stop back exercises if your pain gets worse, and to see a GP for advice. As a rule of thumb from our back pain guides, mild pain up to about 5 out of 10 that settles within an hour and is no worse the next morning is fine. Leg symptoms are the exception: stop any move that sends pain, tingling or numbness further down a leg.

Source 9

"Pilates is the best exercise for back pain"Mixed evidence

Pilates ranks at or near the top in several network meta-analyses, at low certainty, but another ranks tai chi first, and head-to-head trials show no clear winner. The WHO found no type of exercise clearly better. The best exercise for your back is one you'll keep doing.

Sources 21, 23, 7, 25, 3, 5

"Rest until the pain goes"Strong evidence

The NHS advises staying active and carrying on with daily activities, and not staying in bed for long periods. NICE recommends encouraging people to continue with normal activities.

Sources 9, 26

"Pilates is dangerous with a slipped disc"Emerging evidence

The NHS says the type of exercise isn't important as long as it doesn't cause pain or make your back pain worse, and two small trials in people with a slipped disc found Pilates helped. Let your symptoms guide which moves you use.

Sources 13, 10, 11

"Clinical Pilates with a physiotherapist works better than a class"Emerging evidence

Not shown. In a trial, clinical Pilates did no better than general exercise when both were led by physiotherapists, and we found no trial comparing physiotherapists with Pilates instructors. A physiotherapist is the right first step for some people, not for everyone.

Sources 15, 34

"The more sessions a week, the better"Emerging evidence

Three sessions a week weren't clearly better than two in the dose trial's reported results, and more sessions didn't speed up improvement.

Sources 6, 14

"Pilates works by fixing a weak core"Mixed evidence

Pilates helps, but so does other exercise, and part of the benefit may come from the attention and expectation of a supervised programme. A core-strength explanation hasn't been shown.

Sources 6, 4, 3