How Much Exercise Eases Back Pain? 239 Trials Point to a Sweet Spot

Back pain is the world’s leading cause of disability. Hundreds of millions of people live with it.

For long-lasting back pain, doctors usually recommend exercise first. But patients rarely get a clear answer to the obvious next question.

That gap matters, because an unclear plan is an easy plan to abandon.

How much exercise, exactly? How often, for how long, and for how many weeks?

A big new review in the British Journal of Sports Medicine tried to answer that, using 239 randomised trials.

What are the best exercises for lower back pain?

Here’s the honest answer: there may not be a single best one.

Earlier studies have crowned different winners, from Pilates to strength training. This new review suggests those rankings may partly reflect how much exercise people did, rather than which kind.

The NHS advises staying active and doing exercises and stretches. It also lists group exercise and physiotherapy among the treatments for back pain.

So a practical rule is to pick an exercise you can stick with. Then get the dose roughly right, which is where this study helps.

One warning first. Some symptoms need emergency care, the NHS says.

Call 999 or go to A&E if back pain comes with any of these:

• numbness or weakness in both legs
• numbness around your genitals or bottom
• changes in your bladder or bowels

The study: 239 trials of exercise for back pain

Who did it

The review was led by researchers in Germany, with collaborators in the UK, Australia and China. It was funded by the German Federal Ministry of Education and Research.

The team searched six databases and two trial registries up to June 2025. They published a detailed plan in advance.

What they included

They included randomised controlled trials of exercise in adults with chronic low back pain. Nearly all trials, 99%, involved non-specific back pain, with no clear single cause.

In total, there were 239 trials published between 1986 and 2025. Participants were 42 years old on average, and 64% were women.

Six ways to measure dose

Instead of lumping “exercise” together, the team modelled each part of a programme separately.

They then estimated which level of each measure gave the biggest improvement in pain and in disability, meaning how much back pain limits daily life.

They set a small but meaningful improvement at 5 points on a 100-point scale.

The results: moderate doses did the work

Short-term sweet spots

In the first three months, the biggest predicted improvements came at fairly modest doses.

For pain, the model predicted drops of about 26 points on the 100-point scale at the best frequency. That’s well above the 5-point bar, but it comes with very low certainty.

An important caution: each row was modelled on its own. These numbers don’t add up to one tested programme.

Even small doses helped

The smallest doses that still gave a meaningful improvement were surprisingly low.

For pain, one session a week, or 10-minute sessions, already crossed the threshold. For disability, it took about two sessions a week, or around 70 minutes a week.

The uncertainty ranges around the smallest and the best doses overlapped. So several doses may work about equally well.

The benefit faded over time

Most doses stopped showing a meaningful benefit after three months. Few trials followed people for a year or more.

The authors suggest keeping results may need more exercise over time. But that evidence was thin.

Did the type of exercise matter?

The team also ran separate analyses for six groups of exercise.

The best doses varied a little by group. For disability, Pilates, yoga and similar practices peaked at more frequent sessions, while strength and combined programmes peaked at about three a week.

Best session lengths ranged from about 20 to 45 minutes, depending on the type. For pain, 3 to 5 sessions a week and 60 to 120 minutes a week worked best across most groups.

The authors’ conclusion is that how exercise is delivered may matter more than which kind you pick.

Was it safe?

Only 68 trials, 28.5%, reported on side effects. Most reported problems were mild, such as a brief rise in pain or soreness.

Serious problems were rare and mostly unrelated to exercise. Because reporting was patchy, though, the true rate of minor problems may be higher.

Why might more exercise not help more?

Nobody knows for sure, and this review didn’t test mechanisms. One possibility is that exercise calms pain signals and rebuilds confidence in movement.

Those gains may come fairly quickly at modest doses. Beyond that, extra effort may bring more soreness and make it harder to keep going.

That’s informed guesswork, not a finding, so treat it as a hypothesis.

The authors note that higher doses may also raise the chance of minor side effects, without much extra relief.

What earlier research found

Exercise works, modestly

A 2021 Cochrane review pooled 249 trials. Exercise reduced chronic back pain by about 15 points on a 100-point scale, compared with no treatment, usual care or placebo.

It judged that a clinically important benefit, with moderate-certainty evidence. The gain in daily function was smaller.

Exercise versus other treatments

The same Cochrane review compared exercise with other options. Exercise probably eased pain more than education alone, by about 12 points, and more than other physical therapies, by about 10.

It did about as well as manual therapy, such as spinal manipulation.

Dose studied another way

A 2024 review in the Journal of Orthopaedic and Sports Physical Therapy looked at 82 trials. It measured dose as overall energy use, not minutes and sessions.

It also found a curved relationship, where more wasn’t always better. In that review, Pilates stood out, though certainty was very low to moderate.

How it fits together

The big picture is consistent. Exercise helps chronic back pain a bit, at least in the short term.

The new review adds that moderate doses may be enough, and that dose may explain why different studies crown different “best” exercises.

How much should you trust this?

Early. It’s the largest review of its kind, but the trials behind it are weak and the dose comparisons are mostly indirect.

What makes it convincing

• It pooled 239 randomised controlled trials.
• It modelled six separate parts of exercise dose, instead of one lump measure.
• The plan was registered and published before the analysis.
• Data and code are publicly available.
• The authors declared no competing interests.

What makes me cautious

• Certainty was very low to low for almost every estimate.
• 84% of trials had at least one outcome at high risk of bias.
• Only seven trials compared different doses directly.
• Most programmes were short, lasting a median of 8 weeks.
• Benefits mostly faded after three months, and long-term data were sparse.
• Exercise intensity couldn’t be analysed, so “dose” here means time and frequency, not effort.

What this means for you

If you live with chronic back pain, this is encouraging. You probably don’t need punishing daily workouts to feel some benefit.

• Aim for regular, moderate sessions. Around 3 or 4 sessions a week of 25 to 30 minutes is a reasonable target.
• Start small if you need to. Even one or two sessions a week may help.
• Choose something you’ll keep doing. Walking, Pilates, yoga, strength work or a class can all count.
• Keep going after the first few months. Benefits tended to fade, so exercise works best as a habit.
• Expect some soreness at first. A brief rise in pain is common and usually settles.
• Get checked if it isn’t improving. See a GP if pain lasts several weeks despite self-care or stops you doing daily activities.

The NHS has more on back pain, including exercises, treatments and when to get urgent help.

In this short NHS video, a physiotherapist demonstrates a simple Pilates exercise for back pain:

What we still don’t know

1. Does the benefit last? Few trials followed people beyond a year.
2. How hard should you exercise? Intensity wasn’t reported well enough to analyse.
3. Do different doses really differ? Only a handful of trials compared doses head to head.
4. What about sciatica? Only two trials included pain radiating down the leg.
5. Do higher doses cause more side effects? Harms were reported too inconsistently to say.

My take: the right dose is one you’ll keep

What I like about this review is its practical question. It asks how much, not just whether.

I’m cautious about the exact numbers. Most of the evidence is weak, and the best doses were estimated indirectly.

Still, the message is reassuring and useful. Moderate, regular exercise seems to do most of the work, and more isn’t always better.

For chronic back pain, consistency may beat intensity. The best plan is the one you can still follow next month.