Does massage help low back pain?

Low back pain is the most common reason people book a remedial massage, and it is the condition where the evidence is most often overstated. It is also where Australia has published a national care standard that says something most clinics would rather not repeat.

The short version

• The Cochrane review of 25 trials found short-term pain relief only, and its authors wrote that they have "very little confidence that massage is an effective treatment" for low back pain.¹
• Australia's national care standard tells patients that what they do themselves is more likely to help than treatments done to them, and names massage as one of the treatments done to you.²
• Every major guideline that includes massage attaches the same condition: only as part of a package that includes exercise or physical activity.³²
• The type of massage did not decide the outcome. In the one large head-to-head trial, relaxation massage did at least as well as structural massage.⁴
• It is safe. No serious adverse events across 25 trials. The common one is being sore afterwards.¹

What the Cochrane review found

The 2015 Cochrane review pooled 25 randomised trials covering 3,096 people with non-specific low back pain. It graded its own evidence "low" to "very low" throughout, mostly for risk of bias and imprecision.¹

Compared with inactive controls, which means sham treatment, a waiting list or nothing:

Compared with active controls, meaning manipulation, mobilisation, TENS, acupuncture, traction, relaxation, physiotherapy, exercise or self-care education, massage came out slightly ahead for pain at both short-term follow-up (SMD -0.37) and long-term follow-up (SMD -0.40, 95% CI -0.80 to -0.01, which only just clears no effect). For function there was no difference against active controls at any point.¹

Read plainly: massage beats doing nothing in the short term, and is roughly interchangeable with the other things you could do instead. That is the finding, and the review's own summary of it is the sentence quoted above.

It sits inside a wider pattern. An overview of 129 systematic reviews of massage for pain published between 2018 and 2023 found not one conclusion rated high certainty.⁵ The uncertainty is not specific to back pain. It is the state of the field.

What Australia's own care standard says

The Australian Commission on Safety and Quality in Health Care published the Low Back Pain Clinical Care Standard in 2022. Its consumer information puts the point without hedging:

Things that you do yourself to control your pain (like pacing yourself when carrying out physical activity) are more likely to help than treatments that are done to you (like medicines, massage or surgery).

The clinician-facing text is consistent with it, asking clinicians to prioritise "active management strategies (such as physical activity, social connection, healthy sleep habits and use of heat) over passive strategies (such as pain medicines, activity avoidance and massage)".²

The same standard does not rule massage out. It says physical therapies such as heat wraps and massage "may also help to improve function and mobility, but only as part of a treatment package including physical activity, with or without psychological therapy, and only for a short period of time".²

Both halves are true and neither is usually quoted. Massage is a supporting act in this standard, not the treatment, and it is explicitly framed as short-term.

One more line from the standard is worth carrying, because it comes up in the treatment room constantly. Quality statement 3 is titled "Reserve imaging for suspected serious pathology", and asks clinicians to discuss with patients the limited role of imaging in diagnosing low back pain.² A scan that shows disc changes is not, on its own, an explanation for the pain.

Where the international guidelines land

Three bodies, three jurisdictions, one shared condition attached: massage is something you add to movement, not something you substitute for it. No guideline anywhere recommends it on its own.

Does the type of massage matter

Not as much as it is sold. In 2011 a randomised trial put 401 adults with non-specific chronic low back pain into three arms: structural massage, relaxation massage, or usual care.⁴

At 10 weeks, against usual care, the relaxation arm improved by 2.9 points on the Roland Disability Questionnaire (95% CI 1.8 to 4.0) and 1.7 points on symptom bothersomeness (95% CI 1.2 to 2.2). The structural arm improved by 2.5 points (95% CI 1.4 to 3.5) and 1.4 points (95% CI 0.8 to 1.9).

The trial set its own thresholds for a clinically meaningful difference in advance: 2 points on the disability questionnaire and 1.5 points on bothersomeness. Against those thresholds the relaxation arm cleared both and the structural arm cleared one. The authors' conclusion was that there was no clinically meaningful difference between the two.⁴

At 52 weeks, what was still measurable was a benefit on function, not on symptoms, and it came from the relaxation arm. The authors describe it as small.⁴

The remedial or relaxation page works through what this means for choosing between them. The short answer is that the technical label is not the thing deciding your outcome.

So is it worth booking

On this evidence, a defensible position looks like this.

What can be said. Massage relieves low back pain in the short term better than doing nothing, it does so about as well as the other conservative options, it is safe, and a session is a pleasant hour that many people find makes movement easier for a while. Guidelines in three countries allow for it as part of a package.

What cannot be said. That it treats the cause, that it produces lasting change, that the technique used is what makes the difference, or that it beats getting moving. Australia's own care standard says the opposite of that last one.

The most useful thing a session can do for chronic low back pain is make the next few days of activity easier, so the activity happens. That is the mechanism every guideline is pointing at, and it is a smaller claim than the industry makes.

Safety, and when to see a GP first

Across all 25 trials in the Cochrane review there were no serious adverse events. The most commonly reported adverse event was increased pain, in between 1.5% and 25% of participants depending on the trial.¹ Being sore for a day afterwards is normal and is not a sign of anything working or not working.

See a doctor before booking, rather than after, if back pain comes with new numbness or weakness in a leg, any change in bladder or bowel control, unexplained weight loss, fever, or if it followed significant trauma. Massage is not the right first stop for any of those.

Frequently asked questions

Does massage help lower back pain?

In the short term, yes, better than doing nothing. The Cochrane review of 25 trials found short-term improvements in pain and function against inactive controls, but no lasting difference, and it graded the evidence low to very low certainty. Against other conservative treatments, massage performed about the same.

How many massages do I need for back pain?

The back pain evidence does not answer this. The clearest dose finding in the massage literature comes from neck pain, where a Cochrane subgroup analysis favoured eight or more sessions of 30 minutes or longer over about four weeks. Treat that as the best available guide rather than a result proven for backs.

Is remedial massage better than relaxation massage for back pain?

Not in the one large trial that compared them. In 401 people with chronic low back pain there was no clinically meaningful difference between structural and relaxation massage, and the small benefit still measurable at 52 weeks came from the relaxation arm.

What do Australian guidelines say about massage for back pain?

The national Low Back Pain Clinical Care Standard tells patients that things they do themselves are more likely to help than treatments done to them, and it names massage among the latter. It allows massage as part of a treatment package that includes physical activity, and only for a short period.

Can massage fix the cause of my back pain?

There is no evidence for that. Most low back pain is non-specific, meaning no single structure can be identified as the cause, and massage has not been shown to produce lasting structural change. What it can reliably do is reduce pain in the short term, which can make it easier to move.

Is massage safe for back pain?

Yes for most people. No serious adverse events were recorded across the 25 trials in the Cochrane review. Increased soreness afterwards was the common complaint. See a doctor first if there is new leg numbness or weakness, any bladder or bowel change, unexplained weight loss, fever, or recent significant trauma.