Does Music Reduce Pain? The Two Reviews People Quote, and the Outcome They Get Mixed Up

The most quoted evidence on music and pain comes from cancer care, it reports a pain effect of SMD -0.67 at very low certainty, and that combination is exactly the kind of claim this site exists to describe accurately. There is a signal. The confidence in it is poor. Both halves of that sentence have to travel together, and in most retellings only the first half survives.

The reason I wanted this page is that music and pain is the claim I hear repeated most often by people who have never read anything about it. It comes up in care homes, in waiting rooms and on the radio, always in the same shape: music is a natural painkiller. What sits underneath that sentence is a real literature with a real problem, and the problem is not that the studies are few.

Where a claim about effect appears below it carries a tier (Tier 1 supported, Tier 2 promising but limited, Tier 3 practice and anecdote), explained on reading arts in health research.

What the cancer review reports

Eighty one trials, five and a half thousand people, and very low certainty.

The 2021 Cochrane review of music interventions in cancer care pooled 81 trials and 5,576 participants. It reported anxiety at -7.73 STAI-S units and pain at SMD -0.67, and it rated both findings very low certainty¹. On this site both are Tier 2: controlled evidence that exists in quantity but at a certainty rating too low to treat the number as settled.

That pair of facts is the best teaching example in the whole of arts in health, and it is worth sitting with rather than skimming. Eighty one trials. Five thousand five hundred and seventy six participants. Very low certainty. The number of studies did not rescue the certainty rating, because certainty is about how the studies were done rather than how many there were: risk of bias in the individual trials, inconsistency between their results, imprecision, and how directly they answer the question being asked. A reader who has understood why a large pile of studies can still produce a weak conclusion has understood most of what matters about reading this field.

The practical translation of very low certainty is not “ignore it”. It is that the true effect could be meaningfully different from the estimate in either direction, and that a later, better trial could move it. If somebody quotes you the SMD without the certainty rating, they have quoted you half a finding.

The figure that is not about pain

The strongest music figure in this literature measures anxiety before an operation, and it gets recruited into pain arguments constantly.

The 2013 Cochrane review of music for preoperative anxiety pooled 26 trials and 2,051 participants and reported a reduction of 5.72 units on the STAI-S anxiety scale (95% CI -7.27 to -4.17)². That is Tier 1, with one caveat stated every time it is used here: the review predates routine certainty rating, so no certainty grade should be attached to it in either direction. The trial count, the sample and the interval are what it has, and they are enough to make it the cleanest music finding on this site outside stroke rehabilitation.

It is also, unambiguously, an anxiety result. Anxiety and pain travel together in ordinary speech and in ordinary bodies, and they are separate outcomes measured on separate instruments. A reduction on a state anxiety scale before an operation is not a statement about how much anything hurt afterwards, and a page that lets the two blur has done the same thing as the funding bid that puts a spend percentage next to an admissions row. Music before surgery and anxiety covers that review on its own terms.

Calibration: what a Tier 1 pain claim would look like

Compare it with the one place on this site where the label is unarguable.

Rhythmic auditory stimulation after stroke improved gait velocity by 11.34 m/min (95% CI 8.40 to 14.28), from 9 trials and 268 participants, at moderate quality³. Tier 1. Five things printed next to the claim: effect size, interval, certainty, sample and a review you can open. Notice also how narrow it is. It is about walking speed and it says nothing about mood, communication or quality of life, and the discipline of not letting it drift is covered on rhythmic auditory stimulation.

A Tier 1 pain claim for music would need the same five things plus one more: an active comparator. A trial of music against nothing tells you what happens when somebody is given an hour of attention, a device and a reason to lie still. A trial of music against another quiet activity of the same length tells you what the music contributed. Those are different questions, and only the second one supports the sentence people want to say.

Music medicine and music therapy are not the same purchase

A nurse handing someone headphones and a session with a registered therapist are different interventions, and the difference has practical consequences.

Recorded music chosen by a patient, or offered by ward staff, is cheap, requires no referral and can start tonight. That is a genuine advantage and no arts in health site should be sniffy about it. A registered music therapist works differently: assessment, goals for a named person, notes, clinical supervision and a planned ending. In the UK that work sits behind a title protected in law, one of four such titles held on a register that listed 6,103 arts therapists in July 2026⁴, and privately the British Association for Music Therapy gives a recommended floor of from £52.00 upwards for an individual session⁵.

This site has not traced a split of the reviews above into therapist delivered and staff delivered music, so no split is printed here. What follows from that is a reading habit rather than a conclusion: when a study is quoted at you, ask which of the two it tested, because the answer changes what the result is worth to you. Personalised playlists for dementia works through the same distinction in a different setting, and music therapy covers what the clinical version involves.

What is reasonable to take from this

A modest, low risk addition to care, with poor certainty behind the number and no claim to be an alternative to anything.

The wider mapping supports that framing rather than a stronger one. The WHO Europe scoping review gathered over 900 publications, including 200 plus reviews which themselves cover over 3,000 studies, plus 700 plus further individual studies, and it maps a literature rather than pooling it⁶. It is evidence that the field has been studied. It is not evidence that anything in it works, and it carries no tier here because it produces no effect estimate.

If you are living with pain, the useful version of all this is small. Music you actually like, available when you want it, costs almost nothing and carries almost no risk. Whether it does anything to your pain is a question the literature answers weakly, and your own experience of it is not a trial result but it is not worthless either. What matters more than either is this: pain that is new, worsening or unexplained is a clinical matter and belongs with the team treating it today. Music therapy in palliative care covers the setting where these questions come up most often, and where the measuring is hardest.