Hospitals are not quiet. Monitors beep, trolleys pass, equipment hums, and patients waiting for surgery do so in a soundscape designed for staff rather than for them. Over the past two decades, researchers have asked a simple question about that environment: if you give a patient headphones and something they want to listen to, does anything measurable change? The answer has been studied far more rigorously than most people assume — and the researchers themselves are considerably more cautious than the headlines their work generates.
An unusually large body of randomised evidence
For a non-pharmacological intervention, perioperative music has accumulated a striking amount of randomised data. In 2015 The Lancet published a systematic review and meta-analysis by Hole, Hirsch, Ball and Meads that pooled 72 randomised controlled trials covering close to 7,000 adult surgical patients. Three years later, a meta-analysis in the British Journal of Surgery by Kühlmann and colleagues identified 92 randomised trials and pooled 81 of them, with 7,385 participants in total.
Few complementary interventions in any field have been tested on that scale. That does not make the findings definitive, as we will see, but it does mean the question has been taken seriously by surgical and anaesthetic researchers rather than only by music specialists.
What the trials actually reported
The Lancet analysis found that, compared with control conditions, patients who listened to music recorded lower postoperative pain scores (standardised mean difference −0.77), lower anxiety (−0.68), lower use of analgesia (−0.37) and higher satisfaction (1.09). Length of hospital stay, notably, was not significantly different — a null result the authors reported alongside the positive ones.
Kühlmann’s team translated their pooled effects into the scales clinicians actually use. Across the trials they analysed, music interventions were associated with roughly a 21 mm reduction on a 100 mm visual analogue scale for anxiety, and around 10 mm on the equivalent pain scale.
A Cochrane review by Bradt, Dileo and Shim, published in 2013, looked specifically at the period before an operation. Across 26 trials and 2,051 participants, music listening was associated with anxiety scores 5.72 points lower on the State-Trait Anxiety Inventory than standard care. In one of the larger included trials, with 327 patients, music listening compared favourably with midazolam on self-reported anxiety. The review authors suggested music interventions may represent a viable alternative to sedatives for preoperative anxiety in some settings — while stressing, in the same breath, that the evidence should be interpreted with caution.
Timing appears to make a difference
One of the more practical findings concerns when the music is played. Hole and colleagues reported that listening at any point around surgery appeared to be associated with benefit, with a trend favouring the period before the operation rather than during or after it.
Kühlmann’s analysis was more granular and pointed in a similar direction with an interesting split: the largest effect on anxiety came from music delivered preoperatively, while the largest effect on pain came from music delivered after the operation. If that pattern holds in future work, it suggests the intervention may be doing different things at different points in the pathway rather than exerting one general effect.
A finding that complicates the obvious explanation
The intuitive explanation for all of this is distraction: a patient listening to something familiar has less attention available for worry and discomfort. That account is plausible, but it does not fit everything in the data.
Kühlmann’s group reported that music played during general anaesthesia — when the patient is unconscious and cannot be consciously distracted by anything — was still associated with lower postoperative pain scores (mean difference −0.41). The authors took this as evidence that purely psychological mechanisms are unlikely to account for the whole picture. It is a single finding within a heterogeneous literature, and it deserves replication rather than extrapolation, but it is the kind of result that keeps a research question open.
The caveats the researchers insist on
Every one of these reviews carries substantial methodological warnings, and they matter more than the effect sizes.
Blinding is the central problem: a patient knows whether they are wearing headphones, and so does the person recording their pain score. The Cochrane authors rated the overall quality of the evidence as low and noted a high risk of bias across the included trials. Heterogeneity is a second issue — the studies differ in genre, who chose the music, the volume, the duration, the delivery method and the surgical procedure, which makes pooling them a blunt instrument. A third is that outcomes such as pain and anxiety are self-reported and susceptible to expectation.
None of this means the findings are empty. It means the honest reading is that the direction of effect is consistent across a large literature, while the size of the effect is uncertain and probably overstated by the weaker studies.
Why the question keeps being asked
Part of the reason perioperative music has attracted so much attention is pragmatic. It is inexpensive, it is non-invasive, it does not interact with medication, it can be stopped instantly, and it is one of very few elements of a hospital stay that a patient can control themselves. Hole and colleagues argued on exactly those grounds that it could reasonably be offered to patients having an operation, given the low likelihood of harm.
That is a different claim from saying music does the work of clinical care, and the distinction is worth holding onto. Anyone facing surgery has questions that belong with their surgical and anaesthetic team, not with a research summary. What the literature supports is narrower and still interesting: that something as ordinary as a familiar playlist, offered at the right moment, appears to be associated with measurable differences in how people experience a hospital stay.
For related reading on listening and arousal, see our pieces on music and stress and music and emotional balance.
Sources
- The Lancet — Hole, Hirsch, Ball & Meads, Music as an aid for postoperative recovery in adults: a systematic review and meta-analysis (2015). 72 randomised trials, nearly 7,000 patients. DOI: 10.1016/S0140-6736(15)60169-6.
- British Journal of Surgery — Kühlmann, de Rooij, Kroese, van Dijk, Hunink & Jeekel, Meta-analysis evaluating music interventions for anxiety and pain in surgery (2018). 92 randomised trials; 7,385 participants.
- Cochrane Database of Systematic Reviews — Bradt, Dileo & Shim, Music interventions for preoperative anxiety (2013). 26 trials, 2,051 participants. DOI: 10.1002/14651858.CD006908.pub2.
Image credit: Envato Elements.
This article summarises published research and is not medical advice. Decisions about care belong with a qualified healthcare professional.
Created with the help of artificial intelligence.
