Stroke Recovery Needs More Than Scans and Scores

A JAMA Network Open trial found that structured personalised music listening during acute stroke hospitalisation was feasible and linked with better mood and sleep. It is not a miracle intervention. It is a reminder that recovery is not only neurological mechanics.

July 28, 2026
Editorial
After stroke, recovery is not only movement and discharge; mood, sleep and dignity are part of the pathway too.[WHYFRAME] / Shutterstock.com

IPM Take

Stroke care can become brutally mechanical.

Find the clot. Reopen the artery. Stabilise the patient. Measure the deficit. Discharge. Somewhere inside that process is a person whose fear, sleep, mood and sense of self may be treated as secondary because they are harder to code.

The music-listening trial from Milan is small and single-centre. It does not prove that music should become standard stroke therapy. But it does ask a better question than many expensive interventions do: what would acute stroke care look like if well-being was treated as part of recovery, not decoration?

Executive Summary

A randomised clinical trial published in JAMA Network Open examined structured, personalised vocal music listening during acute stroke hospitalisation.

The single-centre study at IRCCS San Raffaele Hospital in Milan randomised 150 patients with acute ischaemic or haemorrhagic stroke to music listening or standard care. Participants were enrolled within 24 to 96 hours of symptom onset and had relatively mild stroke severity, with NIHSS scores below 10.

The intervention consisted of one-hour weekday sessions of personalised Italian vocal music delivered by tablet and headphones until discharge. The study found that the intervention was feasible and acceptable, with a 91% enrolment rate, 6.7% dropout and more than 80% of participants rating acceptability highly.

At discharge, the music group had lower depression and anxiety scores than controls. Music listening was also associated with better self-perceived sleep quality and lower pain or discomfort scores. The trial found no improvement in EQ visual analogue score, and its single-centre design, mild-stroke population and lack of active sham control limit generalisability.

Why it matters

  • Patients / advocates: Emotional recovery is not a luxury after stroke. Anxiety, low mood, pain and sleep disruption affect the way people return to life.
  • Clinicians: The trial supports further testing of a low-cost, low-burden intervention, but it should not be overread as proof of broad clinical effectiveness.
  • Hospitals / providers: Acute stroke units need to think beyond survival and neurological scores if they want recovery pathways to be genuinely person-centred.
  • Public authorities: Low-resource interventions deserve attention when they are feasible, safe and potentially scalable, especially where rehabilitation capacity is limited.

Modern stroke care is very good at what it can see clearly.

A blocked artery. A bleed. A score. A scan. A discharge destination. These are essential. Nobody should romanticise the acute science that saves lives and prevents disability.

But recovery does not live only in the scan.

A person in a stroke unit may be frightened, sleepless, disoriented, sad or suddenly aware that the body has become unreliable. Those experiences are not side notes. They shape participation in rehabilitation, confidence after discharge and the emotional meaning of survival.

The Milan trial is not dramatic medicine. That is part of its value.

Personalised vocal music listening is inexpensive, low burden and easy to imagine in routine care. In this trial, it was feasible, acceptable and associated with better mood and sleep outcomes by discharge. No intervention-related adverse events were reported.

But the limits matter. The trial was single-centre. Participants generally had mild strokes. There was no active sham control, so some effects may reflect attention, expectation or the simple fact of receiving a structured intervention. The findings should support larger studies, not instant policy declarations.

Still, the trial lands in an important place.

Health systems often treat psychological well-being after stroke as something to be picked up later, if there is time, if services exist, if the patient asks loudly enough. That is not person-centred care. It is delayed recognition.

Stroke recovery needs thrombolysis, thrombectomy, rehabilitation and secondary prevention. It also needs a pathway that recognises mood, sleep, pain and fear as part of the outcome.

A playlist will not fix stroke care. But it may expose how narrow stroke care has become.

Source & Evidence