The short version
- Acute-stroke patients given an hour of personalized vocal music each weekday scored 2.68 points lower on depression and 1.86 lower on anxiety than those on standard care, with no adverse events.
- The intervention was cheap and well tolerated: 91% of eligible patients agreed to join, only 6.7% dropped out, and the median patient listened for about 6 hours in total.
- The trial was unblinded and relied on self-report, and it did not measure whether the music changed motor or functional recovery.
Stroke patients who spent an hour each weekday listening to music they had chosen themselves left the hospital less depressed, less anxious and sleeping better than patients who got standard care alone, in a randomized trial of 150 inpatients at San Raffaele Hospital in Milan published in JAMA Network Open.
The design
G. Giacalone and colleagues on the San Raffaele stroke unit randomized 150 people admitted with acute stroke, 75 to each arm, mean age 67.9. The music group received a tablet and headphones loaded with personalized Italian vocal music, chosen with the patient, and listened for one hour every weekday until discharge. The control group received standard stroke-unit care.
The numbers on uptake are part of the result. Of eligible patients, 91% agreed to enroll, and 6.7% dropped out. The median patient accumulated about 6 hours of listening before going home, which gives a sense of how short acute stroke stays are and how little intervention this was.
The results
At discharge, measured on standard self-report scales:
- Depression (HADS-D): 2.68 points lower in the music group.
- Anxiety (HADS-A): 1.86 points lower.
- Insomnia (Insomnia Severity Index): 2.83 points better.
All three differences were statistically strong (p < 0.001). On the EQ-VAS, a single 0-to-100 rating of overall health, there was no difference between groups. There were no adverse events.
The HADS scales run from 0 to 21 for each of depression and anxiety, so a 2.68-point gap is a real but moderate shift. It is not a cure for post-stroke depression, which is common and slows recovery.
The limits
The trial could not be blinded. Patients knew whether they were listening to music, and the outcomes were all things patients reported about themselves. Some of the effect is likely the attention, the tablet and the sense of being given something, and the trial cannot separate that from the music.
The outcomes stop at mood and sleep. The trial did not measure motor recovery, language, or function at three or six months, which are the endpoints stroke neurologists most want moved. It also did not follow patients past discharge, so whether the mood benefit lasted a week is unknown.
And the music was Italian vocal music chosen with each patient. The design says “personalized,” and the result probably depends on it.
Why it matters
That last point connects to a broader body of work. A scoping review of 57 studies published in PAIN Reports this autumn by Joke Bradt of Drexel University, looking at why music eases pain, identified three mechanisms that keep recurring: a positive emotional response, a sense of control from choosing one’s own music, and synchronized movement such as tapping or singing along. “Singing along is more effective than just listening to music,” Bradt told Medical Xpress.
The Milan trial used the first two, emotion and choice, and left the third on the table. Patients listened; they were not asked to sing. Given that the strongest effects in the pain literature come from active music-making, there is an obvious next experiment.
National Geographic reported in August on the Johns Hopkins Center for Music and Medicine and the broader push to prescribe music for neurological conditions including stroke, a field with more enthusiasm than large trials. This one is modest in its claims, cheap to run, and easy to copy, which is most of what makes a hospital intervention spread.
What to watch
A follow-up from the San Raffaele group with outcomes at three months, and a version that adds singing or rhythmic movement to the listening hour. A stroke-ward tablet of a patient’s own music costs almost nothing; whether it earns a place on the discharge checklist depends on whether the mood effect outlives the hospital stay.
