The short version
- When 15 people with dementia took part in both a singing group and a matched conversation group, singing eased the burden on care staff and conversation improved residents’ wellbeing.
- During the sessions themselves, residents were just as engaged and showed the same emotions whether they were singing or talking.
- A much larger UK trial found choir singing did not lift depression in care-home residents with dementia, so the honest read is that company and attention may matter as much as music.
Singing groups are one of the most common activities in dementia care, and one of the most praised. A small Dutch study, published in September in the Journal of Alzheimer’s Disease, asked a question most studies skip: does the music do the work, or would any warm, engaging group activity do the same?
The design
The team, from the University of Amsterdam with colleagues at Amsterdam UMC, Leiden University Medical Center and the Amsterdam ensemble Genetic Choir, ran a randomized cross-over study. Fifteen nursing-home residents with dementia took part in two group activities, each for five weeks, in random order. One was a singing group. The other was a conversation group built to match it in every way except that there was no music.
That matched comparison is the point. Many music studies compare a music group with usual care, which means residents in the music group also get extra company, attention and something new to do. Here both arms got all of that.
Before and after each five-week block, staff rated behavioral symptoms and their own burden with the Neuropsychiatric Inventory, and residents’ wellbeing with a quality-of-life scale for late-stage dementia. During sessions, observers rated how engaged residents were and what emotions they showed.
What they found
Each activity helped in its own way. The singing block reduced the burden reported by professional caregivers. The conversation block improved residents’ wellbeing scores. During the sessions, engagement and observed emotion were the same in both.
The authors’ summary is that “both singing and conversation may benefit people with dementia,” and that neither clearly outperformed the other. The paper is behind a paywall, and the abstract gives no effect sizes, so we cannot say how large any of these changes were.
The limits
Fifteen people is a pilot, and the authors call it one. Its main job was to test whether this kind of design is workable. With so few residents, a real difference between singing and talking could easily be missed, and the two positive results could be chance. The abstract does not say whether the staff doing the ratings knew which activity had just run.
The bigger picture
This is not the only recent study to find music level with its comparison. The UK arm of the MIDDEL trial, published in December 2025, randomized 16 care-home units with 192 residents who had dementia and depressive symptoms. They got group music therapy, recreational choir singing, both, or usual care for six months. Neither music option reduced depression, and residents in the choir arm ended with higher scores on the main depression scale than those without it. The authors concluded their findings “do not support the use” of either to relieve depression in this group. They also note the trial started straight after the COVID-19 lockdowns, with heavy staff turnover.
In Singapore, a two-year trial of about 200 older adults without dementia found that weekly choir singing and a weekly health-education group both cut the odds of screening positive for depression and anxiety, by similar amounts by the end.
None of this says music is useless. A Finnish trial we covered found that music sessions held autobiographical memory steady over 10 weeks. But the pattern suggests that much of what a singing group gives may come from being together and being noticed, which a good conversation group also gives. That is useful news for care homes without a choir leader. It also matters for how music programs are described. A group sing is a lovely way to spend an afternoon, and that is a different claim from treatment. Our explainer on music therapy, music medicine and sound healing sets out where those lines fall.
What to watch next: a full-sized version of the Amsterdam cross-over design, with enough residents to show whether singing adds anything that talking does not, measured by raters who do not know which group a resident was in.
