Let’s start with the fact that there’s a good Cochrane review on music therapy for people with autism spectrum disorder. It included ten studies of 165 participants, all RCTs or controlled trials, that compared music therapy to placebo or standard care. Music therapy did well:
The findings of this updated review provide evidence that music therapy may help children with ASD to improve their skills in primary outcome areas that constitute the core of the condition including social interaction, verbal communication, initiating behaviour, and social-emotional reciprocity. Music therapy may also help to enhance non-verbal communication skills within the therapy context. Furthermore, in secondary outcome areas, music therapy may contribute to increasing social adaptation skills in children with ASD and to promoting the quality of parent-child relationships.
But these were small studies, and the outcomes didn’t overlap between them often, and some people still had questions.
Enter the TIME-A RCT. Just published, it’s a randomized trial of 364 kids in 9 countries. It was assessor blinded, and it had reasonably strict entry criteria. It used a widely accepted primary outcome, the Autism Diagnostic Observation Schedule (ADOS). Outcomes were assessed at 2, 5, and 12 months.
Everyone got “enhanced care”, which was usual care plus parent counseling. The intervention group also got improvisational music therapy, where trained therapists sang or played music with each child, to encourage affect sharing and joint attention.
Most of the kids (86%) completed the primary end point of 5 months, and most completed the last end point (80%). Over the primary intervention period, kids got a median 19 sessions of music therapy in addition to other care.
In the music therapy group, the ADOS social affect scores improved from 14.08 to 13.23. In the control group, they improved from 13.49 to 12.58. No difference. Of the 20 exploratory secondary outcomes, 17 showed no difference as well.
You can scream “one trial!” and you’d be right. But this one trial was larger than all the other trials put together and then doubled. You can scream that it didn’t test every type of music intervention, and you’d be right. But it was pretty pragmatic and well thought out.
Most of those previous trials were limited to one local area and one therapist. How generalizable were they? This was multicenter and multicountry.
In 2017, there were about 7000 music therapists in the US. I’m sure they won’t like these results.
The accompanying editorial is fair. It makes balanced arguments both for and against continued focus on musical therapy. I’m going to be honest and say that I fall into the “doesn’t seem to work” camp now. I say that because recognizing how common ASD is, and how important it is to intervene, I want to see time and money go to things that work, not where they might not.
Regardless, whether we do more trials in this area, or abandon music therapy is up to the ASD research and patient communities. Supporters should be pushing for more research, though, not more money to therapies that lack a good evidence base.