Skip to content
Arts in Health Institute

Music, art, dance and drama in health care: what the trials actually show, what is still only promising, and how people get referred.

Music, art, dance and drama, and the health claims made for them.

£60 an hour for art therapy for anxiety. Is that worth it when the research seems thin?

Does it actually work? · started Mar 2, 2026 · 5 replies Locked

#1mikeatthepiano(Joined Feb 2025 · 19 posts)March 2, 2026, 1:05 pm

I have a consultation booked and a number in my head and I keep going round in circles, so I am going to put it here and let people be blunt at me.

Sixty pounds an hour. She suggests twelve sessions to start with and says that is a starting point rather than a course, so realistically I should be budgeting for more. That is a real amount of money for me. Not impossible, but it is the boiler fund.

The anxiety is not dramatic and I am not in crisis. It is the low grade constant kind that has been there for about three years and has quietly taken things off me, and talking therapy through the NHS I have done twice and it helped both times and then wore off both times.

So my problem is this. I went looking for the evidence on art therapy and anxiety, expecting to find something one way or the other, and I have found a great deal of enthusiasm and almost no numbers. And I do not know whether that means it does not work, or that nobody has bothered to check, or that I am bad at searching. Sixty pounds an hour on the third possibility is fine. On the first one it is not.

#2stillpainting62(Joined Nov 2024 · 26 posts)March 2, 2026, 3:20 pm

You are not bad at searching. I looked too, months ago, and came back with the same nothing.

#3Marion K.(Joined Jul 2024 · 38 posts)March 4, 2026, 9:48 am

I have paid for something with thin evidence and I will tell you how that went, though I am aware it is one person and proves precisely nothing.

I paid for private counselling for eight months after my father died. Not art therapy, but the same shape of decision: a real cost, a person I liked, and a literature I did not read because I did not want to know. It was worth it to me. I cannot tell you whether it was worth it in the sense you are asking about, because I have no idea what would have happened if I had spent the same money on eight months of something else, or on nothing.

What I would say is that the honest version of the question is not "does art therapy work". It is "is this particular hour, with this particular person, better for me than what I would otherwise do with the money and the Tuesday". Nobody can research that and it is still a real question.

#4Miriam HalsteadModerator(Joined Mar 2024 · 208 posts)March 6, 2026, 11:32 am

I cannot settle this for you and I want to say so early rather than after four paragraphs that look like they are building to an answer.

On the fee first, because I can be more useful there than you might expect. There is no art therapy fee figure anywhere on this site, and that is deliberate rather than an oversight. BAAT publishes fee guidance to its members only, so any number you see attributed to BAAT has come from somewhere else and been misattributed. The one firm cost figure held here is the British Association for Music Therapy's recommended floor of "from £52.00 upwards" for an individual session, and note all three things about it: it is a recommended floor rather than an average, it is a recommendation rather than a market rate, and it is music therapy rather than art therapy. So your sixty pounds is not obviously out of line with the neighbouring profession's own floor, and that is genuinely all I can tell you about the price.

On the evidence. Your reading is right and the answer is the third of your three possibilities, not the first. There is no locked effect size on this site for art therapy and anxiety, and where that is the case a claim sits at Tier 3. What Tier 3 means matters enormously here: it is a statement about the literature, not about the world, and it comes in two flavours. One is that the controlled evidence exists and is null. The other is that nobody has looked properly. Art therapy for anxiety is the second kind. An absent file is not a negative result, and I would rather tell you that plainly than let the absence do the work of a verdict.

What I would check instead of a number, since a number is not available.

That she is on the register, which takes under a minute and is free. Art therapist and art psychotherapist are two separate protected titles covering the same profession, so either is fine, and a directory listing is not a register.

What an assessment involves and what comes out of it in writing. Weekly sessions with no assessment and no written aim is the thing I would walk away from, at any price.

What ending looks like, asked before you start rather than in month five.

And what a term of the work totals rather than what an hour costs, along with what is included in it: assessment, any reports, reviews, travel. Twelve sessions at sixty is a number you can compare with other things you might do; sixty an hour on its own is not.

Two limits on me. A register is a floor rather than a ranking, so it establishes qualification and a complaint route and tells you nothing about whether she is any good or right for you. And no individual practitioner is recommended here, by me or anybody. Art therapy covers what the work actually is and is explicit about quoting no effect size, and reading arts in health research is the piece for working out what kind of absence you are looking at.

#5Yusuf A.(Joined Aug 2025 · 11 posts)March 9, 2026, 8:14 pm

There is one controlled trial people will throw at you in this conversation and I would rather you met it here than had it quoted at you badly.

MATISSE, Crawford and colleagues, BMJ 2012. Three arm, rater blinded, pragmatic randomised trial across four UK centres. 417 people randomised: 140 to group art therapy, 140 to activity groups as an active comparator, 137 to standard care alone, with 355 of them, 85 percent, followed up at two years. On the two primary outcomes at 24 months there was no difference between the arms. Global functioning came out at an adjusted mean difference of -0.9 with a confidence interval of -3.8 to 2.1, and symptoms at 0.7 with an interval of -3.1 to 4.6. No secondary outcome favoured art therapy either.

One number belongs next to that result: only 86 of the 140 people randomised to art therapy, 61 percent, attended even one group. So it is a pragmatic test of referral rather than a test of therapy received, which is an honest limitation and not a reason to wave the result away.

Now the part that actually matters for your question. That trial tested referral to twelve months of weekly group art therapy, as an adjunct, for people with established schizophrenia, on those two outcomes. It does not test individual art therapy, it does not test anxiety, and it does not test you. If somebody quotes it at you as proof that art therapy does not work, they have generalised it a very long way past what it examined. And if somebody quotes the cost finding at you, the report says two things at once: total costs in the art therapy arm were lower than in the comparison arms, and referral was still judged not a cost effective use of resources.

So it is a real null result, in a different question. It is the nearest controlled evidence there is and it is not evidence about your sixty pounds.

#6mikeatthepiano(Joined Feb 2025 · 19 posts)April 13, 2026, 6:26 pm

Six weeks later and I still have not decided, which I appreciate is an unsatisfying place to leave a thread.

I went to the consultation. I liked her. She did an assessment, she wrote things down, she talked about what the twelve sessions would be for and what would happen at the end, and every one of the things Miriam said to check was there without me having to ask. She is on the register.

And I am still sitting on it. Because the honest position is that I now know there is no evidence either way, which is not the same as good news, and I know she is properly qualified, which does not tell me it will help. So it comes down to whether I want to spend the boiler fund on a maybe, and that is a question about me and not about art therapy, and I keep discovering that I do not want to answer it.

No conclusion. Sorry. If somebody finds this thread hoping it resolves, it does not, and I think that is genuinely where the subject is rather than where I am.

Locked: 60 days went by without a reply. If something in an old thread matters to you now, start a new one rather than adding to this. And if the question is about a person in front of you, whether a relative should be referred, whether a session is doing harm, that belongs with the clinician who knows them, not with us.

More from Does it actually work?