Arts on Prescription Around the World: What Is Documented in Each Country

Four countries organise arts on prescription in four different ways, and the honest comparison between them is about documentation rather than about effect, because none of them has produced a controlled trial of it. The recurring error in international coverage is not exaggeration but conflation: a participant number from one phase of a programme attached to another phase’s dates, places or claims.

I have no international practice. My work is two care homes and a stroke rehabilitation unit in one part of England, and everything below comes from reading the reports rather than from visiting anything. What reading them in sequence taught me is a small and useful thing: the programmes that publish most are not the programmes that claim most. The Swedish reports are the most detailed documents in this field and among the most cautious, and the numbers that travel furthest internationally are the ones with the least paperwork behind them.

Where a claim about effect appears below it carries a tier (Tier 1 supported, Tier 2 promising but limited, Tier 3 practice and anecdote), explained on reading arts in health research.

The one fact that covers all of it

No randomised controlled trials of arts on prescription exist anywhere in the international literature.

The 2024 systematic review screened 7,805 records and included 25 studies, drawn from programmes in Australia, Denmark, Sweden, the United Kingdom and the United States. Of the 25, 10 were qualitative, 7 mixed methods and 8 quantitative studies using uncontrolled before and after designs. In the review’s own words, no randomised controlled trials were identified in the search. All the quantitative studies were rated medium quality, most commonly for an absence of comparator groups, and exactly one, by Bergman and colleagues, had a treatment as usual comparison. Most interventions ran for 8 to 10 weeks, in community settings, arts venues, GP surgeries, primary healthcare settings and one school. The pooled analysis found a statistically significant improvement in wellbeing, resting entirely on uncontrolled designs¹.

That places every national scheme below at Tier 3, and it is the reason this page compares how the work is organised rather than how well it works. Tier 3 is not a verdict that any of these schemes fails. It means the question is open, and the reasons are structural: these programmes are hard to randomise, they run on short funding, and their money goes into delivery.

Sweden: the best documented, and the easiest to misquote

Region Skåne has run Kultur på recept in at least four distinct phases across fifteen years, and the numbers people quote belong to different ones.

Phase	Dates	Place	Participants
1. Original pilot	1 Nov 2009 to 31 Dec 2010	Helsingborg, one health centre	24 in three groups of 8, 7 and 9
2. Kultur på recept 2.0	Jan 2012 to Dec 2014	Helsingborg	187 enrolled, 123 in the published 12 month follow up cohort
3. Primary care phase	2020 to 2021	Primary care, Region Skåne	not stated in the retrieved source
4. Malmö research study	Sep 2021 to May 2024, 12 groups	Malmö, 18 health centres and 9 cultural institutions	198 referred, 128 started, 112 completed, 16 dropped out, median age 61

The phase 1 report is dated 21 March 2011²; the phase 2 participant figures come from a 2017 account of the programme³; the Malmö figures are from the December 2024 research report⁴.

The trap is real and worth stating twice: 24, 187, 123 and 112 are four different numbers from four different studies spanning fifteen years. Sources routinely quote one figure with another’s dates or another’s place. There is also an interim figure in circulation for the fourth phase, of 15 care units, 9 cultural actors and 63 patients, which is superseded by the final report’s 18, 9 and 112.

What the Swedish material is genuinely good for is describing how a public health system organised the work: which professionals referred, how groups were formed, how long they ran, who dropped out and why. That is a description of a system, not evidence that the system produced health, and all of it is Tier 3.

The United States: a pilot, a statewide programme, and a number that jumped between them

The American scheme is the most quoted internationally and the most consistently misquoted.

The pilot. CultureRx involved 8 cultural organisations in phase I and 12 in phases II to IV.

The figure that travels. More than 300 cultural organisations is real, and it counts organisations signed on with the statewide Art Pharmacy programme for FY24 to FY25⁵. It is not the pilot cohort. Using it as the pilot’s size overstates it about twenty five times over, and that is the single most common error in coverage of this scheme.

The other self reported figure. More than 1,940 prescriptions across phases I to IV, FY20 to FY23, with phase IV alone accounting for over 1,200 issued and redeemed, from 10 of the 12 phase III organisations and over 30 prescribers⁶. This is a programme self report with no evaluation document behind it.

The independent evaluation. The peer reviewed evaluation of the same programme reports much smaller numbers for the phase it covered: 12 cultural organisations, 20 healthcare providers, 414 referrals in phase III, 363 participants, and participant data on 84⁷. Neither the 300 plus nor the 1,940 appears in it.

Where the self reported figures and the evaluation disagree, the evaluation is the citation of record here. That is not an accusation of bad faith. The figures measure different things over different periods, and the difficulty is entirely in how they are used afterwards. The museum and gallery side of this programme is covered on museum prescriptions.

The United Kingdom: an infrastructure without a documented programme

The UK has the most developed referral machinery of the four and the least documented programme.

The route runs through social prescribing link workers attached to primary care networks, with the arts sitting inside a much wider set of non clinical referrals covering debt advice, housing, exercise, befriending and practical help. The national body describes the role and what it is meant to achieve⁸, and the practical detail is on social prescribing explained.

What the UK does not have is a published national cohort in the Swedish manner. Its most quoted scheme is Artlift in Gloucestershire, and the document behind the figures is a cost benefit summary written by a GP in December 2011, which analysed 90 patients out of roughly 500 referred, had no control group, described itself as “a simple observational study”, and stated in its own conclusion that it “does not imply causality”⁹. Its three headline numbers come apart on inspection: the 37% is the months 7 to 12 figure against the report’s own full year figure of 24%; the 27% is a reduction in overall spend rather than in admissions; and the per patient saving divides a numerator computed on 90 people by a denominator computed on 500. The whole trail, with the arithmetic, is on does arts on prescription save money.

So the comparison is not flattering in the direction people expect. The country with the best referral infrastructure is the country leaning hardest on the weakest document.

Australia and Denmark: present in the literature, absent from this page’s figures

Both appear among the countries covered by the 2024 review’s included programmes¹, and this site holds no separate locked figures for either, so none are quoted here.

That is a gap in what has been traced rather than a claim that little is happening in either country. It is worth stating explicitly, because a page comparing countries that silently omits two of them implies an absence that may not exist. What would fill it: a national or regional programme report, or a peer reviewed evaluation, with dates, places, participant numbers and design, in a document a reader can open.

One structural difference is worth noting for Australia and New Zealand, on the therapy side rather than the community side. Creative arts therapies there, covering art, drama, dance movement and play based practice, sit under a membership register rather than a statutory one¹⁰, which is a different arrangement from the UK’s four protected titles. That is about the regulated half of this field rather than about arts on prescription, and it matters when comparing what a practitioner’s credentials mean across borders: see dance movement psychotherapy and training as an art therapist.

What the international picture is actually good for

Two things, and neither of them is proof.

Showing that the model is deliverable. Five countries, multiple health systems, different funding arrangements, and in every case somebody managed to get people from a health setting into a creative activity and keep a group running for 8 to 10 weeks. That is a real finding about feasibility, and feasibility is not nothing.

Showing what an honest programme report looks like. The Swedish reports name their phases, print their dropout numbers, and do not claim causation. Read next to the documents that get quoted most often, they are a lesson in how this could be done everywhere.

What it is not good for is settling whether the model improves health, and no amount of international breadth substitutes for a comparison group. The habit of reading reports as descriptions rather than as evidence is the subject of creative health policy and reports.

The rule for quoting any of this

Never quote a participant number without its phase, its years and its place.

That one rule would prevent almost every error on this page, and arts on prescription sets out what the term means before any country gets involved. Applied to Sweden it stops the four phases merging. Applied to the United States it stops a statewide figure standing in for a pilot. Applied to the UK it keeps the 90 and the 500 apart. And applied to any new scheme that appears, it forces the question of which document the number came from, which is the question that does all the work.

If you are trying to reach a scheme rather than to assess one, how to get referred to an arts scheme covers the ask that works, and who pays for arts on prescription covers why schemes in every one of these countries tend not to outlast their funding. Nothing on this page is a reason to change, delay or decline any part of anybody’s treatment, and no arts scheme substitutes for one.