Arts for Carers and Health Staff: Enthusiasm Running Well Ahead of the Evidence

Arts programmes for health staff and unpaid carers are the fastest growing part of this field and the thinnest evidenced part of it, and no effect size for them is locked in this site’s source document, so none appears on this page. The interesting question is not whether the enthusiasm is misplaced, which is a lazy conclusion, but why this particular corner is so difficult to test.

The staff who join in with my sessions are always the same kind of person and always in the same circumstances. A healthcare assistant sings along from the doorway while she waits for somebody to finish a drink. A nurse who was a chorister once stays for two songs on her way past. What has never happened, in several years across three settings, is a member of staff sitting down for the whole hour, because there is no hour in which they are not needed somewhere else. Any evaluation of a staff arts programme has to reckon with that, and most do not. If I ran a session for staff and measured everybody who came, I would be measuring the people who could get away, which is not the group anybody is worried about.

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.

What is on offer

A fairly consistent set of things, delivered under a fairly consistent set of constraints.

Staff choirs. Lunchtime art sessions and drop in making spaces. Reflective writing groups, sometimes attached to Schwartz style rounds or a similar reflective forum. Photography and exhibition projects using staff work. Commissioned artwork in staff rooms and rest areas. Creative sessions attached to induction or to bereavement support. For unpaid carers, the pattern is different again: carer choirs, craft groups run through carers’ centres, and creative sessions attached to a group the person they care for is already attending.

The constraints are the same almost everywhere. Sessions are short, usually thirty to sixty minutes. Attendance is voluntary. Time is rarely protected, so people attend in a break, before a shift, or in their own time. Funding is short term and frequently charitable. And the person running it is generally an artist or a facilitator rather than anybody clinical, which is worth stating because these sessions are sometimes described in language borrowed from therapy. A registered arts therapy is a different thing, delivered by somebody holding one of four titles protected in UK law, on a register that listed 6,103 arts therapists in July 2026¹.

Why this is unusually hard to evaluate

Four problems, and the first is close to fatal for a voluntary programme.

Attendance selects against the target group. The staff most affected by burnout are, almost by definition, the least able to leave the floor for an hour. A before and after measurement of attendees therefore describes people who had time, which is the opposite of the population the programme exists for. This is not a small bias that a larger sample would fix; it is built into voluntary attendance in an understaffed setting.

Blinding is impossible and expectation is high. Nobody is unaware that they spent an hour singing, and people who sign up for a creative session tend to be people who like creative sessions. Every self reported outcome carries that.

The outcomes are contested. Burnout scales, wellbeing scales and job satisfaction measures all move for reasons unrelated to any programme: a change of rota, a difficult week, a national dispute. Harder outcomes such as sickness absence and retention are noisy at the scale of one department and confounded by everything happening in the organisation.

The comparator is almost never right. A programme compared against nothing answers a question nobody was asking. The interesting comparison is against another protected hour doing something else, because that separates the arts element from the plain fact of being given time away from the floor. Very little work does that.

The neighbouring literature shows what happens when these problems go unaddressed. The 2024 systematic review of arts on prescription, which is about participants rather than staff, screened 7,805 records, included 25 studies, and found no randomised controlled trials at all: 10 qualitative, 7 mixed methods and 8 quantitative studies using uncontrolled before and after designs, rated medium quality most commonly for an absence of comparator groups². Staff programmes are, if anything, less well studied than that.

Why no figure appears here

Because none is locked, and quoting one from a summary would be the practice this site exists to correct.

The rule is that a figure is printed only after somebody has opened the study or review that produced it and recorded its design, sample size, outcome measure, effect size, interval, certainty rating and year. That work has not been done here for arts programmes with staff or carers.

What would change this page: a controlled study with a named outcome, meaning a validated burnout measure, sickness absence or retention, reporting an effect size with its confidence interval, a certainty rating, a sample size and a date. With an active comparator: another protected hour doing something else. And with some handling of the attendance problem, whether by protecting the time so that attendance is not self selected, or by measuring at the level of a department rather than of the people who came.

The label meanwhile is Tier 3, of the nobody has looked properly kind rather than the null controlled evidence kind. That is not a verdict that these programmes do nothing.

Two cautions from the wider literature

Both directly relevant to how a staff programme is likely to be written up.

Volume of studies is not strength of evidence. The 2021 Cochrane review of music interventions in cancer care pooled 81 trials and 5,576 participants and still rated its findings very low certainty³, which is Tier 2. If a staff wellbeing review appears with a large study count, the count establishes nothing on its own.

Whole sample first, subgroup second. The 2018 three arm trial of singing for postnatal depression found no significant effect across the whole sample (p=0.16); the result that circulates is a moderate to severe subgroup at week 6, no longer significant by week 10, and singing did not beat the active comparator of creative play⁴. That trial is not about staff, and its shape is exactly the shape a staff programme evaluation will have: a null overall, a positive subgroup, and a temptation to lead with the second. It is Tier 2 for those reasons.

For scale, and because it is often used to imply strength: the WHO Europe scoping review covers over 900 publications, comprising 200 plus reviews and 700 plus individual studies, with the reviews between them covering over 3,000 studies⁵. Those counts nest rather than add, and a scoping review maps a literature without pooling it, so it carries no tier.

The organisational objection, which deserves a straight answer

Burnout has organisational causes, and an arts session cannot change any of them.

Workload, staffing levels, rotas, whether breaks actually happen, autonomy over how work is done, and whether people can raise a concern without cost: those are the things the occupational literature keeps returning to. None of them is affected by a lunchtime choir.

That matters practically, not just rhetorically. A wellbeing offer made instead of addressing those causes can read to staff as a suggestion that the problem is their resilience, and it can produce resentment rather than relief. I have heard exactly that said about a mindfulness offer in a setting where two posts had been vacant for a year. The same offer made alongside a serious look at the causes lands completely differently, and it is the same offer.

So the honest framing for a manager is: this is a good thing to have and it is not a response to understaffing. Anybody proposing one should be able to say which of those two they are doing.

Unpaid carers are a different problem

The obstacle is not interest and it is not time in the abstract. It is replacement care.

A carer cannot attend anything unless somebody is with the person they care for. A programme that does not build that in reaches only the carers who already have help, which is the group with the least need, and it will nonetheless report good attendance and warm feedback from the people it reached.

Three practical implications follow.

Replacement care has to be in the budget, not assumed. That is the single design decision that determines who the programme is for.

Timing and transport matter more than content. Evening sessions frequently exclude the people they are aimed at, and so do sessions in a building that is difficult to reach at the wrong end of a day.

Sessions attached to something the cared for person is already attending solve the problem structurally: a carers’ group running in an adjoining room during an activity session removes the obstacle rather than working around it. That model is common in care home and day centre settings and it is the one I would look at first.

What a defensible programme looks like

Six things, all of which are within an organisation’s control.

1. Protect the time. An hour in a rota is a different intervention from an hour in a break, and it is also the only version that reaches the people who need it.
2. Say what it is for, in plain words. Because staff would like it, and because a shift with something in it is better than one without.
3. Do not promise clinical outcomes. No reduced burnout, no improved retention, no lower sickness absence, because none of that can currently be supported.
4. Evaluate honestly. Attendance, who came, who did not and why, and what people said. That is a description and it is worth having, provided nobody calls it evidence of effect.
5. Keep it regular. A monthly one off produces a photograph. A weekly fixture produces a habit.
6. Ask the staff who did not come. They are the whole question, and nobody ever surveys them.

The practical arrangements for setting anything up in a care or clinical building, including insurance, safeguarding and who owns it internally, are on starting an arts programme in a care setting. Funding patterns are on who pays for arts on prescription, and the sector’s own reports and case studies are gathered in one place⁶, worth reading as descriptions of practice rather than as evidence of effect, a distinction worked through on creative health policy and reports.

If you are the member of staff or the carer

Nothing here argues against going, and it would be a poor reading of the page to take it that way.

Tier 3 is a statement about the research literature, not about your Tuesday. An hour of singing, drawing or writing with other people is worth having on its own terms, and it does not require a trial to justify it. What the evidence does not support is a claim that it will fix burnout, and what nobody should be doing is offering it to you instead of the things that would.

If you are struggling, and particularly if you are a carer without a break, that belongs with your GP, your occupational health service or a carers’ assessment rather than with a creative session. The wider picture of arts programmes inside health buildings is on arts in hospitals, and in social care on arts in care homes. Nothing on this page is a reason to change, delay or decline any part of anybody’s treatment, and no arts programme substitutes for one.