Arts and arts therapies measurably ease depression and anxiety in children and young people, major meta-analysis finds

A sweeping new analysis of randomised controlled trials suggests that everything from music listening and painting to dance movement therapy and drama can meaningfully reduce symptoms of depression and anxiety in children and young people, with effects that appear comparable to, and in some cases larger than, established talking therapies. The systematic review and meta-analysis, published in eClinicalMedicine, pooled data from 77 randomised trials involving children and young people aged 5 to 24, and represents the most comprehensive attempt yet to quantify how the arts affect mental health in this age group.

The research team, led by investigators at Edge Hill University and the University of Manchester, set out to fill a stubborn gap in the evidence base. While the World Health Organization estimates that one in seven children and young people worldwide live with mental health concerns, and the OECD reports that youth mental health worsened in nine of eleven analysed countries between 2012 and 2022, services remain dominated by talking therapies such as cognitive behavioural therapy. Waiting lists in some systems stretch beyond two years, and many young people never access care at all. Arts-based interventions, the researchers argue, could bridge that gap, provided the evidence supporting them is rigorous enough to convince clinicians and policymakers.

Previous syntheses had either been very broad, such as the 2019 WHO scoping review of more than 3,000 studies on arts and health, or very narrow, focusing on a single art form like music or dance. The new study took a different approach: it included all six major arts modalities, dance and movement, literary arts, media, music, theatre and performance, and visual arts, craft and design, as well as multimodal combinations, and drew exclusively on randomised controlled trials published between 2005 and 2025. The protocol was prospectively registered, and the team searched five major databases, ultimately screening 2,446 records after deduplication and assessing 434 full-text reports.

The results are striking. At the end of intervention, arts-based interventions produced a very large reduction in depression scores, with a standardised mean difference of 1.06, based on 23 comparisons from 22 trials involving 1,655 participants. Anxiety showed a large effect of 0.75 standardised mean differences across 56 comparisons from 48 trials and 3,898 participants. Wellbeing and health-related quality of life improved moderately, with effect sizes of 0.53 and 0.54 respectively. Crucially, the benefits did not evaporate the moment the final session ended: follow-up data collected at least one month after interventions showed sustained reductions in depression (0.51) and anxiety (0.67), though these estimates rest on far fewer trials.

Perhaps the most methodologically interesting aspect of the review is its use of physiological biomarkers as objective checks on self-reported outcomes. Because participants in arts interventions cannot be blinded to what they are receiving, self-report measures are vulnerable to expectation effects. The researchers therefore synthesised data on heart rate, blood pressure and cortisol. Arts-based interventions reduced resting heart rate by an average of 4.53 beats per minute across 29 comparisons from 25 studies, and lowered systolic blood pressure by an average of 4.04 mmHg across 15 comparisons. Cortisol changes were smaller and inconclusive. The authors note that population-level reductions in blood pressure and resting heart rate of this magnitude are typically associated with lower long-term cardiovascular risk, suggesting the interventions may confer physiological as well as psychological benefits.

Subgroup analyses revealed intriguing patterns. Effects on depression were larger when interventions were compared against passive controls such as waitlists or usual care rather than active comparators, a well-known phenomenon in psychotherapy research. For anxiety, theatre and performance interventions and multimodal programmes produced the largest effects, outperforming music, visual arts, dance and literary arts. Children aged 9 to 13 appeared especially responsive, showing larger gains in anxiety, quality of life and blood pressure than other age groups, which the authors suggest may make middle childhood a particularly fruitful developmental window. Notably, there was no clear difference between formal arts therapies delivered by qualified therapists and broader participatory arts activities, although the authors caution that this comparison is constrained by a heavy imbalance in the evidence base, with 64 of the 77 meta-analysed trials involving non-therapy arts interventions.

The findings come with important caveats. Heterogeneity across studies was very high, particularly for depression, where the I-squared statistic reached 97 percent, meaning the pooled effect size should not be read as a universal property of arts interventions. When the team removed three studies with implausibly large effect sizes, the depression estimate fell from 1.06 to 0.58, still statistically significant and moderate to large. Removing studies judged at high risk of bias produced a similar attenuation to 0.70. Only four of the 77 meta-analysed trials were rated at low risk of bias, with most lacking pre-specified protocols, adequate allocation concealment reporting, or any possibility of blinding participants. Funnel plot asymmetry hinted at publication bias for depression, anxiety and heart rate outcomes, and GRADE certainty ratings were downgraded to very low for depression and anxiety and low for wellbeing and quality of life.

There is also a structural limitation the authors are candid about: the included trials spanned enormously different contexts. Brief single-session music listening before a dental procedure was pooled alongside multi-week dance movement therapy for diagnosed depression, and the vulnerable population category bundled together children facing everything from burn dressing changes to autism spectrum disorder to kidnapping trauma. Sample sizes ranged from 12 to 463 participants, and no included study used a clinical diagnostic interview as an outcome measure. The scarcity of follow-up data beyond one month, particularly for wellbeing and quality of life, means the durability of benefits remains largely unknown, although one cited school-based programme, Arts for the Blues, reported positive effects up to six months later.

Despite these limitations, the authors argue the overall pattern of evidence positions arts-based interventions as credible, accessible additions to the mental health support landscape for young people, with average effects broadly similar to those reported for talking therapies in comparable populations. They emphasise, however, that policymakers must distinguish between arts activities with therapeutic intent and formal arts therapies delivered by trained professionals, particularly when designing services, allocating resources and planning workforce training. They also point to the potential role of community-based arts programmes in social prescribing schemes that could relieve pressure on clinical services while reaching communities that traditional services often miss.

The research team calls for larger, higher-quality trials with standardised outcome measures, agreed intervention lengths and follow-up timepoints, and routine incorporation of physiological biomarkers alongside psychological outcomes. Head-to-head comparisons between arts interventions and established treatments such as cognitive behavioural therapy are conspicuously absent from the literature and represent a key next step. For now, the message from this landmark synthesis is that the evidence for arts-based mental health support for children and young people has moved decisively beyond anecdote, and that the creative arts may deserve a formal place in how societies respond to a youth mental health crisis that conventional services alone are struggling to contain.

Subject of Research: Effects of arts-based interventions and arts therapies on mental health outcomes in children and young people

Article Title: The effects of the arts and arts therapies on mental health outcomes of children and young people: a systematic review and meta-analysis

Article References: Cheung, P. S., Rhodes, S., Bahr, E., Lakena, W., Sajnani, N., Fietje, N., & Karkou, V. (2026). The effects of the arts and arts therapies on mental health outcomes of children and young people: a systematic review and meta-analysis. eClinicalMedicine, 101, Article 104261. https://doi.org/10.1016/j.eclinm.2026.104261

Image Credits: AI Generated

DOI: 10.1016/j.eclinm.2026.104261

Keywords: arts therapies, mental health, children and young people, meta-analysis, depression, anxiety, randomised controlled trials, wellbeing, quality of life, biomarkers, music therapy, systematic review