We use cookies on our website to provide you with the best possible user experience. Disabling these cookies may prevent our site from working efficiently. To find out more about our cookies read our privacy policy.

RCT Study design to evaluate benefits of art therapy shows how to make the case for interventions

Nick PollardCrawford MJ et al 2010 The MATISSE study: a randomised trial of group art therapy for people with schizophrenia. BMC Psychiatry 10:65 http://www.biomedcentral.com/1471-244X/10/65 - reviewed by Nick Pollard, Sheffield Hallam University.

While creative activities have traditionally been employed in mental health settings for many years the evidence base supporting them is notoriously sparse. This extensive study - which claims to be "the first large scale pragmatic trial of art therapy for people with schizophrenia", involving over 400 participants over 4 sites, is therefore very welcome. NICE (2009) currently recommends that arts therapies should be offered to people diagnosed with schizophrenia, particularly where negative symptoms may impact on the capacity for verbal expression.

The multisite study aimed to compare group art therapy in conjunction with standard care with referral to attention control treatment in conjunction with standard care or standard care alone. Participants from four centres were assessed to determine the impact on health and social functioning. Additionally, given the sharp emphasis on treatment costs in the present context, comparisons were made for cost effectiveness. Previous studies suggest that engagement in art therapy demonstrates a range of improvements in mental health, social functioning, wellbeing and satisfaction with care for those people who have a diagnosis of schizophrenia. Ethical approval was sought from a REC and the protocol was registered as a controlled clinical trial prior to data collection.

Random assignment

Participants were randomly assigned to the three treatment conditions; those assigned to art therapy attended weekly sessions lasting 90 minutes over 12 months. Art therapy involves the participant and therapist in freely and spontaneously making art. Qualified and experienced art therapists led the groups which were co-facilitated by other staff members or volunteers. Additional support was offered to participants who did not engage. The alternative activity groups ran for a similar period and pattern, and included board game sessions, visits to cafes and places of interest, but did not involve art or craft materials or the use of therapeutic conversation: participants who became distressed were diverted or directed to address their concerns with professionals allocated to their care. Art therapy and activity groups were monitored through supervision and the recording of activities to ensure the observation of treatment fidelity. Participants allocated to standard care were restricted from taking part on arts therapies until after final assessment of the study had been completed.

Participants were assessed at 12 months and further data was to be collected 24 months into the study. A full range of assessment measures has been employed including Global assessment of functioning (GAF) and the Positive and Negative Syndrome Scale to assess health and mental health; Morisky Scale to assess use of psychotropic medication; Euroqol EQ-5 D for health related quality of life, and an adapted form of the Adult Service Use Inventory as a measure for cost data. Further self reporting measures were used to collect data from the participants and data was also collected from the participants' keyworkers, but measures were employed to ensure that data was masked without access to personal identifiers.

However, the selection of outcome measures merits discussion. GAF has been used in many psychiatric studies, but is rated by observing staff, not the person themselves.  Symptom change has been shown to have little relationship with quality of life or function (Kazdin 1999).  The EQ5D is a very short (5 question) scale that has only one question about mental health (about anxiety and depression), compared with other quality of life measures such as the Quality of Life Enjoyment and Satisfaction Questionnaire (Q-LES-Q) (Endicott, Lee, Harrison and Blumenthal 1993) or the various forms of the Short-Form health surveys, e.g. the SF 12 (Ware et al, 2002). The EQ5D may not be sensitive to small changes (Brazier, Roberts, Tuschiya and Busschbach 2004) that for people with severe and complex mental health problems may mean a lot.  These instruments, though internationally popular, may not measure changes that the client group value, and therefore lack validity.

Conclusion: a useful guide

This article reports the establishment of the trial and the methodology to be employed in administering the trial and its data collection. The final follow up interviews were to be collected in September 2010 and the authors set out how they will undertake the analysis to come. The design is a very useful guide to developing proposals for similar studies with regard to evidencing other activity based in interventions. It has clearly benefited from the involvement of a large and multidisciplinary team who have been able to combine their expertise to develop a comprehensive assessment battery and to address a range of the issues which previous smaller studies have not been able to resolve. For future studies, however, some of the instruments may need to be considered in terms of how they reflect the experiences of the participants. 

Additional comment from Sarah Cook: GAF has been used in  many psychiatric studies, but is rated by observing staff, not the person themselves.  Symptom change has been shown to have little relationship with quality of life or function.  The EQ5D is a very short  (5 question) scale that has only one question about mental health (about anxiety and depression). It may not be sensitive to small changes that are valued by people with severe and complex mental health problems.  These instruments, though internationally popular, may therefore not register the benefits of art therapy for this client group.

Brazier, J., Roberts, J., Tuschiya, A., and Busschbach, J. (2004) A comparison of the EQ-5D and SF-6D across seven patient groups. Health Economics. 13: 873–884

Endicott, J., Lee, J., Harrison, W. and Blumenthal, R. (1993) Quality of Life Enjoyment and Satisfaction Questionnaire: A new measure. Psychopharmacology Bulletin. 21, 321-326

Kazdin, A (1999) The meanings and measurement of clinical significance. Journal of Consulting and Clinical Psychology. 67(3) 332-339

NICE (2009) Schizophrenia: Core interventions in the treatment and management of schizophrenia in adults in primary and secondary care. London: author.

Ware J.E., Kosinski M., Turner-Bowker D.M., Gandek B. (2002) How to Score Version 2 of the SF-12® Health Survey (With a Supplement Documenting Version 1). Lincoln, RI: QualityMetric Incorporated.