Reviewing art therapy research: a constructive critique

Abstract

The literature search that informed our review initially yielded 12,122 papers of potential interest, derived from seven databases. After applying a series of filters we arrived at 92 papers on which we base our findings, thoughts and recommendations for future work. Our methodological approach was informed by the systematic review guidance published by the Centre for Reviews and Dissemination (2009), and the Arts Council definition of ‘arts activities’. Hence we considered papers reporting therapeutic arts interventions conducted on 'patients' which included some measurement of a health state. After excluding any research on people less than age 18, we selected studies where participants had active (as opposed to passive) engagement with the therapy/treatment/medium. Only study types which were \nquantitative were included in this review. Rather than simply criticise the execution of the research we applied our own expertise to the process. It was immediately evident that definitions and categories would pose some difficulties as there is much variety in the language used to describe the arts, therapies and treatment. This \nis a problem of indexing, causing the literature search and initial screening to be a laborious process. The most commonly reported art activities were: writing, music, art and dance. The most numerous health condition studied was mental health followed by cognitive function, stress \nand cancer. Most research was carried out in the US and the UK. As a discipline, psychology featured regularly. When arts therapists were involved in the research the descriptions and possible effects of the art medium tended to be better elucidated. Future research into the use of art therapy in healthcare will benefit from a synthesis of approaches that can retain the more robust aspects of, for example, RCTs with the insights that \ncan be derived from qualitative methods.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 6 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.