Developing an Evidence Base for the Use of Art Psychotherapy Within a Perinatal Mental Health Service

Methodology:

I intend to use a wait-list control group, whereby following admission to the unit a patient will follow the recruitment process detailed below, only after the recruitment process has been completed in full would consenting patients be randomised to an intervention group which will begin Art Therapy sessions in due course, or to a wait-list control group for a minimum of two weeks and a maximum of six weeks. The intervention group members will receive the Art Therapy intervention, while the waitlist control group are fully aware of the research and are waiting to receive it at a later date. The waitlist participants will be made fully aware that if they are discharged before therapy has become available to them, or if they reach the maximum term of 6 weeks on the waiting list, then they may not receive the intervention. Participants who reach the maximum term of 6 weeks on the waiting list while waiting for therapy availability will be withdrawn from the research and offered a debrief meeting in order to answer any remaining questions about the research. This maximum six week term is based upon the six week average duration of patient stay within the unit.

I am a qualified Art Psychotherapist and will be using a non-directive approach in order that the clients will lead their interactions with the art materials and with the therapy process.

Art Therapists are governed by the British Association of Art Therapists and the Health and Care Professions council who each set standards of professional therapy delivery. I will establish and facilitate a safe therapeutic provision of weekly individual Art Therapy sessions for patients of the unit.

I will work closely with the unit staff team in order to introduce and create a shared understanding about the therapeutic intentions of the art therapy provision. The additional benefit of employing a non-directive, client centred approach to Art Therapy sessions is that this will ensure that the clients themselves have key influence over the theme and artistic content of therapy sessions. The client presentation will then be analysed using a series of recognised quantitative scales to monitor client progress, therefore minimising the influence of the multiple therapists on the research data and increasing the quality of the data.

There will be an initial assessment period within the intervention. This will consist of one session of 50 minute duration, during which an assessment will be made about the clients interaction with art materials. This will be conducted by utilising the Hyland Moon approach of categorising art materials as high or low structure based on their properties. High structure materials being easy to control versus low structure materials being more fluid in nature and therefore more challenging to work with. The differing art making experiences and client interactions that these varying art materials can produce will be valuable in gaining an insight into the clients functioning, in order to facilitate an assessment of the appropriateness of the intervention within the context of each individuals presentation and diagnoses. Following assessment, both the client and the therapist will have the opportunity to either withdraw or consent to continuation of the intervention. Should the therapist determine following assessment that Art Therapy is not appropriate for the patient, for example in the instance that the patients presentation is so severe that an offering of Art Therapy may pose a risk to the therapist or the patient, or that the level of willingness to engage in the assessment phase (in spite of consent previously given) is so lacking that a continued offering is not sustainable, then the patient will be withdrawn from the research. In this instance, a letter will be provided to the patient, inviting them to return for a re-assessment for Art Therapy sessions, at such point as they feel ready to do so in future after a minimum period of two weeks have elapsed since prior assessment. The length of each clients engagement with the Art Therapy sessions, i.e. number of sessions provided, will vary due to the nature of the population. However, the average stay for a client within the unit is 42 days and they would be eligible for inclusion in the research from the point of admission, subject to consent and recruitment procedures. The intention will be to offer a series of five sessions of therapy following assessment. Should the patient be discharged prior to the end of therapy, the option will be offered of returning to the unit for the remaining sessions as an outpatient. The use of qualitative and quantitative data collection methods will account for capturing the complexity of the intervention. The provision of this intervention within the unit is over and above what the unit's "care as usual" plan is, because an Art therapy service does not currently exist in the setting. The following measurement scales are already in use as standard practice within the perinatal unit, therefore staff are comfortable with their application and credibility within the NHS setting. In order to honour the multidisciplinary approach to working with the patients, I therefore intend to have continuity in measures and will employ all of these scales at intervals throughout the participants engagement in therapy sessions.

Quantitative measures:

The use of quantitative measures within the research will mitigate against bias and the influencer of the researcher on the research findings, by presenting statistical data which will then be analysed using IBM SPSS Statistics software and reviewed by my supervisors. In addition to this, an element of influence will be eliminated via the fact that student therapists will also be delivering the intervention.

Application of scales:

The weekly questionnaires (ReQol10 and Bethlem Mother-Infant Interaction Scale) will be administered by the unit care team as part of their standard care plan, the researcher has requested permission from the patient (within the patient consent form) to access the patients medical records to withdraw the weekly scores from these questionnaires, for anonymous inclusion within the research data. This will avoid patients being asked to complete weekly duplicates of these questionnaires that they will already be completing with the unit care team. The researcher and therapists will already be authorised to have access to patient medical records under the terms of the honorary contract as provided by the unit, in order to record brief therapy notes.

Statistical Analysis:

Statistical analysis of the quantitative data collected will be conducted by the researcher, via IBM SPSS Statistics software, using Repeated Measures One-Way ANOVA to analyse the data collected from weekly questionnaires, alongside t-test and One-Way ANOVA analyses which will be applied to the questionnaires used during the assessment phase and at the end of therapy engagement.

Please see the Participant Timeline diagram and Table of Measures attached for additional clarity regarding administration of measures.

Qualitative measure:

In addition to this, I intend to employ the qualitative measure of an end of therapy Reflect Interview which will be recreated with each consenting participant, using the same set of semi-structured interview questions as written by the researcher and AIR (Audio Image Recording - as developed by N. Springham) piece as a way of commemorating and bringing an end point to the work done during Art Therapy sessions. As this is a qualitative measure and has been tested for feasibility as such, the interview questions are semi-structured in order to allow for the client to fully express their lived experience of therapy. I will undertake collaborative editing with client of the AIR recording, in order to demonstrate that I am working to reduce the bias / outside influence on the final piece, this will be achieved by giving the client an opportunity to state at the end of the interview whether they wish me to remove any sections of audio or add anything in. Data will be collected from the AIR's via transcription and analysis (by the researcher) of the transcripts for recurring words and themes within each patients interview.

In spite of the cohort being relatively small and the research undertaken over a relatively short timescale, a large amount of pre and post intervention data may be produced. This will require statistical analysis in order to attempt to determine statistically significant change in the clients presentation. I intend to code all qualitative data gathered using a 6-step framework of thematic analysis devised by Braun \& Clarke, during which I will code interview transcripts and then identify both semantic and latent themes within the coding. Due to the volume of qualitative data in the form of AIR transcripts that I anticipate being generated during the research, I intend to use professional and encrypted Nvivo software, provided by the University to aid the data analysis process.

The end of trial date will be signified by the completion of the minimum term of one full year of data collection.

In order that the population of participants are randomly, yet evenly allocated into wait-list and treatment groups, a blocked randomisation process will be applied. This will be conducted via Sealed Envelope, a website facility which can be customised to produce a list of participants and their randomly allocated group. In addition, the website will allocate each participant with a randomly generated code consisting of two letters and one number, it is ensured that no participant code within the list can ever be repeated. Participant codes will be used in place of names on all questionnaires, the AIR audio file and all data records in order to preserve anonymity. The allocation concealment mechanism within the randomisation process will be that the randomisation list will be produced and held by the Director of Studies and Supervisor, Jamie Bird.

Each time a participant has completed the introduction and recruitment process and given their consent to be involved in the research, the clinician-researcher will then obtain from the Director of Studies the next available participant code for that individual and establish whether they will be allocated to the wait-list or treatment group. Throughout the study, all participants shall remain anonymous to the Director of Studies and shall not be given any identifying information about participants. A record of participant names and the randomisation log of codes will be recorded in the site file which will be stored in an encrypted electronic format and will only be accessed by the lead clinician-researcher. In addition to this, including the varying size blocked element within the randomisation process adds another layer of allocation concealment as the clinician-researcher will be unable to predict the next allocation outcome. The benefit of this would be that the participants would be randomised and their clinical presentation could be measured in comparison as a control group to isolate the effect of the Art Therapy intervention as the independent variable. Additionally, this approach is considered to be ethically preferable to a control group which denies participants access to an intervention for the entirety of a research study.