WITHDRAWN - Formative evaluation and feasibility trial of a collaborative care mental health intervention for prison leavers

Problem

Prison leavers have high levels of anxiety and depression and associated substance use, self harm, homelessness and relationship difficulties. They receive substantial substance misuse care and support for housing, variable psychotropic medication and sickness certification from GPs, but virtually no specialist mental health care. The Engager intervention aims to improve the mental health of men being released from prison who have anxiety or depression. The collaborative care intervention includes joint working with primary care, engagement to develop trust, mentalisation-based approach to work with rapid changes in emotion, development of a ‘shared understanding and plan’, and through the (prison) gate working. The theoretical model was developed using a realist synthesis approach to bring together case studies, a realist review and focus groups. The aim of this study within the programme was to understand whether the model of Engager intervention is acceptable when delivered in practice, and to refine the intervention and its delivery (manual, training, supervision) for a subsequent RCT.

Approach

The latest guidelines in intervention process evaluation (MRC 2014) were used to refine the intervention using a formative evaluation. We used a realist approach to understand what works, for whom, and in what context. Of the 60 eligible participants recruited in the pilot trial 40 prison leavers were randomised to the intervention and 20 to control . Data for the formative evaluation included: interviews (practitioners=9, participants=30), audio recordings, practitioner records, and health records. Peer researchers supported the research process.

Findings

Engagement post release approached 75% with positive comments from all those interviewed. Practitioners believed in the Engager model, and felt empowered to work in new ways. Role uncertainty, greater familiarity with practiced ways of behaving in challenging situations, and local cultural/practical difficulties, meant practitioner behaviour was at times slow to change. Offenders trusted and engaged with Engager practitioners, against the expectations of some other services, describing their practitioner as a ‘mate with good contacts’. Two major lessons were learnt: firstly that by meeting prisoners at the gate on day of release trust could be established despite only brief meetings in the prison; secondly that the ‘shared understanding’, a type of psycho-social formulation, could be developed over time throughout the intervention rather than being a prerequisite prior to planning the intervention. Other refinements included clarification of roles, development of the supervisor team leader role, practitioner self-care, using crisis management to develop the ‘shared understanding’, and improvements to planning endings.

Consequences

The realist formative evaluation of Engager provided rich information about what was and was not working in the Engager intervention, who it was or was not working for, and in what context/s. This fed in to refinements of the Engager manual, training, and supervision for the main Engager RCT which has now commenced.