Background: Recent research has highlighted that the key modifiable behavioural risk-factors for non-communicable diseases (NCDs) are highly prevalent among prisoner populations internationally, including the United Kingdom (UK) prisoner population. Given the wealth of evidence demonstrating that prison-based peer-led interventions can be effective in modifying health knowledge, attitudes and behaviours among prisoners, it is possible that a peer-led intervention may be used to help modify the highly prevalent NCD risk-behaviours among UK prisoners. However, there is a distinct lack of evidence in the form of well-constructed randomised controlled trials to support such an assertion. Moreover, there is uncertainty as to whether or not a randomised controlled trial would even be possible. This research aimed to address this gap in the literature through conducting a feasibility study to establish if a large-scale definitive study to explore the effectiveness of a peer-led intervention to modify the behavioural risk-factors for NCDs among prisoners is possible.
Methods: This research took place in two prisons in the North of England and utilised a mixed methods exploratory sequential design consisting of two phases; a phase one qualitative exploratory study and a phase two randomised controlled trial feasibility study. The first phase involved 3 focus groups with prisoners and 12 one-to-one interviews with members of staff to gather evidence to inform the development of an appropriate peer-led intervention to modify the behavioural risk-factors for NCDs among prisoners. The phase two quantitative randomised controlled feasibility study utilised a 1:1 randomised controlled design. Eighty participants were recruited in total; 40 were randomised to receive the peer-led intervention and 40 were randomised to the control group. Following delivery of the six-week peer-led intervention to the intervention arm, all participants were followed-up over a three-month period (immediately post-intervention, one-month post-intervention and three-months post-intervention) and asked to complete data collection measures.
Results: Thematic analysis of the phase one qualitative data generated eight overarching themes; non-conducive prison environment, scepticism, positive views towards prison peer-led interventions, peer-led interventions in prison – the downfalls, success dependent on peer, managing risks, prison regime impact, and increasing staff buy-in. Discussion of specific intervention design aspects revealed important aspects relating to format, length of delivery, content and training of peer-workers, all of which were considered carefully in the design of the peer-led intervention. The qualitative findings were used to develop a six-week peer-led group intervention to be delivered during the phase two feasibility study.
For phase two, the recruitment target of 80 participants was achieved within approximately two months. Average attendance to the peer-led intervention was 61%. Intervention fidelity was overall quite good, however there were some issues with delivery of components which required participants to reflect on behaviour diaries. The majority of the intervention arm participants appeared to find the peer-led intervention, including its group format and delivery through prisoner peers, acceptable. However, some areas for intervention improvement were identified. There did appear to be a small degree of contamination to the control group. Retention rates were high over the first two follow-up periods (86.25%), but did decline by the final follow-up time-point (66.25%). The completion of individual data collection measures by those engaging in the trial were high over all of the trial time-points, however there were some difficulties in the completion of the objective measures, particularly over the latter follow-up periods.
Conclusion: This feasibility study confirms that it is possible to undertake a definitive trial exploring the effectiveness of a prison-based peer-led intervention in modifying the behavioural risk-factors for NCDs among prisoners. However, based on the findings of this study, it is recommended that slight alterations to the study procedures and the peer-led intervention should be undertaken before commencing with a definitive trial. Additionally, changes to the prison environment to better aid prisoners in modifying their NCD risk-behaviours are warranted.
