How do we co-produce research in the prison environment?

Jennifer Ferguson, Aisha Holloway, Victoria

Guthrie and Dorothy Newbury-Birch

Introduction

This chapter considers undertaking research in a prison environment. Coproduction has proved essential when working with prison staff and prisoners, and the PRISM-A project (mentioned in detail below) highlighted this. This chapter will talk through some facts around the prison environment, some experience from the PRISM-A project, and end with some useful tips for those wishing to undertake research in a similar environment.

Prison environment

Globally, there are over 10 million people incarcerated, with prisoners bearing a substantial burden of communicable and non-communicable diseases (Fazel and Baillargeon 2011). Since the 1940s there has been an increasing trend in the number of individuals incarcerated in the UK. In August 2018, there were 86,000 people in prison and young offender institutions. The cost of prisons to the Ministry of Justice (MoJ) was £3.954 billion in 2016-2017.

Due to changes in organisation. Her Majesty’s Prison and Probation Service (HMPPS) replaced the National Offender Management Service (NOMS) on 1 April 2017 (Office 2017). Since this time, NOMS has ceased to exist and its functions were split between HMPPS and the MoJ. HMPPS retained responsibility for the operational management of prisons, while responsibility for the commissioning and policy has moved to the ministry. HMPPS is an executive agency and is responsible for the prison estate, including adult prisons (both male and female), youth offender institutions and some immigration removal services.

Prison population

The fact that men are responsible for the majority of offending behaviour, is undisputed. The arrest data is released annually by the MoJ to prove this point. There are 131 male prisons in the UK, and 13 extra contracted prisons (Mah et al. 2018). The prison population is a constantly growing population.

The prison population in general in England and Wales has quadrupled in size between 1900 and 2017; and in Scotland the prison population has almost doubled. Northern Ireland prisons have seen an increase in the prison population since 2000. by 38%; although their prison population is currently at its lowest since 2010 (Library 2018). To place some context to those figures, in England and Wales for every 100,000 of the population, 179 are prisoners (both male and female); Scotland, 166, and Northern Ireland, 98. In terms of male prisoners, in England and Wales there are 348 male prisoners for every 100,000 men in the entire population (in 2016/2017). This is a substantial increase from 1900, where there were 152 for every 100,000 men (Ministry of Justice 2018).

Risky behaviours

As with most organisational transformations, cost have been reduced and the funding from the Ministry to NOMS, now HMPPS, reduced by 13% (from 2009-2010 to 2016-2017) and whilst the prison population has remained broadly stable (between 83,852 and 87,080), NOMS reduced the number of operational staff in public prisons by 30% to manage within the reduced budget. The National Audit Office Report (Office 2017) highlighted that the issues the Ministry are looking out for due to all of the change are: the Prison Estate Transformation Programme due to the number of prisons that have been deemed not fit for purpose; staff recruitment and retention; prison safety and security, which has declined since 2012; and finally the mental health of prisoners due to the number of suicides in the estate.

Safety in prisons is declining and the number of assaults has increased from 29,485 by 13% in one year (NPC 2012). The number of serious assaults has increased by 10% (2,856) and the assaults on members of staff by 23% (8,429). Interestingly, however, the number of suicides has fallen from 115, to 69; however, that is still a high amount of suicides. The prevalence of risky drinking in the prison system is between 59% and 63% whilst the levels of probable dependency are between 36% and 43% (Newbury-Birch et al. 2018). Similarly, there is a high prevalence of drug use. Research in the criminal justice system (in prison in particular) around alcohol is sparse (Sondhi et al. 2016; Newbury-Birch et al. 2018), but a growing area of interest, and a number of studies have started to look at alcohol use disorders and screening and brief interventions in the criminal justice system (Newbury-Birch et al. 2009; Coulton et al. 2012; Newbury-Birch et al. 2014; Newbury-Birch et al. 2016a).

Health and social care

The health of prisoners is now considered a public health issue. This is a new change due to the Health and Social Care Act 2012. Section 15( 1 )(c) requires the Board to commission certain health services “for persons who are detained in a prison or in other accommodation of a prescribed description”.

This caused the public health system to change; in order to deliver the Government’s commitment to "improve the health of the poorest, fastest” (Government. 2012).

Alcohol brief interventions for male remand prisoners: protocol for a complex intervention framework development and feasibility study (PRISM-A)

The PRISM-A study (Holloway et al. 2017), an MRC funded project, took place in 2016 across two prisons, one in the North East of England, and one in Scotland. The aim was to explore the feasibility and acceptability of an ABI for adult remand, male prisoners; to develop an appropriate AB1 for such prisoners; and to prepare a protocol for a multi-centre randomised pilot study. The project was aligned to the early phase of the Medical Research Councils (MRC) framework for the development and evaluation of complex interventions (Craig et al. 2008). The study comprised three stages, using mixed methods, and separate recruitment for prison stakeholders in stage 2.

Stage 1

The aim of stage 1 was to identify the prevalence of self-reported hazardous/ harmful alcohol consumption in adult male prisoners. Prevalence was measured using the gold standard Alcohol Use Disorder Identification Test (AUDIT) (Babor et al. 1989) screening tool. This allows an individual to be categorised as either a harmful, hazardous or dependent drinker by completing 10 screening questions and giving a simple score. The possibility of a potential alcohol use disorder is determined by a score from 0 40 and the categories are split from 0-7, 8-15, 16-19 and 20+. A score of 8 or more suggests is considered a potential alcohol use disorder. The AUDIT and reported views of AB1S delivered in prisons with acceptability of participation in a future ABI study with follow up. Research Assistants (RAs) carried out this element of the study using a questionnaire. The study aimed to recruit 250 adult male remand prisoners in each prison site, aged 18 or over, incarcerated for less than three months, and willing to provide informed consent. Written consent was obtained by the RAs and the questionnaire included the AUDIT, and questions experiences or opinions of receiving alcohol advice or information, their willingness to receive it and whether or not they would be willing to take part in a study involving a follow up element. Questions were read out and answers recorded by the RA on a hard copy.

Stage 2

The aim of stage 2 was to complete some qualitative interviews to first understand the feasibility and acceptability amongst adult remand prisoners of an ABI designed for them. Second, the qualitative work aimed to explore their perspectives in relation to their: beliefs and perceptions about their alcohol use, views regarding the acceptability of receiving an ABI whilst on remand, experiences of engaging with health professionals in relation to their alcohol use, the nature of this and the perceived impact and outcome of it, and finally their perceptions of acceptable alcohol screening, intervention delivery points and techniques, methods of delivery, and by whom.

Twenty male remand prisoners across two study sites were interviewed. Each participant was shown an infographic outlining the key components and the nature of an ABI. A focus group was also held at each site to explore prison stakeholders perceptions of the feasibility and acceptability of an ABI with this population. Focus groups were facilitated again by the RAs and were structured by a topic guide, informed by the qualitative data collected in stage 2.

Stage 3

Stage 3 involved identifying what an adapted intervention mapped self-efficacy enhancing ABI looks like. This was done using data collected and analysed in stages 1 and 2. Intervention mapping was used to refine and develop an existing self-efficacy enhancing ABI (Kok et al. 2016) to reduce reported levels of alcohol consumption in male remand prisoners. This was undertaken by members of the research study group, with activity and input form the Expert Advisory Group.

The results of the PRISM-A project showed that this type of research can be carried out in prisons. However, through the qualitative work, especially with staff and stakeholders, it became apparent that working alongside the staff in prison was essential. More time than was anticipated had to be put into working out the different regimes in each prison, and learning which members of staff needed to be involved in different ways. For example, it could not have been anticipated how vital it would be to work with peer prisoners. Putting the time into working with the relevant staff, as opposed to alongside them, made a significant difference in how the project worked.

Differing prisons in PRISM-A

It is important to note that although all prison establishments are very similar in terms of the environment; regimes differ depending upon what the Governor has established in any given prison. This became very apparent in the PRISM-A study. Some examples of this are included below.

One RA had keys, the other did not, which meant having to be escorted by already busy prison officers. Given the prison environment, and the amount of locked gates and doors, this can hold up research activity significantly. Not to mention, going to the bathroom. Another difference was that one RA could only data collect in the mornings only, the other all day. The RA with keys could only have access to prisoners before 12pm. However, that

Co-producing research in a prison 95 researcher had keys and could potentially survey 15-20 prisoners in this time. The other could have access to prisoners all day, but given the complication of being escorted, could survey the same amount, in double the time. Peer prisoners, often called “Listeners” or “P1DS workers” introduced the study for one RA, and proved very helpful. The peer prisoners had more time to work closely with the RA and the prisoners interacted with them very well, therefore taking an interest in the study. At the other site, peer prisoners were not used as part of the research study, therefore more work had to be done with staff to ensure everyone understood the study and purpose of the RA being there. At the second site, a lot of every day was taken up explaining this again to the new staff on duty.

Another difference between sites was that, at one, all data collection took place on one wing, the first night centre. At the second site, data was collected from various different places on the prison estate. This, unfortunately, was the site that the RA did not have keys, therefore the amount of time spent data collecting amounted to a lot more at the second site. The final difference was that one prison was mostly remand, meaning there was a high turnover of prisoners, which subsequently is brilliant for data collecting with new prisoners. The other site held both remand and sentenced prisoners, meaning more time was spent determining eligibility for the study, as only those who had incarcerated for less than three months could take part in the study.

As described, it is apparent that having keys, having peer prisoners assisting with the project, and being a remand only setting made the data collection more straight forward for one of the RAs. Both prison sites had an extremely strong co-production approach, which was called upon a number of times in the second site. When difficulties arose due to the RAs lack of keys, the Governor was called to assist with any difficulties, and they were soon ironed out.

However, the data collecting element is merely one part of conducting a research study in a prison. As mentioned in more detail below, setting up the project, and the various training also differs across different prison sites, and the RA who had more straightforward circumstances data collecting, had more rigorous training to undertake than the second RA, partly because of having access to keys, but also, simply the difference in demands from the two prisons.