The resettlement of people in prison with personality disorder‌‌

An estimated two-thirds of the prison population in western countries meet personality disorder diagnosis criteria [16]. Personality disorder is a complex mental health condition, characterised by poor relationship skills, issues with identity and, in some cases, a risk of physical harm to others [17] which people likely experience varying severities of difficulties [18]. Within western prison populations, 47% of males have an antisocial personality disorder diagnosis [16], a type of personality disorder characterised by and increased risk of violence and recidivism [19] substance misuse and self-harm [20,21]. Working with individuals with personality disorder has been associated with high levels of burnout due to emotional and relational demands of this work on staff [22].

The Offender Personality Disorder (OPD) Pathway is a joint health and criminal justice initiative commissioned by His Majesty’s Prison and Probation Service and NHS England [23]. The pathway was set up in 2012 to support individuals with a high risk of recidivism, assessed as meeting criteria for personality disorder. It aims to reduce repeat serious offending, improve psychological wellbeing, produce a competent and trained workforce, and use resources efficiently. Evidence for the effectiveness of the OPD pathway is mixed with a national evaluation indicating self-perceived reductions of risk and increased wellbeing but inconclusive quantitative findings in relation to engagement with the pathway and reduced recidivism [24]. People are screened into the OPD pathway using an algorithm [25], designed to assess presence of personality disorder traits and risk, and subsequently receive a formulation to inform sentence planning and suggest engagement with OPD services if appropriate.

Open prisons are a particularly challenging environment for individuals with a likely diagnosis of personality disorder and therefore open prison OPD services known as Pathways Enhanced Resettlement Services (PERS) have been operational since 2019, focusing on individuals with a high risk of “failing” in the short-term, either through reoffending following release or being returned to closed conditions. PERS aim to support people in open prisons, reducing their risk of being returned to closed conditions and preparing for transition into the community, acknowledging that progression for this population may be non-linear. Service users join voluntarily and alongside usual open prison programmes, receive key work sessions with PERS prison officers where they develop a working formulation to anticipate difficulties and agree a management plan. The services have an open-door policy for individuals to ‘drop-in’ and receive ad hoc support and pre-empt crises.

Only one previous study has explored the experiences of PERS service users [26]; it focussed on the service when it was set up as a pilot in one open prison. The study highlighted the importance of developing trusting, supportive relationships with staff as a key mechanism for change. This reflects findings from other studies exploring OPD service user experiences [24,27]. PERS has since been implemented across five open prison sites. We conducted the first qualitative study to interview service users and staff from PERS in open prisons, and the first qualitative study to purposively recruit service users across the national networks of open prisons in England that include a PERS service. We aimed to explore how staff and service users experience PERS and how the service has, or has not, supported individuals to remain in open prison and transition successfully to the community.

Study design

This study utilised a semi-structured qualitative‌‌ design underpinned by a critical realist methodological orientation [28]. We set out to understand what has, or has not, supported individuals to remain in open prison and transition successfully to the community (the mechanisms of PERS) as well as the context in which those mechanisms occur (through capturing multi-perspectives of staff and service users).

Ethical approval

Ethical approval was granted by HMPPS National Research Committee (Ref: 2020−131/ Ref: 2021−180). All participants provided written informed consent.

Procedure and sample

To explore the experiences of staff and service users we conduced semi-structured interviews. We recruited PERS staff from the five operational services, located in 5/11 adult male Category D (open) prisons in England. Between 1^st April 2021 and 30^th July 2021, we recruited ten of twenty staff (two staff members from each service were approached via email to take part in the study), purposively, for diversity of staff roles, levels of experience and clinical and operational staff. All staff that were approached to take part in the study agreed to participate and provided informed consent by completing the consent form and emailing it to the lead author before the interview began. Three groups of PERS service users were eligible to participate: ex-PERS service users in the community, current PERS service users and ex-PERS service users in closed conditions prisons. Potential participants were initially approached by PERS staff, directly or via offender managers of those returned to closed conditions between 7^th May 2022 and 1^st September 2022. Staff then arranged interviews for current service users and passed details of ex-service users living in the community willing to participate to the researcher. It was not possible to approach anyone in closed conditions prisons during the study period due to COVID restrictions and therefore none took part in the study. All participants who were approached by PERS staff took part in the study. Informed consent was obtained in person prior to the interview for current service users who were interviewed in person allowing opportunity to clarify anything before beginning. Individuals in the community completed the consent form and returned it to the lead researcher via email before the interview commenced and at the start of the interview this was verbally re-confirmed. Neither staff nor service users received any compensation for their participation.

To determine an appropriate sample size, the information power framework was utilised [29]. The narrow study aim, dense sample specificity and expected high quality of dialogue within the interviews indicated a smaller sample would be adequate. However, the lack of established theory of PERS and plan to analyse across cases suggested a larger sample would be most appropriate. Taking these criteria into consideration, a sample of 20 (ten staff and ten service users) was considered sufficient to address the research aim. The final sample size obtained was 19 (ten staff and nine service users).

Data collection and processing

Interviews with PERS staff were conducted by the lead author (GM) remotely via Microsoft Teams between April and July 2021 whilst in-person research within prisons was prohibited due to COVID-19 restrictions. The co-authors developed a topic guide (S1), which explored respondents’ job role, and their views on how PERS supports users in open conditions; the interviewer probed potential barriers and facilitators of service users benefitting from PERS support. The topic guide also included questions regarding how COVID-19 impacted service delivery and service user outcomes. Interviews with PERS staff lasted on average 50 minutes.

Interviews with service users were conducted by the lead author in June and July 2022, with current service users interviewed in person at the open prisons and ex-service users interviewed remotely via telephone. We were unable to recruit any ex-service users in closed conditions. The topic guide (S2) developed for service user interviews asked how they had joined PERS, how PERS had been helpful or unhelpful, how COVID had impacted their use of PERS, what they felt they needed from PERS to succeed, their wellbeing whilst working with PERS and, if applicable, their experience of transitioning into the community from PERS. Interviews with service users lasted on average 45 minutes.

All interviews were recorded on an encrypted Dictaphone, transcribed verbatim by the lead author and no one else was present during any interviews aside from the lead researcher and the participant. No repeat interviews were carried out. Field notes were made immediately after each interview and reviewed as the initial stage of the analysis.

Data analysis

We conducted a Reflexive Thematic Analysis (RTA) [30] using NVivo 14 [31] to code interview transcripts. RTA refers to the processes of critical reflection and refinement pursued by the researcher, during which subjectivity and nuance are considered in relation to predominant assumptions and socio-cultural context [32,33]. Reflexivity in this study manifested as iterative examination, adaptation and refinement of themes, initially generated by GM. Braun and Clarke’s [34] six phase process for RTA was used to analyse the interview data, which were coded inductively.

Phase 1 was familiarisation with the dataset. This involved re-listening to the recordings, becoming immersed in the data, and making notes with initial ideas. Phase 2 was coding; working through the dataset and applying labels to segments of text that appeared potentially relevant or meaningful. Phase 3 was generating initial themes, clustering codes into broader patterns and ideas. Phase 4 was developing and reviewing themes, whereby themes are re-assessed and compared with other themes and against the entire dataset. Phase 5 was refining, defining and naming themes. The final phase, 6, was the writing up.

GM carried out phase 1–3 of the RTA. The initial themes were developed, reviewed and refined alongside MF and HJ. Preliminary findings were shared via email with all PERS staff for comments and presented at a staff away day for discussion which aided the interpretation of the data. This supported us to develop to consider emerging themes. It was not possible to re-contact service users to check the themes. We included quotations – selected for their illustrative efficacy – to emphasise key points. Whilst all interviews contributed directly or contextually to our resulting themes, not all were quoted.

Researcher reflexivity

GM, MF and HJ have previously worked as staff in OPD pathway services (although not in PERS) and had prior knowledge and expectations about the experiences of staff and service users which likely influenced the interview questions and the interpretation of data. The interviewer was a female PhD candidate, with some experience conducting qualitative research and of working in OPD services as an assistant psychologist which brought an understanding the OPD context that helped build rapport. Despite the interviewer not being a clinician within the prison, some service user participants may still have been inhibited in their discussions due to the interviewer holding the position of a researcher.

Sample characteristics

No demographic data was collected about gender (for the staff, all service users were residents in male prisons), age or ethnicity of participants in line with the ethics agreement to maintain confidentiality in this sensitive area. Ten staff were interviewed (two from each service). We interviewed five clinical staff (clinical leads and an assistant psychologist), and five operational staff (mental health nurses and prison officers). The mental health nurses were considered part of the operational staff, due to their role within PERS aligning with the operational delivery of the service, such as providing key work sessions. Staff had worked in PERS between three months and two years (the clinical staff average was 13 months, and the operational staff average was 14 months). Three of the clinical staff and three of the operational staff had prior experience of working in other OPD Pathway services.

Nine service users were interviewed (two each from three of the services, one from another and two in the community at the time of interview). Seven were current PERS users and two were ex-PERS users living in the community. The service users had been in open prison between five months and three years. All service users had been in higher category prisons prior to entering open prison, two of the current service users had been in open prison previously (and been returned to closed conditions) and two service users had been recalled from the community during their sentence.

Qualitative findings

We developed three themes that responded to our research aims. Due to the constraints of our ethics permissions, we do not provide further demographic details of service users (denoted as SU after quotes); staff are described as operational (non-clinical staff with professional roles aligned with the prison service, denoted as OS) or clinical (clinical staff with training and professional roles aligned with health care: denoted as CS).

Reflective statement

Staff and service users were willing and open to participate in the study and explore their experiences of PERS and how they felt PERS influences progression. Staff generally had similar views about their experiences of implementing PERS and the factors that influence progression, however, clinical staff tended to focus on the wider systems that service users exist within and the emotional difficulties they experience, whilst operational staff provided more detail on the day-to-day service delivery and the practical support they provide. This is likely to reflect their roles and the type of contact they have with the service users. Some service users expressed their dislike toward authority and directly referenced the interviewer being an authority figure that previously they would not have felt comfortable speaking with. They stated the work they had done with PERS allowed them to be more open with authority figures and therefore take part in these interviews.

Difficult transitions into open prison

Many of the service users’ motivation to engage with PERS was to gain support with the transition into open prison, but some service users initially felt suspicious about what PERS was. The reluctance to engage with PERS supports research exploring factors which may impact the readiness to engage in services [35]. The suspicion and reluctance to engage in PERS highlights potential negative perceptions of the service, which is likely linked to a lack of trust in others [35] and in systems more widely. PERS staff reduced barriers to engaging by providing clear information about the service and a providing a choice about whether to engage. This in turn allowed service users to start to develop trust with PERS staff.

Moving to open prison was described as a difficult transition by all service users interviewed, with many struggled to adapt to the increased amount of freedom. Previous research has reported service users feeling overwhelmed when moving to an open prison [13]. PERS staff were aware of how difficult this transition could be and reported managing expectations as a useful tool to help service users adapt to their new environment. Expectations of treatment have been explored as a factor linked to treatment non-completion in individuals with personality disorder [35] but have not been considered in relation to standard prison transitions.

Relationships with staff

Service users and staff highlighted the positive trusting relationships they experienced working with PERS. Building trusting relationships has been a consistent theme related to change in other OPD pathway services [36,37] and a pilot PERS service [26]. Trusting relationships were seen as a key mechanism for progression within PERS.

Many staff and service users spoke about poor relationships with non-PERS staff in the open prisons, such as feeling misunderstood by non-PERS wing officers. Poor staff-service users relationships have been reported in previous research in open prisons [10,13,14], linked to the precarity of privileges and the potential for staff to remove them. Poor relationships with staff in open prisons are especially concerning for those with a likely diagnosis of personality disorder, who often experience difficulties with interpersonal functioning and authority [38]. A broad aim of the OPD pathway is workforce development [39] with the added benefit of systemic change in the wider environments that OPD services operate within. PERS staff should therefore focus further on providing supervision and support to prison staff in open prisons to help develop better relationships with service users.

Progression through open prison

PERS was felt to be supporting progression through the open prison and preparing people for release into the community. Several barriers to progression were noted by staff and service users. The accessibility of drugs in the open prison estate and the reduced levels of security to monitor this were noted by staff. Low levels of hope and control were also considered as barriers to progression. There are associations between externalising control and offending behaviour [40], with greater internal control associated with hopefulness [41]. Having hope and a sense of internal control are likely to aid people through the difficulties of progressing in open prison and into the community [36]. Increasing levels of hope and control is therefore an important task of PERS and was seen through many service users reporting increased wellbeing, a newfound ability to forgive themselves and the ability to set more realistic goals. PERS is modelling the OPD pathway approach well, which intends to provide support across the various stages of a person’s journey through the Criminal Justice System [42]. This is especially important in relation to the context of open prison, an important stage of progression in many offender journeys that can be difficult to manage [10].

Limitations

PERS staff may have been biased in reporting the positive impact of PERS, as they had an interest in the continuation of the service. Likewise, the sampling of service users through the PERS staff may have biased sampling towards those who felt positively towards the service. It was not possible to recruit ex-service users returned to closed conditions in the current study, and they may they had different, less positive experiences. It was also not possible to report demographic information, including ethnicity and therefore we were not able to comment on how experiences across may have differed across ethnic groups or the representativeness of the sample. This research took place in the pandemic, a unique time which provides critical insights into experiences at this time, though may also limit generalisability. The number of prison releases following the COVID-19 pandemic (March 2020) reduced by 15% compared to the same period in 2019 [43] but this may be attributable to the prison population serving longer sentences, rather than a direct effect of COVID.

Conclusions

Despite this study exploring low security open prisons, it is promising that the provision of individualised support, addressing needs on a person-by-person basis was received positively by service users and staff; while our findings cannot necessarily be generalised beyond this setting, findings may inform approaches across the prison estate. Further consideration should be given to the relationships developed between prisoners and prison staff, when trusting relationships are developed individuals appear more willing to engage with interventions and receive support in a meaningful way.