A demographic and clinical profile of prisoners accessing an assessment and treatment service within the offender personality disorder pathway

ABSTRACT The Offender Personality Disorder (OPD) pathway enables men and women likely to have personality disorder to be identified and treated in prisons, secure hospitals and the community. The treatment services forming part of the OPD pathway aim to improve the psychological health, wellbeing, and relational skills of its target population. However, there is very little literature on the population that fall within the scope of the OPD pathway. This paper examines the clinical and risk profile of a cohort of male prisoners participating in a long term, high secure prison-based assessment and treatment service, the Beacon, which forms part of the national OPD pathway. Demographic information, psychiatric and offending history, personality disorder diagnosis, and details on trauma symptomatology was collected from a cohort of prisoners participating in the Beacon service. The paper considers the personality traits and mental health needs present in the population alongside key trauma-related difficulties. This paper demonstrates that the present cohort of the Beacon prisoners show evidence of complex emotional and interpersonal needs. Their experiences with trauma and associated symptoms were common, especially related to posttraumatic stress. Limitations are discussed.


Introduction
The Offender Personality Disorder (OPD) pathway was jointly commissioned by the previously named National Offender Management Service (NOMS) and the National Health Service England (NHSE), with the overall goals of improving the psychological health and wellbeing of prisoners who are likely to have personality disorders alongside public protection (NOMS & NHSE, 2015).The 'OPD Pathway Programme' reflected a shift away from the Dangerous and Severe Personality Disorder (DSPD) programme in its' emphasis away from intensive treatment for a small number of individuals, towards psychologically informed management of a greater number of individuals meeting high risk, high harm criteria, who also had personality difficulties.Different services were established in a network across England and Wales, recognizing the need for support at every stage of a prisoner's journey through the criminal justice system (Skett & Lewis, 2019).Prisoners are not required to have a formal diagnosis of personality disorder to be eligible for the services.Instead, criteria includes a likelihood of having a personality disorder or related difficulties, alongside a high likelihood of offence repetition, and a clinically justifiable link between the two (NOMS & NHSE, 2015).A recent large-scale evaluation of the Male Offender Personality Disorder Pathway (Moran et al., 2022) adopted a mixed methods approach to identify how the Pathway as a whole was being experienced by offenders and the staff.One of the aims of the evaluation was to compare the outcomes between individuals referred to OPD services and those not referred, and to identify whether there was evidence of costeffectiveness of the Pathway.The research struggled with selection bias and confounding variables which made causal inference difficult and treatment outcomes inconclusive.The qualitative study suggested positive impacts for both offenders and staff (Moran et al., 2022).
Although there is some data on the demographic and clinical profile of this population (Eastman et al., 2019;Hopton et al., 2021;Yeadon et al., 2021), there remains a gap in understanding this population from a risk and trauma perspective.It is likely that the OPD populations have high posttraumatic stress-related needs, especially within custodial services.Prison populations have a high prevalence of co-morbid mental health conditions such as PTSD and personality disorders (Facer-Irwin et al., 2019).This paper aims to begin to address this evidence gap by providing a detailed description of the clinical, risk, and trauma symptom data on a sample of prisoners admitted to one of the current OPD pathway treatment services, the Beacon, located at HMP Garth.

The Beacon
The Beacon opened in January 2014 as a prison-based personality assessment and treatment unit forming part of the newly developing OPD pathway.It is located within a long term, high secure prison and has capacity for 48 residents.It is delivered as a collaborative partnership between an NHS Trust and His Majesty's Prison and Probation Service (HMPPS).The development of the service has been described in detail in previous publications (Heffernan et al., 2016;Jellicoe-Jones & Nathan, 2015).
The Beacon works with individuals meeting the following criteria: • Aged 21 or over.
• Serving custodial sentences of between 3 and 20 years.
• Assessed as presenting a high likelihood of violent or sexual offence repetition, and a high or very high risk of harm to self or others.
• Likely to have a personality disorder or significant personality difficulties.
• Evidence of a clinically justifiable link between personality disorder/ difficulties and risk of offending behaviour.
The overall aims of the Beacon service is consistent with those of the wider OPD pathway; • To promote a safe, positive and empowering environment.
• To improve psychological health and wellbeing, functioning and interpersonal skills.• To reduce the likelihood of repetition of serious offending.
• To enhance the capacity of the workforce.
Each individual accepted onto the Beacon undertakes an initial threemonth assessment period as part of a treatment program of up to 36 months.During this time, each resident has a formulation-based weekly personalized timetable, detailing scheduled activities and treatments.These include psychological, occupational, educational and social interventions and activities, along with community meetings and association periods.The Beacon, in common with other OPD services, has a clinical model that focuses on the role of attachment and trauma, and the importance of current relationships as having the potential to be therapeutically helpful and enabling.This is reflected in the Beacon service's achievement of the Enabling Environment (EE) Award.This award is based on a set of standards which, it is argued, form a common foundation for 'creating and sustaining a positive and effective environment,' which in turn can 'foster productive relationships and promote good mental health' (Royal College of Psychiatrists, 2015).The EE standards are 10 principles that staff and prisoners work together to uphold, on the basis that doing so maximises the potential for an environment to be supportive and enabling for all (Johnson & Haigh, 2011).

Methods
A total of 113 residents were accepted onto the Beacon between the period of September 2015 and the end of April 2019.All residents were invited to participate in a clinical evaluation of the treatment they received; 73 residents (64.6%) provided written informed consent for researchers to access and use their data in an anonymised form.All outcomes collected were pseudoanonymised by providing an ID number.All the residents included in this analysis had been on the Beacon for a period of approximately three months.
The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committees on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008.All procedures involving human subjects were approved by Health Research Authority -IRAS ID 151614, REC 14/NW/1465, and HMPPS with processing by the National Research Committee.

Data extraction
There was a large amount of missing data on the different measures, which is explained in detail in the following sections.The demographic data were extracted from prison reports, which were a part of the referral process for the admission to the Beacon and the residents' Historical Clinical Risk assessments (HCR-20 Version 3 [Douglas et al., 2013]).This included, historical factors (such as childhood adverse experiences, psychiatric history, and substance misuse); and offending history (such as index offences and sentence details).Only 1 resident's information is missing on some of these measures.
The Personality Disorder diagnosis was extracted from the International Personality Disorder Examination (IPDE; [Loranger, 1999]).The well-known IPDE assessment can provide a Personality Disorder diagnosis as defined by the DSM-IV and ICD-10.The IPDE was either completed prior to admission (and available in the prison reports) or within the initial assessment period by the Beacon clinical staff.This information was missing for 14 residents.It is not possible to say whether this information was missing due to an error in data collection or because the individual did not have a Personality Disorder diagnosis.
The trauma symptomatology was collected with the Trauma Symptom Inventory-2 (TSI-2; [Briere, 2011]).This is a 136-item measure targeted at symptomology that may arise as a result of traumatic experiences.This measure was collected by the Beacon psychology staff within the threemonth assessment period.The measure comprises 4 factor scores, 12 clinical scales and 12 associated subscales, and 2 validity scales.The four factors are considered summary measures of complex posttraumatic disturbance.
The respondents of the TSI-2 measure have to reflect on the past 6 months and endorse the symptoms on a zero (never) to three (all the time) scale.Raw scores are converted to T scores based on gender and age and higher T scores reflect greater symptom endorsement.The T scores have clinical cut-offs; scores below 60 are within normal range, 60-64 are problematic and those 65 or greater represent significant clinical concern (Briere, 2011).The two validity scales include the Response Level (RL) scale that measures under-reporting and the Atypical Response (ATR) scale, which aims to detect likely overreporting.The TSI-2 manual recommends a cut-off score for valid protocols be above a T scores of 75 on the RL scale and a raw score of 15 or above on the ATR scale (Briere, 2011).TSI-2 outcomes were available for 41 residents; thus, data was missing for 32 residents.It is unknown whether this was due to refusal to complete the measure or errors in data collection.

Results
Frequencies were calculated for the demographic information, childhood adverse experiences, psychiatric, substance use, offending histories, sentence details, as well as the IPDE and TSI-2 measures.

Demographic information
The mean age of residents at admission to the service was 36.68 years (standard deviation [SD] = 10.19,range 21-66, median 35).The majority of residents were White British (n = 69), with the remaining residents being White Irish (n = 3) or Black Caribbean ethnicity (n = 1).This demographic information reflects 4 years of data collection.As a comparison, HMP Garth in 2019 reported a population breakdown of 76.38% White British, 1.10% White Irish, 5.88% Asian and Asian British ethnicity groups, 9.55% Black and Black British group, 3.3% mixed ethnicity groups, and 0.49% other ethnicity groups (HM Chief Inspector of Prisons, 2019).

Childhood adverse experiences
Information on historical trauma and adverse childhood experiences is displayed in Table 1.A significant majority were reported to have experienced physical abuse (n = 57/72) and just under half were reported to have experienced sexual abuse (n = 32/72).Of these, 24 residents reported experiencing both physical and sexual abuse.Within both categories, the abuse was reported to have most commonly occurred within the familial home, by a family member.Over a quarter of residents were documented to have spent time in children's home (n = 20/72).There was a record of familial mental health difficulties for a minority of individuals (n = 6/72).

Psychiatric history
Table 1 also displays a breakdown of formal psychiatric diagnoses, given at any time prior to admission to the Beacon.Data was available for 73 residents.Seven of the 73 residents also reported previous contact with the Child and Adolescent Mental Health Service as a child.
The most prevalent disorders were within the 'Other' diagnostic categories (n = 20/73).Attention Deficit Hyperactivity Disorder (ADHD) and Posttraumatic-Stress Disorder (PTSD) were diagnosed in 12 and 10 residents, respectively.Five residents were diagnosed with both disorders and appear in both totals.Furthermore, Depressive Disorder was also a common diagnosis.
In all three diagnostic categories, residents' reports of symptom experiences documented exceeded the formal diagnosis given.For example, 35 residents (of 73 total sample) reported experiencing symptoms of major mood disorders in the past, but had not received a formal diagnosis.A further 24 residents (of 73 total sample) had reports of psychosis symptoms, but no formal diagnosis.

Substance misuse
Data regarding past alcohol and substance misuse levels was available for 72 residents (see Table 1).A significant majority of residents were reported to have problems with alcohol or substances (n = 63; this was the same for both categories), with over half the residents (n = 37) being documented as having problems with both.Where substance misuse was present, usage typically commenced in early adolescence, most often through smoking cannabis.Polysubstance misuse was common, including (but not limited to) use of heroin, cocaine, amphetamines and ecstasy.

Offending history
Table 2 displays details of residents' index offences (the offence for which a sentence is currently being served) as well as their sentence type and category.The mean age of residents at the time the index offence was committed was 27.23 (SD = 7, range 15-45).
The majority of residents had committed an index offence that was categorized as violent (n = 62/72).A minority of residents (n = 10) had committed an index offence that was categorized as sexual.Of the violent index offences of murder (n = 14), nine of the cases involved a victim known to the resident.Of the most common sexual index offence of rape (n = 5); all five victims were female and three were under the age of 16.
A minority of seven residents had been convicted of both a violent and sexual offence at the time the index offence was committed.The most serious of these offence groups is the primary offence.Out of seven residents, three had a primary sexual index offence and four had a primary violent offence.

Sentence details
Table 2 shows that the most common sentence type was imprisonment for public protection (IPP) sentences (n = 34/73), a form of indefinite sentence where a minimum tariff was given, but no maximum.Over a third of residents were serving life sentences (n = 25/73).

IPDE assessment
IPDE assessment outcomes were available for 59 residents, and the six most prevalent Personality Disorder diagnoses are displayed in Table 3.According to the IPDE scoring criteria, a probable diagnosis is assigned when the respondent met one criterion less than the number required for a definite diagnosis (Loranger, 1999).The most common diagnosis was given for Antisocial Personality Disorder, with over half of the sample receiving a definite diagnosis (n = 32), followed by Borderline Personality Disorder (n = 24).One resident did not receive a probable or definite diagnosis of any personality disorder.

Trauma Symptom Inventory (TSI-2)
The TSI-2 results are outlined in Tables 4 and 5.The organisation of the factors, clinical scales, and subscales in the tables indicate the composition of each factor.The TSI-2 outcomes were available for 41 participants.However, the ATR scale validity cut-off invalidated five residents' responses.Unfortunately, there were no records to explain whether such high scores were evidence of overreporting symptoms (the function of the ATR scale) or whether the residents' responses were appropriate within their life circumstances and difficulties.However, in order to follow the manual's guidance, these five residents' scores were taken out of the analysis, leaving a total sample of 36 for factor, clinical scales and subscales and 35 for the validity scales.There were no invalid profiles for the RL scale.The mean T scores displayed in Table 4 indicate that the Beacon population on average showed the highest elevations of scores on Intrusive Experiences, almost reaching clinically elevated levels.The Posttraumatic stress factor, Defensive Avoidance, and Tension Reduction Behaviour mean scores were all elevated to a clinically problematic level.The Beacon sample's scores on Depression, Relational Avoidance, and Hyperarousal were approaching problematic levels.
Table 5 presents the frequencies of the TSI-2 scores divided by the clinical ranges.This demonstrates a more detailed view of the Beacon sample's difficulties and provides a more individualised view.The Beacon residents' scores indicate a population that was almost divided equally on most scales between individuals whose experiences are considered within the normal ranges and those who are having clinically problematic or significant trauma symptomatology.
Posttraumatic stress factor scores and their associated clinical scales and subscales were consistently endorsed at a high level by most of the Beacon sample.Notably, almost all of the Beacon sample scored in the normal range on the clinical scales and subscales of Sexual Disturbance, especially Dysfunctional Sexual Behaviour, which was 100% in the normal range.The Beacon sample did not demonstrate significant difficulties with somatisation using the TSI-2, as the majority of the residents (n = 30 out of 36) scored in the normal range on the factor and associated clinical scale and subscales.Further elaboration on these results and their significance will be presented in the Discussion.pathway; therefore, this paper aims to add to the literature and provide an insight into the cohort's history of mental health difficulties and offending histories.Furthermore, it aims to expand knowledge about this population's trauma-related needs.This study shows that this cohort of Beacon residents have clinically significant needs in posttraumatic stress, according to the TSI-2 measure.
The most common sentence in the Beacon cohort was an IPP sentence, which is a higher proportion than other studies have reported (Moran et al., 2022;Yeadon et al., 2021).This might have an impact on the way the residents engage in the therapeutic programme and the environment, because having an indeterminate sentence can be a barrier to finding hope and building trusting relationships (Moran et al., 2022).The high levels of childhood abuse in the Beacon residents' histories are similar to the offenders who are screened into the OPD pathway.Skett et al. (2017) found that out of 32,848 individuals screened, 56% had experienced some form of childhood difficulties.Due to the limited research, it is difficult to compare the Beacon cohort to other OPD population and know what the likely profiles in these services are.The OPD services are available for offenders with a likelihood of personality-related difficulty and complex needs; therefore, it is likely that there will be a variety in clinical and criminogenic need profiles (Yeadon et al., 2021).

Posttraumatic stress and the Beacon cohort
The Beacon sample showed the highest level of clinical distress on scales and subscales of the Posttraumatic Stress factor.Fifteen of the Beacon residents scored on the clinically elevated range and when combined with the number of residents scoring in the problematic range (n = 9), they make up 67% (n = 24) of the subsample of 36 residents.According to the TSI-2 manual, almost all respondents who have elevated scores on this factor have undergone one or more major traumas in their lives.The majority of the Beacon cohort has experienced childhood physical abuse (n = 57 of 72) and sexual abuse (n = 32 of 72) in the past.This is not surprising given childhood trauma is highly prevalent in prison populations (Altintas & Bilici, 2018).
According to the manual, elevated scores on this factor can also suggest the likelihood of a PTSD diagnosis (Briere, 2011).The psychiatric history available for the Beacon cohort shows only a few residents had a formal PTSD diagnosis, which seems unexpected due to the elevated scores on the TSI-2 Posttraumatic stress factor.However, this is representative of the general prison population.Research investigating trauma-related needs found that PTSD symptoms are more prevalent in prison than in the general population (Baranyi et al., 2018;Jakobowitz et al., 2017).PTSD is also likely to be underdiagnosed and unmet in the prison population (Jakobowitz et al., 2017).
Although few residents have a formal PTSD diagnosis, the TSI-2 trauma assessment showed that a high number of residents within the subpopulation that has TSI-2 scores experienced trauma symptoms that would be associated with the diagnosis.For example, Intrusive Experiences scale is consistent with the 'B' group of symptoms in the DSM-IV PTSD criteria and indicates previous experiences of psychological trauma (Briere, 2011).Twenty-five out of 36 residents scored in the above normal range on this scale, making it a significantly relevant need in the cohort, according to the tool.The Defensive Avoidance scale can be consistent with 'C' group of the DSM-IV PTSD criteria (Briere, 2011) and 22 out of 36 residents scored above the normal range.The elevated scores on this scale suggest some residents of the Beacon sample seeks to avoid, suppress, and eliminate painful thoughts or memories by avoiding events or stimuli in their environment that might trigger them (Briere, 2011).The subscale of Anxious Arousal -Hyperarousal is linked with the 'D' group of symptoms in the DSM-IV PTSD criteria and manifests in hypervigilant and irritable behaviours (Briere, 2011).Twenty residents out of 36 scored above the normal range on this subscale, which also indicates the likelihood that more residents would meet criteria for PTSD diagnosis.

Trauma symptoms and therapeutic work
The TSI-2 scale also assessed different non-specific psychological outcomes of traumatic events, which reflects an individual's difficulties in attachments, perceptions of self and others, and dysfunctional behaviours (Ales & Erdodi, 2022).Some of these symptoms are interpersonal in nature and can lead to experiencing difficulties interacting with others (Briere, 2011).Such difficulties in forming and maintaining interpersonal relationships is a common reason for why individuals with Personality Disorders, especially those with trauma histories can find it challenging to engage in treatment (Tyrer et al., 2015).Therefore, the focus on relationships and relational function within the OPD pathway strategy is considered to be a 'key ingredient' (Moran et al., 2022).The Beacon sample's results on some of the TSI-2 clinical scales and subscales offers an insight into the residents' difficulties, which can impact their therapeutic work and relationship building.
The Beacon cohort showed that 16 residents out of 36 had above normal scores on the Dissociations clinical scale, which suggests that they are often distractible, 'spacing out' and feeling out of touch with their emotions (Briere, 2011).Dissociation is a common symptom of PTSD and associated with childhood traumatic experiences (Altintas & Bilici, 2018).The link between childhood trauma and dissociative experiences has also been shown in prison populations (Ruiz et al., 2008).Literature suggests that clinicians have found that dissociating can mean an ongoing changing of perceptions of self and others, which can disrupt building a therapeutic alliance (Pearlman & Courtois, 2005).This is especially the case when trauma has occurred in past interpersonal experiences (Lawson et al., 2020).
Relational work with survivors of pervasive trauma can be difficult.There are persistent mistrust, emotional lability and relational instability that clinicians have to find a way to work with (Pearlman & Courtois, 2005).The Relational Avoidance clinical subscale showed that there were a high number of Beacon residents who experience relational dysfunction.Twenty-one residents out of 36 scored in the above normal range, indicating these residents could be described as keeping people at a distance and avoiding interdependence in relationships (Briere, 2011).The Beacon sample also showed more clinically elevated levels of symptoms on the Reduced Self-Awareness subscale.The detailed frequencies showed that 19 out of 36 residents scored in the above normal range, indicating that over half of this Beacon subsample has a lack of self-knowledge, such as confusion over feelings and thoughts and has difficulties forming goals.According to Briere (2011), these difficulties can appear in the context of dysfunctional personality traits such as chronic inability to access a sense of self.Some of the trauma symptoms are also indicative of dysfunctional behaviour, which can have an impact on the therapeutic environment.The Beacon sample's scores on the Anger clinical scale showed that 19 out of 36 residents scored above normal range, meaning the residents experienced problematic and clinically elevated levels of anger and had aggressive behaviours.According to Briere (2011), those with elevated scores on this scale might have pervasive feelings of annoyance, bad temper and have unexpected angry reactions which can seem inappropriate or an overreaction.PTSD symptoms have been associated with custodial violence (McCallum, 2018), but a systematic review suggested that PTSD symptoms were more likely associated with reactive aggression, such as primarily a result of arousal and reactions to perceived threat (Facer-Irwin et al., 2019).
Anger and other externalising behaviours are generally a way an individual tries to deal with overwhelming internal states (Briere, 2011).These externalising behaviours can make building relationships difficult in a residential setting, where residents are interacting with each other in both therapeutic and living spaces.The potential impact of the atmosphere is also relevant in relation to the creation and maintaining of an Enabling Environment on the Beacon.The EE standards, as mentioned in the introduction, aim to promote the Beacon atmosphere to be a positive and empowering social environment by building healthy relationships.These elevated levels of trauma symptom experiences within this Beacon cohort illustrate the different complexities pervasive trauma can create and that trauma can be a significant factor underlying a wide range of personality disturbances (Yen et al., 2002).
Given that the OPD pathway does not rely on diagnosis, rather it chooses a loose definition to allow for a greater inclusion of individuals who could benefit from the services on the pathway, it is difficult to know what to expect in terms of presentations of needs of the population (Skett et al., 2017).However, the target population are men and women who are presenting a high risk of serious harm to others, which also point to a challenging population with complex needs.According to the National Evaluation's (Moran et al., 2022) qualitative descriptions of the population, early adverse experiences were referenced often as the catalyst to most of the interpersonal and emotion regulation difficulties the individuals' experienced.It was also narratively reported that in addition to personality disorders, some of the participants also experienced symptoms of PTSD.It is the overall challenge of the OPD pathway services to manage and treat a group of individuals who are varied in clinical need and have challenging behaviours of which many relate to experiences of trauma.

Limitations
There are a number of factors which limited the analysis and the scope of this paper.Firstly, information was only reported for those who consented to participate in the research; just over a third of the residents admitted to the Beacon did not consent for their data to be collected and analysed.Additionally, this paper described one cohort within one service at one stage along the OPD pathway.
Information collected from pre-admission documents may be subject to reporting errors.There is often a lack of a standardized approach to reporting across the United Kingdom, so that similar agencies may report information in different ways and with different levels of detail.Missing data were evident; particularly for the TSI-2 assessment.This may be due to the timing of the consent for this study, delays in the conduct of assessments, or a resident's refusal to undertake them.The assessments were also administered and scored by various members of the clinical and research teams over time, which might have introduced some variation in the responses.The depleting sample size and unexplained missing data made it difficult to draw conclusions about the residents across the different variables.Inconsistent data collection and the limitations of administrative data recording were also a significant barrier to the National Evaluation of the male OPD pathway (Moran et al., 2022).
This descriptive paper demonstrates that the present cohort of the Beacon residents show evidence of complex emotional and interpersonal needs.Their experiences with trauma and associated symptoms were common, which has the potential to create difficulties in receiving the clinical intervention.
*(classification based on OASys documented categories of 'none, some, or significant' misuse).

Table 3 .
Number of definite and probable IPDE diagnoses.