1. Introduction

It is well documented that the prevalence of TBI in offender populations is far higher than in the general population [1]. A systematic review of the literature [2] found TBI prevalence rates in offenders to be, on average 46%, a stark contrast to prevalence rates in the general population, which are estimated to be around 12% [3, 4]. It should be acknowledged that there are large discrepancies in estimates for offender populations, likely due to the different definitions of brain injury (i.e., the inclusion or exclusion of non-traumatic brain injuries), the populations studied, sample sizes and methodology. Most studies use self-report questionnaires to measure brain injury, and questions have been raised about the reliability of this with offender populations [5]. Official diagnosis of brain injury in this population is, however, often hard to obtain and self-report measures such as the Brain Injury Screening Index (BISI) [6] have been found to be consistent with information from medical records [7] and the results of neuropsychological tests [6].

Traumatic Brain Injuries are one of the leading causes of death and disability in young people [8, 9]. Incidence of TBI peaks in late adolescence and early adulthood (16–25 years), with young men more likely to engage in activities associated with brain injury risk, such as contact sports or jobs with industrial risk [10]. Most common in all populations are mild TBIs, which are estimated to account for around 80% of all TBIs, followed by moderate TBI accounting for 10% and severe TBI also accounting for 10% [2]. It is likely that mild TBIs are even more common than is captured by the data, since many people, particularly those from offender populations, do not present to hospital for fear of repercussions or a lack of awareness of the potential seriousness.

The effects of brain injury are wide ranging, and whilst common perception of TBI is often associated with physical or mobility issues, the most debilitating difficulties are often the cognitive, behavioral and emotional consequences. Commonly experienced, are somatic (e.g., fatigue [11] and headaches [12]), cognitive (memory, slowed information processing, impaired attention [13]), behavioral (disinhibition, impulsivity, aggression [14] and emotional difficulties [15]. In mild TBI, many of these symptoms will resolve over time, but for some people (especially those with multiple mild TBI) they will endure and result in a range of persistent difficulties that make day-to-day life challenging [16, 17]. Furthermore, research has also shown that TBI-related cognitive and social impairments are associated with higher levels of previous convictions and higher recidivism rates [18, 19].

Frontal and temporal lobe damage is common following the type of brain injury experienced by men in prison [20]. Damage to the frontal lobes affects behavioral and emotional regulation [14, 15], as well as the ability to cope with new or unfamiliar situations [21]. Damage to the temporal lobes can lead to difficulties with language and memory [13]. Diffuse axonal injuries (commonly seen after road traffic accidents) can impair the speed of information processing [22]. Together, these difficulties can result in a range of behavioral, emotional and personality changes, all of which contribute to impaired social cognition, reduced tolerance to frustration and increased aggression. Links between TBI and “uncontrollable anger” have been identified in men with TBI in prison [17]. Williams et al. [19], in their review of the evidence surrounding TBI and crime, conclude “neuropsychological dysfunction is linked to violence, infractions in prison, poorer treatment gains and reconviction” (p. 842). It has further been shown that TBI is associated with offences being committed at a younger age, longer sentences, increased rates of reoffending and more violent crime [19]. Piccolino and Solberg [23] demonstrated that men with TBI were more frequent users of medical and psychological services whilst in custody. Taken together, this evidence provides a clear rationale for providing brain injury interventions to men in prison who have experienced a TBI.

The prevalence of serious mental health problems in offender populations is far higher than the general population [24]. Furthermore, we know that TBI can often lead to a range of neuropsychiatric sequalae, including behavioral, emotional and personality difficulties [16, 17, 25]. Risk factors for developing these problems include low educational attainment, psychosocial stress, previous psychiatric illness, substance misuse and recurrent TBI [26, 27]. These are all risk factors common to men in the criminal justice system, placing them at significant risk of ongoing mental health issues following a TBI.

Schofield et al. [24] studied men in the New South Wales (Australia) criminal justice system. They found that the relative risks of any psychiatric illness in the six months post-TBI were significantly higher for those with mild and severe TBI. The same study also found an association between TBI characteristics (e.g., multiple TBIs, TBI with loss of consciousness) and screening positive for “major depression”. Rates of major depressive disorder (as measured by the Referral Decision Scale—RDS [28]) in those with a TBI were high at 52%. A United Kingdom based study of a male prison population [6] also found higher levels of anxiety and depression (as measured by the Beck Anxiety Inventory (BAI) and Beck Depression Inventory (BDI)) in participants with TBI when compared to those without.

The issue of co-occurring problems for men with TBI in the criminal justice system (CJS) is one that has been repeatedly demonstrated, e.g., [16, 17]. There is a complex intersection between TBI related issues, such as increased aggression, reduced inhibition, personality change and mental health issues, such as anxiety, depression and substance misuse. Given the complexity of their difficulties (cognitive, emotional and behavioral) it is often difficult for men to access standard treatments or offender rehabilitation programs [19]. It is hypothesized that TBI and its associated difficulties contribute to poorer engagement in offender rehabilitation programs and consequently, higher recidivism rates [2, 29, 30].

The high prevalence of TBI in offender populations, and associated high levels of behavioral, personality and emotional issues, has been key to the development of our Brain Injury Linkworker (BIL) service. The BIL service delivers brain injury screening as well as low and high intensity interventions (three and six sessions, respectively). Ramos, Oddy, Liddement and Fortescue [5] described a series of case studies outlining the foundations and successful application of the BIL approach. Given the difficulty that men with ABI (Acquired Brain Injury) in the criminal justice system (CJS) experience in accessing and making use of standard mental health support (e.g., due to cognitive difficulties and lack of adaptations), we sought to understand the impact of our BIL service on the mental health of the men we work with. As such, we measured anxiety (Generalized Anxiety Disorder assessment, GAD-7; [31]) and depression (Patient Health Questionnaire, PHQ-9; [32]) on entry to the service and after intervention. These measures are commonly used by the UK National Health Service (NHS) and therefore widely understood by a range of professionals. We were interested to explore whether the BIL intervention would result in the men who completed their intervention demonstrating improved anxiety and depression scores as measured by the GAD-7 and PHQ-9, respectively. Whilst our interventions are not focused on anxiety and depression specifically, we wanted to explore whether our brain injury interventions (e.g., psychoeducation and compensatory strategies) have an indirect impact on the men’s psychological wellbeing.

2. Materials and methods

3. Results

At the time of writing, 233 men had been screened by the BIL service across South Wales HMPPS between March 2022 and April 2023. Written consent for their data to be used in the service evaluation was obtained for 171 (73%), and the data relating to those who did not consent were excluded. Only 50 (29%) complete datasets (pre- and post-intervention) were available for analysis. The BIL service is delivered in remand prisons (individuals are held in prison prior to sentencing or for short-term sentencing) and approved premises (hostels for prison leavers who are still deemed as being high risk); as a result, individuals are frequently released or relocated with little or no warning. Interventions may not be completed before moving on, and there is often no opportunity to collect their end-of-intervention outcome data.

Table 2 provides a breakdown of the demographics of the individuals who were screened by the BIL service. Of the 171 men that were assessed, the service was not deemed suitable for 24 (e.g., main presenting issue would be better addressed by a different service), and 14 declined treatments. Of the 133 that were offered further intervention, 83 were either relocated to another prison or released back into the community before treatment was completed. Therefore, the sample comprises of the remaining 50 men who completed the intervention. The groups have broadly similar demographic characteristics (Table 2). All groups were of similar ages except those who declined intervention who were slightly older, although this was not statistically significant against the group average. The majority of participants had less than 10 years of formal education (i.e., left school before sitting public exams at sixteen) in comparison to those who completed the intervention, who had a mode of either less than 10 years of education (46% of group) or ten years or greater (46% of group) i.e., stayed in education until sixteen to eighteen years old). The assessed but declined group however, showed the highest number for TBI severity classification (marginally significant at p = 0.023353559) and greater number of blows to the head compared with all other groups.

All data were entered and analyzed using Microsoft Excel (Microsoft Corporation, 2018, Redmond, WA, USA). Paired samples t-tests were carried out to compare the average self-reported anxiety (GAD-7) and depression (PHQ-9) scores before and after the BIL intervention. Results showed a significant decrease in the scores reported on the Generalised Anxiety Disorder assessment (GAD-7) between the start (M = 15.38, SD = 4.64) and end (M = 11.84, SD = 5.15) of the intervention (t (49) = 2.01, p = 0.00000733). Additionally, there was a significant decrease in self-reported depression symptoms as observed by the scores reported on the Patient Health Questionnaire (PHQ-9) between the start (M = 17.08, SD = 5.55) and end (M = 12.42, SD = 6.55) of the intervention (t (49) = 2.01, p = 0.0000125).

The qualitative and clinical significance of the changes observed within individuals were explored through an analyses of individuals’ classifications of anxiety and depression before and after intervention. Classifications for anxiety showed an improvement, with most of the sample being classed as having minimal/no to moderate anxiety at the end of intervention. Results showed that post-intervention 66% of the sample were classed within this range in comparison to only 36% being classed as having mild to moderate anxiety at the start of intervention (χ² (1, 2) = 9.12, p = 0.01, Fig. 1). In summary, participants were observed to have moved from more severe to milder categories of anxiety by the end of the intervention.

Fig. 1.GAD-7 classification movement. The number of individuals in each severity classification for GAD-7 before and after intervention. This figure demonstrates that at the start of intervention 32 (64%) individuals were classed having severe anxiety, in comparison to only 17 (34%) individuals by the end of intervention. GAD-7: Generalised Anxiety Disorder.

Change scores also indicated that most participants either decreased or maintained their levels of anxiety (Fig. 2), further evidencing the movements observed in the classification changes. The average change score was −3.54 (SD = 6.78, Min = −21, Max = 5). A reduction in anxiety levels was observed for 40 (80%) participants. Anxiety decreases ranged from −21 to −1. Ten participants showed an increase in anxiety scores, with an average increase of 2.60 (SD = 1.51), ranging from +1 to +5.

Fig. 2.GAD-7 change scores. Counts of anxiety score changes for the 50 participants in the sample; calculated by subtracting the pre intervention score from the post intervention (pre intervention − post intervention = change score). This figure shows that on average, individual anxiety scores decreased by 4 (SD = 4.99), with a range of −21 to −1. Although, some participant scores increased post intervention, with a range of +1 to +5. GAD-7: Generalised Anxiety Disorder.

Results also showed that there had been movement in classifications for depression, with 62% being classed as having minimal/no depression or moderate depression at the end of the intervention in comparison with no one falling under the minimal/no depression class and 40% classed as having mild to moderate depression at the start of intervention (χ² (1, 3) = 9.87, p < 0.01) (Fig. 3).

Fig. 3.PHQ-9 classification movement. The number of individuals in each classification for PHQ-9 before and after intervention. This figure indicates movement from severe depression, to mild to moderate depression, with 62% of the sample classified on the lower severity ranges at the end of the BIL intervention. PHQ-9: Patient Health Questionnaire.

Individual depression score changes showed an average decrease of 4.66 points (SD = 6.78) by the end of intervention, compared to their pre-intervention scores (Fig. 4). Thirty-nine of the 50 in this sample, (78%) had score decreases post-intervention which ranged from −21 to −1. The remaining 11 (22%) participants showed an average increase of 3.75 (SD = 3.62), with a range of +1 to +11.

Fig. 4.PHQ-9 change score. Counts of depression score changes for the 50 participants in the sample. This figure shows that on average, individual depression scores decreased by 5 (SD = 6.78), with a range of −21 to −1. Although, some participant scores increased post intervention, with a range of +1 to +11. PHQ-9: Patient Health Questionnaire.

4. Discussion

The BIL service has been running in prison and probation services across the UK since 2015. When the service was first developed, the original aim was to provide low-cost, low intensity neurorehabilitation that would support men with ABI while being sustainable within an overstretched criminal justice system. Ramos et al. [5] demonstrated the potential value of a BIL service through a series of detailed case studies. Through our continued work, we observed that the men often reported improvements in mood and a sense of relief at sharing their story. In the words of one of the people supported, “really enjoyed the sessions, making me feel a lot better”. This was further evidenced by observations that our service users were more engaged in the prison regime (e.g., having and maintaining jobs) and attending probation appointments. We wondered if our brain injury interventions were having a wider impact and contributing to improved mental health in these men.

As part of our service evaluation, we looked at men’s self-reported anxiety and depression symptoms, as measured by the GAD-7 and PHQ-9. We found significant improvements in average scores pre- and post-BIL intervention. To understand the clinical significance of this, we looked at the average change scores and found majority of men moved from more severe to milder classifications on both the GAD-7 and PHQ-9 (e.g., from the moderate to mild depression classification). Further research to explore the link between the BIL intervention and improved mood is needed.

Outside of prison, these men often lead unpredictable lives, with difficulties such as homelessness, social isolation and financial constraints making it hard for them to engage in health or other support services. A BIL session is often the first time that a professional has sat down with the men, and either explained that they have a history which is suggestive of brain injury or, if they were already aware that they have a brain injury, has taken the time to discuss what this means for them. The men often expressed a sense of relief at having an opportunity to speak with a professional who understands what they are experiencing and has the time to make sense of it with them: “Grateful for helping me understand myself…100% grateful”. The power of sharing one’s story in the context of ABI has been well documented [34] and the improved GAD-7 and PHQ-9 scores after BIL intervention indicate that feeling listened to and improved understanding of brain injury [35] may contribute to improved mood.

This study is a service evaluation and as such has several limitations: one, is that of the 171 who consented to take part, complete data was only available for 50 men. The group who declined the service (n = 14) were found to have significantly greater TBI severity. Although significance was marginal, and these are small numbers, it is clinically interesting to consider why this might occur. A possible explanation is that more severe TBI is often associated with poorer insight [36] into one’s condition and this may mean the men are less likely to accept support [37]. Alternatively, those with more severe TBI might have more associated difficulties. Future analyses could focus on those that do not complete treatment by exploring other factors such as degree of cognitive impairment, concurrent mental health diagnoses, substance misuse, offence history and length of incarceration. Better understanding these issues would hopefully enable us to adapt our service and make it more accessible to this group.

Another challenge we have been unable to address with these data, is the wide range of difficulties that the men experience, in order to differentiate which may be caused by brain injury, and which may be the result of other issues. There is a significant overlap in symptoms of brain injury and mental health [15, 16, 24]. Slow information processing, low mood and poor initiation, can all be explained by TBI but also by depression. Impulsivity and poor attention can be explained by frontal lobe damage but also by Attention Deficit Hyperactivity Disorder (ADHD). Furthermore, many of the men we work with have concurrent diagnoses of learning difficulties, ADHD or substance misuse. With the delivery of any non-pharmacological intervention, it is often hard to know what exactly is benefiting the individual and how. Furthermore, this study has no control group, therefore it is not possible to ascertain whether these improvements could have resulted from a different intervention or spontaneous remission. In addition, we do not know whether the men were receiving other interventions concurrently and what impact that could have had.

Several studies refer to the role of brain injury in rates of reoffending [5, 15, 38]. The combination of difficulties experienced by men with TBI (cognitive, behavioral, emotional), are likely to result in poorer engagement with the prison regime and rapport with staff, resulting in less preparedness for release back to the community. The likelihood of reoffending in this context is much greater, as they are far less likely to access and engage with the support they need. One of the strengths of our intervention is that, though brief, it focuses on the specific needs of the individual and, as has been demonstrated here, may have indirect impact on issues such as anxiety and depression. The intersection between mental health, TBI and offending provides a clear context for the ongoing delivery and development of the BIL service. Currently very few brain injury intervention services are described [1, 5, 35] but we hope this paper provides a context for future research and service development.

Finally, our evaluation does not allow us to explore the experience of mental health and TBI in the men from Black, Asian and Minoritized Ethnic backgrounds. Although our sample is representative of the diversity in Welsh populations, it is important for us to reflect on what greater effort we can make to deliver services to people from these backgrounds. Research shows that black men in the UK are likely to be treated more harshly by the criminal justice system [39]. We also know people from Black, Asian and Minoritized Ethnic backgrounds are likely to receive less provision of neurorehabilitation than their white counterparts [40]. Taken together, with the evidence on recidivism and brain injury, this highlights that it is imperative that services like the BIL are equally available and accessible to all.

5. Conclusions

This service evaluation demonstrated that men’s self-reported anxiety and depression symptoms improved following the BIL intervention. This is encouraging and consistent with the qualitative feedback from the men we work with and the staff that support them within South Wales HMPPS. We acknowledge, that as a service evaluation project, there are limitations (e.g., no control group, inability to rule out confounding variables) but we hope we have demonstrated the importance and value of specialist support for men with ABI in prison and probation services and provided directions for future research.

Availability of data and materials

The participants of this study did not give written consent for their data to be shared publicly, so the data cannot be made available.

Author contributions

AB, SDSR and EB—designed the service evaluation. EB and RJ—collected the outcome measures and delivered the BIL interventions. AB—clinical supervision. EB and SDSR—analyzed the data. AB and EB—wrote the manuscript. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript.

Ethics approval and consent to participate

Ethical approval for the service evaluation was sought from the National Research Committee of the Ministry of Justice; reference number 2021-252. All participants were provided with an information sheet and consented to their anonymized data being used in the service evaluation and write-up.

Acknowledgment

The authors acknowledge all those involved in the delivery of the BIL service, specifically Elizabeth Wilce, Project Manager, Davina Jones, Head of Policy and Social Change, and Beth Gore, Foundation Research Support Officer. We also thank the staff in South Wales prison and probation services, who continue to support the BIL service and all the service users, for their continued hard work.

Funding

This research received no external funding.

Conflict of interest

The authors declare the following potential conflicts of interest with respect to the publication of this article: The authors are employed by Brainkind (Formally The Disabilities Trust), the charitable organization that developed and delivers the BIL service. This manuscript was written as part of a service evaluation.