Serious mental illness in probation: A review

Abstract

Research into serious mental illness and probation is reviewed. In addition, there is a specific review of the role of specialist mental health probation staff in the United States (US). In the discussion, we compare progress with the care of the seriously mentally ill within probation in Europe and the US. We conclude that the specialist role for probation staff developed in the US has significant advantages which have been well evaluated which should be implemented, in a large multi-centre trial, across Europe.

Review of research on supporting people with serious mental illness progressing to or subject to community supervision

There is a high prevalence of serious mental illness and co-occurring mental health and substance use disorders amongst prison and probation populations when compared to the general population. For example, in both probation and prison the estimates for psychosis are much higher than in the general population (0.7% vs 10%) (see Baranyi et al., 2022; Bebbington et al., 2016; Brooker et al., 2012; Lize et al., 2015; Pluck and Brooker, 2014; Power and McNally, 2022; Rebbapragada et al., 2021; Sirdifield, 2012). Following release from prison, mortality rates amongst ex-prisoners are also high in comparison to the general population (Binswanger et al., 2007) and internationally, continuity of care after release from prison is problematic. Much has been written about the importance and potential benefits of continuity of care for criminal justice and health outcomes, but once they are in the community, even individuals that recognise that they may benefit from engaging with mental health and/or substance misuse services may encounter barriers to doing so such as stigma, challenges around information sharing between stages of the criminal justice pathway, and between criminal justice agencies and their healthcare partners; and a paucity of appropriate mental health provision to meet their needs (Bradley, 2009; Hamilton and Belenko, 2016; Keil et al., 2008; Public Health England, 2018; Sirdifield et al., 2019 Ventura et al., 1998).

The extent of the disconnect between the recognition of needs and the receipt of services post-release from prison is illustrated in a systematic review of interventions to support prisoners diagnosed with mental illness at the transition into the community (Hopkin et al., 2018). Key studies included within the introduction to this paper are firstly Lennox et al. (2012), a UK-based study of prisoners with serious mental illness and in contact with mental health in-reach teams which found that there was some evidence of discharge planning in just 51% of cases released within the study period, and evidence of direct contact between the prison and community mental health teams (CMHTs) in 38% (n = 20) of cases. Just four of these individuals had since been in contact with the CMHT within a month of discharge (Lennox et al., 2012). Secondly, research conducted in the United States (US) which found that 18.9% of ex-prisoners indicated that they needed mental health treatment, but just 35.6% received treatment (Hamilton and Belenko, 2016: 1091). The likelihood of receiving mental health treatment within 3 months of release was increased by several factors including meeting with a case manager prior to release, receiving assistance to access public health insurance and self-identified need to access services (either pre- or post-release).

A total of 14 articles, relating to 13 studies were included in Hopkin et al.’s (2018) review, ten of which were conducted in the US, with the remainder being conducted in the UK (n = 2) and Australia (n = 1). Research included within this review suggests that there is an association between having a mental illness and/or substance use disorder identified in prison, and higher risks of re-offending and reincarceration following release (Baillargeon et al., 2010: 371). However, it is important to note that the reasons for this association are complex and have yet to be fully understood.

Ex-prisoners’ patterns of service use may also be different from those of the general population. A retrospective cohort study of 6046 ex-prisoners pointed to a high rate of emergency department use amongst ex-prisoners released during the study period. Comparing ex-prisoners to the general population of the state in which the study was conducted showed that ‘visits by ex-prisoners were more likely to be related to mental health disorders, substance use disorders and ambulatory care sensitive conditions than were visits by Rhode Island residents of the same age, sex, race and location of residence’ (Frank et al., 2013: 5) – demonstrating the need for continuity of care for these conditions.

Research on several approaches to improving continuity of care for people with serious mental illness is included within Hopkin et al.’s review. For example, work to ensure that prisoners with serious mental illness had Medicaid insurance on release (Morrissey et al., 2016; Wenzlow et al., 2011), and various approaches that dedicated specific resources for pre-release for planning and supporting access to mental health services for those with a serious mental illness during a defined period post-release. These included an intensive case management approach to link prisoners that have a mental illness and/or substance misuse disorder and are likely to be homeless on release with health services, and improve sharing of healthcare records to support continuity of care (Buck et al., 2011); a randomised controlled trial comparing assertive community treatment, forensic specialist intensive case management and treatment as usual (referral to a community mental health centre) (Solomon and Draine, 1995); a state-wide forensic transition program that aimed to support mentally ill ex-prisoners’ access to services during the 3 months post-release (Hartwell and Orr, 1999); a cross-agency intensive community treatment program combining numerous components including pre-release assessment and planning with post-release intensive case management and crisis support and voluntary confinement to a residential site (Theurer and Lovell, 2008); linked trials of a critical time intervention to engage mentally ill prisoners with a range of services on release and provide practical support, versus treatment as usual (Jarrett et al., 2012; Shaw et al., 2017); and a transitional support initiative to support people in custody with high or complex mental health needs as they move into the community (Green et al., 2016). Variation in the eligibility criteria, staffing and approaches for these programs make it difficult to generalise from the results of these papers.

Two systematic reviews were included within Hopkin et al.’s review. Firstly, a systematic review of trials of interventions to improve people’s health whilst they are in prison or during the year after their release (Kouyoumdjian et al., 2015). By our estimation, this review included seven journal articles focussing on adults with severe mental illness in prison and/or community settings. Two of these were included separately in Hopkin et al.’s (2018) paper (Jarrett et al., 2012; Solomon and Draine, 1995). The remaining five were all published in the US. Three were focused on people with serious mental illness in the community (Cosden et al., 2003; Cusack et al., 2010); one was focused on adults in a custodial treatment unit and in the community that had been detained multiple times and had both mental illness and a substance use disorder (Chandler and Spicer, 2006); one was focused on adult women with bipolar type I or II in a state correctional facility (Ehret et al., 2013); and one was focused on men with mental illness and substance abuse disorders in prison and community settings (Sullivan et al., 2007).

Secondly, a meta-analysis on the effectiveness of interventions for people with serious mental illness and criminal justice involvement in terms of their impact on criminal justice and mental health outcomes. This included 25 studies measuring various criminal justice and mental health outcomes. Aggregate effects were calculated for the criminal justice outcomes and the mental health outcomes, and with one exception showed that the interventions reduced continued criminal justice involvement but did not have a significant impact on mental health outcomes (Martin et al., 2012: 4). However, when the outcomes were disaggregated participants were found to be higher functioning and to have significantly fewer mental health symptoms (Martin et al., 2012: 4). Significant positive effects remained for all of the criminal justice outcomes studied apart from convictions and breaches of release conditions (Martin et al., 2012: 5). Although promising, further research is needed as there was a significant amount of variation within the findings.

Whilst the above models of practice are focused on the transition from prison into the community, another approach outlined in a paper included in the Hopkin et al. (2018) review is a collaboration between a probation officer and a mental health treatment program in a community mental health centre. This innovative intervention aimed to ‘ensure high-quality coordination of care, with ongoing feedback between clinical and probation personnel’ (Roskes and Feldman, 1999: 1616). Care was provided by staff at the community mental health centre, and the probation officer was kept informed of progress or non-compliance and could have input into decisions made including changes to a treatment plan. Just 16 people had been referred to the intervention during the first 24 months, 10 of which were still in treatment when the paper was produced.

Another significant innovation within probation practice in the US is specialist probation officers, which are described below.

A review of the specialist practitioner role in the United States

Two decades ago, the Council of State Governments published the Criminal Justice/Mental Health Consensus Project (Council of State Governments, 2002) that described strategies to improve the way that criminal justice authorities in the US respond to people with serious mental illness. Included in this report were a number of strategies for parole and probation supervision, including providing training for parole and probation officers to better supervise people with mental illness, increasing collaboration and communication with mental health organisations, enhancing collaborative discharge planning from prison to the community and reducing caseload sizes for specialised mental health probation and parole officers. The report included several examples from communities across the US that implemented a specialised mental health probation approach in their jurisdiction. These case studies served as innovative examples to help other jurisdictions adapt such a mental health approach to their local community contexts. These case examples also showed that, within the context of the US highly de-centralised criminal justice system, there is significant model variation making it difficult to develop the evidence base for interventions that could improve both mental health and criminal justice outcomes.

Soon after the report from the Criminal Justice/Mental Health Consensus Project, a seminal work by Skeem and colleagues was published (2006). This research study was the first to name five prototypical elements of speciality mental health probation approaches and begin to direct the focus of criminal justice scholars towards developing an evidence-based approach. These five elements were consistent with those proposed by the Council of State Governments (Council of State Governments, 2002) and included: (1) exclusive mental health caseloads, (2) smaller caseloads compared to traditional probation, (3) specially trained officers, (4) integration of internal and external resources (e.g. resource coordination and treatment linkage) and (5) the use of problem-solving strategies to address non-compliance. Although mental health probation interventions are still understudied in the US, to date four studies have examined the efficacy of the 5-component prototypical model advanced by Skeem and colleagues (2006).

Wolff and colleagues (2014) used a quasi-experimental design to compare three groups of people on probation with mental illness. The first group was comprised of people on probation with mental illness who were assigned to traditional probation caseloads in which officers had a caseload of around 130 people. The second group consisted of people with mental illness on probation who were assigned to a pilot caseload in which officers had a reduced caseload of about 30 people. The third group was comprised of people with mental illness on probation who were assigned to specialised mental health caseloads in which probation officers had about 50 cases. The latter group was also subject to fidelity monitoring. Results from the study showed that those on mental health caseloads had better mental health outcomes and all study participants showed increased probation violations; however, those on speciality caseloads had fewer jail days and fewer arrests due to probation violations.

Manchak and colleagues (2014) also used a quasi-experimental design with propensity score matching to compare traditional caseloads to the prototypical mental health probation model (Skeem et al., 2006). This study further elucidated the role of the mental health probation officer noting boundary spanning behaviours, increased frequency of contacts and better relationships with those on their caseloads. Results from this study showed positive treatment outcomes including greater access to mental health treatment and integrated dual disorder (i.e. mental health and substance use) treatment. Results also showed that although speciality probation officers were more likely to be aware of violations, they were less likely to report a violation, which highlights the role of officer discretion in supervision.

A follow-up to the Manchak et al. (2014) study was published by Skeem and colleagues (2017) and used the same sample, intervention and propensity score matching techniques. In this follow-up study, longer-term criminal justice outcomes were examined, including violent or aggressive acts reported after 1 year and arrests after 2 years. In terms of violence, study results showed no difference between the traditional probation group and those in the speciality mental health probation group. However, people with mental illnesses under traditional supervision were more than twice as likely to be re-arrested within 2 years compared with those on speciality mental health caseloads.

The fourth study of prototypical speciality mental health probation was a randomised controlled trial (Van Deinse et al., 2022a) in which people with serious mental illness were randomly assigned to either traditional probation officers or speciality mental health probation officers in one rural and one urban county. Results from this study showed that speciality officers were more likely to address the mental health needs of those on their caseload compared with traditional officers. In addition, people on speciality mental health probation showed greater mental health treatment engagement compared to those supervised on traditional caseloads. In terms of violations, there were no significant differences in technical violations and any type of violations after 1 year; however, within 6 months, more people on the speciality mental health caseloads had violations due to new crime compared to those on traditional caseloads. In addition to the mixed results, there were some challenges related to program fidelity. Specifically, some caseloads were not fully designated for people with mental illness. Rather, there were some mixed caseloads for speciality officers.

Overall, the research supporting speciality mental health probation is promising. Within the US and in other countries with high degrees of de-centralisation and variation in governance of probation systems, wide variability is expected and perhaps welcomed, given the unique implementation contexts within each jurisdiction. However, the growing research evidence around these prototypical approaches using increasingly rigorous research methods (e.g. propensity score matching, randomised controlled trials) shows that the field can and should continue to coalesce the research to establish evidence-based practice for supervising people with serious mental illness on probation and parole.

Discussion and conclusion

The review above is in two parts. In the first section, we mostly report on the findings of a systematic review of serious mental illness at the intersect between prison and the community which includes the use of probation. The second section of the review focuses on rigorous evaluations of the ‘specialist mental health probation practitioner’ in the US.

An important observation is that there is a dearth of research in this area from Europe and, in particular, the United Kingdom, where we could find only one quantitative study examining any intervention, Critical Time Intervention, used when mentally ill prisoners were released (Shaw et al., 2017). Unfortunately, this study found no differences in outcomes at 1 year follow-up and the intervention was as costly as the control condition. In contrast, in the US, the approaches that have been evaluated, include: ensuring Medicaid insurance upon release (Morrissey et al., 2016; Wenzlow et al., 2011); intensive case management in the community (Buck et al., 2011; Theurer and Lovell, 2008); assertive community treatment (Cosden et al., 2003; Cusack et al., 2010; Solomon and Draine, 1995); living in a therapeutic community (Sullivan et al., 2007) and active alliance with a community mental health centre (Roskes and Feldman, 1999). In addition to the interventions tested above, in the US, there is another group of interventions that come under the heading of the ‘specialist practitioner role’ for probation staff which warranted a review of their own. It is clear that, compared to the UK and Europe, there has been far greater investment in research for mentally ill people on probation in the US. This research has revealed a wide range of possible models for the care and treatment of mentally ill people on probation. Why might this be?

One reason is that in 2002 a series of focus groups, held under the aegis of the Council of State Governments, produced a consensus statement on the best approaches to mental health in the criminal justice system. In this large enterprise, there were a number of advisory boards convened, under the following headings: law enforcement, courts, corrections and mental health. The most relevant recommendations for probation are listed below:

• Assign offenders with mental health conditions on probation, to probation officers with specialised training and small caseloads.

• Mental health providers whose clients are on probation, while being careful not to become monitors of compliance, can also assist the individual to understand the consequences of their behaviour in terms of sanctions and can build a collaborative relationship with the specialised probation officers.

• Develop a program that seeks to prevent a probation revocation by offering intensive treatment rather than incarceration for those who violate probation conditions.

(Source: Council of State Governments, 2002)

In England and Wales, a similar initiative was published in 2009 by the Department of Health, the so-called Bradley Report. Lord Bradley’s brief was to conduct an ‘independent review to determine to what extent offenders with mental health problems or learning disabilities could be diverted from prison to other services and what were the barriers to such diversion’. Sadly, this report focused little on probation. In comparison to the US, then, there is no discussion about best practice in Europe and very little evaluation of possible models to inform any such agreement. Indeed, recent survey work undertaken for the Council of Europe (Brooker and Monteiro, 2021) demonstrated that half (53%) of all European countries had a policy framework for mental health and probation compared to over 90% of countries that had a policy for mental health in prisons (94%). In comparison to the US, the Council of Europe survey also revealed other important differences. For example, a study conducted in the US asked a random sample of probation agencies to provide prevalence rates of mental illness among people on their caseloads (Van Deinse et al., 2022b). Of the 179 survey respondents, 33% were able to provide a prevalence rate. Most of the respondents who provided a prevalence rate estimated the figure rather than using agency data to determine the rate. In Europe, 9% of probation agencies were able to provide a prevalence rate. In the same US study, probation agencies were asked whether they implement a mental health probation approach, including speciality officers, embedded treatment, enhanced training and more. Approximately one-fifth of respondents reported having a mental health probation approach. By contrast, one out of 47 Countries/Jurisdictions in Europe has a mental health probation approach. Finally, in Europe, 11% of countries use standardised assessment tools to assess mental health whilst in the US study, 27% reported using a mental health screening tool and 11% reported using a mental health assessment tool. It has been argued in a recent paper published in Europe that probation services should utilise a standardised set of outcomes measures in order to screen the mental health status of all probation clients (Brooker et al., 2022). Overall, it can be concluded that the emphasis on mental illness is far stronger in the US than it is in Europe.

Research methodology is also an important topic to acknowledge here. On the one hand, the research on mental health and probation is enhanced by the rigorous study designs that have been demonstrated in randomised and quasi-experimental designs in the US and in one such study in Europe. On the other hand, study rigour in terms of data collection is limited in the European probation research context given the lack of quantitative research methods (Brooker et al., 2023). Although qualitative methods are important for understanding phenomenon, quantitative research methods in probation research is better suited for examining the impact of interventions and probation approaches on key outcomes such as re-arrest, violations and mental well-being.

In summary, there are four essential recommendations that flow from these data presented here:

1. Establish standardised protocols for screening people on probation for possible mental illness.

2. Develop guidelines that are backed by European authorities that detail the role of specialist officers (similar to the US the US CSG report noted above).

3. Provide funding for specialist officer posts within European probation settings.

4. Engage in rigorous analysis measuring fidelity to the guidelines and the outcomes of adopting these specialist officers.

The time has come to effect change not just in the lives of probationers with a mental illness but also the staff in probation services in Europe who work with impossibly high caseloads and who have limited training in mental illness. We call here for a European multi-centre trial of the mental health specialist role in European probation services. If this is not possible, at the very least, there should be pilot studies across Europe using the highly valuable lessons that have been learnt in the US.