Legal Frameworks

In this section, an overview of forensic services and provisions across four general legal frameworks is provided, including: common law, civil law, the legislation of former Communist countries (Soviet, Soviet-controlled, and the Balkans), and Islamic law. We summarize how they address the main issues of fitness to plead, the mental disorder defense, diminished responsibility, hospital discharge provisions, forensic population composition, and forensic service provision. Additionally, information available in English, French, or Spanish is reviewed to gain insight into issues such as the presence of dedicated forensic institutions, the degree of integration between forensic and general mental health, the centralization of services, the continuity of services at discharge, and the presence of dedicated housing support for the forensic population. It is important to note that the information presented below relies exclusively on academic and nonacademic literature, and is accurate, complete, and up-to-date to the extent that the source documents permitted. Information was readily available for some jurisdictions (e.g., Canada, United Kingdom), but it was scarce for many regions of the world.

Forensic Mental Health Service Provision

Summary

The review above clearly shows the wide variation in legal frameworks around such issues as fitness to plead and criminal responsibility, as well as the absence of clear processing patterns, both within and across large legal systems. No straightforward conclusion about the organization of forensic services and the actual comparability across nations can be made. Despite these limitations, some trends do emerge. Fitness to stand trial is much more common in common-law countries in comparison to civil-law countries, where it is nearly nonexistent, with the exceptions of Germany, Chile, Argentina, Japan, China, and Taiwan. This could be explained by the fact that common law is more adversarial in nature, whereas civil law adopts an inquisitorial approach. The concept of fitness was also available in some former Communist countries. The insanity defense, on the other hand, is a legal concept that is both deeply embedded in countries of civil-law tradition and of common-law tradition, despite being rarely used in England. The Nordic countries of Denmark and Sweden instead sentence criminally responsible offenders to treatment.

In most countries, a defendant found unfit could be either detained in a hospital (either forensic or general) or released, depending on the risk level and on the severity of the offense. Many common-law countries have put in place safeguards to prevent indefinite or arbitrary detention permanently of unfit defendants. After a given amount of time, if found permanently unfit, the defendant would either be released, civilly committed, or submitted to a special hearing.

The population characteristics of the forensic population can also vary in terms of diagnosis and type of index offense. The types of mental disorders that might be considered as a basis for partial or full lack of criminal responsibility (or for a treatment sentence in some cases, including England) were rarely explicitly stated in the law, and most often were at judges’ discretion, which resulted in many countries being ‘on the fence.’ The pattern shows that psychotic illnesses (mainly schizophrenia) are accepted, along with severe affective disorders and organic mental disorders. The inclusion of personality disorders, however, varies widely. Offenders with a personality disorder are an important proportion of forensic patients in England and Wales, the Netherlands, Belgium, Germany, Austria, and Greece, but they are rarely admitted elsewhere. The severity of the offenses committed range from mostly misdemeanors (Israel) to mostly severe violent crimes (the Netherlands). Sexual offenders rarely reach forensic services through a mental disorder defense, with the exception of South Africa.

While the duration of the forensic hospitalization is, in many jurisdictions, unlimited and subject to regular reevaluation, other countries embedded safeguards, either to ensure that forensic patients are detained a minimum amount of time or no longer than a maximum amount of time, often in relation to the level of severity of the offense. While in Italy the minimum length of treatment must be related to the seriousness of the offense, it is the maximum length that is related to the seriousness of the offense in Ireland and most of Australia. The court may also establish the duration, such as in Portugal (maximum), Brazil, and Greece (both minimum). The decisional power to discharge a patient most often rests with the court, but sometimes rests with the treating psychiatrist (for instance, in England and Wales, South Africa, and China) or is subject to ministerial assent (in New Zealand and most of Australia). Independent bodies or mental health tribunals have been put in place in Canada, Ireland, Scotland, Queensland, Belgium, and Israel. In a minority of countries, such as in England and Wales, the courts cede all powers over a mentally disordered offender once they are admitted to a hospital, and the patient is essentially treated like a civilly detained patient.

One of the bases of the organization of forensic services is the extent to which they are integrated with the civil mental health services. In some cases, forensic facilities are entirely distinct from the health system; sometimes, they are under the authority of another department altogether. In other cases, there are virtually no distinction between forensic services and general mental health services and allocation is based on security rather than legal status, mentally disordered offenders being treated on the same wards as aggressive or high-risk nonoffenders, such as in Finland and England. In general, the less populous a country is, the more likely it is that the services will be centralized in one national forensic hospital.

There is an important dearth in the literature when it comes to postdischarge forensic services, including continuity of care and supportive housing. When information is available about the follow-up programs, performance indicators and outcomes are rarely available, which makes drawing conclusions about the actual continuity of care nearly impossible. It is worth noting that the United Kingdom is arguably the jurisdiction where forensic services are the best documented in the literature. Clear pathways of care have already been developed and put in place, putting the country at the forefront of evidence-based policies.

The general portrait of forensic mental health frameworks around the world remains limited by the significant variability in the availability and specificity of information. No studies were identified that could indicate which kind of organization of services is most effective or efficient—an endeavor that would certainly be of interest for future research and require the development of an international consortium. Such massive heterogeneity in the design of forensic systems likely produces major variation in forensic mental health outcomes; however, the nature and magnitude of this variation is not well understood. Some attempts have begun in this direction in relation to long-term forensic patients by the COST team. ¹⁰ Despite being unable to draw conclusions about the superiority of one approach for organizing forensic services over another, key system-level principles can be extracted from this review as well as the wider literature on the organization of mental health services in order to propose a model for the organization of forensic services.

A Balanced Service Continuum

As evidenced in our international review, contemporary forensic mental health systems tend to be aligned with a ‘custodial paradigm’ (McKenna, Furness, Dhital, Park, & Connally, 2014), whereby resources and services are concentrated in institutional settings, such as stand-alone forensic institutions, specialized units in general mental health hospitals, or inpatient services offered in correctional settings. Forensic institutions play a vital role in containing and mitigating public safety risk, enabling justice-involved people with mental disorders access to therapeutic processes and interventions, and facilitating community reintegration, crime desistance, and personal recovery. At the same time, forensic institutions also embody the oppressive features of a ‘total institution’ (Goffman, 1961), whereby detainees (or inpatients) are, to varying degrees, exposed to deprivations of liberty, autonomy, material goods, intimate relations, and security. Forensic institutions are also a very costly means of providing services that could be delivered in community-based settings with equal, if not greater, effectiveness.

Emerging from similar concerns regarding institutional care is the proposal that mental health and substance use systems are most effective when they provide access to a comprehensive and balanced continuum of services and supports. In a balanced care approach, a flexible range of services is primarily provided in communitybased, local settings that span the specialized and nonspecialized sectors and emphasize features such as providing care close to home, tailoring treatment to individual need, supporting the choices of service users, ensuring that care is coordinated between providers and agencies, and offering mobile services (Thornicroft & Tansella, 2013). A balanced care model that provides ready access to evidencebased treatment interventions would prevent some people with mental illnesses from unnecessarily becoming involved in the criminal justice system. It would also ensure that high-quality services reach those with mental health needs who are in the criminal justice system (e.g., arrested by police, detained or incarcerated, on parole or probation) in order to prevent further criminalization (Munetz & Griffin, 2006).

One conceptual framework that is used to portray what balanced care systems look like is the ‘tiered model’ (National Treatment Strategy Working Group, 2008; Organization of Services for Mental Health, 2003; Rush, 2010). A fundamental aspect of the tiered model is that mental health and substance use systems contain logical groupings of services that have comparable levels of intensity, specialization, and restrictiveness. Comparable services can be clustered into tiers that are most appropriate for meeting the needs of distinct groups of people based on the acuity, complexity, and chronicity of their mental health and substance use problems. For instance, highly specialized and intensive (and costly) services are most suitable for people with severe (i.e., highly acute, complex, and chronic) problems, and least appropriate for people experiencing low to moderate levels of distress or impairment. Another feature of the tiered model is that the distribution of services should align with the population distribution of problem severity. Those with lower-level problem severity are more numerous in the population and, as such, the demand (i.e., volume) for less intensive, less specialized services is great. Conversely, severe mental health and substance use problems are less common in the population and the demand for highly specialized, highly intensive services is small. The shape of service systems should correspond with the demand for different service levels.

According to the principles of the tiered model, a balanced forensic mental health care continuum would resemble the figure below.

Drawing from other frameworks (e.g., Kennedy, 2002; Mullen, 2000; O’Dwyer et al., 2011; Thornicroft & Tansella, 2013; World Health Organization, 2003), this conceptual model is designed with six tiers that contain three clusters of custodial-based forensic mental health services (Tiers 4–6) and three clusters of community-based services (Tiers 1–3). Each service cluster would contain evidence-based interventions corresponding with the particular function of that cluster. The menu of interventions available within each cluster will vary between jurisdictions. Compared with other tiered service models designed for general mental health and substance use systems, the model depicted here is weighted more heavily toward institutional services in order to reflect the distinct nature of forensic populations and the risk management function of forensic systems. Nonetheless, striking a balance between custodial- and community-based services is vital for forensic mental health systems (Mullen, 2000), with some suggestion that the investment in community services should be equal to or greater than the amount being spent on custodial care (Thornicroft & Tansella, 2013).

As was revealed by our review of forensic systems, the upper half of the pyramid (Tiers 4–6) parallels with how parts of the world currently distribute custodial populations in low-, medium-, or high-security settings (Pillay, Oliver, Butler, & Kennedy, 2008). Tier 6’s high-security, custodial services are the most restrictive, intensive, specialized, and costly in the forensic mental health system; it also serves the smallest portion of the forensic population. The focus of services in Tier 6 is on stabilizing acute mental health and substance use problems, while also targeting specific criminogenic needs in order to enable a safe transition to a lower security level. In contrast, Tier 4’s low-security, custodial services comprise the least intensive and least specialized services within custodial care settings. Tier 4 services are generally designed to prepare service users for transitioning into the community, for instance building self-management, relapse prevention, and independent living skills. The lower half of the pyramid (Tiers 1–3) provides a continuum of care in the community that is appropriately matched to service users’ level of criminogenic, mental health, and substance use treatment needs. Tier 3’s high-intensity community services likely include supportive and structured living environments (e.g., supported housing, residential facilities) combined with assertive outreach and treatment provided by specialized, multidisciplinary forensic mental health service providers (e.g., forensic assertive community treatment teams). Tier 1’s low-intensity community services require the least forensic specialization and would focus on bridging people to nonforensic systems of care.

Figure 1.1   Balanced Forensic Mental Health Care Continuum

Matching Services to Needs

Integral to the tiered model is the principle that systems are best organized in a manner that allows individuals to access a full range of services according to their level of need (National Treatment Strategy Working Group, 2008; Thornicroft & Tansella, 2003; World Health Organization, 2003). A system that offers services tailored to an individual’s needs, also known as a ‘stepped care’ model, can effectively respond to varying levels of illness chronicity, acuity, and complexity, and is responsive to a person’s changing needs (Bower & Gilbody, 2005; Rush, 2010). Comprehensive and regular assessments of needs that match people with the appropriate services (type and intensity) is an essential feature of a system that is organized around a stepped care approach (Goldman, Thelander, & Westrin, 2000). A welldesigned system provides individuals with the least restrictive, least intensive, least expensive, and least intrusive interventions, with individuals stepped up to higher intensity care or stepped down to lower-intensity care based on current service need (McDaid & Thornicroft, 2005; Von Korff & Tiemens, 2000).

These system-level principles are well aligned with evidence-based practice for correctional interventions (Andrews & Bonta, 2010; Lipsey & Cullen, 2007; Prins & Draper, 2009; Romani, Morgan, Gross, & McDonald, 2012). Of particular relevance is the risk–need–responsivity (RNR) theoretical framework, which posits core principles for correctional services to effectively treat offending behavior (Andrews, Bonta, & Hoge, 1990; Andrews, Bonta, & Wormith, 2011; see review in Chapter 16 of this book). The RNR model indicates that service intensity should match the level of risk for reoffending presented by an offender, with intensive correctional services being reserved for higher-risk offenders (risk principle). Alongside this, services should target and address criminogenic needs—the dynamic risk factors with an established empirical linkage to reoffending (need principle). Furthermore, interventions should be oriented toward cognitive behavioral strategies and the offenders’ learning style, motivation, abilities, and strengths (responsivity principle).

Another applicable theoretical model, developed by Prins and Draper (2009) and subsequently elaborated on by others (Osher, D’Amora, Plotkin, Jarrett, & Eggleston, 2012; Skeem, Manchak, & Peterson, 2011), has applied RNR principles to the management of individuals with mental illness under community corrections supervision. This framework posits that criminogenic needs and functional impairment should determine the intensity of, and degree of coordination between, criminal justice and mental health services. An increase in criminogenic needs and risk levels should be matched by intensive supervision. Similarly, higher levels of functional impairment should be matched with intensive mental health treatment. Concomitantly, increases in service intensity should necessitate greater coordination and integration between mental health and correctional services.

A forensic mental health system that is organized according to the tiered model and the Prins and Draper (2009) framework would match service clusters, containing distinct therapeutic interventions and security measures, to target groups based on their risk and need profile, as is depicted in the table below. Others have proposed similar strategies for distributing resources based on the needs of adults with behavioral health problems who are under correctional supervision (Osher et al., 2012). Additionally, the ‘stratified risk and care’ model described by Pillay and colleagues (2008) outlines a similar way that forensic service users are distributed across security levels according to their level of clinical need and risk for violence and crime.

Custodial forensic services (Tiers 4–6) are most appropriate for forensic service users who are assessed as being at moderate to high levels of criminogenic risk, as well as those with severe mental health and substance use problems. The Prins and Draper (2009) model suggests that persons with higher levels of criminogenic needs accompanied by more serious mental health and substance use problems would be the most likely to benefit from a highly specialized forensic mental health service, with its integration of risk assessment, risk management, and mental health and substance abuse treatment. Community forensic services (Tiers 1–3) may be most appropriate for forensic service users assessed as being at low to moderate levels of criminogenic risk and have low to moderate mental health and substance use treatment needs. The needs of forensic service users who have lower levels of criminogenic, mental health, and substance use needs are likely to be met by services with lower levels of forensic specialization. Those with higher criminogenic needs who are living in the community may be best served by highly specialized forensic services, such as forensic assertive community treatment.

Much of the literature in this area focuses on criminogenic risk and need factors, but a recent body of scholarship has cast doubt on the validity of using criminogenic risk to inform decisions regarding the assignment of service users to different security levels in a forensic mental health care continuum (O’Dwyer et al., 2011). A set of instruments, known as the DUNDRUM quartet, have been developed to inform decisions concerning the placement and movement of service users in the forensic mental health system (Kennedy, O’Neill, Flynn, & Gill, 2010). These instruments rely on a different set of variables, compared to those used by criminogenic need and violence risk appraisal instruments, to assess the appropriateness of security level placement. Although more research is needed on the effectiveness of instruments such as the DUNDRUM quartet to improve system-level processes and outcomes, adjustments to the above table may be warranted, such that the service clusters and security levels (especially in custodial settings) are matched to a target groups’ therapeutic security needs, rather than their criminogenic, mental health, and substance use treatment needs. The latter batch of variables may be more appropriate for matching individuals to appropriate interventions (type and intensity), rather than determining their placement within the forensic mental health care continuum.

Integrating Systems

Forensic service users need to access different systems and subsystems to have their intersecting needs appropriately met. Many obstacles prevent services from achieving optimal levels of integration and coordination, such as fragmented funding mandates. The complexity of the criminal justice, psychiatric, and social problems existing within the forensic mental health population present challenges for coordinating and integrating services for this group—especially since forensic service users are undesired, unwanted, and rejected by other service systems (Wolff, 2002).

Receiving care from different systems, or different components of a system, is very problematic in a population with co-occurring, complex needs, and that has experienced severe social ruptures and interrupted care trajectories. A fundamental principle for effectively organizing mental health and substance use services is that systems should be integrated (Craven & Bland, 2006; Thornicroft & Tansella, 2003). In the forensic mental health context, system integration refers to bridging the service delivery gaps that exist within and between mental health, substance use, criminal justice, corrections, housing, and other essential service sectors. The purpose of systems integration is to encourage seamless service delivery, to promote efficiency, to optimize the use of scarce resources, and to improve outcomes. Systems integration is promoted through structures and processes that facilitate collaborations, partnerships, and communication between professionals and agencies.

Models of systems integration between mental health, criminal justice, and community support services have been developed (Lamberti, Deem, Weisman, & LaDuke, 2011; Osher et al., 2012; Weisman, Lamberti, & Price, 2004). Our review of forensic systems also shows various ways in which different jurisdictions throughout the world have tried to build systems integration in order to meet the needs of forensic service users. One aspect of systems integration focuses on linkages and coordination among the various components of the mental health, substance use, and correctional service continuum. Integration between the custodial and community settings is also essential for the efficient and effective delivery of forensic mental health services. Opportunities for multi-system collaboration and system integration exist throughout the tiered model. For instance, the needs of forensic service users at the lower tiers (e.g., 1 or 2) could be effectively met by the general mental health system, since the need for specialized forensic services may no longer exist (Humber et al., 2011), an approach that has been adopted by many countries. Integrating risk management strategies and transferring forensic skills into the general mental health system, or transitioning forensic service users to the general mental health system when appropriate has a number of advantages (Jansman-Hart et al., 2011), including freeing up space in the forensic system but also avoiding unnecessarily restricting the liberty of forensic service users.

As our international review of frameworks demonstrated, numerous options exist for organizing forensic mental health systems. In addition to population differences, the models vary in the degree to which services are integrated with the other sectors, such as the general, nonforensic mental health system as well as the correctional system. One approach, which we call a ‘specialized’ model, offers specialized forensic services to all forensic services users throughout their care pathway in custody (e.g., forensic hospital) and the community (e.g., forensic outpatient care). In contrast, a ‘mixed’ model provides access to specialized care based on need whereby only some forensic service users (e.g., moderate–high risk) receive specialized forensic services, while others receive hospital- or community-based services in the general mental health system. This model’s focus on tailoring care (specialized or integrated) based on individual need is consistent with the principles of the tiered model. Lastly, in a ‘general’ model, forensic service users receive services either entirely (i.e., hospital and community) or partly (i.e., community only) in the general mental health system—an arrangement that is primarily based on the availability of resources rather than people’s needs. There are theoretical strengths and weaknesses of these different ways of integrating systems, but their relative effectiveness requires further study.

Risk-Averse Society

The concept of risk refers to the probability of an event and the magnitude of its consequences; risk and probabilities can be difficult to grasp and can lead to a variety of responses in society. “Risk events interact with psychological, social and cultural processes to heighten or attenuate public perceptions of risk and related risk behavior” (Kasperson et al., 1988, p. 178). Sociologists argue that modern society is increasingly being structured around risk, with social institutions constituted to protect the population from perceived dangers (e.g., terrorism, climate change, infectious diseases, and financial crises: Beck, 1992). Some very low-risk phenomena, such as violent behavior toward a stranger by a person with a mental illness, can elicit strong public concern and insecurity (Monahan & Steadman, 2012). A fearful and anxious public turns to their government for protection, and, in turn, government officials (e.g., politicians) maintain social order by convincing the public of their capability to manage and control risk. Some politicians strategically use such opportunities to gain popularity and political advantage by promoting punitive and coercive measures (e.g., incarceration) to suppress ‘risky problems’ (Pratt, 2007). These harsh punitive measures are problematic in that they are largely symbolic and undermine the rehabilitative objectives of the criminal justice, forensic, and correctional systems.

Most Westernized or high-income societies are increasingly risk-averse—the current political situation around the world regarding the influx of Syrian immigration in many countries is probably one of the most obvious examples at this time. Closer to our current subject, one only has to think of the ‘Not In My Backyard’ (NIMBY) ¹⁷ phenomenon that persons with mental illness are often up against when trying to find housing in their preferred neighborhood. This risk aversion plays an important role in the organization of forensic mental health services, in particular regarding security levels, discharge planning and decisions for patients as well as legislation and policies pertaining to mentally ill offenders. In Canada, for instance, the Not Criminally Responsible Reform Act, which came into law in July 2014 under the conservative government, created a ‘high-risk accused’ category, mostly based on the severity of the offense, which increased and prolonged the restrictions placed on the subgroup of people found NCRMD. This type of reform is risk aversion-based, not rooted in any evidence, and the product of a penal populist style of governing (Tersigni, 2016). Penal populism is generally understood as a policy framework based on public opinion and fear rather than expert advice or an evidence base and leads to policy and legislation that are tough on crime. The tough on crime approach plays on its appeal for the public to have ‘safe communities’ and reduced violence and criminality regardless of the evidence (Bousfield, Cook, & Roesch, 2014; Cook & Roesch, 2012). A recent simulation study was conducted by applying retrospectively the criteria of the legislation on a sample of individuals found not criminally responsible on account of mental disorder. Results indicate that individuals who would potentially be designated highrisk accused reoffended at similar or lower rates than non-high-risk accused individuals (Goossens et al., 2016), thereby demonstrating that the new legislation may not in fact have the intended effects. Zero risk of course does not exist and can never be ensured in regards to human behavior, but the dangers of the risk-averse culture are the increased stigmatization, marginalization, and intolerance of persons with a mental illness who become involved with the criminal justice system. This, in turn, makes it increasingly difficult to assist individuals who need it most.

Even within the mental health services, some community service providers (e.g., mental health, housing, employment) are reluctant to work with discharged forensic service users who are perceived as being a risk of violence. Community forensic services are, thus, not always available or able to follow up patients that are returning to the community. In Ontario and British Columbia, two of the largest provinces in Canada, the civil health care systems are allowed to refuse to treat forensic patients. As a result, the patients may be staying in the forensic system for longer than necessary and thus occupy forensic beds that are otherwise greatly needed (Crocker, Nicholls, Seto, Roy, et al., 2015c).

The Perceived Dichotomy Between Recovery and Public Safety—Fuel for Penal Populism

The generally poor understanding of the relationship between mental illness and violence in the general public is partly due to the fact that, historically, the research has not always been clear about this relationship (Monahan, 1992) and researchers have not effectively communicated what is known about justice-involved mentally ill persons in general (Crocker et al., 2015c). It may also be partially reinforced by an emphasis on a deterministic biomedical model of mental illness that can create dualistic thinking between individual recovery and public safety, in the eyes of the public and politicians. In fact, when it comes to violence and mental illness, there is a general tendency to view recovery and public safety as opposites, when the evidence has shown that it is through rehabilitation and recovery that one can achieve a reduction in risk (Dvoskin, Skeem, Novaco, & Douglas, 2012). The risk–need–responsivity model research also shows that excessive intervention that is disproportionate to risk can actually have the inverse effect of increasing risk (Andrews, 2012) and, consequently, decreasing public safety. Research has clearly shown that punitive policies and coercive practices tend to, in the long term, exert a negative effect on public safety. Some of the best examples were observed in the United States through the tough on crime legislation of the 1990s (Blumstein, 2012). Traditionally, conservative governments have argued for more incarceration/punitive approaches, while liberals tend to promote crime prevention strategies based on social development (Blumstein, 2012). The resurgence of conservative, right-wing governments around the globe will continue to pose challenge for forensic mental health services in striving for recovery-oriented approaches despite increasing pressures to be punitive, often fueled by highly mediatized cases to use punitive and coercively driven approaches. It will become increasingly important to develop clear messaging around short-, medium-, and long-term outcomes and programs that reduce recidivism rates in forensic populations and improve public safety in order to guide policy-making and reduce stigma. Future research could explore what are the most effective strategies for educating policy-makers and the general public around justice-involved persons with mental illness (Crocker et al., 2015c).

Forensic Mental Health Stigma

Stigma is a social process that aims to exclude, reject, shame, and devalue groups of people on the basis of a particular characteristic (Weiss, Ramakrishna, & Somma, 2006). It is conveyed by the widespread perception among the general public that individuals with mental illnesses are dangerous (Jorm, Reavley, & Ross, 2012; Mestdagh & Hansen, 2014; Parcesepe & Cabassa, 2013; Pescosolido, 2013; Schomerus et al., 2012). It manifests when people with mental illness feel as though they are devalued and discredited members of society, which can lead to hopelessness, poor self-esteem, disempowerment, reduced self-efficacy, poor treatment adherence, and decreased quality of life (Livingston & Boyd, 2010) and it ultimately leads to numerous structural barriers for accessing social and economic opportunities, such as employment, education, and housing, ultimately restricting the rights of people with mental illness and hindering their recovery.

Forensic mental health service users face substantial stigmatization, which results from intersecting psychiatric and criminal labels as well as other forms of marginalization (e.g., poverty, racism). Roskes and colleagues (1999) identified ‘double stigma’—the co-occurring statuses of being mentally ill and a convicted criminal—as a barrier to community reintegration. Similarly, in a study with ex-offenders who had co-occurring mental health and substance use problems, Hartwell (2004) provided evidence to suggest that problematic community reintegration is compounded by the intersection of these three statuses (mentally ill, addicted, ex-con), which she called ‘triple stigma.’ A small body of research has investigated stigma in the forensic mental health system (Margetić, Aukst-Margetić, Ivanec, & Filipčić, 2008; West, Vayshenker, Rotter, & Yanos, 2015; West, Yanos, & Mulay, 2014; Williams, Moore, Adshead, McDowell, & Tapp, 2011). The results of one study indicate that receiving specialized forensic services is associated with increased exposure to social stigma, such as being perceived by others as potentially violent. Additionally, forensic service users commonly experience instances of structural stigma, such as being refused access to services (e.g., housing, day programs) on account of the ‘forensic’ label (Livingston, Rossiter, & Verdun-Jones, 2011). These barriers are in addition to those that are directly related to mental illness and other forms of social and economic disadvantage.

Stigma is a major challenge for many mental health service users, including those who are in the forensic system. Obtaining employment, housing, and necessary services is made more difficult through the interlocking effects of multiple statuses and labels that combine to create oppression, disempowerment, and marginalization. To protect themselves from stigmatizing experiences, forensic service users may engage in behaviors, such as social withdrawal, that are unproductive for recovery from mental illness and desisting from crime. They may also be blocked from accessing vital resources (e.g., housing, social network) that are required for living a prosocial, healthy life in the community. Although more research is needed, the existing evidence suggests that stigma is influenced by the policies and practices of forensic mental health systems. The fact that forensic services users are exposed to greater levels of social and structural stigma is a factor that decision-makers should be aware of so that they can work toward mitigating such negative effects of the forensic system. At an individual level, this might include making programs available to help service users cope with negative effects of stigma, or, at a structural level, this would involve rectifying policies and practices that arbitrarily restrict forensic service users’ access to social and economic opportunities.

Providing Increased Ward Safety—The Safewards Model

Challenging behaviors such as verbal and physical aggression, self-harm, suicide, and medication refusal are relatively common occurrences in forensic settings (Nicholls, Brink, Greaves, Lussier, & Verdun-Jones, 2009). These behaviors place patients and staff at risk of serious harm and hamper rehabilitative efforts. Various ‘containment’ strategies have been used, from chemical and mechanical restraints to seclusion and special observation by staff in an attempt to manage these behaviors or improve outcomes (Bowers et al., 2015). Whereas some health care professionals defend these practices as necessary measures for protection, others have described the practice of seclusion and restraint as dangerous violations of human dignity and rights (van der Schaaf, Dusseldorp, Keuning, Janssen, & Noorthoorn, 2013).

The Safewards model is one of the most recent approaches for reducing and even eliminating seclusion and restraints in civil psychiatric settings (a). The reciprocal relationship between conflict and containment are at the center of the model, whereby conflict triggers containment and containment procedures trigger conflict. In this model, conflict refers to the risk behaviors that service users engage in that could threaten their own safety and the safety of others. Containment refers to the prevention strategies that seek to minimize harmful outcomes or prevent conflict events from occurring at all. The model is constructed around the following six domains, which identify the main influences on conflict and containment rates: the staff team, the physical environment, outside hospital, the patient community, patient characteristics, and the regulatory framework (Bowers, 2014). The staff team domain includes the internal structure of the ward, which is composed of rules of service user conduct, daily and weekly routine (i.e., what happens and where), and the overall ideology asserted by the staff to highlight the purpose of the ward and what it offers to service users. The physical environment domain refers to the built environment’s quality (i.e., better-quality environments evoke greater respect for patients, more comfort, and greater care) and complexity (i.e., more difficult to observe environments make supervision by the staff harder). The outside hospital domain highlights the stressors from outside the hospital relating to the service users’ friends, family, or home. The patient community domain pertains to the collective values, beliefs, and behavior among service user groups and their relevance for producing contagion or discord. Moreover, the patient characteristics domain expresses a large variety of patient characteristics that can give rise to conflict behavior (i.e., symptoms, personality traits, demographic features). Finally, the regulatory framework domain refers to the external structure of the ward and includes the constraints on the service users’ behavior dictated largely from outside the ward itself (i.e., Mental Health Acts, coerced detention, national policies, hospital policies).

With the aforementioned factors in mind, the Safewards approach seeks to reduce conflict as well as reduce the use of seclusion and restraints through workforce training, service user involvement, debriefing techniques, senior management commitment to change, and the use of audits to inform practice (Bowers et al., 2015). The model also enables the creation of a list of interventions that could enhance the staff modifiers (i.e., actions of the staff as individuals or teams) that can influence the frequency of conflict and containment, thereby reducing conflict and containment rates (Bowers et al., 2015).

A body of research supports the domains in the Safewards model. Eggert et al. (2014) examined the person–environment interaction effects of environmental design on ward climate, safety, job satisfaction, and treatment. With regards to treatment outcomes, as perceived safety increased so did patient discharges, although only to a small degree. This result suggests that, when people feel safe in the treatment environment, treatment is more likely to have a positive impact, resulting in earlier discharge from the hospital. In another study, conducted by van der Schaaf et al. (2013), the ‘presence of an outdoor space,’ ‘special safety measure,’ and a large ‘number of patients in the building’ increased the risk of being secluded. On the other hand, design features such as ‘total private space per patient,’ a higher ‘level of comfort’ and greater ‘visibility on the ward’ decreased the risk of being secluded. Some support was also found in a literature review by Alexander and Bowers (2004) on ward rules, providing evidence that service users were in fact calmer and less disruptive on wards with clear, consistent rules, and clear roles for staff. Finally, service users’ perceptions about the social climate of the ward were found to have a significant relationship with their satisfaction with forensic services. However, the variables most strongly associated with satisfaction with forensic services were their perceptions about the nature of therapeutic relations with staff (Bressington, Stewart, Beer, & MacInnes, 2011). Moreover, in a study of participant observation coupled with interviews of 131 staff collected over 38 months on four acute and two chronic wards, Katz and Kirkland (1990) concluded that violence was in fact more common in wards with unclear staff functions. Violence was less frequent in wards with strong psychiatric leadership, clear staff roles, and events that reflect the Safewards model’s main principles.

Bowers et al. (2015) conducted a large-scale clustered randomized controlled trial over a three-month period to test a series of interventions to increase safety and reduce coercion. The results indicate that simple interventions aiming to improve staff relationships with patients can reduce the frequency of conflict and containment. Furthermore, relative to the control intervention, when conflict events occurred, the Safewards intervention reduced the rate of conflict events by 15 percent and containment events by 26 percent (Bowers et al., 2015). Price, Burbery, Leonard, and Doyle (2016) used a service evaluation incorporating a nonrandomized controlled design to analyze the effects of Safewards on conflict and containment between and within wards. Informal feedback sessions with staff were conducted to explore views on the acceptability of the interventions and the adherence to the interventions was measured. The benefit of using the Safewards intervention in a between-ward analysis by measuring conflict and containment rates failed to reach significance. However, there was a significant association between ward, conflict, and containment. Furthermore, overall views of the interventions were valued more highly on female wards than on male wards. Staff from female wards reported increased confidence in their role, increased psychological understanding of patient behavior, and reduced fear of patients. On the other hand, in male wards, the perception was that the interventions were only of use for patients who were receptive to care and that many of the patients were resistant to engagement with nursing staff at any level beyond having their basic needs met. Hence, interviews of patients and staff about the causes of patient violence gave strong support to the importance of positive appreciation, emotional regulation, teamwork skill, technical mastery, moral commitments, and effective structure (Bensley, Nelson, Kaufman, Silverstein, & Shields, 1995; Bond & Brimblecombe, 2004; Finnema, Dassen, & Halfens, 1994; Lowe, 1992; Spokes et al., 2004). In sum, the Safewards model is showing some promising results for a safer future for patients and staff in forensic mental services and merits further empirical inquiry.

Rehabilitation—The Good Lives Model

Offender rehabilitation has been a topic of continuous debate, with a profound shift in attitudes in the last few decades (Ward & Brown, 2004). As discussed earlier in this chapter, the risk-need-responsivity (RNR) model (Andrews & Bonta, 2006) has been a prevailing approach to treatment of offenders in Canada, as well as other parts of the world (e.g., Australia, the United Kingdom, and New Zealand), for the last three decades. As its underlying assumption, the model highlights that offenders are bearers of risk for recidivism, and that the primary aim of offender rehabilitation should be to reduce this recidivism risk. Although meta-analyses provide support for the efficacy of this model in reducing recidivism among general and sexual offenders (Andrews et al., 1990; Andrews & Dowden, 2005; Hanson et al., 2002), some researchers have argued that this evidence is insufficient to conclude that this treatment-based program is in fact effective. More specifically, Ward and Stewart (2003) argue that the RNR model does not pay enough attention to the individual and the idiosyncratic goals and preferences of service users.

The model proposed by Ward and his colleagues is highly influenced by the area of positive psychology rather than the cognitive behavioral/social learning framework of the RNR oriented programs. This theory of rehabilitation is known as the good lives model (GLM) and it represents a strength-based approach to offender rehabilitation. The model promises to enhance the effectiveness of current efforts by addressing limitations of the risk management approach and helping reduce recidivism rates (Andrews, Bonta, & Wormith, 2011; Lindsay, Ward, Morgan, & Wilson, 2007; Ward & Brown, 2004; Ward & Gannon, 2006; Whitehead, Ward, & Collie, 2007). The dual focus of GLM is to help offenders to live better by promoting personally important goals and equipping them with the skills, values, and attitudes necessary to reduce and manage their likelihood of committing further crimes (Ward & Gannon, 2006).

GLM identifies the primary ‘human goods’ that ultimately lead to an individual’s sense of happiness, such as healthy living and functioning, knowledge, excellence in agency, inner peace, friendship, spirituality, and creativity (Ward & Gannon, 2006). The achievement of these primary goods is highly dependent on internal factors, such as skills and abilities, and external factors, such as opportunities and support networks, unique to the individual (Barnao et al., 2010). An individual seeking inner peace may, for instance, turn to substance use if they are unable to attain this ‘human good’ in more socially acceptable ways. The GLM integrates personal preferences into treatment planning through the completion of a comprehensive assessment to gain an understanding of how each offender conceptualizes a ‘good life.’ On the basis of that information, a treatment plan is designed to increase the person’s capacity to achieve their version of a ‘good life’ using socially acceptable and legal means. By focusing on providing offenders with the necessary internal and external conditions for meeting their human needs in more adaptive ways, the assumption is that they will be less likely to harm themselves or others.

Proponents of GLM argue that it offers a unique approach to rehabilitating mentally ill offenders. Specifically, GLM conceptualizes people with mental illness as being no different from others insofar that they, like everybody, are attempting to achieve primary goods and live worthy lives. However, the presence of mental illness negatively impacts the salience of primary goods and compromises the means (i.e., cognitive, psychological, and social skills) by which they are obtained—all of which are further impacted by external variables (e.g., level of distress, financial resources: Barnao, Robertson, & Ward, 2010). Under the GLM framework, interventions and programming in forensic settings can establish new normative means by which individuals seek (and obtain) primary goods by assisting to increase an individual’s capabilities and transition antisocial means to ones that are more socially acceptable (Barnao et al., 2010). However, the author suggests that for some cases, medical-modeled care remains the most realistic approach to achieving primary goods (e.g., inner peace: Barnao & Ward, 2015). Introducing GLM into forensic settings is timely given recent trends toward more holistic approaches to rehabilitation as well as efforts to reduce the negative effects of mental illness and criminal identities. GLM is also in keeping with human rights perspectives of offender rehabilitation for its concentration on service users as self-determining agents, rather than objects of risk management.

The most commonly cited criticism of GLM is its lack of strong empirical support. Recent programs using GLM have demonstrated positive results. For instance, Lindsay et al. (2007) reported that the GLM approach was an effective way of motivating sexual offenders with intellectual disabilities and encouraging them to engage in the difficult process of changing entrenched maladaptive behaviors. Furthermore, they also found that utilizing the principles of this model enabled therapists to make progress with particularly intractable cases (Lindsay et al., 2007). Additional research, principally with sex offender populations, has also confirmed the principles underlying the GLM model (Barnett & Wood, 2008; Harkins & Woodhams, 2012; Whitehead et al., 2007; Willis & Grace, 2008). Research with non-mentally ill offender populations demonstrate improved treatment engagement and therapeutic alliance, reduced dropout rates, and improved outcomes as a result of incorporating GLM principles into treatment programs (Barnao, Ward, & Casey, 2015; Ware & Bright, 2008). Early research with forensic populations has found that GLM helped forensic service users discover their motivation for change through supporting the development of personally meaningful goals (Barnao et al., 2015) and has allowed practitioners to integrate various treatment theories into their work with patients (Barnao & Ward, 2015). Bouman, Schene, and de Ruiter’s (2009) study of the short- and long-term effects of subjective well-being among forensic service users in the community found that satisfaction with health and fulfillment of life goals were associated with decreased offending (self-reported) in the short term. Moreover, in the long term, satisfaction with health and general life satisfaction were associated with reduced reconvictions for violence among those who were assessed as high risk. These findings support the notion that fulfillment of primary goods is associated with reduced recidivism as well as desistance. In sum, GLM has demonstrated preliminary effectiveness in enhancing treatment engagement, fostering desistance, and paying increased attention to environmental contexts. GLM has the promise to supply forensic mental health practitioners with the tools needed to engage difficult patient populations, and to strengthen the capacity of forensic service users to overcome tremendous challenges in their lives.

Providing Forensic Expertise Upstream in Mental Health

Research has shown that as many as three-quarters of individuals entering forensic services had previously received some form of psychiatric service (Crocker et al., 2015a). Studies have also reported patterns of individuals’ unsuccessful attempts to obtain the help of health care providers during times of crisis and immediately preceding the commission of violent acts (Stanton & Skipworth, 2005). This points to an opportunity for violence or criminality prevention strategies to be put in place. To do so, forensic expertise and experience must be shared upstream by bringing the risk assessment and management knowledge to community-based and civil psychiatric services and programs in order to target risk factors and potentially prevent criminal behavior and violence. This is consistent with a key recommendation of a large Canadian group of stakeholders who gathered in Montreal in 2014 to establish a set of priorities for research and knowledge transfer in the field of mental health, justice, and safety (Crocker et al., 2015c). When forensic mental health and general mental health services work in different subsystems of care, this knowledge transfer is likely to be more onerous. Targeted and well-supported knowledge transfer strategies must, therefore, be implemented to share the forensic expertise. Knowledge exchange is a two-way dialogue and exchange of information between those who receive and use knowledge and those who generate and transfer it (Mitton, Adair, McKenzie, Pattent, & Perry, 2007). Effective knowledge transfer strategies are meant to draw upon existing relationships, resources, and networks to the maximum extent possible, while at the same time building new resources as needed by users. Crocker et al. (2015) suggested that current knowledge exchange initiatives in forensic mental health remain insufficient for the systematic use of empirically based risk assessment methods; only through a collaborative process of elaboration and implementation of these tools will they be widely implemented and used (Scullion, 2002). Stronger partnerships between forensic and general mental health services could increase the cross-pollination of evidence-based risk assessment and management strategies with the aim of preventing the involvement of people with mental illnesses, especially those at risk for criminality and aggressive behavior, in the criminal justice, forensic, and correctional systems. In turn, these partnerships can provide increased opportunities for general adult mental health services to infuse forensic mental health services with recognized evidence-based practices such as integrated concurrent disorder (severe mental illness and substance use disorder) treatment strategies (Mueser, Noordsy, Drake, & Fox, 2003) and progressive therapeutic approaches (e.g., recovery-oriented). In turn, this would lead to increased continuity of care between the systems.

Patient Engagement/Service User Involvement

Over the past decade, there has been a strong push internationally toward models of mental health service delivery that are patient-centered and recovery-oriented (Mental Health Commission of Canada, 2012; U.S. Department of Health and Human Services, 2009). Patient engagement refers to the active participation and meaningful involvement of patients in a range of activities and decision-making processes in the health care system (Tambuyzer, Pieters, & Van Audenhove, 2011). It aims to provide patients with self-determination and control over health care decisions, and to move away from paternalistic health care practices toward systems of care that support patients’ choices and acknowledge the value of their lived experiences (Forbat, Hubbard, & Kearney, 2009; Liberman & Kopelowicz, 2005; Tambuyzer, Pieters, & Van Audenhove, 2011).

In many ways, forensic mental health systems may seem to be unsuitable places for patient engagement. Certain characteristics of forensic services, such as the public safety and involuntary orientation, complicate processes aimed at building engagement and facilitating power sharing. The real potential for dangerous situations to occur means that containment and control practices are prioritized in forensic settings, and patient engagement activities may feel unsafe and uncomfortable for forensic mental health service providers. The characteristics of some forensic service users, such as antisocial personality disorder, criminal or violent histories, susceptibility to mental decompensation, poor illness insight, treatment nonadherence, suicidality, and risk of aggression or violence, can present serious impediments to adopting patient engagement strategies in a forensic mental health hospital (Green, Batson, & Gudjonsson, 2011).

For the aforementioned reasons, patient engagement historically has not been prioritized in the forensic mental health system. Nevertheless, a growing number of scholars and practitioners, who recognize the value of patient engagement practices and recovery principles, are wrestling with the issues associated with incorporating them into forensic mental health settings (Drennan & Alred, 2012; Gudjonsson, Webster, & Green, 2010; McKenna et al., 2014; Simpson & Penney, 2011). The service user movement in some jurisdictions, such as the United Kingdom, is particularly strong and has made substantial gains in the forensic system (Bowser, 2012; Spiers, Harney, & Chilvers, 2005). In order to support patient engagement at a systems level, organizations must place value on the lived experiences of service users and meaningfully involve them in the planning and delivery of services. Forensic service users would be supported to collaborate in their own care and to help others in similar situations (e.g., co-facilitation, peer support). They can also be integrated into quality improvement processes, involved in the hiring and training of staff, and engaged as peer researchers (Bowser, 2012; Livingston, Nijdam-Jones, & Team P.E.E.R., 2012; Livingston, Nijdam-Jones, Lapsley, Calderwood, & Brink, 2013).

Studies in this area are demonstrating that, despite the challenges, forensic service users can be engaged in ways that are consistent with recovery-oriented, patient-centered care approaches (Livingston, Nijdam-Jones, & Brink, 2012; Staley, Kabir, & Szmukler, 2013). Enhancing patient engagement in a custodial setting has the potential to enhance forensic service users’ experiences of care (Livingston et al., 2013) and improve safety (Polacek et al., 2015). Greater research is needed to better understand how patient engagement and related approaches (e.g., recovery) influence forensic mental health processes (e.g., risk management) and outcomes (e.g., recidivism).