Show Summary Details
Page of

(p. 83) Mental Health Courts 

(p. 83) Mental Health Courts
Chapter:
(p. 83) Mental Health Courts
Author(s):

David DeMatteo

, Kirk Heilbrun

, Alice Thornewill

, and Shelby Arnold

DOI:
10.1093/med-psych/9780190844820.003.0005
Page of

date: 12 November 2019

The demonstrated success of drug courts led to the development of other problem-solving courts, including mental health courts. Although drug courts and mental health courts serve different (albeit overlapping) populations of offenders, they are based on the same underlying treatment-oriented philosophy. Like all problem-solving courts, mental health courts seek to address the causes of criminal behavior and focus more on rehabilitation than punishment. Specifically, mental health courts focus on mental health symptoms and disorders that may play a role in offending.

In this chapter, we first review the overrepresentation of individuals with mental illness in the criminal justice system, which was the impetus for the development of mental health courts. Next, we describe the history and current state of mental health courts in the United States. A substantial portion of this chapter is dedicated to summarizing the research on mental health courts. Although there is considerably less research on mental health courts than on drug courts, the available research is promising in some respects while lacking in others. Finally, we conclude the chapter with a discussion of innovative mental health court practices and the future of mental health courts.

Mental Health in the Criminal Justice System

As briefly described in Chapter 1, individuals with mental illness are significantly overrepresented in the criminal justice system. Although statistics vary widely, a conservative estimate is that nearly 25% of all individuals in the U.S. correctional system, which includes offenders in jails, prisons, and on community supervision (probation, parole), exhibit symptoms of mental illness (Baillargeon, Binswanger, Penn, Williams, & Murray, 2009; Diamond, Wang, Holzer III, Thomas, & Cruser, 2001; Ditton, 1999). When limited to incarcerated populations, more than 50% of all inmates in jails and state prisons have some form of mental illness (Kim, Becker-Cohen, & Serakos, 2015). A survey conducted by the Bureau of Justice Statistics indicated that, as of mid-2005, 65% of state prisoners, 45% of federal prisoners, and 64% of jail inmates reported experiencing mental health symptoms (James & Glaze, 2006). Furthermore, 49% of state inmates, 40% of (p. 84) federal inmates, and 60% of jail inmates reported symptoms of a mental health disorder included in the American Psychiatric Association’s (1994) Diagnostic and Statistical Manual of Mental Disorders (DSM-IV), with some meeting criteria for personality and developmental disorders (James & Glaze, 2006).

With respect to specific mental health symptoms and disorders, the Bureau of Justice Statistics survey showed that approximately 20% of state and federal inmates had a depressive disorder, 12% exhibited symptoms of mania, 8% were diagnosed with an anxiety disorder, 7% had posttraumatic stress disorder, 6% were diagnosed with a personality disorder, and 5% had schizophrenia (James & Glaze, 2006). These rates of mental illness are considerably higher than the rates found in the general population. Furthermore, inmates suffering from mental illness are more likely to have co-occurring homelessness, previous incarceration, and drug involvement (Cloyes, Wong, Latimer, & Abarca, 2010).

Several large U.S. correctional facilities house more individuals with mental illness and substance use disorders than many inpatient psychiatric treatment centers (Ford, 2015; McCuan, Prins, & Wasarhaley, 2007; Torrey, Kennard, Eslinger, Lamb, & Pavle, 2010). In fact, the three largest providers of mental health services in the United States are Cook County Jail, Riker’s Island, and the Los Angeles County Jail (Gavett, 2012). Unfortunately, most jails and prisons do not have sufficient resources or trained staff to address the needs of offenders who have mental health symptoms and disorders. Also, as the prevalence of mental illness tends to increase with the age of the correctional population, it is expected that the percentage of inmates with mental illness will increase as the U.S. correctional population grows older (Chiu, 2010; Sterns, Lax, Sed, Keohane, & Sterns, 2008). Furthermore, of the inmates with mental illness, only one-third of state prisoners and one-sixth of jail inmates reported receiving mental health treatment while incarcerated (James & Glaze, 2006). Indeed, only 15.1% of prisoners with a mental health problem were given a prescription for psychiatric medication, and only 12% received mental health counseling (James & Glaze, 2006).

These statistics highlight the prevalence of mental health problems within the U.S. correctional population, and they also demonstrate that the criminal justice system lacks adequate treatment and case management services for offenders with mental health problems. The concerns regarding mental illness among justice-involved populations are exacerbated by the recent prison overcrowding phenomenon. The United States incarcerates more people per capita than any other country in the world, and the U.S. correctional population quadrupled in size since the mid-1980s, with more than 2.3 million individuals currently incarcerated and several million more under correctional supervision in the community (Wagner & Rabuy, 2017; Wilper et al., 2009). Overcrowding has made it challenging for correctional facilities to provide even basic services, such as (p. 85) safety and security, and thus many correctional facilities are simply not sufficiently equipped to provide adequate mental health treatment (Kim et al., 2015).

Mental Health Courts: Development and Overview

Mental health courts were developed to better address the needs of the increasing number of offenders with mental illness who come into contact with the criminal justice system (Griffin & DeMatteo, 2009). In these courts, legal professionals (e.g., judges, attorneys, probation/parole officers) and community professionals (e.g., treatment providers, social services staff, case managers) work collaboratively to address offenders’ needs. Mental health courts provide participants with community-based mental health treatment and other services aimed at reducing their symptoms of mental illness (Redlich, Steadman, Monahan, Robbins, & Petrila, 2006; Steadman, Davidson, & Brown, 2001). By addressing the needs that are most likely to lead to future offending, mental health courts seek to interrupt the cycle of arrest, conviction, incarceration, release, and rearrest that is often characteristic of offenders with mental illness (Castellano & Anderson, 2013).

The mental health court team also strives to help offenders with mental illness reintegrate into the community through intensive judicial monitoring and the provision of case management and other social services, including job training, employment, and housing support (Bazelon Center for Mental Health Law, 2003; Denckla & Berman, 2001; Steadman et al., 2001). Although the operations of mental health courts vary, underlying most mental health courts is the belief that facilitating access to mental health treatment and providing judicial monitoring will ameliorate clinical symptoms and improve functioning, which will ultimately reduce criminal recidivism (Farole, 2006; Griffin & DeMatteo, 2009). We return to a discussion of the underlying philosophy of mental health courts later in this chapter.

A precursor of the mental health court model was developed by Judge Evan Goodman in 1980 in Marion Country, Indiana (Coons & Bowman, 2010). Judge Goodman developed a program in which offenders with mental illness were diverted from jail to a nearby hospital for a psychiatric evaluation. A hospital recreation room was converted into a makeshift courtroom so that court proceedings could be held in the same building as the hospital’s psychiatric unit. Prior to court proceedings, the offender would be assessed by a psychiatrist, and the court would then require eligible participants to complete outpatient treatment. A forensic team would monitor participants’ treatment compliance and progress. This court was ultimately dismantled in the early 1990s, and such models were not used until several years later when significant events in Florida triggered the rapid development of mental health courts throughout the United States (Castellano & Anderson, 2013).

(p. 86) In 1997, the Florida legislature charged a special task force with investigating the relationship between mental illness and involvement in the criminal justice system. The task force found a growing incarceration rate for low-level offenses committed by offenders with mental illness, a dearth of psychiatric assessments for such offenders once they were taken into custody, and the inadequate provision of mental health treatment after release to the community. In response to this investigation, the Florida legislature allocated $1.5 million for the development of the country’s first mental health court in Broward County (Castellano & Anderson, 2013; Orr et al., 2009).

The Broward County Mental Health Court, which is a division of the county criminal court, handles cases involving individuals who have been charged with nonviolent, misdemeanor offenses and identified as mentally ill or developmentally disabled. It is a voluntary preadjudication program that diverts offenders into treatment in lieu of a trial or other disposition. The court has several goals: (a) enhance communications and interactions between the criminal justice system and mental health system, (b) ensure that offenders with mental health involvement are not housed in jail, (c) balance the rights of offenders with public safety considerations, (d) divert offenders who are mentally ill and charged with minor offenses into community-based services, and (e) monitor the delivery of services and treatment.

By early 2005, less than 10 years after the development of the first mental health court, there were 90 functional adult mental health courts in the United States (Redlich et al., 2006), with several other courts in advanced stages of development. At that time, mental health courts could be found in 34 states, primarily in the western (37%) and southern (37%) regions of the United States (Council of State Governments, 2005). Although there were fewer mental health courts in the Midwest (15%) and northeast (11%) (Council of State Governments, 2005), Ohio had more mental health courts than any other state (Redlich et al., 2006). At that time, mental health courts were serving more than 7,500 offenders, with an average of 36 cases per court; there was a wide range of caseloads among mental health courts, with some courts handling as few as 3 cases and others as many as 852 cases (Redlich et al., 2006). Mental health courts varied in terms of the types of cases accepted, with 49% primarily managing misdemeanor cases, 14% primarily handling felony cases, and the remainder handling both misdemeanor and felony cases (Redlich et al., 2006).

By 2008, the number of mental health courts in the United States more than doubled to 182, with many others in advanced stages of development (Erickson, Campbell, & Lamberti, 2006; Griffin & DeMatteo, 2009; Thompson, Osher, & Tomasini-Joshi, 2007). This time period also saw the development of juvenile mental health courts (Cocozza & Shufelt, 2006; Thompson et al., 2007) and the adoption of the mental health court model by other countries, including Canada, (p. 87) the United Kingdom, and Australia (Griffin & DeMatteo, 2009). By 2011, there were an estimated 250 mental health courts in the United States (Council of State Governments Justice Center, 2011), and the National Drug Court Resource Center (2014) reported that there were 414 operational mental health courts in the United States by 2014. A recent publication from the National Drug Court Institute indicates that the number of mental health courts in the United States increased by 36% between 2009 and 2014, and mental health courts are projected to be the fasting growing type of problem-solving court over the next several years (Marlowe, Hardin, & Fox, 2016).

Goals and Philosophies of Mental Health Courts

Most mental health courts were developed in communities that had a preexisting drug court, and the mental health courts drew from the therapeutic jurisprudence philosophy that underlies drug courts (Griffin & DeMatteo, 2009). As such, mental health courts adhere to the premise that the criminal justice system can simultaneously protect public safety and provide needed mental health treatment and other services to offenders with specific needs (Odegaard, 2007; Rottman & Casey, 1999; Wexler, 2001). By providing mental health treatment services combined with intensive judicial supervision, mental health courts adhere to the therapeutic jurisprudence philosophy.

Although the specific operational protocols of mental health courts vary among jurisdictions (see Erickson et al., 2006), the basic goal/purpose of most mental health courts is the same: to divert offenders with mental health treatment needs into specialized criminal justice processing that involves the provision of judicially supervised community-based mental health treatment and other needed services (Boothroyd, Pothyress, McGaha, & Petrila, 2003; Moore & Hiday, 2006). By diverting offenders with mental health needs to treatment providers, mental health courts seek to ameliorate mental health symptoms that could contribute to criminal behavior and, as a result, minimize the “revolving door” pattern of the criminal justice system whereby the same offenders are frequently rearrested (Griffin & DeMatteo, 2009). Additional goals of many mental health courts include (1) reducing the criminalization of individuals with mental illness by preventing those offenders from being incarcerated inappropriately, (2) reducing the hospitalization rate of offenders with mental illness, (3) helping correctional facilities run more smoothly by diverting offenders with mental health needs from standard criminal justice processing that may lead to incarceration, and (4) saving money and resources for the criminal justice and mental health systems (Casey & Rottman, 2005; Griffin & DeMatteo, 2009).

(p. 88) Organization and Operation of Mental Health Courts

Mental health courts come in three models: preadjudication, postadjudication, and probation-based (Griffin & DeMatteo, 2009). Under the preadjudication model, an offender is not required to plead guilty prior to enrolling in a mental health court program. Rather, the court defers prosecution, and, in exchange, the offender agrees to comply with mandated mental health treatment (Almquist & Dodd, 2009). In the postadjudication model, a guilty plea or conviction is required prior to enrolling in a mental health court program, but the final disposition of the case is deferred until the offender completes the treatment program (Griffin & DeMatteo, 2009). Some mental health courts that follow this model allow for the expungement of an offender’s criminal record upon successful completion of the mandated treatment program (Almquist & Dodd, 2009). Finally, under the probation-based model, an offender pleads guilty, is convicted, and receives a sentence of probation, and the mental health court imposes mandated treatment as a requirement of probation (Griffin & DeMatteo, 2009).

Several researchers have examined the prevalence of each of the three operational models of mental health courts. One study on early mental health courts found that of the eight courts that existed at the time, two operated under a preadjudication model, three operated under a probation-based model, and three operated under a postadjudication model (Griffin, Steadman, & Petrila, 2002). Even within these models there is considerable variability in how these courts operate. A 2003 national survey of 20 mental health courts found that half of the courts required a plea prior to participation in the program, and more than one-third provided for dismissal of criminal charges or record expungement upon successful completion of the program (Bazelon Center for Mental Health Law, 2003). A later study examining mental health courts found that of seven courts, six used the postadjudication model and one used the preadjudication model (Redlich, Steadman, Monahan, Petrila, & Griffin, 2005). In addition, a 2005 study of a mental health court in Clark County, Washington found that the court initially operated under the preadjudication model, but adopted the postadjudication model after receiving funding from the Substance Abuse and Mental Health Services Administration (Herinckx, Swart, Ama, Dolezal, & King, 2005).

Mental Health Court Staff

Mental health courts use a nonadversarial, multidisciplinary team approach to facilitate the effective supervision and treatment of participants. The team typically consists of some combination of the following: judge, defense attorney, prosecutor, probation officer, parole officer, case manager, and representatives (p. 89) from community-based treatment providers (Almquist & Dodd, 2009). The composition of the team varies by court. O’Keefe (2006) reported that a Santa Barbara, California, mental health court team consisted of judges, district attorneys, public defenders, probation officers, police officers, representatives from substance use and mental health treatment providers, a case manager, and an “intensive care team,” which provided services to participants at their homes and focused on addressing participants’ basic needs. Another study on a Washington mental health court found that the team consisted of a judge, attorneys for the defense and prosecution, probation officers and staff, case managers, and mental health providers (Herinckx et al., 2005). Two other Washington mental health courts (Seattle and King Country) had teams composed of judges, clinical social workers, prosecutors, probation advisors, defense attorneys who worked closely with social workers, and court coordinators (Trupin & Richards, 2003). In addition to attorneys, case managers, and a judge, a Brooklyn mental health court team included a project director, clinical director, forensic director, and resource coordinator, along with a psychiatrist and social workers (O’Keefe, 2006). A mental health court team in Pennsylvania (Allegheny County) consisted of a judge, assistant district attorney, public defender, clinical social workers, forensic case managers, and a liaison staff member to work with probation (Ridgely et al., 2007). Thus, despite some common components, the staffing in mental health courts can vary considerably by jurisdiction.

Several researchers have hypothesized that the quality of the relationships between mental health court participants and court staff plays a role in the participants’ success, but there are very little data on this topic. In their meta-analysis, Sarteschi, Vaughn, and Kim (2011) concluded that positive relationships with mental health court staff are associated with greater participant success in mental health court programs, but we did not find any other empirical research on this important topic. The role of the judge and court staff in drug courts has received a good deal of attention (see Marlowe, Festinger, & Lee, 2004), and this area should be explored further in mental health courts, particularly because the relationships between mental health court staff and participants likely vary widely by court. Identifying the types of relationship that are associated with higher levels of participant success could help establish a gold standard for mental health courts moving forward.

Mental Health Court Referral Process

Referrals for mental health courts can be made by prosecutors, defense attorneys, judges, jail staff, and family members of offenders (Almquist & Dodd, 2009). A 2006 evaluation of a Brooklyn mental health court found that 44% of participant referrals came from defense attorneys, 30% from competency hearings, 10% from the district attorney’s office, 10% from other judges, and 5% percent (p. 90) from other specialty courts (O’Keefe, 2006). An early study of four mental health courts found that most referrals came from county jail staff members (Goldkamp & Irons-Guynn, 2000).

The time between referral and enrollment in mental health court varies considerably among courts. Steadman and colleagues found that although 14% of participants were enrolled immediately, the average wait-time between referral and enrollment in the seven mental health courts ranged from 0 to 45 days (Steadman, Redlich, Griffin, Petrila, & Monahan, 2005). They also reported that, of the referrals received, 14% of offenders were accepted on the same day as the referral and 52% percent of all the referred offenders were accepted. The court rejected referrals for various reasons, including failure to meet mental health disorder criteria (30%), nature of the offender’s criminal history (19%), lack of agreement from the prosecution (18%), and incompetence or instability of the offender (7%).

Some courts are quite efficient in enrolling referred participants. The mental health court in Broward County, Florida, frequently heard cases and was often able to admit participants on the same day they were referred (Petrila, Poythress, McGaha, & Boothroyd, 2001). However, a study examining a Brooklyn mental health court found the average postreferral wait-time was 2–3 months (O’Keefe, 2006). One study found that potential mental health court participants were typically not waiting for enrollment into the court from a jail cell because those charged with more minor crimes were usually released even without posting bail (Hiday, Moore, Lamoureaux, & de Magistris, 2005). Another study examined the length of time between initial arrest and either mental health court enrollment or criminal disposition for offenders with and without mental illness (Redlich, Liu, Steadman, Callahan, & Robbins, 2012). In that study, Redlich and colleagues found that the average length of time between arrest and mental health court enrollment was 70 days, the average length of time between arrest and disposition of offenders with mental illness going through standard criminal justice processing was 76 days, and the average length of time between arrest and disposition for offenders without mental illness was 37 days.

Linking Mental Health Court Participants to Treatment

Mental health courts also differ in how they connect participants with treatment services. Participants are typically linked with treatment providers by case managers who work within the mental health court. A national survey of 20 mental health courts found that courts tend to connect participants with treatment programs by reconnecting participants with previously or currently used treatment providers or directly referring participants to community treatment providers (Bazelon Center for Mental Health Law, 2003; see Almquist & Dodd, 2009). A small percentage of these courts simply provided participants (p. 91) with information about providers and encouraged participants to seek treatment, without any meaningful coordination of treatment (Bazelon Center for Mental Health Law, 2003). A Santa Barbara mental health court provided each participant with a case manager who connected the participant with treatment providers and also provided assistance with transportation, housing, job training, and substance use issues (Cosden, Ellens, Schnell, Yamini-Diouf, & Wolfe, 2003).

The specific treatments provided to participants vary by court and depend on participants’ clinical needs and the availability and resources of the community treatment providers. Most mental health courts do not have in-house treatment providers or provide funding for external treatment services. As such, mental health courts often seek to develop positive relationships with treatment providers to ensure participants receive adequate treatment (Almquist & Dodd, 2009).

Encouraging Participant Engagement: Incentives, Rewards, and Sanctions

To encourage participants to engage in treatment, mental health courts typically use a behavioral modification scheme based on incentives/rewards and sanctions. Incentives and rewards are often used to motivate participants to comply with treatment requirements, and they typically include verbal recognition, gift certificates, reduction in the length or intensity of mandated treatment, and a reduction in the required number of judicial status hearings (Griffin & DeMatteo, 2009). For instance, a 2008 study evaluating a Utah mental health court found that compliant participants were rewarded with verbal praise and placed on a “Rocket Docket,” which allowed them to be first in line at judicial status hearings so they could leave court early (Van Vleet, Hickert, Becker, & Kunz, 2008). Another study found that a mental health court in Arkansas used a wide array of incentives to encourage participant engagement with treatment, including verbal praise and applause, certificates of program completion, and fewer required court appearances (Ferguson, McAuley, Hornby, & Zeller, 2008). The provision of positive reinforcement in the form of incentives and rewards is intended to promote the continued compliance of mental health court participants.

When participants fail to comply with treatment requirements, mental health courts will often impose sanctions that could include verbal reprimands, an increase in the required number of status hearings, mandatory community service, jail time, and, in extreme cases, expulsion from the mental health court program (Goldkamp & Irons-Guynn, 2000; Griffin et al., 2002). In a 2003 survey of sanctions used by 20 mental health courts, 64% imposed jail time, 36% amended participants’ treatment plans, 27% reprimanded noncompliant participants and increased the number of required court appearances, and 18% contemplated (p. 92) expelling participants who failed to comply (Bazelon Center for Mental Health Law, 2003). A Utah mental health court sanctions noncompliant participants by requiring them to perform community service (Van Vleet et al., 2008). The goal of punishing participant noncompliance is to shape participants’ behavior to be more compliant with program requirements.

Early mental health court research by Griffin et al. (2002) found that mental health courts rarely used jail time and termination from the program as sanctions for noncompliance. A 2006 national survey evaluated the use of jail time sanctions by 90 mental health courts throughout the United States (Redlich et al., 2006). Although 92% of the courts reported a willingness to use incarceration as a sanction for program noncompliance, the majority of the courts reported using such sanctions very infrequently; only 2% of the courts used jail time as a sanction for more than 50% of their cases, 39% used jail time in 5–20% of their cases, and 33% used jail time in less than 5% of their cases (Redlich et al., 2006). Other investigators found that three of the four mental health courts they studied used jail as a sanction (Callahan, Steadman, Tillman, & Vesselinov, 2013). They also identified several factors that were associated with being sent to jail as a sanction, including being charged with a drug offense (as opposed to a crime against people), heavier drug involvement, a reputation of noncompliance, and frequency of in-custody hearings (Callahan et al., 2003).

Mental health court judges may be hesitant to use jail time as a sanction for participant noncompliance because incarceration can be detrimental to individuals with mental health disorders. Almquist and Dodd (2009) noted that incarceration could be highly stressful to such individuals and lead to further deterioration of their mental and emotional functioning, which could in turn lead to increased noncompliance. Indeed, research has shown that judges may be reluctant to impose jail time as a punishment when it appears that the offender’s noncompliance was due to mental illness (Griffin et al., 2002). Moreover, incarcerating a mental health court client disrupts any progress the client is making in the program.

Nature of Mental Health Court Participation

As with other problem-solving courts, the decision to enter a mental health court must be made voluntarily by the offender. Although some mental health court participants do not perceive mental health programs to be coercive, some participants exhibit unawareness of the nature and purpose of the court. In a 2002 study of the Broward County, Florida, mental health court, participants reported experiencing minimal coercion when deciding whether to participate in the program (Poythress, Petrila, McGaha, & Boothroyd, 2002). However, a 2003 study of that same court found that only 28% of participants were given information about the goals and nature of the mental health court (Boothroyd et al., 2003). Boothroyd (p. 93) et al. (2003) also reported that 54% of participants reported understanding that participation was voluntary, but more than half of this group reported that the voluntary nature of the court was not explained to them prior to their initial hearing. Another investigator evaluated participants’ perceptions in 10 mental health courts and found that most participants did not feel coerced into participation (Redlich, 2005). Notably, participants who knowingly and voluntarily enrolled in mental health court and who were made aware of the nature and goals of the program were more likely to succeed, whereas participants who felt coerced into the program were less likely to adhere to the program requirements (Redlich, 2005).

Mental Health Court Trends

Some studies have examined the difference between “first-generation” and “second-generation” mental health courts. The research conducted by Griffin and colleagues focused on eight early programs that constituted the first group of mental health courts in the United States (Griffin et al., 2002). These courts, which were established around the same time and shared common characteristics, including a focus on accepting defendants with misdemeanor charges, have been categorized as “first-generation” mental health courts (Redlich et al., 2005). Redlich et al. (2005) found that second-generation mental health courts are characterized by an increase in acceptance of offenders with felony charges, which in turn has led to an increase in the use of the postadjudication model. In addition, second-generation courts more frequently use jail as a sanction and provide increased supervision of participants.

Another study by Redlich and colleagues identified factors associated with the number of participants seen by mental health courts (Redlich et al., 2006). They found that the number of mental health court clients was associated with the length of time a court had been in existence (with more established courts having more participants), fewer initial status reviews, and more types of community supervision offered. These findings have important implications for mental health courts that are seeking to increase (or decrease) the number of participants they serve.

At least one study has examined mental health courts from a cost–benefit perspective. Ridgely et al. (2007) examined whether mental health courts are less expensive than their traditional court counterparts and found that an Allegheny County (Pennsylvania) mental health court exhibited a trend toward greater long-term net savings. Unfortunately, we were not able to find any large-scale studies evaluating the costs and benefits of mental health courts. Given the current economic climate, demonstrating that mental health courts are both effective and cost-effective should be a priority.

(p. 94) Mental Health Court Participants

Types of Crimes

There is significant variability among mental health courts in terms of the types of crimes that are deemed acceptable for admission into the court. Some courts only accept offenders with misdemeanor charges, others only accept offenders with felony charges, and some courts accept both misdemeanor and felony offenders (Almquist & Dodd, 2009). A national survey conducted in 2003 found that 10 of 20 existing mental health courts only accepted offenders charged with misdemeanors, while the other 10 courts accepted offenders charged with either misdemeanors or felonies (Bazelon Center for Mental Health Law, 2003). Of 87 mental health courts evaluated in a 2006 survey, 40% only accepted offenders charged with misdemeanors, 10% only accepted those charged with felonies, and 50% accepted offenders charged with misdemeanors or felonies (Redlich et al., 2006).

Whether a mental health court accepts offenders charged with misdemeanor or felony offenses is often contingent on the jurisdiction in which the court is situated. Some mental health courts, for example, function within a municipal court and therefore will primarily oversee offenders with misdemeanor charges, while other mental health courts operate within trial courts and can accept felony cases. Even without jurisdictional limitations, policy concerns may dictate the types of crimes that a mental health court is willing to oversee. Some funding entities, for example, have preferences regarding the acceptance of individuals with certain types of charges. There may also be preferences among the various stakeholders involved in mental health courts. The protection of public safety is a high priority, and it may lead some mental health courts to limit enrollment to offenders charged with misdemeanor offenses (Almquist & Dodd, 2009).

Early mental health courts tended to focus on offenders with misdemeanor charges, but an increasing number of courts are accepting individuals charged with felony offenses (Almquist & Dodd, 2009). Research on first-generation mental health courts found that seven out of eight early mental health courts primarily accepted individuals with misdemeanor charges, and many even imposed restrictions regarding the types of misdemeanors that would be accepted (Griffin et al., 2002). A 2003 study found that the first mental health court in the United States (Broward County, Florida) only accepted offenders with nonviolent misdemeanor charges, violation of an ordinance, or traffic offenses (Boothroyd et al., 2003). A later study found that, by 2005, several mental health courts that had initially only accepted offenders charged with misdemeanor offenses had begun accepting some offenders charged with felonies (Redlich et al., 2005). This study also found that several second-generation mental health courts accepted (p. 95) offenders charged with felonies, with some focusing exclusively on offenders with felony charges (Redlich et al., 2005).

Mental health courts are becoming increasingly more likely to accept individuals who have been charged with violent offenses. Accepting offenders charged with violent crimes is unusual among most types of problem-solving courts. However, a 2003 study found that of 20 mental health courts that were evaluated, only 4 utilized bright-line exclusions for offenders with a history of violence (Bazelon Center for Mental Health Law, 2003). Another 2003 study found that a mental health court in Santa Barbara, California, accepted individuals with a history of violence if the district attorney and mental health court staff determined that the individual was no longer a threat to the general public (Cosden et al., 2003). An evaluation of a Brooklyn mental health court that was developed in 2002 found that although the court initially only accepted individuals with nonviolent charges, the court later broadened its acceptance criteria and began enrolling individuals with violent felony charges on a discretionary basis; by mid-2004, 39% of the participants in the mental health court had been charged with a violent crime (O’Keefe, 2006). It is worth noting that some mental health courts require victim permission for an offender to participate in the program if the underlying offense was violent.

In a review of 20 studies, Honegger (2015) identified substantial variability in mental health court eligibility criteria. Of the mental health courts examined, some only accepted individuals with misdemeanor charges (Cosden, Ellens, Schnell, &Yamini-Diouf, 2005; Frailing, 2010; Hiday & Ray, 2010; Hoff, Baranosky, Buchanan, Zonana, & Rosenheck, 1999; Keator, Callahan, Steadman, & Vesselinov, 2013; Luskin, 2013; Moore & Hiday, 2006; Steadman, Redlich, Callahan, Robbins, & Vesselinov, 2011; Van Vleet et al., 2008), while others accepted individuals with both misdemeanor and felony charges (Boothroyd et al., 2003; Boothroyd, Mercado, Poythress, Christy, & Petrila, 2005; Christy, Poythress, Boothroyd, Petrila, & Mehra, 2005; Dirks-Linhorst & Linhorst, 2012; Hiday, Wales, & Ray, 2013; Trupin & Richards, 2003). A few of the courts included in Honegger’s (2015) review only accepted certain types of felonies (Sneed, Koch, Estes, & Quinn, 2006; Steadman & Naples, 2005). The lack of uniform eligibility criteria across courts is one of the many reasons it is challenging for researchers to conduct empirical investigations comparing mental health courts.

Mental Health Diagnosis

Mental health courts also vary in terms of the procedures used to determine if an individual is an appropriate candidate (i.e., has the requisite mental health issues) for diversion into the court. Most mental health courts only accept individuals who exhibit signs of a serious mental illness (Almquist & Dodd, 2009). (p. 96) However, some courts make acceptance contingent on an offender satisfying diagnostic criteria for a specific mental health diagnosis, while other courts accept participants by confirming a previous mental health diagnosis (Erickson et al, 2006; Griffin et al., 2002). The wide variability in eligibility criteria related to an offender’s mental health function raises some concerns relating to equal protection (see Chapter 9, for more discussion on this issue), and it also contributes to the difficulties in empirically evaluating mental health courts.

A 2005 study examining the mental health diagnostic requirements of 90 mental health courts found that 37% only accepted individuals with a diagnosed DSM-IV Axis I disorder, 21% accepted individuals with a mental illness that was “serious and/or serious and persistent,” and 16% indicated having some requirements regarding the types of mental illness required to be exhibited by participants; the remaining 26% had no specific requirements regarding mental health diagnoses (Criminal Justice/Mental Health Consensus Project, 2005). One study found that the most common diagnoses of participants from seven mental health courts were schizoaffective disorder, schizophrenia, bipolar disorder, depressive disorders, and other mood disorders (Steadman et al., 2005). The most prevalent diagnoses of participants in the Broward County, Florida, mental health court were depressive disorder (25%), bipolar disorder (24%), and schizophrenia (17%) (Boothroyd et al., 2005).

Mental health court participants often have co-occurring substance use and mental health disorders. Research on a mental health court in Santa Barbara, California, found that 83% of the participants had co-occurring substance use and mental health disorders (Cosden et al., 2003). Another study found that 59% of participants in a mental health court in Anchorage, Alaska, had co-occurring disorders (Ferguson et al., 2008). Finally, a 2007 study of a San Francisco behavioral health court (which is similar in structure and function to a mental health court) found that 56% of participants were diagnosed with both a severe mental health disorder and a substance use disorder (McNiel & Binder, 2007). Including participants with co-occurring disorders has implications for the types of community-based treatments and services that are needed to adequately address participants’ needs.

In her comprehensive review of 20 mental health court studies, Honegger (2015) found substantial variability in terms of the nature and specificity of required mental health diagnoses for mental health court participants. One court excluded offenders with personality disorders or developmental disabilities (Herinckx et al., 2005), while other courts accepted offenders with developmental disorders, intellectual disability, and any other DSM-IV-TR Axis I disorders (Sneed et al., 2006). Some courts required that participants be diagnosed with major depressive disorder, bipolar disorder, or schizophrenia (Frailing, 2010; Luskin, 2013). Again, as with some of the other differences among mental health (p. 97) courts, differences relating to eligible mental health disorders make it challenging to compare mental health courts.

Demographic Characteristics

Some research suggests that there is an overrepresentation of female and Caucasian participants in mental health courts considering the proportion of those two groups within the broader criminal justice population. One study examined referrals made to seven mental health courts and found that referred individuals were more likely to be Caucasian, female, and older (Steadman et al., 2005). Another study found that female and Caucasian participants were slightly overrepresented in a Community Resource Court in North Carolina (Hiday et al., 2005). A more recent comprehensive review suggests that this trend may be changing. Specifically, a review of 18 mental health court studies by Sarteschi et al. (2011) found that most mental health court participants were Caucasian males in their mid-30s.

The Effectiveness of Mental Health Courts

As with most other interventions with justice-involved populations, participation in mental health courts results in a number of important outcomes. The shared goal of all problem-solving courts is to reduce recidivism among offenders who have historically been overrepresented and undertreated in the criminal justice system. However, each type of problem-solving court also has an interest in outcomes that are specific to the type of court. Whereas drug courts are primarily interested in the impact of the courts on criminal recidivism and relapse to drug use, mental health courts are primarily interested in how participation in the program affects recidivism and mental health.

The effectiveness of mental health courts can be measured by looking at a number of process and outcome variables. In terms of process, an important question is how quickly mental health court participants begin receiving services after they are enrolled in the court. In terms of outcome, there are important questions related to the effects of mental health courts on participants’ mental health, connecting participants to behavioral health services, participants’ quality of life, and criminal recidivism.

In the sections that follow, we examine the available research on mental health courts. As will be evident, the research on mental health courts is considerably less developed than the research on drug courts. Unlike the drug court research, there are very few meta-analyses, systematic reviews, and large-scale studies. With that said, mental health courts have received much more research attention than all of the other types of problem-solving courts (with the exception of drug (p. 98) courts). Taken as a whole, the body of research on mental health courts is promising in some respects, but it also highlights several areas in which much more research is needed.

Mental Health Outcomes

Some mental health court research has focused on the clinical outcomes related to participant treatment. Indeed, one philosophy underlying mental health courts is that recidivism rates can be reduced by ameliorating mental health symptoms (Honegger, 2015). Many studies have evaluated changes in mental health symptoms using various psychometric measures, including the Addiction Severity Index (ASI; McLellan, Luborsky, Woody, & O’Brien, 1980), the Behavior and Symptom Identification Scale (BASIS-32; Eisen, 1996), the Brief Psychiatric Rating Scale (BPRS; Overall & Gorham, 1962), the Colorado Symptom Index (CSI; Shern et al., 1994), and the Global Assessment of Functioning (GAF; Hall, 1995). Using well-established measures with acceptable psychometric properties, as opposed to locally developed measures with little validation research, improves the validity and reliability of the findings.

The available research on clinical outcomes among mental health court participants also reflects different methodological approaches. Honegger’s (2015) review of 20 mental health court studies revealed that 8 of the 20 studies evaluated the impact of mental health courts on mental health functioning, with one study utilizing an experimental design, six studies using quasi-experimental designs, and one study using a pretest/posttest design. Given the importance of research methodology in terms of the types and strength of inferences that can be drawn, the methodologies commonly used to evaluate mental health courts will be discussed in more detail later in this chapter.

Early studies on mental health court outcomes produced mixed results. For example, Trupin and Richards (2003) found that mental health court participants exhibited improvement in scores on the GAF scale, which is a measure of psychological impairment. Cosden et al. (2003) compared outcomes of mental health participants with a control group and found that mental health court participants exhibited significant reductions in emotional distress and drug use issues, but that both groups exhibited improvements in alcohol problems, with no significant differences between the two groups in this area. A later study by Cosden et al. (2005) found that mental health court participants exhibited improvement on the GAF scale, but they also found no significant differences in GAF scores and BASIS-32 scores between mental health court participants and offenders involved in standard criminal court processing. Around that same time, Boothroyd et al. (2005) used a quasi-experimental design to compare (p. 99) mental health court participants to offenders going through standard criminal justice processing. They administered the BPRS to individuals from both groups during initial hearings and at 1, 4, and 8 months after the initial court appearance, and results revealed no significant between-group differences in BPRS scores across time points (Boothroyd et al., 2005). Finally, a more recent meta-analysis of eight studies that examined mental health outcomes was unable to calculate a valid effect size because there was insufficient homogeneity among the studies (Sarteschi et al., 2011).

Several researchers have focused on the use of emergency psychiatric services as a proxy for mental health court outcomes. The results of these studies were mixed. Steadman and Naples (2005) compared mental health court participants from six mental health courts (three pre-booking and three post-booking) with nondiverted offenders. In this study, the researchers tracked emergency room visits as a proxy for mental health symptoms and found that mental health court participants (31.3%) had significantly more emergency room visits than the nondiverted offenders (23.9%) over a 1-year period following diversion. This study also found that mental health court participants (31.7%) were hospitalized at a significantly higher rate than the nondiverted offenders (18.7%) over the same 1-year period (Steadman & Naples, 2005). Overall, mental health court participants were linked to community-based services at higher rates than the nondiverted offenders. However, Steadman and Naples (2005) suggested that one explanation for these results is that mental health court participants were better connected to mental health professionals who would encourage the use of emergency services when necessary.

Other studies, however, have not found differences between mental health court participants and offenders in standard criminal justice processing. For example, Luskin (2013) compared psychiatric hospitalization rates for mental health court participants and offenders in standard criminal justice processing after 6 months of enrollment. Results revealed that the rate of psychiatric hospitalizations decreased significantly from the first hearing to the 6-month follow-up, but there was no significant difference in hospitalization rates between mental health court participants and nondiverted offenders (Luskin, 2013). Another study found no significant differences in emergency room visits or contact with crisis services between mental health court participants and nondiverted offenders (Keator et al., 2013). However, when compared with the nondiverted offenders, Keator et al. (2013) found that mental health court participants were more likely to receive treatment from inpatient providers or drug detoxification facilities.

Finally, some research has looked at other clinical outcomes, such as substance use, among mental health participants. For example, in their 2005 study, Cosden et al. evaluated changes in substance use behaviors among mental health (p. 100) court participants using the ASI. Results revealed that both mental health court participants and nondiverted offenders exhibited significant improvements in ASI scores over time but that the mental health court participants exhibited greater improvements (Cosden et al., 2005).

When taken as a whole, it is apparent that the research evaluating the mental health outcomes of mental health court participants is still in its early stages. The relatively minimal extant literature provides conflicting evidence regarding the degree to which mental health courts improve the mental health functioning among participants. There is not enough evidence at this point to determine whether mental health courts are in fact superior to standard criminal justice processing in terms of improving mental health functioning. Consistent with Honegger’s (2015) conclusion, we believe that additional data are needed to adequately evaluate the impact of mental health courts on participants’ mental health functioning.

Connection to Behavioral Health Services

Underlying the development of mental health courts is the idea that they reduce criminal recidivism by increasing the connection to behavioral health services and ensuring participant adherence to treatment. As such, some research has evaluated the effectiveness of mental health courts using outcome variables operationalized under this theory, including the number of participant referrals to mental health providers, the number of participants receiving treatment, and the quantity of services provided. Rather than focusing on the outcome of participation in mental health courts, this research focuses on the ability of mental health courts to connect participants to needed services and the process they use to make such connections.

Some studies have examined the extent to which mental health courts provided improved access to psychiatric services (see Honegger, 2015, for a brief description of these types of studies). For example, Cosden et al. (2005) assessed mental health court participants and nondiverted offenders at baseline and 6, 12, 18, and 24 months after participation, and they found that mental health court participants utilized mental health services at significantly greater rates than did nondiverted offenders. A similar study found that mental health court participants exhibited a 17% increase in the utilization of behavioral health services and used mental health services at a scientifically higher rate than the nondiverted offenders (Boothroyd et al., 2003). In more recent research, Keator et al. (2013) evaluated the number of “therapeutic treatment episodes” (operationalized as individual therapy, group therapy, case management, and medication management) received by mental health court participants and (p. 101) nondiverted offenders, and they found that mental health court participants were more likely to receive treatment services. They also found that mental health court participants received treatment more quickly and received a greater number of treatment hours (Keator et al., 2013). Finally, in their study of 94 participants in a South Dakota mental health court, Sneed et al. (2006) found a 5% increase in the number of participants who reported receiving mental health services from the initial intake to the 12-month follow up.

Some studies have examined the ability of mental health courts to effectively connect participants to certain types of mental health services. For example, in early research, Steadman and Naples (2005) found that mental health court participants received treatment services significantly more than did nondiverted offenders. Specifically, the mental health group received significantly more medication than the nondiverted group (81.5% vs. 73.7%) and significantly more mental health counseling (62.2% vs. 56.8%) (Steadman & Naples, 2005). In another study, investigators compared the amount of services received 1 year before enrollment in mental health court with the amount of services received 1 year after enrollment in mental health court, and they found that participants received significantly more hours of case management and medication management and significantly more days of mental health treatment following enrollment in mental health court (Herinckx et al., 2005). In more recent research, Luskin (2013) evaluated outcomes from four mental health courts in Indiana and found no significant increase in the rate of general outpatient or individual counseling services at a 6-month follow-up (Luskin, 2013). However, when compared with a nondiverted sample, the mental health court participants received significantly more outpatient treatment and different types of treatment, including individual counseling, group counseling, and case management services (Luskin, 2013). Luskin (2013) explained that the differences observed between mental health court clients and nondiverted offenders at the 6-month follow-up resulted from improvements in treatment provision among the mental health court group and deterioration of treatment provision in the nondiverted group.

However, not all research supports the ability of mental health courts to effectively connect participants with needed services. For example, in their evaluation of two mental health courts in Washington (Seattle and King County), Trupin and Richards (2003) found no significant differences between mental health court participants and nondiverted offenders with respect to service utilization. Notable, though, is that King County participants received significantly more mental health treatment post-enrollment compared with the rates of service they received pre-enrollment (Trupin & Richards, 2003).

In related research, Cosden et al. (2005) examined the relationship between the quantity of treatment services provided and the psychosocial outcomes of mental health court participants. Although they did not find statistically (p. 102) significant results, they highlighted a trend whereby individuals with lower GAF scores (which indicates more severe psychological symptoms and impairment) received more hours of treatment services (Cosden et al., 2005). These results suggest that at least some mental health courts are providing higher doses of treatment services to participants who have greater mental health needs, which is consistent with the Risk-Need-Responsivity Model (Andrews & Bonta, 2010b).

Quality of Life

In addition to seeking to ameliorate mental health symptoms by connecting participants to needed treatment services, mental health courts also seek to improve participants’ perceived quality of life. With that said, we could only identify one study that explicitly studied quality of life among mental health court participants. Cosden et al. (2005) evaluated participants’ perceived quality of life using Lehman’s Quality of Life-Short Form (QOL-SF), and they found significant increases in scores (indicating better quality of life) among both mental health court participants and nondiverted offenders. However, they also found that mental health court participants exhibited significantly greater improvements in life satisfaction scores over time when compared with the nondiverted offenders (Cosden et al., 2005). The results should be interpreted cautiously, however, because the mental health court group received more intensive treatment services than the comparison group, which creates a confound that could skew the results. As is evident, there is a notable lack of research on the impact of mental health courts on participants’ quality of life, and additional research is needed before we can make firm conclusions about the effectiveness of mental health courts in this domain.

Recidivism

Because reducing criminal recidivism is an explicit goal of all problem-solving courts, it is not surprising that most mental health court research has examined recidivism rates. For the criminal justice system, reducing recidivism is a high priority, and demonstrating that mental health courts effectively reduce recidivism in a cost-effective way can provide considerable leverage with funding sources. Although mental health courts are also interested in other outcomes (e.g., mental health functioning, quality of life), recidivism is by far the most heavily emphasized goal.

(p. 103) Fifteen of the 20 studies reviewed by Honegger (2015) examined the relationship between mental health court participation and recidivism rates. In these studies, recidivism was variably operationalized as new arrests, new convictions, number of days in the community post reentry, or number of days incarcerated after mental health court participation. Thirteen studies that examined recidivism found a statistically significant relationship between mental health court participation and lower recidivism rates (see Dirks-Linhorst & Linhorst, 2012; Frailing, 2010; Herinckx et al., 2005; Hiday & Ray, 2010; Hiday et al., 2013; Hoff et al., 1999; McNiel & Binder, 2007; Moore & Hiday, 2006; Sarteschi et al., 2011; Steadman et al., 2011; Steadman & Naples, 2005; Trupin & Richards, 2003 [Note: Trupin and Richards (2003) consisted of two studies]). One of the studies reviewed by Honegger (2015) found that there was a statically significant relationship between mental health court participation and higher recidivism rates (see Cosden et al., 2005), and the final study found no significant difference in recidivism rates between a mental health court sample and a comparison sample (see Christy et al., 2005).

Sarteschi and colleagues (2011) conducted a meta-analysis focused on evaluating the degree to which mental health courts effectively reduce recidivism. After evaluating 18 peer-reviewed and non–peer-reviewed studies, they reported a medium effect size (−.054) for the relationship between mental health court participation and reduced recidivism (Sartechi et al., 2011). There were, however, some notable limitations in the meta-analysis. Most of the included studies consisted of nonrepresentative samples with disproportionate numbers of Caucasian males. Furthermore, the majority of the studies were quasi-experimental and did not control for differences between groups. It is important to note that many studies of mental health courts use quasi-experimental designs because mental health court stakeholders are likely reluctant to agree to random assignment of offenders to mental health court or standard criminal justice processing. Although quasi-experimental designs limit the nature and strength of the inferences that can be drawn, they can still provide valuable data.

Steadman et al. (2011) conducted one of the only multisite investigations of the effects of mental health courts on recidivism. They evaluated four mental health courts using a longitudinal design over 18 months with large sample sizes (447 mental health court participants and 600 nondiverted offenders), and they accounted for the differences between the two groups using propensity score matching. There was a significant difference between the two groups in recidivism rates during the 18-month period following reentry, with mental health court participants being significantly less likely to be rearrested than nondiverted offenders (49% vs. 58%). In addition, during this 18-month period, mental health (p. 104) court participants were incarcerated for significantly fewer days on average than the nondiverted offenders (82 days vs. 152 days) (Steadman et al., 2011).

In earlier research, investigators examined the frequency and nature of rearrests among mental health court participants and a comparison group of nondiverted offenders (McNiel & Binder, 2007). Results revealed that mental health court participants spent more time in the community prior to rearrest for nonviolent and violent charges, and they also exhibited a decrease in rearrest rates for nonviolent and violent offenses (26% lower and 55% lower than the comparison group, respectively). Also notable is that mental health courts seemed to produce a lasting impact, as the results remained consistent at an 18-month follow-up assessment (McNiel & Binder, 2007).

Several other quasi-experimental studies and pretest/posttest design studies have found a significant relationship between mental health courts and decreased recidivism rates. Specifically, studies have found decreased rearrest rates (Dirks-Linhorst & Linhorst, 2012; Herinckx et al., 2005; Hiday & Ray, 2010; Hiday et al., 2013; McNiel & Binder, 2007; Moore & Hiday, 2006), fewer new bookings (Trupin & Richards, 2003), fewer days of incarceration (Frailing, 2010; Hoff et al., 1999), a decrease in the commission of minor offenses (Moore & Hiday, 2006), and an increase in days spent in the community prior to rearrest (Steadman & Naples, 2005). In a randomized control trial, Cosden et al. (2013) randomly assigned 235 offenders to a mental health court or comparison group, and they found that mental health court participants had significantly fewer convictions than the comparison group (46% vs. 60%). However, they also found that mental health court participants were more likely to be convicted for violating probation requirements, while offenders in the comparison group were more likely to be convicted of a new crime (Cosden et al., 2013).

Research has also focused on identifying factors that may be associated with lower recidivism rates among mental health court participants. Studies have shown that successful graduation from a mental health court program is associated with lower recidivism rates (Burns, Hiday, & Ray, 2013; Dirks-Linhorst & Linhorst, 2012; Hiday & Ray, 2010; Kubiak, Tillander, Comartin, & Ray, 2012; Moore & Hiday, 2006; Snedeker, Beach, & Corcoran, 2017). Other research has shown that mental health court participants of Hispanic ethnicity were less likely to recidivate than Non-Hispanic participants (Burns et al., 2013) and that participants entering mental health court with more serious offenses are more likely to have lower recidivism rates (Hoff et al., 1999). One study found that lower recidivism rates were associated with fewer days of arrest and incarceration prior to mental health court enrollment, a lack of substance abuse issues, steady housing, and a diagnosis of bipolar disorder (as opposed to a diagnosis of schizophrenia or major depressive disorder) (Steadman et al., 2011). In a recent study, Snedeker et al. (2017) found that graduating from the program, having (p. 105) stable and consistent mental health treatment, and participating in a program that offered incentives are associated with lower odds of reoffending, increased time to new reoffense, and fewer criminal charges. Hiday and Ray (2010) compared recidivism rates of mental health court participants and an equally eligible group of offenders who went through standard criminal justice processing. Two years after completion, recidivism rates decreased in both groups, but the greatest reduction was exhibited by mental health court participants who completed the program (Hiday & Ray, 2010). Participants who completed the mental health court program had the lowest recidivism rate (25%) and the fewest arrests, while mental health court participants who did not complete the program had the highest recidivism rate (55%) (Hiday & Ray, 2010).

Other research has identified factors that are associated with more negative recidivism outcomes for mental health court participants. For example, participants who fail to complete a mental health court program are more likely to recidivate (Hiday et al., 2013), as are participants who have a substance abuse disorder (Hiday et al., 2013; Reich, Picard-Fritsche, Cerniglia, & Hahn, 2014) and those with higher rates of arrest (Reich et al., 2014) or more days in jail (Burns et al., 2013). Participants with a history of engaging in more serious criminal activities are also more likely to recidivate (Moore & Hiday, 2006), as are participants who are younger (Hiday et al., 2013; Reich et al., 2014; Trawver, 2013) and who are not Caucasian, have less education, and rely on disability benefits (Trawver, 2013). In other research using a pretest/posttest design, investigators found that mental health court participants who were arrested after enrollment had a significantly higher number of prior days in jail and more previous arrests (Case, Steadman, Dupuis, & Morris, 2009).

One study examined the relationship between the principles of therapeutic jurisprudence and recidivism rates within mental health courts (Redlich & Han, 2014). Specifically, the researchers examined whether certain therapeutic jurisprudence principles were present at enrollment, including procedural justice, knowledge of the nature of the mental health court, and perceived voluntariness of participation in the mental health court program. They found that higher levels of therapeutic jurisprudence principles were associated with higher rates of graduation from mental health court programs (Redlich & Han, 2014). Although this association became statistically nonsignificant when the researchers included mediating factors, Redlich and Han (2014) nonetheless concluded that an indirect relationship remained: the presence of therapeutic jurisprudence principles led to fewer rearrests, decreased incarceration times, and increased participant compliance, which subsequently increased the likelihood of successful completion of the mental health court program.

Overall, the available research suggests that mental health courts have promise in terms of reducing recidivism rates for offenders with mental illness. Although (p. 106) the research in this area is somewhat limited and some of the methodologies have notable shortcomings, the body of research is encouraging and growing. After discussing some of the more notable limitations in the research, we offer recommendations for researchers moving forward.

Methodological and Research-Based Limitations

It is important to note that many studies on mental health courts have significant methodological limitations. For example, much of the research on mental health courts is marked by high rates of participant attrition (e.g., Cosden et al., 2005), incomplete data regarding mental health (e.g., Herinckx et al., 2005), and nonrepresentative samples that did not adequately represent the composition of the criminal justice population (e.g., Herinckx et al., 2005; Sneed et al., 2006; Trupin & Richards, 2003). As such, the research findings from these studies should be interpreted with caution.

Furthermore, as with research on other problem-solving courts and other criminal justice interventions, the majority of studies on mental health courts have not used experimental designs, which limits the nature and strength of the inferences that can be drawn. Rubin (2008) created a hierarchy of evidence that set systematic reviews and meta-analyses at the highest level, followed by large-scale replications of experimental studies, experimental studies, and quasi-experimental studies. Only two studies on mental health courts have used experimental designs (Cosden et al., 2003, 2005), with the majority of mental health court research studies using quasi-experimental designs or pretest/posttest designs. Of the quasi-experimental studies identified in Honegger’s (2015) review, only four used techniques aimed to mitigate the influence of nonrandom assignment—two studies used propensity score matching (McNiel & Binder, 2007; Steadman et al., 2011) and two used covariates (Boothroyd et al., 2003, 2005). We only found one meta-analysis (Sarteschi et al., 2011) and two comprehensive literature reviews (Heilbrun et al., 2012; Honegger, 2015).

As may be gleaned from the available research, conducting methodologically rigorous research on mental health courts is quite challenging for several reasons. First, using experimental designs with mental health court participants raises several ethical and legal concerns. There are many reasons why judges, defense attorneys, prosecutors, and offenders would resist randomly assigning offenders to either a mental health court or standard criminal justice processing. However, without an experimental design, the ability to draw causal inferences regarding the effects of mental health courts is limited. Second, federal funding has allowed for the relatively rapid development of mental health courts throughout the United States, and this rapid growth has made it challenging for researchers (p. 107) to keep up (Wolff & Pogorzelski, 2005). As such, the development of mental health courts is continuing even though we still lack a firm understanding of their effects. Third, the majority of longitudinal mental health court research has been conducted over relatively short time periods. Because some serious mental health disorders are chronic, it is important to understand the long-term impacts of mental health courts on mental health and other outcomes. Finally, as with most research, some nonmeasured variables could impact the validity of mental health court research, including the experience and motivation of the mental health court staff, participant understanding of his or her own mental health disorder, and the personality and working styles of the judge and other court staff (Wolff, 2000).

The variability of mental health courts across jurisdictions is a major reason why it is difficult to draw broad conclusions about the effects of mental health courts. Mental health courts lack uniformity in structure and are idiosyncratic in operation (Erickson et al., 2006). As outlined earlier, variability is found in the following aspects of mental health courts: (1) mental health court eligibility criteria (misdemeanors only, felonies only, both), (2) psychiatric diagnosis criteria (e.g., Does the court require a DSM diagnosis? Does the court only accept participants with bipolar disorder, schizophrenia, or major depressive disorder? Does the court exclude personality disorders? Does the court accept individuals with co-occurring disorders?), (3) types and quality of services provided, and (4) measured outcomes (i.e., Is success measured by fewer days incarcerated? Fewer convictions? Improved mental health functioning? Increased use of mental health services?). The variability among mental health courts makes it difficult for researchers to effectively compare programs or make broad conclusions about mental health courts (Watson, Hanrahan, Luchins, & Luriggio, 2001).

Participant attrition and noncompliance also present challenges for researchers. Moore and Hiday (2006) reported that one-third of the mental health court participants failed to complete the program due to noncompliance or a decision to opt out. Many studies indicated that participants were “noncompliant” but did not operationalize the term. There are different types of noncompliance (e.g., failing to meet program requirements vs. a new arrest), so the lack of specificity regarding the nature of the noncompliance limits the conclusions that can be drawn.

The demographic characteristics of study samples are also a source of concern. Only 4 of the 20 studies included in Honegger’s (2015) review evaluated mental health courts with samples that were representative of the demographic composition of a justice-involved population. Honegger (2015) speculated that this demographic disparity could be accounted for by the following factors: (1) studies were conducted in geographic areas with a largely Caucasian population, (2) there were higher opt-out rates among racial and ethnic minorities compared (p. 108) with Caucasian participants, and (3) racial and ethnic minorities exhibited disproportionate ineligibility factors that precluded mental health court enrollment. Some researchers have expressed concern that many mental health court studies evaluate samples of participants that are homogenous and fail to accurately represent the offender population that the courts were designed to target (Wolff & Pogorzelski, 2005). Conducting studies with representative samples is an important step toward understanding the potential impact of mental health courts and, in turn, justifying the development and implementation of these courts across the country.

The lack of representativeness among study samples is obviously a concern for researchers, but it also highlights a larger concern: mental health courts do not appear to be enrolling participants who are reflective of the larger correctional population. As previously noted, research suggests that mental health court participants are predominantly Caucasian males in their mid-30s (Sarteschi et al., 2011). Given the demographic makeup of the correctional population in the United States, this finding is concerning. Racial and ethnic minorities are overrepresented in the U.S. criminal justice system (Rosich, 2007; Sabol & Minton, 2008), yet they appear to be considerably underrepresented in mental health courts. This suggests that mental health courts are failing to enroll participants from the target offender population.

To evaluate behavioral health outcomes, much of the extant research focuses on the number of therapy hours or case management meetings provided to mental health court participants, with relatively little focus on the specific types of services provided. Research should consider whether the provided interventions are evidence-based and whether they were designed to target criminogenic risk factors (Keator et al., 2013). It is also important to examine whether mental health court participants received interventions designed to targeted problem-solving and criminal thinking (as opposed to general supportive therapy). Because behavioral health treatments are so varied, simply evaluating the number of treatment hours provided will inevitably leave many questions unanswered. To better understand the relationship among treatment, mental health outcomes, and recidivism rates, the nature and quality of the treatment must be considered.

Recommendations for Future Research

The research reviewed in this chapter provides reason to be cautiously optimistic about mental health courts. There is growing evidence that mental health courts reduce recidivism, but research on the ability of these courts to (p. 109) connect participants to needed treatment services and improve mental health outcomes is mixed. Given the number of mental health courts in the United States (400+), and the projection that mental health courts will be the fastest growing type of problem-solving court over the next several years (Marlowe et al., 2016), it is important that we have a clear understanding of the effects of these courts.

A key recommendation for researchers is to use rigorous methodology. As reviewed in Chapter 4, several researchers have conducted randomized controlled trials in drug courts, so there is precedent for conducting rigorous research in a problem-solving court. If an experimental design is not possible, researchers should use statistical methods to account for the differences between nonrandomized groups, such as propensity score matching or the inclusion of covariates. Furthermore, given concerns related to participant attrition, researchers should use intent-to-treat designs as the primary method of analysis, with protocol analyses being considered a secondary or parallel approach.

There are a number of other suggestions for future researchers. First, mental health courts should be compared to other diversion programs designed for offenders with mental health disorders (Wolff & Pogorzelski, 2005). Such research would provide an important comparative benchmark regarding the effects of mental health courts. Second, research should examine the relationship between the nature and quality of the provided services and various participant outcomes (e.g., mental health functioning, quality of life, recidivism). In their comprehensive quantitative review, Sarteschi et al. (2011) found that some studies suggested that there may be a relationship between the quality of the treatment services and the success of mental health court participants. Unfortunately, there is insufficient empirical evidence to support this claim at present. Examining the services provided to mental health court participants is complicated because mental health courts typically have limited control over the quality of the treatment provided to participants. As such, the court itself may not be responsible if a participant’s mental health symptoms do not improve with the treatment. Third, more longitudinal and follow-up research is needed to examine the longer term effects of mental health courts. Fourth, research should examine the previously discussed demographic disparity in which mental health courts are primarily enrolling groups of participants who are not reflective of the correctional population. Fifth, research should examine whether mental health courts are cost-effective compared to standard criminal justice processing. Sixth, once more outcome research is conducted, researchers should conduct dismantling studies to examine which specific aspects of mental health courts are responsible for their effects.

(p. 110) Finally, more research is needed to examine the complex relationship among mental illness, criminogenic risk factors, and recidivism. Research suggests that offenders with and without mental illness share the same types of criminogenic risk factors (Andrews, Bonta, & Wormith, 2006; Bonta, Law, & Hanson, 1998). However, “Despite mounting evidence for a more complex understanding of recidivism among persons with mental illness, mental health court research has customarily assumed that clinical and psychosocial factors precipitate criminal behavior, and that improvement in symptomology reduces recidivism” (Honegger, 2015, p. 486). Indeed, Skeem, Manchak, and Peterson (2011) have asserted that providing treatment through mental health courts may not be the most effective way to reduce recidivism. As such, future research should consider using a more nuanced conceptual model that takes into account the relationship among mental health symptoms, criminogenic risk factors, and recidivism (see Skeem & Monahan, 2011).

Next Steps

Since the first mental health court was established in Florida more than 20 years ago, these courts have become a firmly established intervention in the criminal justice system. With more than 400 operational mental health courts and many more in advanced stages of development, it is safe to assume that mental health courts will continue to serve certain offenders with mental health involvement. Despite the proliferation of these courts, important questions remain about their utility, effectiveness, and how well they serve the intended population of offenders. As noted in previous sections of this chapter, we need a better understanding of the effects of these courts on recidivism and mental health functioning, among other things. There is also a need for cost–benefit research that examines whether mental health courts make fiscal sense compared to other interventions for offenders with mental illness.

In addition to further developing the research base, consideration should be given to how the mental health court model can be adapted or expanded. As with drug courts, mental health courts serve a fraction of the offenders with mental illness, so identifying ways to broaden the reach of mental health courts would allow them to serve more eligible offenders. One way to broaden the reach of mental health courts is to use less conservative eligibility criteria. As previously noted, some mental health courts require a formal psychiatric diagnosis for an offender to be eligible. Requiring a formal diagnosis as a prerequisite to eligibility may be too restrictive. There are likely many offenders with mental health symptoms that do not rise to the level of severity needed to satisfy diagnostic (p. 111) criteria, but who would nonetheless benefit from the services offered as part of mental health court. A loosening of the eligibility criteria, without compromising public safety, would be an effective way to expand the reach of mental health courts.

As mental health courts continue to be developed and refined, it will be important for these courts to address both criminogenic needs and behavioral health needs. For these courts to be effective, there cannot be a sharp distinction between behavioral health interventions and criminogenic interventions. Indeed, interventions for behavioral health needs and criminogenic needs often overlap, and eligible offenders may need help in the areas of mental health, substance use, family, housing, and jobs. There is a subgroup of offenders whose offending behavior is driven primarily by behavioral health symptoms, but the majority of offending behavior results from both behavioral health needs and criminogenic needs (Andrews et al., 2006; Bonta et al., 1998). Accordingly, mental health courts can be most effective by addressing both sets of needs.

Another possibility involves combining mental health courts with other types of problem-solving courts, an approach being considered in Delaware. Given the proliferation of problem-solving courts in Delaware over the past 20 years, Delaware Supreme Court Chief Justice Leo E. Strine, Jr. appointed the Criminal Justice Council of the Judiciary to review all of the problem-solving courts in Delaware to help determine how to increase their impact (Criminal Justice Council of the Judiciary, 2016). Delaware has a range of problem-solving courts, including truancy court, human trafficking court, gun court, reentry court, drug court, veterans treatment court, and mental health court. Among other things, this Council concluded that Delaware would benefit from creating a unified statewide treatment court that encompasses all of the other problem-solving courts, including mental health courts (Criminal Justice Council of the Judiciary, 2016). The Council asserted that the needs of offenders often span multiple problem-solving courts, and a single-source approach would enhance the likelihood that these multiple needs are addressed. Such an approach would also allow for the consolidation of treatment providers into a single unit, which would improve communication and the coordination of treatment services (Criminal Justice Council of the Judiciary, 2016). A statewide approach may be feasible in Delaware, which is a relatively small state geographically, and larger states could create several umbrella problem-solving courts spread out geographically throughout the state.

Finally, mental health courts may benefit from use of the restorative justice model. Restorative justice focuses on the rehabilitation of offenders through reconciliation with victims and the community. As mental health courts have become more willing over the years to accept offenders charged with violent (p. 112) offenses, victim involvement in mental health courts has increased. Restorative justice has been shown to have several benefits over standard criminal justice processing. In a meta-analysis, Latimer, Dowden, and Muise (2005) found that restorative justice programs were more effective than traditional nonrestorative approaches in terms of victim and offender satisfaction, restitution compliance, and recidivism. Adding in a restorative justice component may therefore boost the effectiveness of mental health courts in reducing criminal recidivism.