Racial Variation in Mental Health Court Experiences and the Associations of These Experiences With Recidivism

Abstract

Little is known about how mental health court (MHC) experiences (including satisfaction and sanctioning, incentivizing, and life change events) differ by race and how this variation affects recidivism among MHC participants. This study examined how MHC experiences and recidivism differ between White (n = 170) and Black (n = 132) participants from four different MHCs in the United Sates. Negative binomial regression models, conducted separately for the two groups, indicated that life changes were associated with decreased arrests and that court sanctions were associated with increased number of arrests for both White and Black participants. Higher program satisfaction was positively associated with recidivism among Black participants only. To provide equitable services for people convicted of a crime with mental illness, professionals should acknowledge racial variation in the experiences of MHC participants (both within the MHC and the communities in which they are situated) and how these experiences relate to recidivism.

Keywords

mental health court recidivism criminal justice racial variation

Introduction

The overrepresentation of people of color in the U.S. criminal justice system is well-known (Alexander, 2012; Tonry, 2010). In addition to these disparities, overrepresentation of people with mental illness in the criminal justice system is a long-standing problem (Prins, 2014; Torrey et al., 2014). Research has demonstrated that mental health courts (MHCs) can function to reduce recidivism and connect people with mental illness to treatment (Honegger, 2015; Lowder et al., 2018; Sarteschi et al., 2011). However, there is a dearth of research on how and for whom desired outcomes are brought about. In particular, it is not yet clear whether racial disparities exist in MHCs similar to what is seen in other parts of the justice system. In addition, little research has directly examined how variation of MHC experiences by race affects future criminal behaviors of MHC participants. The primary goal of this study is to examine how MHC experiences and outcomes differ by race. The MHC experiences include program satisfaction, life changes after MHC enrollment, and receipt of sanctions and incentives; recidivism is the outcome of interest.

MHCs

MHCs are specialized programs for defendants with mental illness that seek the adjudication of criminal charges and municipal code violations by using a problem-solving model. MHCs divert people charged with a crime into the community to receive necessary treatment and support services instead of incarceration (National Institute of Justice, n.d.). Although these courts may differ somewhat in structure, objectives, and function by jurisdiction (Wolff & Pogorzelski, 2005), the essential elements of MHCs include a screening process to determine who can participate in the program, multidisciplinary planning and administration from dedicated personnel (e.g., judge, prosecutor, defense counsel, treatment providers), clear terms of participation and informed choice to participate, and confidentiality safeguards. In addition, MHCs aim to increase participants’ access to evidence-based treatments and services, monitor participants’ adherence to court conditions, and employ sanctions and incentives to promote adherence to these conditions (Thompson et al., 2007).

Conceptual Framework

Figure 1 depicts the framework guiding our analyses. In summary, MHCs frequently employ sanctions and incentives as a method of encouraging behaviors and engagement in services that are aimed at initiating and sustaining positive life changes among their participants (Linhorst et al., 2010; Redlich et al., 2006). How and to what extent participants experience these sanctioning, incentivizing, and life events will shape their perceptions of satisfaction with the MHC (Canada & Watson, 2013; Redlich & Han, 2014). The extent to which positive life changes are imparted will influence whether MHC participants return to criminal behavior and incarceration (Han, 2019; Ward & Maruna, 2007). Importantly, these processes are not immune to the personal and systemic biases that permeate American society and that contribute to discrimination and to differential experiences based upon race/ethnicity status (Bonilla-Silva, 1997). For example, implicit biases against people of color may lead to dissimilar application of sanctions and incentives to MHC participants. This unequal treatment can lead MHC participants who are people of color to develop lower perceptions of satisfaction with the court and to receive fewer benefits from MHC participation (i.e., positive life changes) than White participants. Moreover, variation in recidivistic outcomes that are played out largely in the community may also reflect the larger racial/ethnic disparities seen in policing, law enforcement, and America’s court system (Brewer & Heitzeg, 2008).

Figure 1:

Mental Health Court Processes and Community Outcomes Pervaded by Bias and Discrimination

Racial and Ethnic Discrimination

Considerable disadvantage and hardship can be imposed upon others by persons who, simply by their position in the bureaucratic structure, are allowed to exert wide discretion in whether to unleash the force of the justice system on those identified as transgressing (or suspected of transgressing) the law. For example, Olson (2016), in his research on race/ethnicity and the use of solitary confinement in American prisons, discusses the work of Lipsky (2010) to highlight the substantial power prison staff have in terms of whether to impose severe sanctions on individuals convicted of a crime. Such professionals are termed “street-level bureaucrats” by Lipsky (2010) and have “considerable discretion in determining the nature, amount, and quality of benefits and sanctions provided by their agencies” (p. 13). The sanctioning of individuals convicted of a crime in prison is but one point in the criminal justice system wherein decisions are made—often by single actors or small groups of people who have been granted broad discretionary authority—as to whether and to what degree to expose Americans to consequential restrictions and mandates. At many of these points, people of color are subjected to these impositions at higher rates and more harshly than Whites (e.g., Bailey et al., 2019; Epp et al., 2013; Kochel et al., 2011).

Focal concerns theory is put forth to explain why criminal justice actors may discriminate against people of color. Focal concerns theory states that criminal justice authorities’ imposition of sanctions and punishment is informed by their perceptions of the blameworthiness of people suspected or convicted of a crime, their beliefs about how best to protect the community, and by their understanding of the availability of resources and agency constraints (see Steffensmeier et al., 1998). Given often limited knowledge of their individual characteristics and their circumstances, authorities rely on cognitive heuristics (i.e., stereotypes) to determine the culpability of the person being scrutinized. These stereotypes are socially constructed and often reflect the beliefs that people of color possess more negative internal traits, are more dangerous, and more criminal than Whites (e.g., Steen et al., 2005). Contributing to racialized social systems (Bonilla-Silva, 1997), “focal concerns, thus, are the vehicles by which social context intrudes on sanctioning decisions and by which substantive rationalities are brought to bear within formalistic social-control regimes” (Lin et al., 2010, p. 763).

Race/Ethnicity and MHC Experiences

MHCs frequently attempt to modify the behavior of participants through the use of sanctions and incentives (Callahan et al., 2013; Thompson et al., 2007). The basic premise underlying the use of these sanctions and incentives is that they will, as part of the cadre of services offered through MHCs, promote engagement in services that facilitate meaningful improvements in the lives of MHC participants. That is, “. . . mental health courts are based on a therapeutic philosophy that focuses on linking individuals convicted of a crime to community-based treatments and resources and offering incentives and sanctions that encourage pro-social change . . .” (Dollar et al., 2018, p. 33). Recent research conducted with data from the MacArthur MHC Project underscores the importance of imparting such positive life changes among MHC participants, as these changes were found to be associated with reduced recidivism (Han, 2019). Moreover, such changes—often reflective of mental health recovery (e.g., improved relationships with families, increased empathy toward others, and greater understanding of mental illness and illness management)—are seen as important goals by MHC participants (Canada & Ray, 2016; Eschbach et al., 2019; Yuan & Capriotti, 2019).

Honegger’s (2015) review of MHCs highlights the underrepresentation of people of color in the MHC literature, in general, and posits this may be due to research being conducted in locations that have populations that are overwhelmingly White, people of color declining to participate in MHCs or MHC research, and/or people of color being found ineligible for these courts at disproportionate rates. However, research that has examined racial differences in the MHC context suggests some disparities. For example, Steadman and colleagues (2005) found that eligible Black adults were less likely than eligible Whites to be referred to MHC; though, after referral, there was no influence of race (White vs. Black) in acceptance into the court. McNiel and Binder (2007) found that White individuals convicted of a crime were less likely than a group of people categorized as not White to enroll in the MHC program. Research has also identified racial variations such that White participants appeared more likely to comply with court orders compared with a group of MHC participants categorized as not White (Redlich et al., 2010), which may contribute to the lower rates of graduation from MHCs observed among members of the latter group (Dirks-Linhorst & Linhorst, 2012; Hiday & Ray, 2010). In sum, previous studies examining racial disparities in MHCs have emphasized pre-enrollment processes or graduation rates without exploring the court experience, its impact on justice outcomes, and how these relationships may differ by race/ethnicity. Indeed, Dirks-Linhorst and Linhorst (2012) recommend that further study is necessary to better understand the mechanisms of MHCs by exploring the effect diverse racial/ethnic backgrounds have on program outcomes.

MHC Experiences and Recidivism

In the present study, MHC experience refers to program satisfaction, life changes as a direct benefit of the program enrollment, and the receipt of sanctions and incentives among MHC participants. While MHC studies show some positive outcomes (Han, 2016; Hiday & Ray, 2010; Redlich & Han, 2014), there is a paucity of research on what programmatic experiences are associated with these outcomes. Thus, researchers suggest that evaluation studies need to investigate the factors associated with reported outcomes to fully understand mechanisms of success and failure (Han & Redlich, 2015).

Research within other problem-solving court contexts may be instructive in this regard. For example, one meta-analytic study reveals that harsh sanctions, including incarceration, actually increase the possibility of recidivism for drug court participants and that intermediate community sanctions (e.g., boot camps, electronic monitoring, house arrest) have no effects on future arrest or drug use in drug courts (Gendreau et al., 2000). Similarly, Marlowe et al.’s (2003) review of the literature led them to argue that punitive strategies have little impact on drug use or future arrest among drug court participants. In contrast, a systematic review of 55 drug court evaluation studies found that clear incentives are an essential contributor to successful outcomes (Wilson et al., 2006). Little is known about these relationships in the MHC context and how they may differ by race.

Several studies have begun to investigate participants’ experiences while in MHC. MHC participants are known to have more satisfaction with court procedures compared with people with mental illness who were charged with a crime and processed through the traditional court system (Poythress et al., 2002; Stefan & Winick, 2005). In addition, MHC participants report more positive beliefs that procedures of the MHC were just when compared with people experiencing involuntary hospital admission or traditional court (Poythress et al., 2002; Wales et al., 2010). Moreover, MHC participants feel more respected and have more positive feelings about the program than court-ordered outpatients convicted of a crime in civil court (Munetz et al., 2014). These findings may help to explain why MHC participants have significantly improved outcomes (e.g., better global functioning, more life satisfaction, less psychological distress, less drug and substance use) compared with traditional court participants (Cosden et al., 2005).

Research by Han (2019) is one of the few studies to examine how life changes attributed to participation in MHC are related to recidivism. Han found receipt of sanctions positively associated with recidivism while positive life changes among MHC participants were found negatively associated with recidivism. As a result, the author suggested that MHCs should focus on intermediate outcomes such as improving relationships with others and symptom management to reduce recidivism among MHC participants. Importantly, the author did not examine whether the race of MHC participants differentiated these intermediate and recidivism outcomes.

Studies of MHCs show outcomes of reduced recidivism, but explication of the mechanisms that support these outcomes is lacking. Such gaps preclude replication of effective MHC processes. The purpose of this study is to examine how MHC experiences are related to recidivism and to test whether these experiences and their relationships with recidivism differ by race. Given prior research, we hypothesize the following:

Hypothesis 1: Black participants will experience fewer positive life changes, less satisfaction with the court, and more sanctions and fewer incentives than White participants.

Hypothesis 2: Positive MHC experiences (i.e., positive life changes, satisfaction, and more incentives and fewer sanctions) will be negatively associated with recidivism.

Hypothesis 3: The associations among MHC experiences and recidivism will differ based upon race.

Method

Data Source and Study Sample

Data for the current study were obtained from the MacArthur MHC project (see Redlich et al., 2010). The MacArthur MHC project is a multi-site and prospective research project that involved participants from four MHCs (i.e., San Francisco, Santa Clara, Minneapolis, and Indianapolis), selected based on a national survey of MHCs in the United States (n = 90). To be eligible for the MacArthur MHC study, the courts were required to have a large caseload (n > 200), to have a lengthy history of operation (more than 4 years), and to accept participants with a wide range of charge (e.g., misdemeanor and felony; Redlich et al., 2006).

Participants’ socio-demographic information, MHC experiences (e.g., receipt of sanctions and incentives), clinical characteristics (e.g., psychiatric symptoms, substance use), and crime-related data (e.g., crime severity, age of arrest) were self-reported via survey. Study participants were surveyed twice, at baseline entry in the court and 6 months later. Ninety-one out of 448 participants in the MacArthur project (20.31%) failed to complete the follow-up survey. Testing indicated those who completed both surveys did not differ from those who did not in terms of age, gender, education, the presence of a serious mental illness (SMI), crime severity, and age at first offense. A higher percentage of Black respondents completed both surveys than completed only the baseline survey. Only study participants who completed the baseline and follow-up survey were included: 170 White and 132 Black participants. Other demographic groups were excluded from the current study due to their small sample size (e.g., Latinx and biracial individuals).

Measures

MHC Experiences

The main independent variables for the current study included program satisfaction, life changes, receipt of court sanctions, and receipt of court incentives. Data to operationalize MHC experiences were collected via the 6-month follow-up survey. The MHC Satisfaction Scale (MHCSS) was adapted from the Mental Health Statistics Improvement Program (MHSIP) survey (J. A. Carlson et al., 2011; Jerrell, 2006). The revised MHCSS includes eight statements concerning participants’ feelings toward and experiences within the MHC (e.g., “I liked being in the mental health court in the past 6 months,” “I felt free to complain at the mental health court”). A 5-point Likert-type scale was used, with 1 = strongly agree to 5 = strongly disagree. Each score was reverse coded, with a higher score indicating more satisfaction and then summed for a total MHC satisfaction score. In this study, Cronbach’s alpha of the MHCSS was .88.

The Scale of Life Changes included six statements asking what changes participants made in their life as a direct benefit of being in the MHC in terms of their own life, family relationships, mental health symptoms, employment and education, and legal involvement (e.g., “I am better able to control my life”; “I am getting along better with my family”; “I do better in school and/or work”). A 5-point Likert-type scale was used, with 1 = strongly agree to 5 = strongly disagree. Each score was reverse coded and then summed so that a higher total score indicates more positive life changes. In this study, Cronbach’s alpha of this measure was .83.

Sanctions were measured by asking, “From the things a mental health court can do, which of the following sanctions or penalties have actually happened to you in the past 6 months?” Sanctions consisted of (a) having to go to the MHC more often and see the judge; (b) doing community service; (c) having to see a psychiatrist, psychologist, case manager, or probation officer more often; (d) getting a lecture from the MHC judge; (e) having privileges taken away, like getting to leave the state or having your curfew restricted; and (f) going back to jail. Each sanction was measured by a dichotomous question (yes = 1, no = 0) and summed to provide a total number of sanctions MHC participants received during the past 6 months.

Incentives were measured by asking, “From the things a mental health court can do, which of the following things have actually happened in the past 6 months to you to help keep you on track?” Incentives consisted of (a) having to attend court less often, (b) praise from the judge or others in the court, (c) having the judge say good things to you, (d) having a case manager or probation officer write a good report, and (e) getting a gift certificate. Each incentive was measured by a dichotomous question (yes = 1, no = 0) and summed to provide a total score for incentives experienced in MHC.

Recidivism

The dependent variable, recidivism, was considered as the number of times the study participant was arrested for a new offense within the 12 months following MHC enrollment, according to arrest records from the Federal Bureau of Investigation and county justice systems. Returns to jail for violations of community supervision during MHC participation were coded as a sanction.

Demographic Variables

Demographic variables were collected at baseline entry to the court and included age (in years) and gender (male or female). Respondents who self-identified as either White or Black were included in the study. Due to small sample sizes, other racial groups, Latinx, and respondents who self-identified multiple racial categories were excluded.

Major Risk Factors for Recidivism

Research on criminal risk has identified a set of “Central Eight” risk factors that are strongly predictive of recidivistic behavior patterns (for a review, see Bonta & Andrews, 2016). Of the Central Eight risk factors, Low Educational and Employment Attachment, Lack of Marital Supports, Substance Use, and Prior Criminal Behavior are included here. Education was dichotomized as 0 to 11th grade (0) and high school graduate or higher (1). Employment status was indicated as either (0) unemployed, (1) employed part-time, or (2) employed full-time during past 6 months. We dichotomized marital status: (1) married or cohabiting with a significant other or sexual partner versus (0) widowed, divorced, separated, or never married. As a continuous variable, drug use was measured in number of days by asking individuals at MHC enrollment: “In the past 30 days, how often have you used illegal or non-prescribed drugs?”

Variables reflecting prior criminal behavior included the age of first arrest and most serious crime. The age of the first arrest was a continuous variable and measured by asking, “How old were you when you were first arrested?” Then, reflective of recent research on early-start/late-start offending among people with mental illness (see Matejkowski et al., 2017), age of first arrest was recoded to indicate presence of an arrest prior to age 18. In addition, utilizing data from the baseline survey, we identified the participant’s most serious crime with a single-item question: “Tell me the most serious offenses you have ever been arrested for.” Crime severity was originally coded on a 10-point severity scale (i.e., 1 = Murder/Manslaughter, 2 = Rape/Sodomy, 3 = Assault, 4 = Robbery, 5 = Kidnapping/Arson, 6 = Other crimes against a person, 7 = Sex crimes, 8 = Property crimes, 9 = Drug crimes, and 10 = Minor crimes) using the objective crime severity categories developed by the U.S. Department of Justice (Wolfgang et al., 1985). Due to insufficient numbers in certain categories (e.g., rape [n = 2]; kidnapping/arson [n = 1]), the severity of crime was reversed and recoded into three groups (1–3 = minor crime, 4–7 = moderate crime, 8–10 = severe crime). Accordingly, a higher score indicates a higher level of crime severity.

Minor Risk Factors for Recidivism

Minor criminal risk factors are a broad set of individual and environmental characteristics that, when compared with major risk factors, have a weaker association with criminal behavior than the Central Eight risk factors. While not directly related to criminal behavior, these risk factors still have an important role to play in crime prevention efforts—as indirect, “upstream” targets for crime prevention and as characteristics of people involved with crime and their environments that should be considered to improve engagement in treatment (for a review, see Matejkowski & Conrad, 2019). Minor risk factors included in the current study reflected poverty and mental health status.

Poverty was operationalized through data indicating receipt of public benefits and homelessness. Data on receipt of public benefits (e.g., Medicaid, Social Security, or Social Security Disability Insurance) indicated whether participants received those benefits during the past 6 months (yes = 1, no = 0). In addition, homelessness was a continuous variable and measured by asking “During the past 6-month period, approximately how many nights were you homeless?”

Mental health status was operationalized with a measure of psychiatric symptomatology and with a variable indicating the presence of an SMI. Psychiatric symptoms were measured at 6 months following court enrollment using the Colorado Symptoms Index (CSI; Conrad et al., 2001). The measure includes 15 statements concerning psychological or emotional difficulties (e.g., “In the past month, how often did you hear voices, or hear or see things that other people did not think were there?”). A 5-point Likert-type scale was used, with 0 (not at all) to 4 (at least every day). These values were then summed for a total CSI score. The CSI has been used with large number of people convicted of a crime with mental illness and has been found to be sensitive and internally consistent (Draine et al., 2005). In this study, Cronbach’s alpha for the CSI was .85.

We also dichotomized diagnoses obtained from county court and MHC records. An indicator of SMI was created by grouping schizophrenia spectrum (38.74%), other Axis I psychosis (5.3%), bipolar (24.83%), and depression (13.91%) versus all other diagnosis (17.22%). This diagnostic criterion alone has been used extensively in the literature to define SMI (e.g., Matejkowski & Ostermann, 2015; McAlpine & Mechanic, 2000) and indicates a high level of interference with social and occupational functioning.

Data Analysis

We compared sample characteristics between White and Black respondents using independent-samples t tests for continuous variables and chi-square tests for categorical variables and provided corresponding effect sizes (Cohen’s d; Cohen, 1988) with confidence intervals (CIs). We used regression models to identify factors associated with recidivism. Due to many study participants having no arrests or only one arrest (76.90%) and the resultant non-normal distribution of our dependent variable (e.g., Kurtosis = 8.61), we conducted negative binomial regression (Allison, 2010; Hilbe, 2011). A log-likelihood test also confirmed that negative binomial regression models with our sample had better model fit than Poisson regression models (p < .001). We tested for potential clustering of the data among the four MHCs. Results of this testing (intraclass correlation coefficient = .014) indicated negligible clustering effects and therefore, no need for multilevel modeling (Hox et al., 2017).

Descriptive analyses found the drug use variable had approximately 10% missing observations (n = 32). Missingness tests indicated mean imputation for White and Black samples separately was appropriate (Li, 2013). Finally, multicollinearity was examined using a conservative cutoff point of the variance inflation factor (VIF < 4.0; Fox, 1991). No variable violated this rule. Power analyses indicated that a sample size of 223 would yield a desired statistical power of 0.8 (Cohen, 1988). Thus, the sample size (N = 302) is sufficient to identify meaningful effects. We conducted all statistical analyses using STATA, version 15.

Results

Table 1 presents sample characteristics. The mean age of the sample was 38.04 (SD = 0.61). The majority of the sample was male (57.24%), not married or cohabiting (74.26%), had at least a high school diploma (61.72%), and received public benefits (52.15%). In terms of crime-related factors, the mean number of arrests was 0.87 (SD = 0.08) and approximately 56% of the sample had an arrest prior to age 18. The most common serious crime category included property and drug crimes (46.18%), followed by violent crimes (25.58%).

Table 1:

Sample Characteristics

	M (SE) or Frequency (%)			Effect size^a [CI]	t or χ²	p
Variables	All (N = 302)	White (n = 170)	Black (n = 132)	Effect size^a [CI]	t or χ²	p
Socio-demographic variables
Age (years)	38.04 (0.61)	37.66 (0.84)	38.55 (0.88)	0.09 [1.14, 0.31]	0.72	.453
Gender				−0.17 [−0.40, 0.05]	2.27	.134
Male	174 (57.24)	92 (53.49)	82 (62.12)
Female	130 (42.76)	80 (46.51)	50 (37.88)
Education				−0.52 [−0.75, −0.29]	4.53	<.001
0–11th grade	116 (38.28)	47 (27.49)	69 (52.27)
High school graduate and higher	187 (61.72)	124 (72.51)	63 (47.73)
Marital status				−0.22 [−0.45, 0.01]	3.42	.056
Married or cohabiting	78 (25.74)	51 (29.82)	27 (20.45)
Other	225 (74.26)	120 (70.18)	105 (79.55)
Public benefit				0.26 [0.03, 0.49]	5.06	.024
Received	158 (52.15)	80 (46.51)	78 (59.54)
Not received	145 (47.85)	92 (53.49)	53 (40.64)
Drug use (days)	3.87 (0.52)	3.22 (0.62)	4.75 (0.88)	0.19 [−0.03, 0.42]	1.46	.097
Employment status				−0.36 [−0.59, −0.13]	10.25	.006
Unemployed	233 (77.15)	122 (71.76)	111 (84.09)
Part-time	43 (14.24)	26 (15.29)	17 (12.88)
Full-time	26 (8.61)	22 (12.94)	4 (3.03)
Homelessness (days)	9.25 (1.63)	7.48 (1.87)	11.60 (2.86)	0.15 [−0.08, 0.38]	1.25	.211
Colorado Symptom Index (higher, severe)	13.33 (0.31)	13.18 (0.36)	13.53 (0.51)	0.07 [−0.16, 0.29]	0.58	.561
Serious mental illness				−0.08 [−0.31, 0.15]	0.49	.485
No	52 (17.22)	27 (15.88)	25 (18.94)
Yes	250 (82.78)	143 (84.12)	107 (81.06)
Crime-related variables
Number of arrests 12-month post MHC enrollment	0.87 (0.08)	0.70 (0.09)	1.08 (0.13)	0.28 [0.05, 0.51]	2.43	.017
Arrest prior to age 18				0.37 [0.14, 0.60]	9.83	.002
No	168 (55.63)	108 (63.53)	60 (45.45)
Yes	134 (44.37)	62 (36.47)	72 (54.55)
Severity of crime				0.20 [−0.03, 0.43]	3.78	.287
Minor crimes	28 (9.2)	20 (11.76)	8 (6.11)
Property and drug crimes	138 (46.18)	80 (47.06)	59 (45.04)
All other crimes against a person	57 (18.94)	31 (18.24)	26 (19.85)
Murder, rape, assault	77 (25.58)	39 (22.94)	38 (29.01)
MHC experience variables
Program satisfaction (higher, more)	29.51 (0.41)	28.69 (0.56)	30.58 (0.60)	0.27 [0.04, 0.49]	2.27	.023
Life changes (higher, positive)	18.86 (0.33)	18.22 (0.47)	19.69 (0.46)	0.26 [0.03, 0.48]	2.21	.028
Court sanction (higher, more)	1.10 (0.08)	0.97 (0.10)	1.27 (0.13)	0.22 [−0.01, 0.45]	1.91	.057
Court incentive (higher, more)	2.38 (0.08)	2.11 (0.10)	2.74 (0.12)	0.47 [0.23, 0.70]	4.01	<.001

Note. SE = standard error; CI = confidence interval; MHC = mental health court.

Cohen’s d (Cohen, 1988).

Bivariate statistics indicate that, within the 6-month follow-up period, Black participants had a greater number of arrests than White counterparts (1.08 vs. 0.70, t = 2.43, p = .017). A higher percentage of Black than White participants had been arrested before the age of 18 (54.55% vs. 36.47%, χ² = 9.83, p = .002). Furthermore, Black participants had a lower level of educational attainment (χ² = 4. 53, p < .001) and a higher percentage received public benefits (χ² = 5. 06, p = .024) when compared with White participants. Black participants also had less stable employment compared with White participants (unemployed: 84.09% vs. 71.76%, χ² = 10.25, p = .006). Regarding MHC experiences, Black participants reported a higher level of satisfaction with the MHC program (30.58 vs. 28.69, t = 2.27, p = .023), more positive life changes (19.69 vs. 18.22, t = 2.21, p = .028), and receipt of more court incentives (2.74 vs. 2.11, t = 4.01, p < .001) when compared with White participants. Receipt of sanctions was similar across both groups.

Table 2 shows the results of negative binomial regression models. The results of Model 1 (the full sample; N = 302) indicate positive life changes and receipt of public benefits were associated with reduced recidivism, while program satisfaction and court sanctions were both associated with increased recidivism (all ps < .05). The full model also indicates that being Black was associated with increased number of rearrests (β = −.38, 95% CI = [−0.75, −0.01], p = .046). To further examine racial variation in factors associated with recidivism among Black and White MHC participants, we conducted two separate negative binomial regression models, Models 2 and 3. The results of Model 2 (White participants only; n = 170) indicate that life changes were negatively related to rearrests (β = −.07, 95% CI = [−0.13, −0.01], p = .027), while court sanctions was positively associated with rearrests (β = .38, 95% CI = [0.18, 0.58], p < .001). In Model 3 (Black participants only, n = 132), the two significant factors found in Model 2 (i.e., life changes and court sanctions) remained significant and with the same directions (p < .05). In addition, level of program satisfaction of Black participants was found to be positively associated with rearrests (β = .06, 95% CI = [0.20, 0.11], p = .004).

Table 2:

Results of Negative Binominal Regression Models

	Model 1 (N = 302)			Model 2 (n = 170)			Model 3 (n = 132)
Variables	β	95% CI	p	β	95% CI	p	β	95% CI	p
Race (White = 1)	−.38	[−0.75, −0.01]	.046	—	—	—	—	—	—
Other socio-demographic variables
Age (years)	−.01	[−0.03, 0.01]	.252	−.01	[−0.04, 0.02]	.500	−.01	[−0.03, 0.01]	.310
Gender (female = 1)	.12	[−0.25, 0.49]	.528	.50	[−0.10, 1.11]	.103	−.17	[−0.70, 0.36]	.530
Education (high school or higher = 1)	−.08	[−0.45, 0.29]	.675	−.13	[−0.373, 0.46]	.656	−.005	[−0.47, 0.46]	.985
Marital status (married or cohabiting = 1)	−.06	[−0.46, 0.34]	.765	−.12	[−0.71, 0.47]	.683	−.01	[−0.58, 0.57]	.986
Public benefit (received = 1)	−.43	[−0.78, −0.08]	.017	−.34	[−0.92, 0.24]	.248	−.56	[−1.03, −0.08]	.022
Drug use (days)	.01	[−0.02, 0.03]	.639	−.004	[−0.04, 0.03]	.830	.02	[−0.01, 0.04]	.234
Employment status (reference = unemployed)
Part-time	−.20	[−0.74, 0.35]	.478	.17	[−0.68, 1.02]	.700	−.45	[−1.25, 0.35]	.267
Full-time	−.41	[−1.18, 0.35]	.289	.001	[−0.96, 0.96]	.999	−1.52	[−3.67, 0.63]	.165
Homelessness (days)	.002	[−0.003, 0.01]	.411	.01	[−0.003, 0.02]	.196	.001	[−0.01, 0.01]	.903
Colorado Symptom Index (higher, severe)	−.03	[−0.07, 0.001]	.054	−.05	[−0.12, 0.01]	.096	−.02	[−0.06, 0.02]	.294
Serious mental illness (yes = 1)	.13	[−0.34, 0.59]	.587	.41	[−0.39, 1.21]	.316	−.03	[−0.62, 0.56]	.923
Crime-related variables
Arrest prior to age 18 (yes = 1)	.01	[−0.38, 0.40]	.968	.06	[−0.56, 0.69]	.840	−.12	[−0.64, 0.40]	.652
Severity of crime (reference = minor crime)
Property and drug crimes	−.02	[−0.69, 0.64]	.948	−.04	[−0.95, 0.87]	.929	.29	[−0.83, 1.41]	.607
All other crimes against a person	−.07	[−0.83, 0.68]	.852	−.17	[−1.23, 0.88]	.746	.28	[−0.95, 1.50]	.659
Murder, rape, assault	−.22	[−0.95, 0.51]	.555	−.30	[−1.34, 0.75]	.576	.14	[−1.04, 1.31]	.821
Mental health court experience variables
Program satisfaction (higher, more)	.04	[0.01, 0.08]	.009	.03	[−0.03, 0.08]	.350	.06	[0.02, 0.11]	.004
Life changes (higher, positive)	−.07	[−0.11, −0.03]	<.001	−.07	[−0.13, −0.01]	.027	−.07	[−0.13, −0.02]	.011
Court sanction (higher, more)	.29	[0.17, 0.41]	<.001	.38	[0.18, 0.58]	<.001	.21	[0.07, 0.36]	.004
Court incentive (higher, more)	−.03	[−0.17, 0.12]	.722	−.03	[−0.27, 0.22]	.837	−.03	[−0.22, 0.16]	.740
Constant (b)	.72	[−0.84, 2.29]	.366	.44	[−1.89, 2.77]	.710	.20	[−1.99, 2.39]	.859

Note. Model 1: Full sample; Model 2: White participants only; Model 3: Black participants only. CI = confidence interval.

Discussion

Our hypothesis that Black MHC participants would experience less desirable MHC experiences than White participants was not supported. Contrary to intermediate outcomes predicted by focal concerns theory, Black participants reported significantly more positive life changes, higher program satisfaction, and more court incentives than White participants. Findings were in mixed support of our hypotheses that positive MHC experiences would be negatively associated with recidivism and that those associations would differ by race. Regression analyses indicated positive life changes and court sanctions were related with recidivism as predicted. However, program satisfaction was positively related (for Black participants only), and receipt of incentives was unrelated, with recidivism. Models also indicate little difference among Black and White MHC participants in terms of how their experiences were related to recidivism; however, being Black was associated with recidivism.

While Black participants were observed to have more negative experiences in the community during the study observation period (i.e., rearrests) than Whites, experiences within the MHC itself appeared more positive. This suggests that the MHCs in this study may have served as relatively safe havens in which discriminatory practices were minimized. In terms of focal concerns theory, the opportunity for discrimination arises when decision makers lack sufficient information to accurately assess threat and blameworthiness of an individual and instead rely on (often) negative stereotypes to fill this information void (Steffensmeier et al., 1998). It may be that the intensive supervision and treatment monitoring provided through the multidisciplinary MHC team provides MHC professionals sufficient and accurate information on which to base their assessments and that this knowledge obviates the need to rely on such unreliable heuristics and results in more equitable treatment. However, as MHCs are community-based service providers, they must contend with local law enforcement authorities who act without the benefit of such information, are susceptible to relying upon negative stereotypes about people of color, and who therefore arrest people of color at higher rates than Whites (Kochel et al., 2011). Thus, the ultimate effectiveness of MHCs at reducing recidivism may be hamstrung by the systemic bias present in a community’s law enforcement system.

Race, MHC Experiences, and Recidivism

Sanctions and Incentives

Overall, observations comport with the recommendation that incentives should be doled out within MHCs at a higher rate than sanctions (Blandford et al., 2015). Indeed, among the total sample, participants received on average 2.38 incentives compared with receipt of an average of 1.10 sanctions. The recommendation to emphasize incentives is based upon the empirical literature indicating that rewards are more effective at shaping behavior than punishments (e.g., Wodahl et al., 2011). However, this does not appear to be the case in the current study as receipt of incentives was not associated with arrests within 12 months of entering the MHC. This null finding may be due to the ratio of incentives to sanctions falling below the recommended level of four to one (Gendreau, 1996; Wodahl et al., 2011).

The observation that, for all participants, MHC sanctions were positively related to recidivism comports with prior research (Gendreau et al., 2000). While severity of sanctions was not examined in the current study, prior research has indicated that over-reliance on harsh sanctions in response to problematic behavior should be avoided. If severe sanctions are too often applied, either avoidance of the aversive stimulus (i.e., the MHC) or ceiling effects that diminish the sanction’s ability to shape behavior can quickly result (Marlowe & Kirby, 1999; Marlowe & Wong, 2008). Furthermore, severe sanctions run the risk of habituating problem-solving court participants to severe punishments including incarceration. That is, once accustomed to sanctions, punishment (and the threat of punishment) becomes less of a deterrent to engage in problematic behaviors (Marlowe & Kirby, 1999). Whether the relationship observed between sanctions and recidivism is the result of these conditioning effects or is simply a reflection of the MHCs under study not managing recurring criminal behaviors among participants is unclear. If the latter, then MHCs should establish processes for assessing risk for criminal behavior among their participants and provide services and supports targeted to address identified criminogenic needs as well as processes for developing strategies to remediate environmental conditions that contribute to or perpetuate these needs.

Positive Life Changes

Black participants reported more positive life changes than White participants. Experiencing more positive life changes was associated with reduced number of rearrests for both Black and White participants. This may be considered unsurprising as the positive life changes measured in this study reflect important theoretical domains in rehabilitation for people who are justice involved. For example, the survey tool measuring life changes asks participants, as a direct result of being in the MHC, whether they are getting along better with their family and whether they are doing better in school and/or work. As referenced in the “Method” section, these items reflect two of the eight central risk factors subsumed under the Risk–Needs–Responsivity (RNR) model (Bonta & Andrews, 2016). The survey instrument also asks respondents whether their MHC participation has allowed them to better control their life, to more effectively deal with daily problems and whether their mental health symptoms are less disruptive to their lives. These items can be considered to reflect the “primary human goods” of agency, pleasure, and inner peace as put forth by Ward et al. (2007) and synthesized in their Good Lives Model (GLM) of rehabilitation. This model suggests the achievement of these primary goods will accompany reduction in criminal behaviors. That life changes reflective of both models were related to rearrests may explain why other indicators of criminogenic need (i.e., educational, employment, and marital status; drug use) were observed to be unrelated to recidivism.

While these models have sometimes been viewed as at-odds with each other (Andrews et al., 2011; Ward et al., 2012), research in community corrections finds support for combining components of both the GLM and RNR model (Matejkowski et al., 2015). Furthermore, this observation suggests that MHC participants could benefit from courts establishing methods to systematically implement combinations of these two approaches. However, these hybrid implementations should accompany evaluation as to their effectiveness at achieving desired outcomes.

Program Satisfaction

Black MHC participants reported higher levels of program satisfaction than White participants. Program satisfaction was positively related to rearrests among Black participants. Previous research has shown a positive relationship between consumer satisfaction with behavioral health services and outcomes and that this relationship is mediated frequently by increased use of services (for a review, see M. J. Carlson & Gabriel, 2001). To gain a better understanding of service engagement in the MHC context, future research is needed to examine whether these relationships are maintained and to what extent they are moderated by race among MHC participants.

Higher program satisfaction among Black participants may be the result of this group receiving relatively high incentives through the court. Black participants reported receiving on average 2.74 incentives during a 6-month period compared with 2.11 among White participants (p < .001). Indeed, post hoc analysis revealed a positive correlation between incentives received and program satisfaction (r = .42; p < .001). As such, MHCs that aim to improve satisfaction among the clients they serve should consider more frequent use of incentives to reinforce desired client behaviors.

The positive relationship between program satisfaction and recidivism observed among Black participants was unexpected. If an individual is satisfied with being a member of an MHC, it follows that they would avoid behaviors that may result in their expulsion from the court (i.e., behaviors that result in rearrest). It could be that this relationship is simply a measurement artifact. Satisfaction was measured at 6 months following entry to the MHC whereas arrest data were collected a full 6 months longer, for a total of 12 months following program entry. Absent precise data on date of arrest, it is not possible to determine whether arrest occurred prior to or following administration of the program satisfaction survey.

The relationship between satisfaction and arrest may also arise from arrest events leading to more opportunities to interact with program staff. Several of the items used to assess program satisfaction focus on how staff treat MHC clients: “Staff at the mental health court encouraged me to take responsibility for how I live my life,” “Staff at the mental health court believed that I could grow, change and recover.” In a problem-solving court context, staff are likely to spend time with clients who are facing challenges in the community (such as being arrested), to find methods to overcome these challenges. The staff attention and support that may follow from arrest could be contributing to the positive relationship observed between arrest and program satisfaction among Black participants in the current study. For example, two recent studies highlight the importance of staff and social support for MHC participants. Participants in a Sacramento MHC (Yuan & Capriotti, 2019) identified benefiting from “drop-in activities” (p. 463) provided at their local mental health agency and further suggested that the caseloads of their MHC caseworkers be reduced so that participants could spend more time with them. Other qualitative analyses with MHC participants reported on by Canada and Gunn (2013) suggest that social support from MHC staff and peers in treatment was an important enabler of behavior change toward recovery. If our observation of the relationship between satisfaction and recidivism is due to underlying social support, then MHCs that aim to increase program satisfaction and social support for their clients should consider increasing opportunities for staff–client interactions, at times other than to respond to problematic behaviors. Future research should also explore the effects on post-program recidivism of “alumni groups,” post-graduation continuing care and support plans, and other efforts to keep MHC graduates linked with supportive others.

Potential Differential Treatment in the Community

Race remained a significant predictor of arrest in the presence of covariates for recidivism. Therefore, findings of increased justice involvement among Black participants can also be reflective of research that has shown people of color to be at increased risk for surveillance in the community by law enforcement and for punitive action by the courts. For example, Black drivers (Epp at al., 2013; Smith & Petrocelli, 2001) and pedestrians (Gelman et al., 2007; Goel et al., 2016) are more likely to be stopped and questioned by police officers. However, whether this disproportionate frequency of stops results in increased arrest of people of color has been brought into question (Tillyer & Engel, 2013). Nevertheless, research reported in the introductory literature review suggests that a portion of the higher-than-average number of arrests among Black participants may be due simply to discrimination by law enforcement. It should be noted some important covariates were not available for testing and given the amount of variation in rearrest explained by our model (pseudo-R² = .078), differential treatment cannot be confidently assumed. Thus, future research examining racial/ethnic disparities in rearrest among MHC participants should include a more robust set of theoretically relevant as well as empirically informed individual and structural predictors of crime.

Implications for Policy and Practice

The reliance of “street level bureaucrats” (Lipsky, 2010, p. 13) on racial/ethnic stereotypes to inform their discretionary decision making hypothesized by focal concerns theory suggests that training of these professionals can reduce discriminatory practices. For example, to avoid over-policing of racially diverse communities, implicit bias training for police officers is becoming more common (James, 2018). While little research has been conducted to evaluate the impact of such trainings on the behaviors of law enforcement, there appears to be potential for such training to impart recognition among participants of the importance of neutral, fair, and unbiased policing (James, 2018; Madon et al., 2017). Implicit bias training could therefore also be applied in MHC settings to promote equitable application of incentives and sanctions, which, as this study demonstrates, are related to participant satisfaction and program efficacy.

Given that MHCs are situated in the community and the success of these programs are based in no small part on the decisions of local law enforcement to arrest and incarcerate program participants, MHCs should develop and maintain relationships with law enforcement authorities that support the aim of reduced incarceration. For example, given appropriate information-sharing authorizations, cooperative relationships among MHCs and local law enforcement that provide the latter a more accurate picture of MHC clients could (as focal concerns theory would predict) reduce law enforcement’s reliance on racialized heuristics when making arrest decisions. Such arrangements could support appropriate and racially equitable diversions to crisis centers, rather than jail, of those MHC participants who come to the attention of law enforcement due to complications arising from a psychiatric crisis (Watson et al., 2010).

As communities of color tend to be over-policed (Austin et al., 2016) and many MHCs provide housing services (Felton, 2015), an aim to increase integration of residential accommodations among MHC participants should be considered. Scattered-site supported housing (as opposed to clustered housing in low-income neighborhoods) could reduce the odds of legal entanglements among MHC participants by reducing unnecessary surveillance and potential for arrest and incarceration. In addition, scattered-site supported housing could enhance community integration and expand community supports for this population (Quilgars & Pleace, 2016).

Given the findings on the relationship between program satisfaction (and the likelihood of this variable serving as a proxy for social support; see “Program Satisfaction” section) and recidivism, developing social supports in the community is critical for promoting and sustaining gains with MHC participants (Canada & Gunn, 2013). These efforts should be an integral component of MHCs, wherein various formal and informal prosocial supports are fostered. Alumni programs, often used in drug courts (Huddleston, 2005), may be one useful component of these efforts and may serve to prevent criminal behaviors among MHC participants and enhance their quality of life in the community.

Limitations and Conclusion

Shortcomings of the MacArthur MHC study, and therefore of the current study, have been identified elsewhere (e.g., Han & Redlich, 2015; Redlich & Han, 2014) and include a caveat regarding generalization of findings to all MHCs in the United States. The data set also did not allow for analysis of racial and ethnic groups, other than Black participants, who may also experience MHCs differently from Whites. In addition, the absence of data on important treatment, community, and demographic factors to control for in multivariable models must be acknowledged. Although the study included objective arrest records from the Federal Bureau of Investigation and county agencies, not all criminal behavior would be captured through a documented arrest. As such, these administrative records may under-identify participants’ criminal involvement. Also, the 6-month follow-up survey precludes capture of the full range of the participants’ MHC experiences. Analyses would have been benefited from qualitative interviews that incorporated the voice of participants and their perceptions of fairness, procedural justice, and quality of incentives and severity of sanctions experienced in MHC.

Despite these limitations, the study provides novel insights into the workings of MHCs. It is the first-known test of racially based differential treatment within MHCs and of how these experiences are related to recidivism. Findings support prior research identifying the limitations of sanctions to improve behavior. While the positive relationship between receipt of sanctions and recidivism observed in the current study may reflect MHC management practices aimed at more criminally oriented clients, the effectiveness of sanctions to promote desired behaviors appears to be a dubious endeavor with MHC clients. Findings also signal the importance of addressing important theoretically based and empirically supported needs domains encompassed in the study’s “life changes” instrument. This instrument assessed areas of improvement that are congruent with those identified by the GLM and RNR models of rehabilitation as critical to reducing criminal behavior. As such, MHCs should establish policies and procedures for assessing these needs domains and providing services in areas where needs are identified.

Finally, MHCs should consider the real possibility that heightened arrest rates among some of their members may simply be based on skin tone. Increased supervision of racially diverse communities and disparate treatment of persons of color by law enforcement is not uncommon. Black MHC participants reside within these communities and are likely to experience such treatment. As such, thorough consideration by the MHC team of the circumstances surrounding the arrest of MHC participants who are people of color should be conducted prior to their basing any additional sanctions or program dismissal on that arrest. Collaborations among MHCs and local law enforcement may help to reduce such events from occurring in the first place.

Footnotes

ORCID iDs

Jason Matejkowski

Sungkyu Lee

Woojae Han is an assistant professor in the School of Social Welfare at Soongsil University. His research focuses on alternative criminal justice system and community rehabilitation for offenders with mental illness.

Jason Matejkowski is an associate professor of social work in the School of Social Welfare at the University of Kansas. His research focuses on policies and services involving adults with mental illness who are involved with the criminal justice system or who are homeless.

Sungkyu Lee is an associate professor in the School of Social Welfare at Soongsil University. His research focuses on mental health service utilization and community integration of individuals with mental illness who are involved with the criminal justice system.

References

Alexander

(2012). The new Jim Crow: Mass incarceration in the age of colorblindness. The New Press.

Allison

(2010). Fixed effects regression methods in SAS. Statistics and data analysis. http://www2.sas.com/proceedings/sugi31/184-31.pdf

Andrews

D. A.

Bonta

Wormith

J. S.

(2011). The Risk-Need-Responsivity (RNR) model: Does adding the Good Lives Model contribute to effective crime prevention? Criminal Justice and Behavior, 38, 735–755. https://doi.org/10.1177/0093854811406356

Austin

Eisen

Cullen

Frank

(2016). How many Americans are unnecessarily incarcerated? Brennan Center for Justice. http://www.brennancenter.org/sites/default/files/publications/Unnecessarily_Incarcerated.pdf

Bailey

C. A.

Galicia

B. E.

Salinas

K. Z.

Briones

Hugo

Hunter

Venta

A. C.

(2019). Racial/ethnic and gender disparities in anger management therapy as a probation condition. Law and Human Behavior, 44(1), 88–96.

Blandford

A. M.

Fader-Towe

Ferreira

Greene

(2015). Developing a mental health court: An interdisciplinary curriculum—Handbook for facilitators. Council of State Governments, Justice Center. https://www.courtinnovation.org/sites/default/files/media/document/2018/Developing%20a%20Mental%20Health%20Court_Handbook%20for%20Curriculum%20Facilitators.pdf

Bonilla-Silva

(1997). Rethinking racism: Toward a structural interpretation. American Sociological Review, 62, 465–480. https://doi.org/10.2307/2657316

Bonta

Andrews

D. A.

(2016). The psychology of criminal conduct. Taylor & Francis.

Brewer

R. M.

Heitzeg

N. A.

(2008). The racialization of crime and punishment: Criminal justice, color-blind racism, and the political economy of the prison industrial complex. American Behavioral Scientist, 51(5), 625–644. https://doi.org/10.1177/0002764207307745

10.

Callahan

Steadman

H. J.

Tillman

Vesselinov

(2013). A multi-site study of the use of sanctions and incentives in mental health courts. Law and Human Behavior, 37, 1–9. https://doi.org/10.1037/h0093989

11.

Canada

K. E.

Gunn

A. J.

(2013). What factors work in mental health court? A consumer perspective. Journal of Offender Rehabilitation, 52, 311–337. https://doi.org/10.1080/10509674.2013.801387

12.

Canada

K. E.

Ray

(2016). Mental health court participants’ perspectives of success: What key outcomes are we missing? International Journal of Forensic Mental Health, 15, 352–361. https://doi.org/10.1080/14999013.2016.1230155

13.

Canada

K. E.

Watson

A. C.

(2013). “Cause everybody likes to be treated good”: Perceptions of procedural justice among mental health court participants. American Behavioral Scientist, 57, 209–230. https://doi.org/10.1177/0002764212465415

14.

Carlson

J. A.

Sarkin

A. J.

Levack

A. E.

Sklar

Tally

S. R.

Gilmer

T. P.

Groessl

E. J.

(2011). Evaluating a measure of social health derived from two mental health recovery measures: The California Quality of Life (CA-QOL) and Mental Health Statistics Improvement Program Consumer Survey (MHSIP). Community Mental Health Journal, 47, 454–462. https://doi.org/10.1007/s10597-010-9347-8

15.

Carlson

M. J.

Gabriel

R. M.

(2001). Patient satisfaction, use of services, and one-year outcomes in publicly funded substance abuse treatment. Psychiatric Services, 52, 1230–1236. https://doi.org/10.1176/appi.ps.52.9.1230

16.

Cohen

(1988). Statistical power analysis for the behavioral sciences. Lawrence Erlbaum.

17.

Conrad

K. J.

Yagelka

J. R.

Matters

M. D.

Rich

A. R.

Williams

Buchanan

(2001). Reliability and validity of a modified Colorado Symptom Index in a national homeless sample. Mental Health Services Research, 3, 141–153. https://doi.org/10.1023/A:1011571531303

18.

Cosden

Ellens

Schnell

Yamini-Diouf

(2005). Efficacy of a mental health treatment court with Assertive Community Treatment. Behavioral Sciences & the Law, 23, 199–214. https://doi.org/10.1002/bsl.638

19.

Dirks-Linhorst

P. A.

Linhorst

D. M.

(2012). Recidivism outcomes for suburban mental health court defendants. American Journal of Criminal Justice, 37, 76–91. https://doi.org/10.1007/s12103-010-9092-0

20.

Dollar

C. B.

Ray

Hudson

M. K.

Hood

B. J.

(2018). Examining changes in procedural justice and their influence on problem-solving court outcomes. Behavioral Sciences & the Law, 36, 32–45. https://doi.org/10.1002/bsl.2329

21.

Draine

Blank

Kottsieper

Solomon

(2005). Contrasting jail diversion and in-jail services for mental illness and substance abuse: Do they serve the same clients? Behavioral Sciences & the Law, 23(2), 171–181. https://doi.org/10.1002/bsl.637

22.

Epp

Maynard-Moody

Haider-Markel

D. P.

(2013). Pulled over: How police stops define race and citizenship. University of Chicago Press.

23.

Eschbach

L. A.

Dalgin

R. S.

Pantucci

(2019). A three stage model for mental health treatment court: A qualitative analysis of graduates’ perspectives. Community Mental Health Journal, 55, 590–598. https://doi.org/10.1007/s10597-018-0346-5

24.

Felton

(2015). Stable housing is key to mental health court success. Community Housing Network. https://communityhousingnetwork.org/stable-housing-is-key-to-mental-health-court-success/

25.

Fox

(1991). Regression diagnostics. SAGE.

26.

Gelman

Fagan

Kiss

. (2007). An analysis of the New York City Police Department’s “Stop-and-Frisk” policy in the context of claims of racial bias. Journal of the American Statistical Association, 102(479), 813–823. https://doi.org/10.1198/016214506000001040

27.

Gendreau

(1996). The principles of effective intervention with offenders. In Harland

(Ed.), Choosing correctional options that work (pp. 117–130). SAGE.

28.

Gendreau

Goggin

Cullen

F. T.

Andrews

D. A.

(2000). The effects of community sanctions and incarceration on recidivism. Forum on Corrections Research, 12, 10–13.

29.

Goel

Rao

J. M.

Shroff

(2016). Precinct or prejudice? Understanding racial disparities in New York City’s stop-and-frisk policy. The Annals of Applied Statistics, 10, 365–394. https://doi.org/10.1214/15-AOAS897

30.

Han

(2016). Impact of community treatment and neighborhood disadvantage on recidivism in mental health courts. University at Albany, State University of New York.

31.

Han

(2019). Life changes matter more than satisfaction or sanctions/incentives: An examination of mental health court experience factors associated with arrest. International Journal of Forensic Mental Health, 18, 376–388. https://doi.org/10.1080/14999013.2019.1588434

32.

Han

Redlich

A. D.

(2015). The impact of community treatment on recidivism among mental health court participants. Psychiatric Services, 67, 384–390. https://doi.org/10.1176/appi.ps.201500006

33.

Hiday

V. A.

Ray

(2010). Arrests two years after exiting a well-established mental health court. Psychiatric Services, 61, 463–468. https://doi.org/10.1176/appi.ps.61.5.463

34.

Hilbe

J. M.

(2011). Negative binomial regression. Cambridge University Press.

35.

Honegger

L. N.

(2015). Does the evidence support the case for mental health courts? A review of the literature. Law and Human Behavior, 39, 478–488. https://doi.org/10.1037/lhb0000141

36.

Hox

J. J.

Moerbeek

Schoot

(2017). Multilevel analysis: Techniques and applications. Routledge.

37.

Huddleston

(2005). Drug courts: An effective strategy for communities facing methamphetamine. U.S. Department of Justice.

38.

James

(2018). The stability of implicit racial bias in police officers. Police Quarterly, 21, 30–52. https://doi.org/10.1177/1098611117732974

39.

Jerrell

J. M.

(2006). Psychometrics of the MHSIP Adult Consumer Survey. The Journal of Behavioral Health Services & Research, 33, 483–488. https://doi.org/10.1007/s11414-006-9033-3

40.

Kochel

T. R.

Wilson

D. B.

Mastrofski

S. D.

(2011). Effect of suspect race on officers’ arrest decisions. Criminology, 49, 473–512. https://doi.org/10.1111/j.1745-9125.2011.00230.x

41.

(2013). Little’s test of missing completely at random. The Stata Journal, 13, 795–809. https://doi.org/10.1177%2F1536867X1301300407

42.

Lin

Grattet

Petersilia

(2010). “Back-end sentencing” and reimprisonment: Individual, organizational, and community predictors of parole sanctioning decisions. Criminology, 48, 759–795. https://doi.org/10.1111/j.1745-9125.2010.00201.x

43.

Linhorst

D. M.

Dirks-Linhorst

P. A.

Stiffelman

Gianino

Bernsen

H. L.

Kelley

B. J.

(2010). Implementing the essential elements of a mental health court: The experiences of a large multijurisdictional suburban county. The Journal of Behavioral Health Services & Research, 37, 427–442. https://doi.org/10.1007/s11414-009-9193-z

44.

Lipsky

(2010). Street-level bureaucracy: Dilemmas of the individual in public service. Russell Sage.

45.

Lowder

E. M.

Rade

C. B.

Desmarais

S. L.

(2018). Effectiveness of mental health courts in reducing recidivism: A meta-analysis. Psychiatric Services, 69, 15–22. https://doi.org/10.1176/appi.ps.201700107

46.

Madon

N. S.

Murphy

Sargeant

(2017). Promoting police legitimacy among disengaged minority groups: Does procedural justice matter more? Criminology & Criminal Justice, 17, 624–642. https://doi.org/10.1177%2F1748895817692849

47.

Marlowe

Kirby

(1999). Effective use of sanctions in drug courts: Lessons from behavioral research. National Drug Court Institute Review, 2, 1–31.

48.

Marlowe

Wong

C. J.

(2008). Contingency management in adult criminal drug courts. In Higgins

S. T.

Silverman

Heil

S. H.

(Eds.), Contingency management in substance abuse treatment (pp. 334–354). Guilford Press.

49.

Marlowe

D. B.

DeMateeo

D. S.

Festinger

D. S.

(2003). A sober assessment of drug courts. Federal Sentencing Reporter, 16, 153–157. https://doi.org/10.1525/fsr.2003.16.2.153

50.

Matejkowski

Conrad

(2019). Minor risk factors. In Morgan

(Ed.), The SAGE encyclopedia of criminal psychology (pp. 896–898). SAGE.

51.

Matejkowski

Conrad

Ostermann

(2017). Does early onset of criminal behavior differentiate for whom serious mental illness has a direct or indirect effect on recidivism? Law and Human Behavior, 41, 68–79. https://doi.org/10.1037/lhb0000231

52.

Matejkowski

Ostermann

(2015). Serious mental illness, criminal risk, parole supervision, and recidivism: Testing of conditional effects. Law and Human Behavior, 39, 75–86. https://doi.org/10.1037/lhb0000094

53.

Matejkowski

Severson

M. E.

Manthey

T. J.

(2015). Strategies for postrelease supervision of individuals with serious mental illness: Comparing specialized community corrections officers to those not serving on a specialized team. Journal of Offender Rehabilitation, 54, 520–537. https://doi.org/10.1080/10509674.2015.1076106

54.

McAlpine

D. D.

Mechanic

(2000). Utilization of specialty mental health care among persons with severe mental illness: The roles of demographics, need, insurance, and risk. Health Services Research, 35, 277–292.

55.

McNiel

D. E.

Binder

R. L.

(2007). Effectiveness of a mental health court in reducing criminal recidivism and violence. American Journal of Psychiatry, 164, 1395–1403. https://doi.org/10.1176/appi.ajp.2007.06101664

56.

Munetz

M. R.

Ritter

Teller

J. L. S.

Bonfine

(2014). Mental health court and assisted outpatient treatment: Perceived coercion, procedural justice, and program impact. Psychiatric Services, 65(3), 352–358. https://doi.org/10.1176/appi.ps.002642012

57.

National Institute of Justice. (n.d.). Adult mental health courts. https://www.crimesolutions.gov/PracticeDetails.aspx?ID=34

58.

Olson

J. C

. (2016). Race and punishment in American prisons. Journal of Public Administration Research and Theory, 26(4), 758–768. https://doi.org/10.1093/jopart/muw026

59.

Poythress

N. G.

Petrila

McGaha

Boothroyd

(2002). Perceived coercion and procedural justice in the Broward mental health court. International Journal of Law and Psychiatry, 25, 517–533. https://doi.org/10.1016/S0160-2527(01)00110-8

60.

Prins

S. J.

(2014). Prevalence of mental illnesses in U.S. state prisons: A systematic review. Psychiatric Services, 65, 826–872. https://doi.org/10.1176/appi.ps.201300166

61.

Quilgars

Pleace

(2016). Housing First and social integration: A realistic aim? Social Inclusion, 4(4), 5–15. https://doi.org/10.17645/si.v4i4.672

62.

Redlich

A. D.

Han

(2014). Examining the links between therapeutic jurisprudence and mental health court completion. Law and Human Behavior, 38, 109–118. https://doi.org/10.1037/lhb0000041

63.

Redlich

A. D.

Steadman

H. J.

Callahan

Robbins

P. C.

Vessilinov

Ozdogru

A. A.

(2010). The use of mental health court appearances in supervision. International Journal of Law and Psychiatry, 33, 272–277. https://doi.org/10.1016/j.ijlp.2010.06.010

64.

Redlich

A. D.

Steadman

H. J.

Monahan

Robbins

P. C.

Petrila

(2006). Patterns of practice in mental health courts: A national survey. Law and Human Behavior, 30, 347–362. https://doi.org/10.1007/s10979-006-9036-x

65.

Sarteschi

C. M.

Vaughn

M. G.

Kim

(2011). Assessing the effectiveness of mental health courts: A quantitative review. Journal of Criminal Justice, 39, 12–20. https://doi.org/10.1016/j.jcrimjus.2010.11.003

66.

Smith

M. R.

Petrocelli

(2001). Racial profiling? A multivariate analysis of police traffic stop data. Police Quarterly, 4, 4–27. https://doi.org/10.1177/1098611101004001001

67.

Steadman

H. J.

Redlich

A. D.

Griffin

Petrila

Monahan

(2005). From referral to disposition: Case processing in seven mental health courts. Behavioral Sciences & the Law, 23, 215–226. https://doi.org/10.1002/bsl.641

68.

Steen

Engen

R. L.

Gainey

R. R.

(2005). Images of danger and culpability: Racial stereotyping, case processing, and criminal sentencing. Criminology, 43, 435–468. https://doi.org/10.1111/j.0011-1348.2005.00013.x

69.

Stefan

Winick

B. J.

(2005). A dialogue on mental health courts. Psychology, Public Policy, and Law, 11, 507–526. https://doi.org/10.1037/1076-8971.11.4.507

70.

Steffensmeier

Ulmer

Kramer

(1998). The interaction of race, gender, and age in criminal sentencing: The punishment cost of being young, Black, and male. Criminology, 36, 763–798. https://doi.org/10.1111/j.1745-9125.1998.tb01265.x

71.

Thompson

Osher

Tomasini-Joshi

(2007). Improving responses to people with mental illnesses: The essential elements of a mental health court. Justice Center, The Council of State Governments. https://www.bja.gov/Publications/MHC_Essential_Elements.pdf

72.

Tillyer

Engel

R. S.

(2013). The impact of drivers’ race, gender, and age during traffic stops: Assessing interaction terms and the social conditioning model. Crime & Delinquency, 59, 369–395. https://doi.org/10.1177/0011128710389583

73.

Tonry

(2010). The social, psychological, and political causes of racial disparities in the American criminal justice system. Crime and Justice, 39, 273–312. https://doi.org/10.1086/653045

74.

Torrey

E. F.

Lamb

H. R.

Eslinger

D. F.

Biasotti

M. C.

Fuller

D. A.

(2014). The treatment of persons with mental illness in prisons and jails: A state survey. http://www.treatmentadvocacycenter.org/storage/documents/treatment-behind-bars/treatment-behind-bars.pdf

75.

Wales

H. W.

Hiday

V. A.

Ray

(2010). Procedural justice and the mental health court judge’s role in reducing recidivism. International Journal of Law and Psychiatry, 33, 265–271. https://doi.org/10.1016/j.ijlp.2010.06.009

76.

Ward

Mann

R. E.

Gannon

T. A.

(2007). The Good Lives Model of offender rehabilitation: Clinical implications. Aggression and Violent Behavior, 12, 87–107. https://doi.org/10.1016/j.ijlp.2010.06.009

77.

Ward

Maruna

(2007). Rehabilitation. Routledge.

78.

Ward

Yates

P. M.

Willis

G. M.

(2012). The Good Lives Model and the Risk Need Responsivity model: A critical response to Andrews, Bonta, and Wormith (2011). Criminal Justice and Behavior, 39, 94–110. https://doi.org/10.1177/0093854811426085

79.

Watson

A. C.

Ottati

V. C.

Morabito

Draine

Kerr

A. N.

Angell

(2010). Outcomes of police contacts with persons with mental illness: The impact of CIT. Administration and Policy in Mental Health and Mental Health Services Research, 37, 302–317. https://doi.org/10.1007/s10488-009-0236-9

80.

Wilson

D. B.

Mitchell

MacKenzie

D. L.

(2006). A systematic review of drug court effects on recidivism. Journal of Experimental Criminology, 2, 459–487. https://doi.org/10.1007/s11292-006-9019-4

81.

Wodahl

E. J.

Garland

Culhane

S. E.

McCarty

W. P.

(2011). Utilizing behavioral interventions to improve supervision outcomes in community-based corrections. Criminal Justice and Behavior, 38, 386–405. https://doi.org/10.1177/0093854810397866

82.

Wolff

Pogorzelski

(2005). Measuring the effectiveness of mental health courts: Challenges and recommendations. Psychology, Public Policy and Law, 11, 539–605. https://doi.org/10.1037/1076-8971.11.4.539

83.

Wolfgang

M. E.

Figlio

R. M.

Tracy

P. E.

Singer

S. I.

(1985). The national survey of crime severity. Bureau of Justice Statistics, U.S. Department of Justice. https://www.ncjrs.gov/App/Publications/abstract.aspx?ID=96017

84.

Yuan

Capriotti

M. R

. (2019). The impact of mental health court: A Sacramento case study. Behavioral Sciences & the Law, 37, 452-467. https://doi.org/10.1002/bsl.2421