Misdemeanor Arrestees With Mental Health Needs: Diversion and Outpatient Services as a Recidivism Reduction Strategy

Abstract

Individuals with mental illnesses who are arrested for criminal activity cycle between criminal justice and mental health systems at disproportionately high rates. Studying recidivism of this population has been difficult due to separate system data bases. This study compared recidivism outcomes of 102 adults with mental illness who were arrested for a misdemeanor offense. One group had a diagnosed mental illness (n = 58) and the other group was diagnosed with co-occurring mental health and substance abuse disorders (n = 44). As a condition of their personal recognizance bond, both groups voluntarily agreed to stabilize on medication and report to community-based outpatient mental health clinic. Participants in both groups had fewer rearrests and fewer days in jail in the 12 months following discharge from diversion relative to the 12 months prior to diversion participation. Outpatient mental health service utilization following 24 hr in jail seems to be a viable means of reducing recidivism among accused misdemeanant defendants.

Keywords

recidivism mental illness dual diagnosis jail diversion mental health services

Introduction

Over the past four decades, criminologists have tracked the steady increase in the rates of offenders with mental illness in the criminal justice system. Some researchers have concluded that the magnitude that mental health deinstitutionalization has had on correctional facility growth remains empirically uncertain at best (Kim, 2016). However, various historical and theoretical explanations suggest that the criminalization of mentally ill individuals, particularly those individuals with substance abuse problems, have played a significant role (Fisher, Silver, & Wolff, 2006; Ringhoff, Rapp, & Robst, 2012; Slate, Buffington-Vollum, & Johnson, 2013). Many mental health institutions have narrowed the type of clients eligible for treatment, such that some institutions receive funding to treat major psychotic and affective disorders only. Significantly less mental health institutions offer services to treat dual diagnosis or co-occurring mental health and substance use disorders, and most do not accept persons with criminal records. As a result, the criminal justice system intervenes in the lives of a sizable number of mentally ill individuals, with and without substance abuse problems.

Jails are not necessarily the most therapeutic environments for stabilization, and many jails lack adequate mental health treatment or support (Gonzales & Davak, 2006). Jail environments have been shown to exacerbate mental health symptoms (Cloyes, Wong, Latimer, & Abarca, 2010). The totality of the situation demonstrates a substantial need for innovative methods in dealing with individuals with mental illness and co-occurring substance abuse disorders who have been arrested for less serious offenses.

Offering alternatives to incarceration has been explored in some jurisdictions that include substance abuse treatment, involuntary outpatient commitment programs (Center for Health Care Services, 2006; Grantham, 2011), specialized probation, mental health court, and/or a residential treatment facility (Castillo & Alarid, 2011). While these services have made a difference with convicted offenders, there remains a substantial gap in pre-adjudication services for mentally ill defendants accused of less serious offenses.

Partnerships have formed between mental health organizations and local jail detention facilities to offer outpatient treatment and supervision services for clients shared by both systems. Within this structure, defendants must appear at the next court date for the pending case and not commit any new crimes. In the justice system, pre-adjudication options for defendants include short-term detention without criminal charges (White, Goldkamp, & Campbell, 2006), mental health court, deferred probation supervision, and assertive community treatment after release from county jail (McCoy, Roberts, Hanrahan, Clay, & Luchins, 2004). However, once a mentally ill defendant is facing criminal charges, many defendants remain in jail until case disposition to stabilize on medication before starting supervision.

In the mental health system, community-based services range from community mental health centers to outpatient clinics to provide services for persons with mental illnesses without the need for jail incarceration or commitment to a state hospital. Outpatient clinics provide treatment services for indigent persons with mental illnesses who are at-risk of further criminal justice involvement. We argue that the goals of the mental health system and the criminal justice system are interconnected, especially with regard to medication compliance and recidivism. This study examined compliance and recidivism for defendants who were arrested for a misdemeanor offense, and given the choice, as a condition of their personal recognizance bond, to stabilize on medication through an outpatient mental health clinic rather than stabilizing in jail. Defendants with a diagnosed mental illness were compared with dual diagnosed defendants who had co-occurring mental illness and a substance abuse problem.

Offender Recidivism With Mental Health Diagnosis Compared With Co-Occurring Mental Health/Substance Use

Risk of recidivism, or return to criminal behavior, has been extensively studied in criminal justice. Among a multitude of factors considered, previous criminal history was one of the best predictors of recidivism (Bonta, Law, & Hanson, 1998), followed by procriminal attitudes, associating with criminal peers, presence of dysfunctional relationships, employment/educational challenges, and substance abuse (Canales, Campbell, Wei, & Totten, 2014). With the exception of antisocial personality disorder, the presence of a major mental health disorder, by itself, was not a recidivism risk factor, when a meta-analysis was conducted on studies between 1959 and 1995 comparing offenders with and without a diagnosed mental illness (Bonta et al., 1998). However, the presence of other risk factors such as alcohol and drug abuse puts mentally disordered offenders at higher risk for recidivism (Bonta et al., 1998) and for treatment noncompliance (Herbeck, Fitek, Svikis, Montoya, Marcus, & West, 2005).

More recent studies that have directly compared recidivism rates of mentally ill offenders with offenders with co-occurring disorders have all found that the co-occurring disordered offenders had higher recidivism rates or were more likely to be rearrested earlier (Castillo & Alarid, 2011; Hartwell, 2004; McReynolds, Schwalbe, & Wasserman, 2010; Modestin & Wuermle, 2005; Swanson, Holzer, Ganju, & Jono, 1990; Wallace, Mullen, & Burgess, 2004; White et al., 2006). For example, data were collected from over 300 mentally ill offenders who were sentenced to community supervision. Offenders with a dual diagnosis of mental illness/alcoholism were more likely to recidivate earlier and to be rearrested for a violent offense than mentally ill offenders without an alcohol problem (Castillo & Alarid, 2011).

In sum, when the recidivism of mentally ill individuals were compared with individuals with co-occurring disorders, the studies sampled from convicted offenders previously housed in psychiatric hospitals or correctional facilities. The only exception to this was the Castillo and Alarid (2011) study that chose offenders in community correctional programs managed by adult probation. Studies that directly assess recidivism in diversionary programs are examined below.

Jail Diversion Programs for Mentally Ill Offenders

Jail diversion programs for persons with mental illnesses began in the early 1990s and have since expanded to 560 programs in nearly every state in the United States (Case, Steadman, Dupuis, & Morris, 2009). Jail diversion can be either pre- or post-booking programs. Pre-booking programs allow police officers to divert offenders with mental illnesses in lieu of making an arrest. Pre-booking diversion typically does not result in any charges filed. Post-booking diversion programs identify and release appropriate individuals after they have been arrested, but they vary in the time defendants spend in jail before release. Given the intent of the police to file a criminal charge, the goal of post-booking programs are to screen and divert eligible defendants as soon as possible from jail until their next court date. Defendants in the post-booking program had generally more serious histories of crime, more serious use of alcohol and drugs, and were less able to socially function than defendants in pre-booking programs (Lattimore, Broner, Sherman, Frisman, & Shafer, 2003).

A literature review of diversion programs for persons with mental illnesses under the criminal justice system yielded program descriptions (Hartford, Carey, & Mendonca, 2006; Ryan, Brown, & Watanabe-Galloway, 2010; Steadman et al., 2001), and a study that measured staff perceptions of pre-trial programs for individuals with mental illness across North Carolina (Tanner, Wyatt, & Yearwood, 2008). The number of offenses and rearrests were reduced while on supervision, and program staff believed that defendants benefited more from these services over traditional bail procedures. There was consensus that diversion improved access to services, reduced jail time, and helped the defendants appear on court dates which further reduced issuing warrants for failure to appear (Tanner et al., 2008).

One large multisite evaluation was funded in the late 1990s by the U.S. Department of Health and Human Services, Substance Abuse and Mental Health Services Administration (SAMHSA). Three pre-booking sites were selected in Memphis, TN; Montgomery County, PA; and Multnomah County, OR. Post-booking sites included Phoenix, AZ; Connecticut, and Lane County, OR (Broner, Lattimore, Cowell, & Schlenger, 2004; Case et al., 2009; Lattimore et al., 2003; Shafer, Arthur, & Franczak, 2004; Steadman & Naples, 2005). Each study site shared a common quasi-experimental, longitudinal panel study design which consisted of non-random assignment of participants to diversion or no jail diversion. Although similar eligibility criteria was used, comparison groups were quite different than the diverted group at many sites. Data were collected at baseline, 3 months, and 12 months.

Steadman and Naples (2005) found that individuals who accessed community-based mental health services spent less time in jail without posing a threat to the safety or welfare of the general population in the community. Diverted participants experienced less arrests 1 year after program completion than when compared with the year before program intake (Steadman & Naples, 2005). However, in general, there were no differences in the number or incidence of arrest with diverted compared with non-diverted defendants at most sites. However, the odds of rearrest after 12 months increased for diverted participants in Oregon. The validity of these results are limited not only because of the non-equivalent treatment and control groups, but the “relatively small differences in treatment received by the diverted and non-diverted groups” (Broner et al., 2004, p. 537).

Other than the SAMHSA multisite study which began nearly 20 years ago, only one other empirical evaluation of diversion for mentally disordered offenders could be located. The study found the rate of recidivism for diversion to be no different than with similar individuals who completed mental health court (Boccaccini, Christy, Poythress, & Kershaw, 2005). In summary, the available diversion research is mixed as to whether diversion programs reduced future recidivism (Steadman & Naples, 2005) or whether there were few differences (Boccaccini et al., 2005; Broner et al., 2004). Not only have a small number of studies been conducted with diversion programs for mentally ill offenders, but there has been no study of diversion program outcomes comparing offenders with a major mental health disorder and offenders with a dual diagnosis of mental health and substance use disorder. This study fills that gap in the literature with recidivism outcomes and number of days spent in jail of pre-adjudicated clients, and compared them to the same client’s behavior 1 year before program enrollment. We hypothesized that clients who were involved in the mental health clinic program as a jail diversion would have reduced rearrest rates 1 year following their involvement, compared with rearrest rates prior to clinic involvement. This study also compared defendants with a diagnosed mental illness with dual diagnosed defendants with a mental illness and a substance abuse problem. We hypothesized that defendants with a diagnosed mental illness would have lower recidivism rates than dual diagnosed defendants.

Methods

Mental health service data were merged with county-level arrests and jail admission data from the same jurisdiction prior to de-identifying the data and providing to the researchers. This jurisdiction had a municipal jail which temporarily held up to 150 recently arrested persons for up to 48 hrs and a larger county jail that held over 4,000 people. Site visits were conducted by the researchers at the municipal jail, the county jail, and the mental health outpatient program, known as the Community Reintegration Program (CRP), to understand the referral process and treatment services offered.

Program Description

The CRP began in March 2010 as a post-booking diversionary outpatient treatment program for offenders with mental illness who were arrested for a misdemeanor offense within the county, but had not yet been convicted. Clients were referred to CRP primarily by the Magistrate’s Office during the pre-trial stage, but they could also be referred by police officers or the mental health court (Rubin, Alarid, & Rodriguez, 2014). Individuals were checked through a state-wide database to determine if they had previously been diagnosed with a mental illness by a licensed practitioner. If a diagnosis was not present but a mental health issue was suspected, the individual was further screened by a licensed psychiatrist using the Diagnostic and Statistical Manual of Mental Disorders (4th ed., text rev.; DSM IV-TR; American Psychiatric Association [APA], 2000), which was the most recent version available at the time. The hope was that within 24 hrs of being in municipal detention, the individual would agree to accept outpatient services. Qualifying for a “mental health personal recognizance bond” typically required that the individual lived within the county (a homeless shelter qualified), could provide names and contact information for two references, was not currently on probation or parole supervision, and was competent. Once individuals obtained a personal recognizance bond, they were transported by law enforcement personnel to the outpatient clinic on the next business day. At the CRP, the staff designed an individualized treatment plan (Rubin et al., 2014).

The CRP handled 50 to 55 clients at a time and ensured the client remained mentally stable, attended his or her next court date, and paid any court costs. The length of CRP depended on the speed of the court docket, which ranged between 3 and 12 months with an average of 155 days. The CRP staff reviewed progress of each client twice per week. The CRP staff consisted of one Program Director, one Licensed Professional Counselor who practiced cognitive-behavioral therapy, and two case managers/social workers, each with a caseload of 25 to 28 clients. Following the resolution of their misdemeanor case (cases were either dismissed or resulted in a guilty plea), clients were transferred out of the CRP program (Rubin et al., 2014).

Data

De-identified county-level service utilization data were extrapolated for all clients (n = 102) who entered the CRP program from the program’s inception in March 2010 to December 31, 2010. CRP mental health program data were matched with county-level criminal justice arrest data by program staff prior to releasing the data to the authors. Mental health program data included the program entry date, program end date, primary mental health diagnosis, secondary mental health diagnosis (if applicable), diagnosed substance abuse disorder, and Global Assessment of Function score. Four years of data were provided by the county jail that included dates of arrest, offense at arrest, and the booking and release dates each person spent in jail between March 1 2009 and April 1, 2013. Using each client’s CRP start and end date, the arrest data, number of offenses, and time spent in jail were calculated during two different time periods for that individual client: 12 months before the client entered CRP, and 12 months following termination from CRP. These time periods were rigidly adhered so that a pre–post comparison design could be used with the study participants. Arrests and jail time outside of these time frames were not included in the final data set so that each data collection period was consistent regardless of when the client entered the CRP program and how long they spent in the program. The number of days each client spent in CRP was measured by calculating the days between the start and ending dates.

Two outcome variables included program compliance and rearrest. Program compliance was measured as “compliant” (1) or “non-compliant” (0). Compliance was measured at the point when a client successfully terminated from the pre-trial program to a lower level of service. Program compliance meant that a client agreed to take medication, attend all court dates, and/or fulfill any counseling requirements. A non-compliant client was terminated unsuccessfully if one or more of the following occurred: the client did not show up for the next court date, refused medication or treatment services, and/or was rearrested for a new crime during the pre-trial period. Rearrest was measured by tracking each client for up to 12 months after CRP program completion for the presence (1) or an absence (0) of a new arrest.

In addition to these two dichotomous dependent variables, the number of days spent in jail 12 months before CRP were compared with 12 months after CRP ended, and whether the CRP program reduced the likelihood of rearrest 1 year after the client’s CRP end date, when compared with that same client 1 year before CRP admission.

Sample Description

The socio demographic characteristics of 102 clients enrolled in CRP can be found in Table 1. Available variables included gender, age, race/ethnicity, education level, annual income, marital status, and living arrangements. The majority of the clients were male (64%). Regarding race/ethnic background, people who identified themselves as Hispanic were highest (63%), followed by White/Caucasian (18%) and Black/African Americans (16%). The number of Hispanic clients is representative of the demographic composition of the area from which the sample was drawn. According to U.S. census data, 63.2% of the population identified themselves as persons of Hispanic or Latino origin (U.S. Census Bureau, 2014). Because of the high number of Latino clients, this variable was collapsed into non-Hispanic (0) and Hispanic (1).

Table 1.

Sample Demographic and Clinical Characteristics (%).

Characteristic	Mental disorder only (n = 58)	Dual diagnosis (mental and substance abuse disorders) (n = 44)	Total N = 102 (%)
Sex/gender
Male	37 (64.9)	28 (63.6)	65 (64.4)
Female	20 (35.1)	16 (36.4)	36 (35.6)
Race/ethnicity
Hispanic	36 (62.1)	29 (65.9)	65 (63.7)
White, non-Hispanic	9 (15.5)	9 (20.5)	18 (17.6)
Black, non-Hispanic	10 (17.2)	5 (11.5)	15 (14.7)
Other (Pacific Islander, Asian, Native-American)	3 (5.2)	1 (2.3)	4 (3.9)
Age at enrollment
18-39 years	35 (61.4)	23 (52.3)	58 (57.4)
40-61 years	22 (38.6)	21 (47.7)	43 (42.6)
Marital status
Never married	32 (56.1)	20 (46.5)	52 (52.0)
Divorced/separated	19 (33.4)	17 (39.5)	36 (35.3)
Married	6 (10.5)	6 (14.0)	12 (11.7)
Living arrangement (χ² = 3.93)*
Cohabitating/relative/significant other	39 (67.2)	21 (47.7)	60 (58.8)
Lives on own/in a facility/homeless	19 (32.8)	23 (52.3)	42 (41.2)
Annual income
$0-$10,000	50 (86.2)	32 (72.7)	80 (80.3)
$10,001-$20,000	5 (8.6)	9 (20.5)	14 (13.7)
$20,001 and higher	3 (5.2)	3 (6.8)	6 (6.0)
Education
Did not complete high school	27 (48.2)	15 (35.7)	42 (42.9)
High school graduate and above	29 (51.8)	27 (64.3)	56 (57.1)
Primary mental health diagnosis
Bipolar/mood disorder	37 (63.8)	36 (81.8)	73 (71.6)
Schizophrenia	18 (31.0)	7 (15.9)	25 (24.5)
Anxiety or post-traumatic stress disorder	3 (5.2)	1 (2.3)	4 (3.9)
GAF scores (functioning ability)
31-40 = serious impairment	31 (55.4)	17 (38.6)	48 (48.0)
41-50 = some functioning impairment	18 (32.1)	23 (52.3)	41 (41.0)
51 and above = functional	7 (12.5)	4 (9.1)	11 (11.0)

Note. GAF = Global Assessment of Functioning.

p < .05.

Age was a continuous variable, whereby individuals ranged from 18 to 60 years old. About 57% of the sample was under age 40 at the time they entered the program. About half the sample had never been married (51%) followed by individuals who were either divorced (16%) or separated (15%). Only 12% of participants were married. This variable was collapsed into two categories for analysis: 0 = single, divorced, and separated; 1 = married. Education was treated as a continuous variable in the analysis. About 43% of clients had not finished high school, which is representative of the education level of people in the criminal justice system nationwide. However, many clients in this program did not have full-time employment and relied on disability support or held part-time jobs. This is indicative of the 80% of program participants who reported an income of less than $10,000 a year. The low income level was related to why only 11% of the sample were able to support themselves.

About 65% of the sample lived with family, and an additional 15% lived in a community facility. About 8.5% reported being homeless. In summary, the sample consisted predominantly of Hispanic males between ages 40 and 60 who had never been married, earned less than $10,000 a year, and were dependent on someone else for daily living arrangements.

Clinical Characteristics

As shown in Table 1, about seven out of 10 individuals were diagnosed with bipolar disorder (71%) followed by schizophrenia (23%). About 59% of the sample had a second mental health diagnosis. The most common secondary diagnosis was drug/alcohol dependence/abuse. A smaller number of clients had an additional diagnosis of anxiety/panic disorder, dissociative disorder (7%), post-traumatic stress disorder/adjustment disorder (7%), attention deficit hyperactivity disorder (3%), mood disorder (2%), or dementia (1%). In summary, out of 102 clients, 58 (56.9%) had a mental disorder only with no evidence of any kind of substance dependency/abuse, whereas the remaining 44 clients had a dual diagnosis of mental illness and drug/alcohol dependence/abuse diagnoses.

The Global Assessment of Functioning (GAF) score is a subjective test administered by trained mental health professionals and is a valid measure of an individual’s ability to function (Colwell, Villarreal, & Espinosa, 2012). The GAF is used for juveniles and adults with mental health needs. GAF scores range from 0 (complete inability to function) to 100 (high-level functioning ability). GAF scores in this study originated from previous mental health treatment records administered by trained licensed professional counselors at the same center for health care services. Nearly half (48%) of the sample scored between 31 and 40, which indicates that an individual has impairment in communicating or major impairment in work, school, family relations, judgment, thinking, or mood. GAF scores between 41 and 50 comprised 41% of the sample, and the remaining 11% scored 51 and 60, which indicated occasional panic attacks or moderate difficulty in social, occupational, or school functioning (APA, 2000).

Data Analysis Plan

Variables in the data were screened for multicollinearity and they met the assumptions for multivariate normality. A series of t tests for paired samples for each pre- and post-behavior were conducted using SPSS Version 22. A t test is an appropriate measure to use for the whole group when the group means under comparison have the same unit of analysis and time period. A t test can determine whether the group means differed before and after participation in the CRP program (Warner, 2012).

Regression analysis determined what variables would predict program compliance and post-program rearrest. Logistic regression is an accepted technique for a dichotomous dependent variable. Simultaneously considering variables allows variables to compete with each other, such that the strongest ones more clearly emerge to predict outcomes. In logistic regression, the dependent variable is transformed by taking the natural logarithm of the variable as an odds ratio of each event (Warner, 2012).

Results

Three separate paired t tests were performed on the entire sample of 102 to compare paired variable means: 1 year prior to CRP admission versus 1 year after the CRP termination/end date. The paired variables were the number of arrests per 12 month period, the number of offenses committed per 12 month period, and the number of days spent in jail per 12 month period. The results of these t tests can be found in Table 2, along with their individual variable means and standard deviations. Table 2 showed that the number of arrests for the entire group decreased from an average of .68 to .27 and that this difference was significant. The decreased number of arrests was more pronounced for the dual diagnosis group than for the group with no substance abuse problems. The number of offenses committed significantly decreased from 1.02 to 0.36 in the 12 months after the client left the CRP program than prior to CRP admission.

Table 2.

Before, During, and After Program Exposure.

	Mental disorder only	Dual diagnosis	Both groups	Both groups	Both groups
	M (SD)	M (SD)	M (SD)	Paired-sample t (before vs. after)	Cohen’s d	Effect size
12 months before CRP
Number of arrests	0.53 (.88)	0.86 (.93)	0.68 (.91)	3.75*	0.459	0.22
Number of offenses committed	.86 (1.62)	1.23 (1.41)	1.02 (1.54)	4.04*	0.492	0.24
Number of days in jail	11.91 (.28)	22.84 (40.85)	16.63 (34.39)	2.91*	0.375	0.18
During CRP program
Number of days in program	149.10 (138.78)	163.77 (96.35)	155.43 (121.96)
Compliant (1 = yes)	0.50 (.50)	0.61 (.49)	0.55 (.50)
12 months after CRP ended
Number of arrests	0.21 (.74)	0.36 (1.01)	0.27 (.87)
Number of offenses committed	0.26 (.82)	0.50 (1.41)	0.36 (1.12)
Number of days in jail	3.74 (14.81)	7.84 (24.76)	5.49 (19.66)

Note. CRP = Community Reintegration Program.

p < .01.

The number of days spent in jail by the entire group decreased significantly from an average of 16.6 days over 12 months to only 5.5 days. During the year before CRP, the dual diagnosis group averaged about 23 days in jail, whereas the mental health only group averaged only 12 days. During the 12 months after CRP ended, the number of days was reduced about the same amounts relative for both groups—to 8 days for dual diagnosed clients and 4 days for clients without a substance abuse problem.

Cohen’s d and the effect sizes were calculated for each of the paired before and after measures. Dunlop, Cortina, Vaslow, and Burke (1996) recommend to use the original means and standard deviations for each of the paired samples (as opposed to the pooled standard deviation). While the effect sizes will be lower, they account for possible correlation between variables that may occur with a paired sample. The effect sizes indicate a small difference.

Binary logistic regression was used to examine what variables predicted client program compliance in one model. Note that program compliance was a proxy measure of program success. The overall compliance rate of the program was 55%, and it was slightly lower for dual diagnosis clients (43%) compared with clients without a substance abuse diagnosis (57%), but this differences was not statistically significant.

The second model examined predictor variables of rearrest within the 12-month period after CRP program termination. In both regression models, there were no significant differences by gender, race/ethnicity, age, education level, income, and marital status. Given the number of cases (n = 102), a more parsimonious model of five variables was included that could potentially account for behavior during CRP and the likelihood of arrest after CRP. These variables were previous criminal history (whether a client had or had not been arrested in the 12 months before CRP), current living arrangement (lack of family support vs. living with relatives/significant other), GAF score of functionality, dual diagnosis, and number of days under supervision in the CRP. The results of both logistic regression analyses are shown in Table 3.

Table 3.

Predictors of Program Compliance and Post-Program Recidivism.

	Program compliance (1 = compliant)			Rearrest within 12 months after program (1 = yes)
Predictor variables	B	SE	Odds ratio	B	SE	Odds ratio
Dual diagnosis (1 = yes)	.35	.48		.97	.64
Prior arrests 12 months before CRP (1 = yes)	.39	.49		.75	.63
Lives with family/sig. other (1 = yes)	.46	.48		−.06	.60
GAF score (continuous)	.12*	.06	1.13	−.07	.07
Number of days in CRP (continuous)	.01**	.00	1.01	−.01**	.00	.99
Nagelkerke R² (model)	.31			.26

Note. GAF = Global Assessment of Functioning; CRP = Community Reintegration Program. Odds ratios were reported for only the significant variables in each model.

p < .05. **p < .01.

In the first model, the number of days spent in the program had the highest Wald score (14.58) which indicated that this variable contributed the most to the model. As a person spent more time in the program, the odds of completing successfully/compliance during the program increased. More specifically, the odds ratio for this variable was 1.01, which meant that when all other variables were held constant, for each day a client spent in the program, the odds of being compliant as opposed to non-compliant increased by 1.01. The GAF score of functionality was also significant. As the GAF score increased by one point, the likelihood that the client would comply and complete the program successfully increased by 1.13 times. Prior arrests, dual diagnosis, and living with family did not predict program completion.

In the second model, only the number of days in CRP predicted post-program behavior; except in this instance, the less time a client spent in CRP, the more likely he or she was to be rearrested within the 12-month period after CRP. Prior arrests before CRP, living arrangements, dual diagnosis, and the GAF score did not significantly predict rearrest. We failed to find support for our hypothesis that mentally ill only participants would have lower rearrest rates than dually disordered participants. In a separate analysis (not in Table 3), we tested whether program compliance had any bearing on post-rearrest. We found that the way the client exited the CRP (compliant or non-compliant) had no significant effect on whether or not the client was rearrested.

Discussion

As a post-booking diversion program, the CRP coordinated a continuum of care and collaborated between local criminal justice agencies (police, courts, and county jail) and the mental health system. Clients voluntarily entered the outpatient program within 24 hr of arrest for a non-violent misdemeanor in lieu of staying in jail until their next court date.

This study found that the compliance rate was slightly lower for dual diagnosis clients (43%) compared with clients without a substance abuse diagnosis (57%), but the difference was not statistically significant. Clients with a mental health problem were no more or less likely than clients with a dual diagnosis to be rearrested within 12 months of leaving the CRP program. This may be due to the fact that dual diagnosed clients did not receive drug and alcohol treatment during the pre-trial process which may have helped reduce recidivism. While there was one licensed counselor who provided outpatient cognitive-behavioral group therapy, the CRP program focused primarily on medication adherence and stability. There were no significant differences by gender, race/ethnicity, age, education level, annual income, and marital status in either the program compliance variable and in the post-program rearrest outcome.

This study found that time spent in the CRP program predicted post-program rearrest. Rearrest during the program was one reason for early discharge from the diversion program, but a new arrest that occurred during the program was completely independent from the rearrest after the client left the CRP program. There was a significant decrease in the number of arrests and offenses during the 12-month period after the client left the CRP program than in the 12 months prior to CRP admission. The time in the CRP program likely functioned to ensure medication adherence and stabilize the client, while also ensuring that the client attended each court hearing during the pre-trial process. The number of days spent in jail significantly decreased following CRP involvement. These findings support previous research (Steadman & Naples, 2005; Tanner et al., 2008) that indicate that diversion programs were effective at reducing further involvement in the criminal justice system. However, our results were different from Broner et al. (2004) who admitted that their finding of no difference in rearrest may have been due to lack of comparability between groups and the difficulty of finding a comparison site.

While access to treatment services are mixed across various diversion programs (Broner et al., 2004), structured diversion programs that have the support and involvement of the courts are generally viewed as more successful than unstructured diversion programs (Broner, Mayrl, & Landsberg, 2005). The CRP may have indeed functioned as a “boundary spanner” between local correctional facilities and mental health agencies as it supervised and treated clients while the criminal courts decided whether to prosecute, release, or dismiss the charges (Steadman, 1992).

This study also found that the GAF score and time spent in the outpatient program were both significantly related to increasing compliance and overall success. These findings are consistent with GAF functionality being important predictors in program compliance (Herbeck et al., 2005). However, GAF scores did not predict future rearrest.

Limitations of the study must be mentioned, to include that the findings were restricted to pre-trial misdemeanants, and may not apply to post-plea or post-adjudication courts or diversion programs that contain exclusively felons. A second limitation of the findings was that while we can state that changes in arrest and jail days occurred pre- and post-, the data do not allow us to state unequivocally that these changes were due to the intervention. Third, the outcome variables were limited to program compliance and post-program rearrest. While both variables are sound and accepted outcomes in criminal justice, we were not able to measure outcomes typical in the mental health system such as quality of life (homelessness, employment, or economic independence), improvement in functioning (change in GAF scores), or client satisfaction with services (Barry & Crosby, 1996; Lehman, 1996; Ruggeri, Gater, Bisoffi, Barbui, & Tansella, 2008). Effective programs ideally have prominent leadership, continuous training, and positive staff attitudes toward consumer-centered treatment (Whitley, Gingerich, Lutz, & Mueser, 2009). In turn, treatment in community-based centers can also lead to changed outlooks on mental illness, and even greater satisfaction and general well-being for persons with mental illness (Ruggeri et al., 2008; Whitley et al., 2009).

Cooperation between different agencies and constituencies can bridge the gaps and lead to more effective implementation; it can also develop goals that balance the criminal justice system’s public safety concerns and individuals’ rights to mental health services. Community-based mental health centers can provide a balance between treatment rights and the need for public safety in a manner that satisfies legal requirements (Lamb, Weinberger, & Gross, 1999). These services, though, are not meant for all mentally ill offenders or to make them no longer dangerous; instead, the focus lies on stabilizing the mental illness and maintaining functioning and control (Lamb et al., 1999).

This study showed that a voluntary, community-based diversion clinic can be potentially effective at addressing court-ordered treatment in lieu of jail at the pre-trial stage if arrested for a misdemeanor crime. Directions for future research may wish to examine the utility of outpatient clinics that also provide substance abuse treatment for dual diagnosis clients.

Footnotes

Acknowledgements

The authors acknowledge assistance by the Center for Health Care Services for data procurement. Views expressed are solely that of the authors and do not represent views of the Center for Health Care Services.

Authors’ Note

The findings were presented at the annual conference of the Academy of Criminal Justice Sciences, March 20-23, 2013, in Dallas, Texas.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Financial support for data collection and analysis were provided by a collaborative faculty research grant from the University of Texas at San Antonio.

References

American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th ed., text rev.). Washington, DC: Author.

Barry

M. M.

Crosby

(1996). Quality of life as an evaluative measure in assessing the impact of community care on people with long-term psychiatric disorders. British Journal of Psychiatry, 168, 210-216.

Boccaccini

M. T.

Christy

Poythress

Kershaw

(2005). Rediversion in two postbooking jail diversion programs in Florida. Psychiatric Services, 56, 835-839.

Bonta

Law

Hanson

(1998). The prediction of criminal and violent recidivism among mentally disordered adult offenders: A meta-analysis. Psychological Bulletin, 123, 123-142.

Broner

Lattimore

P. K.

Cowell

A. J.

Schlenger

W. E.

(2004). Effects of diversion on adults with co-occurring mental illness and substance use: Outcomes from a national multi-site study. Behavioral Sciences & the Law, 22, 519-541.

Broner

Mayrl

D. W.

Landsberg

(2005). Outcomes of mandated and non-mandated New York City jail diversion for offenders with alcohol, drug, and mental disorders. The Prison Journal, 85, 18-49.

Canales

D. D.

Campbell

M. A.

Wei

Totten

A. E.

(2014). Prediction of general and violent recidivism among mentally disordered adult offenders: Test of the Level of Service/Risk-Need-Responsivity (LS/RNR) instrument. Criminal Justice and Behavior, 41, 971-991.

Case

Steadman

H. J.

Dupuis

S. A.

Morris

L. S.

(2009). Who succeeds in jail diversion programs for persons with mental illness? A multi site study. Behavioral Sciences & the Law, 27, 661-674.

Castillo

E. D.

Alarid

L. F.

(2011). Factors associated with recidivism among offenders with mental illness. International Journal of Offender Therapy and Comparative Criminology, 55, 98-117.

10.

Center for Health Care Services. (2006). Providing jail diversion for people with mental illness. Psychiatric Services, 57, 1521-1523.

11.

Cloyes

K. G.

Wong

Latimer

Abarca

(2010). Time to prison return for offenders with serious mental illness released from prison. Criminal Justice and Behavior, 37, 175-187.

12.

Colwell

Villarreal

S. F.

Espinosa

E. M.

(2012). Preliminary outcomes of a pre-adjudication diversion initiative for juvenile justice involved youth with mental health needs in Texas. Criminal Justice and Behavior, 39, 447-460.

13.

Dunlop

W. P.

Cortina

J. M.

Vaslow

J. B.

Burke

M. J.

(1996). Meta-analysis of experiments with matched groups or repeated measures designs. Psychological Methods, 1, 170-177.

14.

Fisher

W. H.

Silver

Wolff

(2006). Beyond criminalization: Toward a criminologically informed framework for mental health policy and services research. Administration and Policy in Mental Health and Mental Health Services Research, 33, 544-557. doi:10.1007/s10488-006-0072-0

15.

Gonzales

Davak

(2006). Out of jail and into treatment. Behavioral Healthcare, 16, 24-26.

16.

Grantham

(2011). Right place, right time, right approach: Texans collaborate to build a model jail diversion and crisis mental health system. Behavioral Healthcare, 31, 14-19. Retrieved from http://www.behavioral.net/node/19772?page=0

17.

Hartford

Carey

Mendonca

(2006). Pre-arrest diversion of people with mental illness: Literature review and international survey. Behavioral Sciences & the Law, 24, 845-856.

18.

Hartwell

(2004). Triple stigma: Persons with mental illness and substance abuse problems in the criminal justice system. Criminal Justice Policy Review, 15, 84-99.

19.

Herbeck

D. M.

Fitek

D. J.

Svikis

D. S.

Montoya

I. D.

Marcus

S. C.

West

J. C.

(2005). Treatment compliance in patients with comorbid psychiatric and substance use disorders. The American Journal on Addictions, 14, 195-207.

20.

Kim

(2016). Psychiatric deinstitutionalization and prison population growth: A critical literature review and its implications. Criminal Justice Policy Review, 27, 3-21.

21.

Lamb

H. R.

Weinberger

L. E.

Gross

B. H.

(1999). Community treatment of severely mentally ill offenders under the jurisdiction of the criminal justice system: A review. Psychiatric Services, 50, 907-913.

22.

Lattimore

P. K.

Broner

Sherman

Frisman

Shafer

M. S.

(2003). A comparison of prebooking and postbooking diversion programs for mentally ill substance-using individuals with justice involvement. Journal of Contemporary Criminal Justice, 19, 30-64.

23.

Lehman

A. F.

(1996). Measures of quality of life among persons with severe and persistent mental disorders. Social Psychiatry and Psychiatric Epidemiology, 31, 78-88.

24.

McCoy

M. L.

Roberts

D. L.

Hanrahan

Clay

Luchins

D. J.

(2004). Jail linkage assertive community treatment services for individuals with mental illness. Psychiatric Rehabilitation Journal, 27, 243-250.

25.

McReynolds

L. S.

Schwalbe

C. S.

Wasserman

G. A.

(2010). The contribution of psychiatric disorder to juvenile recidivism. Criminal Justice and Behavior, 37, 204-216.

26.

Modestin

Wuermle

(2005). Criminality in men with major mental disorder with and without comorbid substance abuse. Psychiatry and Clinical Neuroscience, 59, 25-29.

27.

Ringhoff

Rapp

Robst

(2012). The criminalization hypothesis: Practice and policy implications for persons with serious mental illness in the criminal justice system. Best Practices in Mental Health, 8, 1-19.

28.

Rubin

Alarid

L. F.

Rodriguez

M. J.

(2014). A partnership to develop a utilization-focused evaluation plan for a criminal justice diversion program. Psychiatric Services, 65, 8-10.

29.

Ruggeri

Gater

Bisoffi

Barbui

Tansella

(2008). Determinants of subjective quality of life in patients attending community-based mental health services. The South-Verona outcome project 5. Acta Psychiatrica Scandinavica, 105, 131-140.

30.

Ryan

Brown

C. K.

Watanabe-Galloway

(2010). Toward successful postbooking diversion: What are the next steps? Psychiatric Services, 61, 469-477.

31.

Shafer

Arthur

Franczak

(2004). An analysis of post-booking jail diversion programming for persons with co-occurring disorders. Behavioral Sciences & the Law, 22, 771-785.

32.

Slate

R. N.

Buffington-Vollum

J. K.

Johnson

W. W.

(2013). The criminalization of mental illness: Crisis and opportunity for the justice system (2nd ed.). Durham, NC: Carolina Academic Press.

33.

Steadman

H. J.

(1992). Boundary spanners: A key component for the effective interactions of the justice and mental health systems. Law and Human Behavior, 16, 75-87.

34.

Steadman

H. J.

Naples

(2005). Assessing the effectiveness of jail diversion programs for person with serious mental illness and co-occurring substance use disorders. Behavioral Sciences & the Law, 23, 163-170.

35.

Steadman

H. J.

Stainbrook

K. A.

Griffin

Draine

Dupont

Horey

(2001). A specialized crisis response site as a core element of police-based diversion programs. Psychiatric Services, 52, 219-222.

36.

Swanson

J. W.

Holzer

C. E.

Ganju

V. K.

Jono

R. T.

(1990). Violence and psychiatric disorder in the community: Evidence from the epidemiological catchment area surveys. Hospital & Community Psychiatry, 41, 761-770.

37.

Tanner

Wyatt

Yearwood

D. L.

(2008). Evaluating pretrial services programs in North Carolina. Federal Probation, 72, 18-27.

38.

U.S. Census Bureau. (2014). State & county quickfacts. San Antonio, TX. Retrieved from http://quickfacts.census.gov

39.

Wallace

Mullen

P. E.

Burgess

(2004). Criminal offending in schizophrenia over a twenty-five year period marked by deinstitutionalization and escalating comorbid substance abuse. American Journal of Psychiatry, 161, 716-727.

40.

Warner

R. M.

(2012). Applied statistics: From bivariate through multivariate techniques (2nd ed.). Los Angeles, CA: Sage.

41.

White

M. D.

Goldkamp

J. S.

Campbell

S. P.

(2006). Co-occurring mental illness and substance in the criminal justice system: Some implications for local jurisdictions. The Prison Journal, 86, 301-326.

42.

Whitley

Gingerich

Lutz

W. J.

Mueser

K. T.

(2009). Implementing the illness management and recovery program in community mental health settings: Facilitators and barriers. Psychiatric Services, 60, 202-209.