CJ Discussion

Effectiveness of a Short-Term Mental Health Court: Criminal Recidivism One Year Postexit Virginia Aldigé Hiday, Heathcote W. Wales, and Bradley Ray Online First Publication, June 17, 2013. doi: 10.1037/lhb0000030

CITATION Hiday, V. A., Wales, H. W., & Ray, B. (2013, June 17). Effectiveness of a Short-Term Mental Health Court: Criminal Recidivism One Year Postexit. Law and Human Behavior. Advance online publication. doi: 10.1037/lhb0000030

Effectiveness of a Short-Term Mental Health Court: Criminal Recidivism One Year Postexit

Virginia Aldigé Hiday North Carolina State University

Heathcote W. Wales Georgetown University Law Center

Bradley Ray Indiana University–Purdue University Indianapolis

This article investigated criminal recidivism 1 year postexit from a mental health court (MHC), which has, unlike prior MHCs studied, relatively short periods of court supervision. It benefits from a federal pretrial services agency that screens all arrestees for mental illness and dedicates a specialized supervi- sion unit (SSU) to provide supervision and services while on pretrial release to all screened positive, including MHC participants. We compared criminal activity prior to key arrest with criminal activity post court disposition in MHC participants (N � 408) and MHC-eligible mentally ill arrestees in SSU (N � 687) receiving the same supervision and services while controlling for possible confounders. The proportion of MHC participants arrested was significantly lower in the year after MHC exit and significantly lower than that of the comparison group. They also averaged fewer rearrests and had a longer time to rearrest. MHC graduates made the greatest gains and accounted for the recidivism differences between MHC participants and the comparison group. This study adds to the accumulating evidence of the effectiveness of MHCs in reducing recidivism among offenders with severe mental illness.

Keywords: mental health courts, recidivism, arrests, mandated treatment, specialty courts

Mental Health Courts (MHCs) are one of the many new pro- grams that aim to reduce criminal offending of persons with mental illness by diverting them from the criminal justice system into the community mental health system (Broner, Lattimore, Cowell, & Schlenger, 2004; Hiday & Wales, 2013). They follow the drug court model in structure, having (a) a separate docket, (b) one or two dedicated judges who preside at regular status hearings, (c) dedicated prosecution, (d) a nonadversarial team approach involving consensus decisions by criminal justice and mental health professionals, (e) voluntary participation of defendants, and (f) dismissed charges or avoidance of incarceration, depending on whether the defendant enters pre- or postadjudication, after suc- cessful completion of mandated treatment (Moore & Hiday, 2006; Redlich, Steadman, Monahan, Robbins, & Petrila, 2006).

MHCs vary in some details, but there are many commonalities. Defense and prosecuting attorneys do not dispute guilt or inno- cence. Rather, they work as a team with judges, criminal justice personnel, mental health practitioners, and other providers to find treatment and services, and to allot encouragement and sanctions

that will address the underlying causes of each participant’s of- fending while protecting the public. Team members recognize that setbacks are common; thus, they offer second chances and help participants try again; but they enforce compliance and maximize motivation to change with graduated sanctions from increased monitoring to overnights in jail (Hiday, Moore, Lamoureaux, & deMagistris, 2005; Redlich et al., 2006).

Studies that examine the effect of MHCs on their main goal of reduced criminal recidivism are accumulating. They all report that mentally ill defendants who go through MHC are less likely to offend than before entering the MHC (Christy, Poythress, Boothroyd, Petrila, & Mehra, 2005; Dirks-Linhorst & Linhorst, 2012; Frailing, 2010; Herinckx, Swart, Ama, Dolezal, & King, 2005; Hiday & Ray, 2010; Moore & Hiday, 2006; Palermo, 2010; Steadman, Redlich, Callahan, Robbins, & Vesselinov, 2011; Tru- pin & Richards, 2003), and when compared with mentally ill defendants in traditional criminal court, they are no more likely to reoffend (Christy et al., 2005; Cosden, Ellens, Schnell, Yamini- Diouf, & Wolfe, 2003; Dirks-Linhorst & Linhorst, 2012; Frailing, 2010; McNiel & Binder, 2007; Moore & Hiday, 2006; Steadman et al., 2011; Trupin & Richards, 2003). The question of whether MHCs do better than traditional criminal courts, that is, whether they reduce criminal recidivism beneath the recidivism level of mentally ill defendants who go through regular criminal adjudica- tion, has not been satisfactorily answered because of methodolog- ical shortcomings. Some studies had no comparison group (Heri- nckx et al., 2005; Hiday et al., 2005; Hiday & Ray, 2010; Palermo, 2010), and comparison groups in other studies consisted of those from the same jurisdiction who were not referred to MHC or did

Virginia Aldigé Hiday, Department of Sociology, North Carolina State University; Heathcote W. Wales, Georgetown University Law Center; Bradley Ray, School of Public and Environmental Affairs, Indiana University–Purdue University Indianapolis.

Correspondence concerning this article should be addressed to Virginia Aldigé Hiday, Department of Sociology, Box 8107, North Carolina State University, Raleigh, NC 27695-8107. E-mail: ginny_hiday@ncsu.edu

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an Ps

yc ho

lo gi

ca l

A ss

oc ia

tio n

or on

e of

its al

lie d

pu bl

is he

rs .

T hi

s ar

tic le

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

not opt into MHC, thus producing unknown selection bias (Dirks- Linhorst & Linhorst, 2012; Frailing, 2010; McNiel & Binder, 2007; Steadman et al., 2011; Trupin & Richards, 2003).

Some comparison groups received unknown, minimal, or no treatment, leaving unanswered the question of whether mental health treatment or participation in MHC itself made the difference (Christy et al., 2005; Dirks-Linhorst & Linhorst, 2012; Frailing, 2010; McNiel & Binder, 2007; Moore & Hiday, 2006; Steadman et al., 2011; Trupin & Richards, 2003). We do not know whether it is the mental health treatment and services integral to MHC (but that could be provided by other programs) that account fully for the observed reductions in recidivism, or whether what goes on in MHC (the legal procedures, monitoring, sanctions, and support) constitutes a contributory cause.

A third question has not been satisfactorily addressed: Do the positive effects of MHC continue after defendants exit the court? Follow-up in almost all recidivism studies began after MHC entry and continued for a limited follow-up time (�1 year.); hence, defendants were being monitored by and receiving services, treat- ment, and support from MHC for all or most of the follow-up period, given that average time in MHC is approximately 1 year or more (Cosden et al., 2003; Dirks-Linhorst & Linhorst, 2012; Frailing, 2010; Herinckx et al., 2005; Redlich et al., 2010; Ross- man et al., 2012). Most recidivism studies, thus, leave unknown MHC effects after court exit. Only four studies examined criminal recidivism for a substantial time after MHC exit (Dirks-Linhorst & Linhorst, 2012; Frailing, 2010; Hiday & Ray, 2010; McNiel & Binder, 2007). All reported a continuation of MHC effects in this postexit period without its supports, services, treatment, and mon- itoring. Although these four studies avoided some of the limita- tions of earlier studies and represented socioeconomically different jurisdictions (a suburb, a medium-sized city, a rural county with a university town, and an urban inner city), one had no comparison group (Hiday & Ray, 2010), two had no prior measure (Dirks- Linhorst & Linhorst, 2012; McNiel & Binder, 2007), one followed only graduates of the court (McNiel & Binder, 2007), and none controlled for treatment and services. The questions of whether a MHC can continue to effect reduced criminal recidivism after court exit, whether participation in MHC itself impacts recidivism above the impact of treatment and services, and whether the reductions are any greater than those experienced by comparable persons with mental illness in traditional criminal court remain to be answered.

Study Overview

In this study, we attempted to answer these three questions with data on a large sample of mentally ill defendants in a misdemeanor MHC and comparable mentally ill defendants in the same juris- diction, both of whom were released pretrial and received the same supervision, case management, and services from the same in- house agency and community providers. The only intervention difference between MHC and traditional criminal court defendants was the added participation in MHC.

The Setting

The study MHC, serving the District of Columbia and located in its main criminal court building, has high caseloads, providing a

large sample size. It accepts competent severely mentally ill ar- restees, charged with misdemeanors, who have no pending charge or conviction of a dangerous or violent felony in the prior 5 years. It monitors participants for treatment and behavior compliance at required monthly status hearings. Upon consecutive 4 to 6 months of substantial compliance with court mandates (extended on occa- sion to seven or eight months if a participant is making progress), participants graduate and have their charges dismissed. Those who repeatedly fail to meet the court’s treatment and behavioral man- dates are returned to regular criminal court for prosecution. Every participant is represented by an attorney in all MHC appearances and in the ongoing negotiation (with participant present) with the prosecutor (U.S. attorney). The study MHC is different from MHCs reported in the literature in having a relatively short period of court supervision (4 to 6 months) and more limited court involvement in monitoring case managers and providers (such monitoring occurs only in progress discussion during court hear- ings with the participant present and in policy/organization dis- cussion during quarterly stakeholder meetings).

Another difference is that this study’s MHC, as part of the District of Columbia criminal courts, has the benefit of a federal pretrial services agency that screens all arrestees for severe mental illness and dedicates a specialized supervision unit (SSU) to pro- vide supervision and services (case management and linkage to mental health service agencies, and drug testing and treatment for those dually diagnosed) while on pretrial release to those screened positive, including MHC participants. Screening occurs the morn- ing after arrest, followed in the afternoon by arraignment and a pretrial release hearing. The judge orders pretrial release to SSU for those screened positive for severe mental illness who pose minimal public safety or disappearance risk. Some defendants with mental illness are missed in the initial screening and are later placed with SSU when case managers or counsel detect mental illness. Others who are initially deemed disappearance risks or remanded to the mental hospital for competency evaluation and restoration may later be released under SSU supervision. When a SSU case manager receives a MHC-eligible defendant, s/he is expected to refer that person to MHC shortly thereafter. Though not a formal requirement for MHC, the U.S. attorney will not sign a deferred prosecution agreement for official MHC entry until a defendant has been linked with a community mental health agen- cy; thus, case managers refer most clients only after their beginning reception of services from a community mental health provider.

Method

In our quasi-experimental design, we compared 1-year recidi- vism post court exit of MHC participants (N � 408) with that of other MHC-eligible defendants under SSU supervision on pretrial release (treatment-as-usual, N � 687) who received the same services and supervision by the same agency, while controlling for possible confounders. We hypothesized that MHC participants would have lower recidivism than the comparison group in the year post MHC exit/case disposition on all measures. Furthermore, we hypothesized that MHC completers would have lower recidi- vism on all measures than those who did not complete MHC because completers get a “full dose” of MHC (Moore & Hiday, 2006). To test this hypothesis, we compared MHC completers

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an Ps

yc ho

lo gi

ca l

A ss

oc ia

tio n

or on

e of

its al

lie d

pu bl

is he

rs .

T hi

s ar

tic le

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

2 HIDAY, WALES, AND RAY

(graduates who fulfilled their individualized mandates) with those who did not complete MHC because of ejection for noncompliance with court mandates or their own decision to opt out because they disliked aspects of the court.

Data Collection

We used administrative data of the pretrial services agency for all misdemeanor arrestees screened positive for severe mental illness who did not have a recent (current or past 5 years) violent or dangerous felony charge1 or current domestic violence charge, who were released pretrial to SSU during the first 2 years of the operation of the District of Columbia’s Mental Health Diversion Court (the MHC), whose cases were not disposed of by another diversion court, and who had complete data (October 2007 through November 2009; N � 1,095, 408 of whom participated in the study MHC). We excluded from the MHC sample defendants sent back to traditional criminal court at their first MHC hearing (n � 37), defendants who did not show up for their first MHC hearing (n � 3), and MHC participants who had administrative closures because of death, sickness, or other (n � 6). The data included sociode- mographic, clinical, and District of Columbia arrest information from 1 year before the key arrest that brought each of them into the sample through 1 year after disposition of the key arrest in MHC or traditional criminal court. Pretrial services agency data also included FBI national arrest data on prior arrests outside the district, but these were not used because of not having comparable postarrest data. Records kept by the MHC judges on processing MHC participants supplemented the pretrial services agency data. Informed consent was not necessary because all data were deiden- tified. The two universities’ and the pretrial services agency’s institutional review boards and the MHC judges approved the research.

Measures

We used four measures to compare postexit recidivism of MHC participants with other MHC-eligible arrestees in SSU: (a) any arrest, (b) number of arrests, (c) any felony arrest in the 1 year following key arrest disposition (for MHC graduates, key arrest disposition is when they exit MHC), and (d) time to arrest follow- ing key arrest disposition. We excluded from postexit recidivism measures rearrests that occurred after key arrest but prior to court exit/disposition. These arrests occur prior to our follow-up period and MHCs often forgive them as slip-ups (Hiday et al., 2005; Ray, Dollar, & Thames, 2011). We employed them as predictors of postexit recidivism because they indicate noncompliance with MHC mandates and continued offending. We did not count key arrest in prior arrests. To predict recidivism in multivariate models that control for possible confounders, we employed logistic regres- sion and used the Cox proportional hazards model to predict time to rearrest.

Results

Sample Description

Table 1 presents the characteristics of our sample. The mean age of both groups is 41 years. The MHC group has equal numbers of

males and females, but the SSU group has significantly more males (63.3% vs. 50.0%; �2 � 18.69, p � .001, Cramer’s V � 0.13). Both groups are overwhelmingly African American (�90%), with no significant difference between them. A positive test for illegal drugs (drug use) at first check-in with the SSU case manager is high for both groups, but the SSU group has signifi- cantly more individuals with a positive drug test (63.9% vs. 53.9%; �2 � 10.64, p � .001, Cramer’s V � 0.09). Almost all members of both groups were arrested in the year before their key arrest (99.3% SSU, 98.3% MHC). Most members of both groups had only one arrest in the year before the key arrest (61.0% SSU, 72.3% MHC), but more SSU participants had two or more prior arrests and they averaged significantly more arrests in the year prior to the key arrest (1.7 vs. 1.4, t � 4.72, p � .001, d � 0.32, 95% CI [0.27, 0.38]).

Table 2 presents dispositions of sample members’ cases. Just less than half of the SSU comparison group received a guilty verdict (45.9%); most of the other cases were dismissed, often as part of a plea agreement in another case, and a small proportion received a not-guilty verdict. Close to three-fifths of MHC partic- ipants graduated (58.3%), that is, they completed the mandatory period of supervision, treatment, and services, and had their cases dismissed by the MHC. The noncompleters were as likely to have their cases dismissed in traditional criminal court as they were to be found guilty (21.1%, 19.9%), and a few were found not guilty (0.7%). Overall, MHC cases were more likely than SSU cases to result in dismissal or acquittal (78.9% vs. 54.1%), whereas MHC

1 The MHC exclusion criteria are no current or pending violent or dangerous charge, and no violent or dangerous felony conviction in the past 5 years. Conviction information was not available in the data set. Instead, we used a broader exclusion of no violent or dangerous felony charge that yielded a more conservative comparison with MHC participants.

Table 1 Sample Characteristics

SSU MHC

M SD M SD

Age 40.7 11.6 41.4 11.0 Sexb� N % N %

Male 435 63.3 204 50.0 Female 252 36.7 204 50.0

Race Non-White 639 93.0 367 90.0 White 48 7.0 41 10.0

Drug use at check-inb�

Yes 439 63.9 220 53.9 No or unknown 248 36.1 188 46.1

M SD M SD

Prior arrests (1 year)a� 1.7 1.0 1.4 0.8 Prior arrests count (1 year)b� N % N %

0 prior arrests 5 0.7 7 1.7 1 prior arrest 419 61.0 295 72.3 2 or more prior arrests 263 38.3 106 26.0

Total 687 408

Note. MHC � mental health court; SSU � specialized supervision unit. a t test. b �2. � p � .001.

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an Ps

yc ho

lo gi

ca l

A ss

oc ia

tio n

or on

e of

its al

lie d

pu bl

is he

rs .

T hi

s ar

tic le

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

3MENTAL HEALTH COURT EFFECTIVENESS

noncompleters and SSU controls had approximately the same rates of dismissal and acquittal (52.4%, 54.1%, respectively).

Table 3 presents offense categories for key arrests by most serious offense charged. There is a significant difference between MHC participants and the comparison group (�2 � 31.22, p � .001, Cramer’s V � 0.17) that is driven by the MHC group having more with drug charges and the comparison group having more with person offenses. The most common person offenses in fre- quency order were simple assault, assault on police (defined broadly to include resisting arrest or impeding a law enforcement officer), and attempted threats to do bodily harm.

Recidivism

Table 4 presents recidivism in the year after MHC exit/case disposition. The majority of each group was not rearrested, but MHC participants were significantly less likely to be arrested in the follow-up year than persons in the SSU (27.5% vs. 37.3%; �2 � 11.04, p � .001, Cramer’s V � 0.10). Most rearrests were misdemeanors; only small proportions of either group were ar- rested for felonies (�10%). Even fewer were rearrested for violent felonies (�2%). The distribution of number of arrests in the bottom part of Table 4 shows a significant difference between MHC and SSU defendants (�2 � 15.64, p � .001, Cramer’s V � 0.12). Almost three-fourths of MHC participants had no rearrest,

and just over three-fifths of SSU defendants had no rearrest. Among those arrested, most members of the SSU and MHC groups had only one arrest (22.1%, 19.4%, respectively), with number of arrests steadily declining thereafter.

Table 5 presents persons arrested in each group after the key arrest for two overlapping time periods: (a) between key arrest and SSU placement, and (b) between key arrest and case disposition. Only a few of either group (4.9% SSU, 2.5% MHC) were arrested after key arrest before a traditional criminal court judge placed them with SSU; and just over one fifth was arrested in each group between key arrest and case disposition (23.3% SSU, 21.1% MHC). The number or persons arrested for felony offenses was much less (�5%), and the number of those arrested for violent felonies was even smaller (1.0% SSU, 0.7% MHC). Some of each group who failed to appear for court hearings or SSU appointments were issued bench warrants that resulted in arrest, representing 17% of SSU arrests between key arrest and disposition and 13% of MHC arrests during this period. There was no significant differ- ence between the SSU and MHC groups on arrests during either period.

Theoretically, there should have been no chance of arrest be- tween key arrest and SSU placement. Screening for severe mental illness occurred in lockup the morning after an arrest, and those screened positive were supposed to get pretrial release with SSU placement that same day, or were to continue detention in jail. In any case, there should have been no time at risk for arrest. Screening, however, did not identify all persons with severe men- tal illness; thus, many were placed with SSU only later, when their mental illness was detected. Many were not screened because they were released on their own recognizance when brought in. Some identified as mentally ill were detained because of previous failure to appear or were remanded to the mental hospital for competency evaluation and restoration. There was, thus, considerable time between key arrest and SSU placement; only 34.9% and 13.2% of the SSU and MHC samples, respectively, were placed with SSU within 1 week. As can be seen in Table 5, average days from key arrest to SSU placement ranged from 47.5 to 126.3, with those arrested during this period having higher averages. Average time from key arrest to case disposition ranged from 105.8 days to

Table 2 Sample Dispositions

SSU MHC

N % N %

SSU SSU–Guilty 315 45.9 — — SSU–Dismissed 354 51.5 — — SSU–Not guilty 18 2.6 — —

MHC–Completer 238 58.3 MHC–Noncompleters

MHC–Guilty — — 81 19.9 MHC–Dismissed — — 86 21.1 MHC–Not guilty — — 3 0.7

Total 687 408

Note. MHC � mental health court; SSU � specialized supervision unit.

Table 3 Sample Offense Categories on Key Arrest

SSU MHC

N % N %

Offense categorya�

Person 205 29.8 70 17.2 Property 124 18.0 80 19.6 Drugs 205 29.8 164 40.2 Bail reform 62 9.0 25 6.1 Public order 57 8.3 49 12.0 Other 34 4.9 20 4.9

Total 687 408

Note. MHC � mental health court; SSU � specialized supervision unit. a �2. � p � .001.

Table 4 One-Year Postexit Outcomes

SSU MHC

N % N %

Percent arresteda� 256 37.3 112 27.5 Any felony rearrest 67 9.8 33 8.1 Any violent felony rearrest 13 1.9 4 1.0 Number of arrestsa�

0 431 62.7 296 72.5 1 152 22.1 79 19.4 2 65 9.5 23 5.6 3 27 3.9 7 1.7 4 10 1.5 3 0.7 5� 2 0.3 0 0.0

Total 687 408

Note. MHC � mental health court; SSU � specialized supervision unit. a �2. � p � .001.

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an Ps

yc ho

lo gi

ca l

A ss

oc ia

tio n

or on

e of

its al

lie d

pu bl

is he

rs .

T hi

s ar

tic le

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

4 HIDAY, WALES, AND RAY

374.5 days, again with those arrested having higher averages. There is great variation in these time periods, as indicated by the exceptionally large standard deviations. Time between key arrest and SSU placement ranged up to 684 days, and time between key arrest and case disposition ranged up to 1,115 days.

MHC participants had significantly more days than the SSU comparison group in both time periods (key arrest to SSU place- ment: 91.7 vs. 51.4 days, t � 8.57, p � .001, d � 0.54, 95% CI [�3.88, 4.96]; key arrest to disposition: 237.7 vs. 140.8 days, t � 10.11, p � .001, d � 0.63, 95% CI [�8.49, 9.75]). Over half of the comparison group was placed in SSU within 30 days, but it took twice as long for half of MHC participants to be placed in SSU. At 90 days, 86.2% of the comparison group was placed in SSU, but still the MHC group lagged, with only 65.9% placed in SSU. Similar lags occurred in reaching case disposition. Almost half of the SSU comparison group reached disposition within 90 days, but only 15.0% of MHC participants did so. At 180 days, 75.5% of the SSU group and 43.1% of MHC members reached disposition, and at 1 year, 93.2% of the SSU group and 82.1% MHC participants reached disposition.

To see whether noncompleters accounted for MHC arrests prior to case disposition, Table 6 presents the same arrest information as Table 5, but only for MHC participants who had one or more arrests, broken down by court completion status. There were few

persons arrested among either completers or noncompleters be- tween key arrest and SSU placement. More were arrested by disposition. Noncompleters were significantly more likely to be arrested by then (33.5% vs. 12.2%, �2 � 27.16, p � .001, Cram- er’s V � 0.26), but rearrest did not automatically lead to MHC termination. Of those arrested before leaving MHC (n � 61), approximately two-fifths (39.3%) remained in MHC and graduated (not shown in table). Rearrest between key arrest and disposition also did not increase the likelihood of a guilty verdict in traditional criminal court. The proportions found guilty among those rear- rested and not rearrested before disposition were almost identical (SSU: 46.9% vs. 45.5%; MHC: 35.1% vs. 35.9%, not shown in table).

Table 7 presents the pre–post arrest comparison for MHC par- ticipants by completers and noncompleters and for MHC eligibles in SSU. All groups experienced significant declines in average number of arrests, dropping from an average of over one arrest in the year before key arrest to less than one arrest in the year after case disposition. MHC completers had the largest average decline (�1.11), even though they had the smallest average number of prior arrests, and had the smallest proportion rearrested (17.6%). But even most MHC noncompleters were not rearrested in the year following their case disposition (58.8%).

Multivariate Analyses

We used logistic regression to discern whether any participation in MHC affected recidivism in the year after MHC exit/case disposition (more than treatment as usual in the SSU group), controlling for possibly confounding variables (including individ- ual characteristics significantly different between the two groups). We first modeled the effect of MHC participation on the odds of rearrest. We then sequentially added three sets of possible con- founders: (a) the socioeconomic variables of age, gender, and race; (b) the criminological factors of illegal drug use and number of prior arrests; and (c) the process variable of any pre-disposition arrest. The first model in Table 8 shows that MHC participants were 36% less likely than the comparison group (SSU) to be rearrested in the year following their MHC exit (p � .001). In each subsequent model with the added confounders, MHC participation maintained its significant effect. Even in the last model that con-

Table 5 Persons Arrested After Key Arrest Date

No arrest One or more arrests

SSU MHC SSU MHC

N % N % N % N %

Time period Arrest to SSU placement 653 95.1 398 97.5 34 4.9 10 2.5 Arrest to disposition 527 76.7 322 78.9 160 23.3 86 21.1

Average days M SD M SD M SD M SD

Arrest to SSU placement 47.5 60.8 91 85.3 126.3 131.1 120.4 101.8 Arrest to disposition 105.8 105.2 201.2 130.4 255.6 198.9 374.5 205.2

Felony arrest N % N % N % N %

Arrest to SSU placement 682 99.3 407 99.8 5 0.7 1 0.2 Arrest to disposition 654 95.2 390 95.6 33 4.8 18 4.4

Note. MHC � mental health court; SSU � specialized supervision unit.

Table 6 Persons Arrested After Key Arrest by Mental Health Court Outcome

One or more arrests

Completer Noncompleter MHC

N % N % N %

Time period Arrest to SSU placement 5 2.1 5 2.9 10 2.5 Arrest to disposition 29 12.2 57 33.5 86 21.1

Average days M SD M SD M SD

Arrest to SSU placement 88.2 48.4 152.6 135.6 120.4 101.8 Arrest to disposition 406.5 217.7 358.2 200.2 374.5 205.2

Note. MHC � mental health court; SSU � specialized supervision unit.

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an Ps

yc ho

lo gi

ca l

A ss

oc ia

tio n

or on

e of

its al

lie d

pu bl

is he

rs .

T hi

s ar

tic le

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

5MENTAL HEALTH COURT EFFECTIVENESS

trolled for sociodemographic characteristics, illegal drug use, num- ber of prior arrests, and any arrest between key arrest and dispo- sition, MHC participants were one fourth less likely to be rearrested. In this last model, four other variables showed signif- icant effects: illegal drug use, number of arrests in the year prior to key arrest, and any arrest between key arrest and disposition raised the odds of rearrest in the year following MHC exit/case disposi- tion. Increased age lowered the risk.

Table 9 presents the results of logistic regression predicting rearrest with MHC participants separated into completers and noncompleters (the suppressed or comparison category is the SSU group as in the previous table). One can see that MHC completion made an even larger difference in lowering the odds of arrest after court exit than did MHC participation. The first model shows that MHC completers were approximately two-thirds less likely to be rearrested than SSU sample members; MHC noncompleters were not significantly different in rearrest from SSU sample members. By the final model with all confounders included, MHC com- pleters were still approximately 51% less likely to recidivate. As in Table 8, older sample members were less likely to be rearrested, and each arrest in the year prior to key arrest and any arrest between key arrest and disposition increased the likelihood of rearrest post MHC exit/case disposition. The final model shows that illegal drug use became nonsignificant, as graduation and pre-disposition arrest subsumed drug use at first check-in, suggest- ing that illegal drug users who were able to change their behavior and graduate were able to maintain that change after leaving MHC.

To examine differences in time to rearrest after court exit/ disposition, we used the Cox proportional hazards model, a type of survival analysis that allows us to model the time expected to elapse before an event while controlling for all predictor variables from the final models in the logistic regression analyses. In pre- dicting time to first rearrest in the year after court exit/disposition, we found that both MHC participation and MHC completion are significant (hazard rate � 0.814, 0.546, respectively; models not shown). The survival analysis also generated a “life table” that calculates the probability that a terminal event (rearrest, in this case) will occur at specified time intervals.

Figure 1, a graphic presentation of the life table, shows the estimated cumulative probability of rearrest for MHC participants and MHC eligibles in SSU. The cumulative survival rate across the y-axis illustrates the proportion of sample members who had not yet been rearrested after court exit/disposition. As shown in the figure, SSU eligibles not in MHC had a significantly shorter time to rearrest. The distance between the curves is the estimated

postdisposition positive effect of MHC at any time point. For example, at 9 months, 24.2% of the comparison group had been rearrested compared with only 17.6% of MHC participants.

Figure 2 presents the same graphic presentation, but with MHC participants divided into completers and noncompleters. These results on time to postexit arrest reaffirm what our previous anal- yses have shown: It was MHC completers who accounted for the differences in recidivism between MHC participants and SSU eligibles. We see the MHC completers have a longer time to postexit arrest. At Month 9 postexit/disposition, 11.8% of MHC completers had been arrested compared with 23.9% of MHC noncompleters and 24.2% of SSU eligibles.

Discussion and Conclusions

This study adds to the accumulating evidence of the effective- ness of MHCs in reducing recidivism among offenders with severe mental illness. One year after participants exited the court, the proportion arrested was significantly lower than in the year before their MHC entry, and significantly lower than that of other mis- demeanor defendants with severe mental illness who had compa- rable services but who were processed in traditional criminal court. MHC participants also averaged fewer arrests and had a longer time to rearrest than the comparison group. Only a few of either group were rearrested for a felony, and there was no significant difference between the two groups in felony rearrests.

Although the comparison group was similar to MHC partici- pants in having severe mental illness, and in age, race, and offense level and type, it was not fully comparable. Its members were significantly more likely to be males and illegal drug users, and to have more prior arrests. Nonetheless, members of the comparison group averaged significantly fewer arrests in the year after dispo- sition of their cases than in the year before their key arrests. This study, thus, shows prepost reduction of arrests not only for MHC but also for a pretrial release diversion program with case man- agement, linkage to treatment and services, and in-house drug testing and treatment for those with illegal drug use.

This MHC’s graduation rate among those leaving was midway in the range reported by other studies (31% to 80%; see Dirks- Linhorst & Linhorst, 2012; Herinckx et al., 2005; Hiday & Ray, 2010; McNiel & Binder, 2007; Moore & Hiday, 2006; Redlich et al., 2010). Like these other studies that examined MHC outcomes by completion, this study found that it was participants who completed or graduated from MHC who made the greatest gains.

Table 7 Differences in the Number of Arrests One Year Before and After Mental Health Court/Case Disposition

Proportion rearrested (%)

Arrests before Arrests after

N M SD M SD M 95% CI SD t

MHC 408 27.5 1.38 0.82 0.38 0.73 �1.00 0.90, 1.09 1.01 19.83�

Completers 238 17.6 1.32 0.76 0.21 0.48 �1.11 1.01, 1.23 0.86 19.97�

Noncompleters 170 41.2 1.46 0.88 0.64 0.93 �0.82 0.65, 1.00 1.17 9.15�

SSU 687 37.3 1.67 1.08 0.60 0.96 �1.07 0.98, 1.17 1.26 22.20�

All clients 1095 34.0 1.57 0.99 0.52 0.88 �1.05 0.97, 1.11 1.18 29.31�

Note. MHC � mental health court; SSU � specialized supervision unit. � p � .001.

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an Ps

yc ho

lo gi

ca l

A ss

oc ia

tio n

or on

e of

its al

lie d

pu bl

is he

rs .

T hi

s ar

tic le

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

6 HIDAY, WALES, AND RAY

T ab

le 8

L og

is ti

c R

eg re

ss io

n P

re di

ct in

g A

ny R

ea rr

es t

M od

el 1

M od

el 2

M od

el 3

M od

el 4

B (S

E )

E xp

b (9

5% C

I) B

(S E

) E

xp b

(9 5%

C I)

B (S

E )

E xp

b (9

5% C

I) B

(S E

) E

xp b

(9 5%

C I)

M H

C �

0. 45

(0 .1

4) � � �

0. 64

(0 .4

9, 0.

83 )

� 0.

41 (0

.1 4)

� � �

0. 66

(0 .5

1, 0.

87 )

� 0.

25 (0

.1 4)

� 0.

78 (0

.5 9,

1. 03

) �

0. 29

(0 .1

5) � � �

0. 75

(0 .5

6, 1.

00 )

A ge

0. 00

(0 .0

1) 1.

00 (0

.9 9,

1. 01

) �

0. 01

(0 .0

1) �

0. 99

(0 .9

8, 1.

01 )

0. 00

(0 .0

1) �

1. 00

(0 .9

8, 1.

01 )

Fe m

al e

� 0.

24 (0

.1 3)

0. 78

(0 .6

0, 1.

02 )

� 0.

25 (0

.1 4)

0. 78

(0 .5

9, 1.

02 )

� 0.

27 (0

.1 4)

0. 76

(0 .5

8, 1.

01 )

W hi

te �

0. 34

(0 .2

5) 0.

71 (0

.4 3,

1. 16

) �

0. 25

(0 .2

6) 0.

78 (0

.4 7,

1. 29

) �

0. 19

(0 .2

7) 0.

83 (0

.4 9,

1. 40

) D

ru g

us e

(c he

ck -i

n) 0.

47 (0

.1 4)

� �

1. 61

(1 .2

2, 2.

12 )

0. 38

(0 .1

5) � �

1. 46

(1 .0

9, 1.

94 )

N um

be r

of pr

io r

ar re

st s

0. 42

(0 .0

7) � � �

1. 52

(1 .3

3, 1.

73 )

0. 37

(0 .0

7) � � �

1. 44

(1 .2

6, 1.

65 )

A ny

pr e-

di sp

os iti

on ar

re st

1. 22

(0 .1

6) � � �

3. 40

(2 .5

1, 4.

62 )

� 2

lo g

lik el

ih oo

d �

2 13

86 .8

7� � �

13 81

.9 3�

� �

13 25

.8 7�

� �

12 63

.2 2�

� �

N ot

e. M

H C

� m

en ta

l he

al th

co ur

t. �

p �

.0 5.

� �

p �

.0 1.

� � �

p �

.0 01

T ab

le 9

L og

is ti

c R

eg re

ss io

n P

re di

ct in

g A

ny R

ea rr

es t

M H

C C

om pl

et er

s an

d N

on co

m pl

et er

M od

el 1

M od

el 2

M od

el 3

M od

el 4

B (S

E )

E xp

b (9

5% C

I) B

(S E

) E

xp b

(9 5%

C I)

B (S

E )

E xp

b (9

5% C

I) B

(S E

) E

xp b

(9 5%

C I)

C om

pl et

er �

1. 02

(0 .1

9) � �

0. 36

(0 .2

5, 0.

52 )

� 0.

98 (0

.1 9)

� �

0. 38

(0 .2

6, 0.

55 )

� 0.

77 (0

.2 0)

� �

0. 47

(0 .3

2, 0.

69 )

� 0.

71 (0

.2 0)

� �

0. 49

(0 .3

3, 0.

73 )

N on

co m

pl et

er 0.

16 (0

.1 7)

1. 18

(0 .8

4, 1.

66 )

0. 20

(0 .1

8) 1.

22 (0

.8 7,

1. 73

) 0.

25 (0

.1 8)

1. 29

(0 .9

0, 1.

84 )

0. 13

(0 .1

9) 1.

14 (0

.7 9,

1. 66

) A

ge 0.

00 (0

.0 1)

0. 99

(0 .9

9, 1.

01 )

0. 00

(0 .0

1) �

0. 99

(0 .9

8, 1.

01 )

0. 00

(0 .0

1) �

0. 99

(0 .9

9, 1.

01 )

Fe m

al e

� 0.

26 (0

.1 4)

0. 77

(0 .5

9, 1.

01 )

� 0.

25 (0

.1 4)

� 0.

78 (0

.5 9,

1. 02

) �

0. 26

(0 .1

4) 0.

77 (0

.5 8,

1. 02

) W

hi te

� 0.

21 (0

.2 6)

0. 81

(0 .4

9, 1.

33 )

� 0.

16 (0

.2 6)

� 0.

85 (0

.5 1,

1. 42

) �

0. 12

(0 .2

7) 0.

89 (0

.5 2,

1. 50

) D

ru g

us e

(c he

ck -i

n) 0.

34 (0

.1 5)

� 1.

41 (1

.0 6,

1. 87

) 0.

27 (0

.1 5)

1. 31

(0 .9

8, 1.

76 )

N um

be r

of pr

io r

ar re

st s

0. 41

(0 .0

7) � �

1. 51

(1 .3

3, 1.

72 )

0. 36

(0 .0

7) � �

1. 44

(1 .2

6, 1.

64 )

A ny

pr e-

di sp

os iti

on ar

re st

1. 17

(0 .1

6) � �

3. 21

(2 .3

6, 4.

37 )

� 2

lo g

lik el

ih oo

d �

2 13

59 .4

8� �

13 55

.2 4�

� 13

07 .4

7� �

12 51

.6 5�

�

N ot

e. M

H C

� m

en ta

l he

al th

co ur

t. �

p �

.0 5.

� �

p �

.0 01

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an Ps

yc ho

lo gi

ca l

A ss

oc ia

tio n

or on

e of

its al

lie d

pu bl

is he

rs .

T hi

s ar

tic le

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

7MENTAL HEALTH COURT EFFECTIVENESS

Graduates had a significantly smaller proportion arrested, fewer arrests, and a longer time to rearrest than noncompleters; and it was graduates who accounted for the recidivism differences be- tween MHC participants and the SSU comparison group.

This study went further than the four other postexit MHC studies, in that both the MHC and comparison groups were under court-ordered monitoring by the same pretrial services agency, and

were provided the same case management, drug testing and treat- ment, and mental health services linkage from the same case managers. Multivariate analyses, controlling for differences be- tween the MHC and comparison groups (sex, prior arrests, and illegal drug use), showed positive effects of MHC above and beyond case management, treatment, and services. Furthermore, the positive effects of this study’s MHC occurred even with a

MHC

SSU

Months Post MHC Exit/Case Disposition

C um

ul at

iv e

Su rv

iv al

R at

Figure 1. Time to rearrest for MHC and SSU.

Months Post MHC Exit/Case Disposition

C um

ul at

iv e

Su rv

iv al

R at

Noncompleters

SSU

MHC Completers

Figure 2. Time to rearrest for SSU, MHC completers, and noncompleters.

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an Ps

yc ho

lo gi

ca l

A ss

oc ia

tio n

or on

e of

its al

lie d

pu bl

is he

rs .

T hi

s ar

tic le

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

8 HIDAY, WALES, AND RAY

much shortened period of MHC participation or “dose” compared with other MHCs (average of 4 to 6 months vs. � 1 year).

This study, like other studies examining reoffending of persons with and without mental illness, found that illegal drug use and number of prior arrests were positively associated with rearrest (Bonta, Law, & Hanson, 1998; Hiday & Wales, 2013; Steadman et al., 2011). Defendants who use illegal drugs and have a history of more arrests represent the higher risk cases. Although riskier, they should not be excluded from diversion programs for mentally ill offenders because research has shown that many higher-risk cases can be successes in completing the programs and in not reoffend- ing (Marlowe et al., 2009; Rempel & DeStefano, 2001). The lower-risk cases, those who did not use illegal drugs and had one or no arrest in the year prior to key arrest, were less likely to reoffend (77% SSU, 89% MHC), but the majority of illegal sub- stance users in the current study were not rearrested (60% SSU, 65% MHC). Even a major proportion of those with three or more arrests in the year prior to key arrest did not reoffend (62% SSU, 50% MHC). Indeed, such higher-risk cases that succeed bring a greater benefit to cost ratio to a diversion program than lower-risk cases (Rossman et al., 2012).

A best practice for preadjudication diversion programs is to release mentally ill detainees as soon as possible from jail into the community (Osher, Steadman, & Barr, 2003; Steadman, McCarty, & Morrissey, 1988). Early release helps avoid incarceration’s negative effects and allows arrestees to begin receiving the treat- ment and services that it is hoped set them on a new path to avoid reoffending (Osher et al., 2003; Redlich, Liu, Steadman, Callahan, & Robbins, 2012; Steadman et al., 1988). In this study’s judicial district, pretrial release into the community occurs the day after key arrest for most persons charged with misdemeanors. Our data, however, indicated that it took longer to get a placement with the SSU for persons with severe mental illness (averaging 57.8 days). Delay in placement occurred because most such persons were released before screening, not identified at the initial screening, detained in jail because of previous failure to appear, or sent for competency evaluation to the mental hospital. More sensitive testing to discern severe mental illness and competency initially would seem desirable but would also take longer for all detainees and require more expertise. Hence, the benefits of such screening would need to be balanced against greater expense and possible delayed pretrial release, causing longer incarceration periods.

One criticism of MHCs, and one reason that defendants may decline an offer to enroll, is that MHCs extend the time during which defendants are under court supervision for mandated treat- ment and behavioral change (Redlich et al., 2012). In our study, case disposition took approximately three months longer, on av- erage, for MHC participants than for the comparison group; but the 4 to 6 months required to complete MHC was only partially responsible for MHC participants’ longer time. The comparison group’s earlier placement in SSU to obtain treatment and services accounted for almost half of the difference. One could contend that MHC participants are compensated for their longer average time from key arrest to disposition by their greater average time, with no supervision in the year after MHC exit/case disposition result- ing from their reduced recidivism. If their reduced recidivism continues, the compensation will be even greater.

Limitations

One might argue that the reduced recidivism of our study’s MHC, as that of most diversion programs, is caused by “cherry- picking” defendants who have characteristics perceived to produce compliance with treatment and behavioral mandates, and who are thus more likely to have reduced offending (Wolff & Pogorzelski, 2005). Indeed, almost all diversion programs have proportionately more older, White, and female participants than in the general jail population (Luskin, 2001; Naples, Morris, & Steadman, 2007). Our study’s MHC and SSU followed this pattern, with larger proportions being female and older than among all arrestees in the District. The larger proportion of females in SSU than in the jail population is to be expected, given that females in the criminal justice system have higher rates of mental illness than males in the criminal justice system (Steadman, Osher, Robbins, Case, & Samuels, 2009; Teplin, 1996). However, the proportion of females in this study’s MHC was even higher than the proportion in SSU, suggesting that factors other than mental illness accounted for their participation. It is possible that differential referral and certifica- tion practices were operative in selection not just of females but all MHC referrals from SSU. Case managers and calendar judges may have differentially referred to MHC those perceived as cooperative and of low public safety risk, which made them easier to link with community treatment providers in anticipation of the U.S. attor- ney’s unwritten standards for acceptance. The voluntary nature of MHC may also have introduced selection bias because defendants motivated to change and willing to comply with mental health treatment are more likely to accept the offer of MHC enrollment. Likewise, defense attorneys may have influenced selection, urging MHC only on those clients whom they thought most willing to change.

Another possible bias stems from a more overt unwritten stan- dard of the U.S. attorney that required no illegal drug use or an indication of motivation to control its use. Although some MHCs enroll substance abusers before they have clean screenings and work with them to become drug free, the U.S. attorney of this study’s MHC approved substance users temporarily, but required negative drug tests for at least two weeks before signing the deferred prosecution agreement for official MHC admission. Knowing this, SSU case managers and defense attorneys may have refrained from offering the possibility of MHC to SSU defendants who continued to test positive for one or more illegal drugs. Our control for illegal drug use at first check-in did not fully remove this bias because it failed to discriminate between those who did and did not become drug free.

Without random assignment, selection bias can easily occur and can call into question whether it is the intervention or sample selection that brings about an effect. Despite the likelihood of selection bias in this study’s MHC sample, stemming from referral and acceptance practices, it should not negate this study’s finding of positive effects of MHC for three reasons. First, our controls mitigated some of its effects in the multivariate analyses. Second, not all voluntary acceptances imply motivation to change. Our qualitative observations indicated that many defendants chose MHC believing they could con the court, especially given its short duration. Of these, some were converted by their experiences in MHC to put forth the effort to maintain new behavioral patterns. Third, and most importantly, without the monitoring, services, and

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an Ps

yc ho

lo gi

ca l

A ss

oc ia

tio n

or on

e of

its al

lie d

pu bl

is he

rs .

T hi

s ar

tic le

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

9MENTAL HEALTH COURT EFFECTIVENESS

treatment provided by the SSU, and the additional monitoring, supports, and sanctions provided by the MHC, even defendants desiring change and cherry-picked to be likely successes probably would not have been able to succeed at not reoffending, given their untreated mental illness, addictions, and established behavior pat- terns. Although we lacked data on socioeconomic status, resi- dence, living conditions, and resources, it was apparent in our court observations that a large proportion of MHC participants lived in impoverished conditions with inadequate resources, fac- tors that would also reduce their chances of not reoffending.

Selection bias likely affected differences not only between the MHC and comparison groups but also between completers and noncompleters through attrition. Noncompleters left MHC by opt- ing out voluntarily because they were no longer motivated to cooperate with treatment and make the mandated behavioral changes, or they were sent back to traditional criminal court involuntarily because of noncooperation and noncompliance. Completers who graduated and received a full dose of MHC got that full dose because they were willing and did cooperate with treatment and follow court mandates. Thus, motivation to change introduced bias in attrition. Nonetheless, it is possible that those motivated to change would not have been as likely to have their motivation sustained without both the MHC and SSU programs.

We did not have data on diagnosis, number of symptoms, or symptom type, or on time incarcerated (not at risk of arrest) in the follow-up period. Inability to control for incarceration time should not have impacted our findings because this study’s jurisdiction placed most misdemeanor offenders and nonviolent felony offend- ers on pretrial release the day after arrest. Further, as we have seen, few were rearrested for felonies and even fewer rearrested for violent felonies. Those found guilty of key arrest charges and incarcerated would reduce time at risk for comparison groups, making our findings for the impact of MHC and graduation con- servative. Inability to control for clinical variables should not have impacted our findings because multiple studies have shown that it is social or criminological factors, not clinical variables, which predict reoffending (Bonta et al., 1998; Hiday & Wales, 2013). A more important limitation was our inability to control for certain criminological factors, especially education, employment, and homelessness.

Our recidivism measures only included arrests in the District of Columbia. With the exception of the one multisite study (Stead- man et al., 2011), all MHC studies have employed rearrest data only from the city, county, or state of the single MHC under study (Christy et al., 2005; Cosden et al., 2003; Frailing, 2010; Hiday & Ray, 2010; McNiel & Binder, 2007; Moore & Hiday, 2006) or did not report the geographic scope of rearrest data (Broner, Lang, & Behler, 2009; Cosden, Ellens, Schnell, & Yamini-Diouf, 2005; Dirks-Linhorst & Linhorst, 2012; Herinckx et al., 2005). It is likely that some offending occurs beyond the borders of local or state jurisdictions. FBI national data indicated that 17.8% of our MHC and SSU sample was arrested outside of the District of Columbia in the year prior to key arrest. We can expect that a similar proportion of arrests occurred outside the district in the follow-up; thus, our recidivism rates probably represent an undercount, just as our 1-year prior arrests do. Unfortunately, we did not have access to FBI national data for the follow-up. To the extent that both MHC and SSU groups had similar proportions with arrests outside the District, recidivism comparison between the two groups should

not have been affected. The SSU group, however, had significantly more arrests outside the district in the year prior to key arrest (19.7% of SSU vs. 14.7% of MHC, p � .039, �2 � 4.276). Accordingly, it is likely that the undercount of post arrests outside the district is greater for the SSU group, making the differences reported in this study between MHC and SSU conservative.

A final limitation of this study is that it is of only a single court, which limits generalization. On the other hand, this MHC pos- sesses the 10 essential elements of a MHC postulated by the Council of State Governments (Almquist & Dodd, 2009). Then again, it is different from other MHCs in having a much shortened time for successful completion, 4 to 6 months as opposed to a year or more. A full dose of this study’s MHC is much less than a full dose of all other MHCs in the literature, except for one. That one MHC had no more than one hearing and no monitoring by the MHC for almost two-thirds of participants (Boothroyd, Poythress, McGaha, & Petrila, 2003). Nonetheless, the current study, as other studies, found that MHC participation was associated with reduced criminal recidivism, suggesting that regular monitoring and sup- port by the court for at least several months is one of the compo- nents of MHC that makes a difference in inducing change in offending behavior. Future research should attend to what other MHC components are mechanisms that make a difference in recidivism. Some studies have shown that MHCs have high levels of procedural justice (Poythress, Petrila, McGaha, & Boothroyd, 2002; Wales, Hiday, & Ray, 2010) and reintegrative shaming (Ray et al., 2011) that researchers suggest are important mechanisms in getting persons to obey the law (Braithwaite, 1989; Tyler, 2006; Wales et al., 2010). No study, however, has yet linked these process mechanisms empirically to reduced offending among MHC participants. Most empirical research on MHCs has yet to address questions of what it is about MHC participation that produces change, much less answer them.

References

Almquist, L., & Dodd, E. (2009). Mental health courts: A guide to research-informed policy and practice. New York, NY: Council of State Governments Justice Center. Retrieved from http://consensusproject .org/jc_publications/mental-health-courts-a-guide-to-research-informed- policy-and-practice

Bonta, J., Law, M., & Hanson, K. (1998). The prediction of criminal and violent recidivism among mentally disordered offenders: A meta- analysis. Psychological Bulletin, 123, 123–142. doi:10.1037/0033-2909 .123.2.123

Boothroyd, R. A., Poythress, N. G., McGaha, M., & Petrila, J. (2003). The Broward mental health court: Process, outcomes, and service utilization. International Journal of Law and Psychiatry, 26, 55–71. doi:10.1016/ S0160-2527(02)00203-0

Braithwaite, J. (1989). Crime, shame, and reintegration. New York, NY: Cambridge University Press.

Broner, N., Lang, M., & Behler, S. A. (2009). The effect of homelessness, housing type, functioning, and community reintegration supports on mental health court completion and recidivism. Journal of Dual Diag- nosis, 5, 323–356. doi:10.1080/15504260903358801

Broner, N., 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 and the Law, 22, 519–541. doi:10.1002/bsl.605

Christy, A., Poythress, N. G., Boothroyd, R. A., Petrila, J., & Mehra, S. (2005). Evaluating the efficiency and community safety goals of the

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an Ps

yc ho

lo gi

ca l

A ss

oc ia

tio n

or on

e of

its al

lie d

pu bl

is he

rs .

T hi

s ar

tic le

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

10 HIDAY, WALES, AND RAY

Broward County mental health court. Behavioral Sciences & the Law, 23, 227–243. doi:10.1002/bsl.647

Cosden, M., Ellens, J., Schnell, J., & Yamini-Diouf, Y. (2005). Efficacy of a Mental Health Treatment Court with assertive community treatment. Behavioral Sciences & the Law, 23, 199–214. doi:10.1002/bsl.638

Cosden, M., Ellens, J. K., Schnell, J. L., Yamini-Diouf, Y., & Wolfe, M. M. (2003). Evaluation of a mental health treatment court with assertive community treatment. Behavioral Sciences & the Law, 21, 415–427. doi:10.1002/bsl.542

Dirks-Linhorst, P. A., & Linhorst, D. M. (2012). Recidivism outcomes for suburban mental health court defendants. American Journal of Criminal Justice, 37, 76–91. doi:10.1007/s12103-010-9092-0

Frailing, K. (2010). How mental health courts function: Outcomes and observations. International Journal of Law and Psychiatry, 33, 207–213. doi:10.1016/j.ijlp.2010.06.001

Herinckx, H. A., Swart, S. C., Ama, S. M., Dolezal, C. D., & King, S. (2005). Rearrest and linkage to mental health services among clients of the Clark County mental health court program. Psychiatric Services, 56, 853–857. doi:10.1176/appi.ps.56.7.853

Hiday, V. A., Moore, M., Lamoureaux, M., & deMagistris, J. (2005). North Carolina’s mental health court. Popular Government, 70, 24–30. Re- trieved from http://sogpubs.unc.edu/electronicversions/pg/pgspsm05/ article3.pdf

Hiday, V. A., & Ray, B. (2010). Arrests two years after exiting a well- established mental health court. Psychiatric Services, 61, 463–468. Retrieved from http://ps.psychiatryonline.org/article.aspx? articleid�101356

Hiday, V. A., & Wales, H. W. (2013). Mental illness and the law. In C. S. Aneshensel, J. C. Phelan, and A. Bierman (Eds.), Handbook of the sociology of mental health (2nd ed., pp. 563–582). New York, NY: Springer. doi:10.1007/978-94-007-4276-5

Luskin, M. L. (2001). Who is diverted? Case selection for court-monitored mental health treatment. Law & Policy, 23, 217–236. doi:10.1111/1467- 9930.00111

Marlowe, D. B., Festinger, D. S., Arabia, P. L., Dugosh, K. L., Benasutti, K. M., & Croft, J. R. (2009). Adaptive interventions may optimize outcomes in drug courts: A pilot study. Current Psychiatry Reports, 11, 370 –376. Retrieved from http://link.springer.com/article/10.1007/ s11920-009–0056-3

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. doi:10.1176/appi.ajp.2007.06101664

Moore, M. E., & Hiday, V. A. (2006). Mental health court outcomes: A comparison of rearrest and rearrest severity between mental health court and traditional court participants. Law and Human Behavior, 30, 659– 674. doi:10.1007/s10979-006-9061-9

Naples, M., Morris, L. S., & Steadman, H. J. (2007). Factors in dispro- portionate representation among persons recommended by programs and accepted by courts for jail diversion. Psychiatric Services, 58, 1095– 1101. doi:10.1176/appi.ps.58.8.1095

Osher, F., Steadman, H. J., & Barr, H. (2003). A best practice approach to community reentry from jails for inmates with co-occurring disorders: The APIC model. Crime & Delinquency, 49, 79–96. doi:10.1177/ 0011128702239237

Palermo, G. B. (2010). The Nevada mental health courts. International Journal of Law and Psychiatry, 33, 214–219. doi:10.1016/j.ijlp.2010.06 .002

Poythress, N. G., Petrila, J., McGaha, A., & Boothroyd, R. (2002). Per- ceived coercion and procedural justice in the Broward mental health

court. International Journal of Law and Psychiatry, 25, 517–533. doi: 10.1016/S160-2527(01)00110-8

Ray, B., Dollar, C. B., & Thames, K. M. (2011). Observations of reinte- grative shaming in a mental health court. International Journal of Law and Psychiatry, 34, 49–55. doi:10.1016/j.ijlp.2010.11.008

Redlich, A. D., Liu, S., Steadman, H. J., Callahan, L., & Robbins, P. C. (2012). Is diversion swift? Comparing mental health court and tradition criminal justice processing. Criminal Justice and Behavior, 39, 420– 433. doi:10.1177/0093854811432424

Redlich, A. D., Steadman, H. J., Callahan, L., Robbins, P. C., Vessilinov, R., & Ozdgru, A. A. (2010). The use of mental health court appearances in supervision. International Journal of Law and Psychiatry, 33, 272– 277. doi:10.1016/j.ijlp.2010.06.010

Redlich, A. D., Steadman, H. J., Monahan, J., Robbins, P. C., & Petrila, J. (2006). Patterns of practice in mental health courts: A national survey. Law and Human Behavior, 30, 347–362. doi:10.1007/s10979-006- 9036-x

Rempel, M., & DeStefano, C. D. (2001). Predictors of engagement in court-mandated treatment: Findings at the Brooklyn treatment court, 1996 –2000. Journal of Offender Rehabilitation, 33, 87–124. doi: 10.1300/J076v33n04_06

Rossman, S. B., Willison, J. B., Mallik-Kane, K., Kim, K., Debus-Sherrill, S., & Downey, P. M. (2012). Criminal justice interventions for offenders with mental illness: Evaluation of mental health courts in Bronx and Brooklyn, New York. Washington, DC: National Institute of Justice. Retrieved from http://www.ncdsv.org/images/UI_ CJInterventionsOffendersWithMHEvalsOfMHCourtsInBronxBrooklynNY_ 2-2012.pdf

Steadman, H. J., McCarty, D. W., & Morrissey, J. P. (1988). The mentally ill in jail: Planning for essential services. New York, NY: Guilford Press.

Steadman, H. J., Osher, F. C., Robbins, P. C., Case, B., & Samuels, S. (2009). Prevalence of serious mental illness among jail inmates. Psy- chiatric Services, 60, 761–765. doi:10.1176/appi.ps.60.6.761

Steadman, H. J., Redlich, A., Callahan, L., Robbins, P. C., & Vesselinov, R. (2011). Effect of mental health courts on arrests and jail days: A multisite study. Archives of General Psychiatry, 68, 167–172. doi: 10.1001/archgenpsychiatry.2010.134

Teplin, L. A. (1996). Prevalence of psychiatric disorders among incarcer- ated women. 1. Pretrial jail detainees. Archives of General Psychiatry, 53, 505–512. doi:10.1001/archpsyc.1996.01830060047007

Trupin, E., & Richards, H. (2003). Seattle’s mental health courts: Early indicators of effectiveness. International Journal of Law and Psychiatry, 26, 33–53. Retrieved from http://www.floridatac.org/files/document/ Trupin_IJLP_Jan03.pdf

Tyler, T. R. (2006). Why people obey the law. Princeton, NJ: Princeton University Press.

Wales, H. W., Hiday, V. A., & Ray, B. (2010). Procedural justice and the mental health court judge’s role in reducing recidivism. International Journal of Law and Psychiatry, 33, 265–271. doi:10.1016/j.ijlp.2010.06 .009

Wolff, N., & Pogorzelski, W. (2005). Measuring the effectiveness of mental health courts: Challenges and recommendations. Psychology Public Policy and Law, 11, 539–569. doi:10.1037/1076-8971.11.4.539

Received June 14, 2012 Revision received January 31, 2013

Accepted February 8, 2013 �

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an Ps

yc ho

lo gi

ca l

A ss

oc ia

tio n

or on

e of

its al

lie d

pu bl

is he

rs .

T hi

s ar

tic le

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

11MENTAL HEALTH COURT EFFECTIVENESS

View publication statsView publication stats

https://www.researchgate.net/publication/239732869