Among justice-involved people with serious mental illness and co-occurring substance use disorders, those who repeatedly commit misdemeanors are perhaps the most difficult to effectively divert into services from the criminal justice system. Despite extensive criminal histories, with today’s overcrowded jails they face relatively little jail time. Offered a choice between a few days in jail or 12 to 24 months of court supervision, they often serve the jail sentence on recommendation of defense counsel.
In 2002, the New York City Mayor’s Office partnered with the Center for Alternative Sentencing and Employment Services to develop a strategy for engaging this population in services. This partnership lead to the development of EXIT, a jail diversion program for justice-involved people with mental illness who are processed through Manhattan’s Criminal Court.
At arraignment, a forensic clinical coordinator screened referred individuals for serious mental illness and program eligibility standards: nonviolent misdemeanor instant offense, at least three prior misdemeanor convictions, and a possible 5 to 30 day jail sentence on the current charge.
Rather than divert people into a lengthy period of court supervision, EXIT emphasized voluntary access to services through a required three-hour Mandated Treatment Assessment Session (MTAS), which was conducted by staff at the program’s office immediately following sentence. The goals of the MTAS were to: 1) assess and address the participant’s immediate needs, including food, shelter, and clothing; 2) outline short- and medium-term goals the participant could pursue through nonmandated case management services; 3) explain the potential benefits of program engagement; and — if the individual accepted services — 4) establish mutually agreed-upon expectations, including means for maintaining contact, level and frequency of contact, and service goals.
After completing the MTAS, an individual could elect to participate in nonmandated case management services to address identified needs. The program coordinated services among various providers, and maintained as-needed contact with participants to ensure sufficient community supports necessary for stability and the reduction of risk for rearrest. Core program elements were drawn from identified best practices, focusing heavily on strengths-based engagement combined with intensive case management. EXIT established a strong commitment to consumer involvement at all stages of program planning, implementation, evaluation, and promotion. Two peer specialists were employed to serve as escorts to appointments and to provide other supportive services to participants and staff, including case consultation, as full members of the treatment team.
EXIT’s high engagement–low coercion model provided a path from the court to community-based treatment with minimal judicial oversight and no probation or parole monitoring. Beyond reporting completion of the MTAS, the program was not obligated to provide status updates on participants to the court.
As shown in Table 1 (below), bipolar, schizophrenia spectrum, and depressive disorders were about equally distributed among defendants who entered the program with a diagnosis. There were 31 of 173 (18 percent) individuals who could not specify a diagnosis, but were admitted to the program based on signs of mental illness apparent to clinical staff during screening.
EXIT participants were a needs-intensive group. In addition to serious mental illness, 87 percent reported current substance use and approximately half were homeless.
The largest number of participants (57) entered the program due to arrest for a property-related offense, followed by possession of a controlled substance (47). Although screenings comprised only 11 percent women, women were admitted to the program at a rate comparable to their male counterparts (43 percent, compared to 41 percent of all men screened). The average age of participants at intake was 39 years.
The EXIT program experienced increased levels of criminal justice buy-in over the life of the program as evidenced by the high utilization rate among judges. All but 23 of the 196 defendants found eligible were released to the program. This is significant given the initial reticence on the part of some judges to release defendants to the program due to concerns that the three-hour MTAS did not constitute a sufficiently stringent sanction. Moreover, judges expressed concern that the program’s voluntary case management model would neither allow for judicial oversight nor provide a compelling reason for participants to remained engaged with services.
Ninety-seven percent of defendants court ordered to complete the MTAS fulfilled their obligation to the court. Of the 168 defendants who completed the MTAS, 120 (71 percent) had subsequent nonmandated in-person contact with program staff. Two-month retention was at 54 percent, with 21 percent remaining engaged with the program for a minimum of six months. For those who remained engaged for a minimum of eight months, program contacts averaged approximately three per month.
A snapshot of 90 EXIT participants was selected for the purpose of analyzing conviction patterns. Participants with felony convictions in the 12 months before or after the MTAS were excluded, since it was expected that far fewer days at liberty would decrease their likelihood of reconviction on misdemeanor charges. EXIT participants with open cases were also excluded from the analysis. Nine individuals were excluded, leaving a cohort of 81. Across the cohort, there was an 18 percent reduction in the aggregate number of convictions in the year following program engagement compared to the year before, representing a decrease from 261 convictions to 214 convictions in the 12-month pre- versus post-MTAS periods [t(80) = 2.09, p=.039].
To determine whether participation in post-MTAS case management services had any effect on recidivism, the 81 participants were divided into three subgroups:
Groups were defined based on an analysis of case management engagement patterns across the entire sample pool. Of the 81-member cohort, 24 subjects (29.6 percent) had no contact, 25 (30.9 percent) had between one and nine contacts, and 32 (39.5 percent) had at least 10 post-MTAS case management contacts.
While all groups experienced a reduction in the aggregate number of convictions in the post- versus pre-MTAS period, the cohort with 10 or more post-MTAS case management contacts (Group 3) experienced the largest decline (24 percent, compared to 18 percent and 11 percent for Groups 2 and 1, respectively). Further analysis revealed that in the post-MTAS year this same Group 3 cohort comprised the highest number and percentage of individuals with no convictions (11, or 34 percent of cohort, representing 52.4 percent of the 21 subjects across
all groups with zero convictions in the post-MTAS year).
Based on the EXIT program data, the chronic patterns of both re-conviction and transient service engagement long associated with people with serious mental illness who repeatedly commit misdemeanors can be interrupted through nonmandated engagement in services. It also suggests that the program services provided by EXIT were viable and responsive to individual needs, as evidenced by the number of participants who remained engaged in program services for periods up to and exceeding six months, and as confirmed through consumer feedback.
The presumption that mandated engagement would have yielded lengthier program tenure rates is tempered by several considerations. First, the aggregate and cohort conviction rate decline suggest that retention drop off is not necessarily indicative of undesirable outcome. Drop off could have reflected more positive alternatives such as reduced reliance on EXIT resulting from the fulfillment of immediate service needs or successful transition to permanent providers. Also compelling is the possibility that retention rates may have been increased with enhanced staffing as opposed to imposition of mandate. For example, during the program’s second year, when it was fully staffed, the minimum six-month retention rate of 35 percent approximated the three month rate averaged over the life of the program (36%).
EXIT demonstrates that people with mental illness who repeatedly commit misdemeanor offenses can engage voluntarily and remain engaged in services beyond any court mandate, with significantly reduced recidivism as an outcome.
Recommended citation: Foley, G., & Ruppel, E. (2008). The EXIT program: Engaging diverted individuals through voluntary services. Delmar, NY: CMHS National GAINS Center.
Mental Health Diagnosis |
Number |
Percent |
Schizophrenia/Schizoaffective Disorder |
38 |
22 |
Bipolar Disorder |
37 |
21 |
Depressive Disorder |
36 |
21 |
Anxiety |
2 |
1 |
Two or More Diagnoses |
29 |
17 |
Diagnosis Unavailable |
31 |
18 |
Intake Arrest Charge |
||
Property Crime |
57 |
33 |
Possession of Controlled Substances |
47 |
22 |
Theft of Services |
12 |
7 |
Trespassing |
12 |
7 |
Disorderly Conduct |
12 |
7 |
Forgery Crimes |
8 |
5 |
Criminal Tampering/Criminal Mischief |
6 |
3 |
Criminal Possession of a Weapon |
5 |
3 |
Other/Unknown |
14 |
8 |
Gender |
||
Male |
150 |
87 |
Female |
20 |
12 |
Other |
3 |
1 |
Race/ Ethnicity |
||
Non-Hispanic African American |
108 |
63 |
Hispanic |
35 |
20 |
Non-Hispanic Caucasian |
28 |
16 |
Other |
2 |
1 |
Age Range |
||
18-29 years |
32 |
18 |
30-39 years |
51 |
29 |
40-49 years |
66 |
39 |
50-59 years |
23 |
13 |
60+ years |
1 |
1 |
1EXIT Project Director, New York City Mayor’s Office of the Criminal Justice Coordinator, New York, New York
2 Research Associate, Center for Alternative Sentencing and Employment Strategies, New York, New York
| Gains@prainc.com | The CMHS National GAINS Center | GainsTAPA@prainc.com |
| 800.311.GAIN | 866.518.TAPA | |
Funded by the Center for Mental Health Services of the Substance Abuse and Mental Health Services Administration < Privacy Policy > For Alternative Access to Web Documents : Email gains@prainc.com  USA.gov is the U.S. government's official web portal to all federal, state and local government web resources and services. |
||