Expert Panel Meeting
Caterina Gouvis Roman, Elizabeth Cincotta McBride, and Jenny W. L. Osborne
Presented at the Evidence-Based Practice for Justice-Involved Individuals: Housing
Expert Panel Meeting
June 1, 2005
Bethesda, MD
The Expert Panel Meetings were Sponsored by: The National GAINS Center for Systemic Change for Justice-Involved Persons through the Center for Mental Health Services, SAMHSA.
Acknowledgement & Disclaimer: The views and opinions expressed in this publication are those of the authors, and do not necessarily represent the official position or policies of the National GAINS Center (operated by Policy Research Associates, Inc.) or the Center for Mental Health Services (CMHS).
This paper presents an overview of housing models used for persons with mental illness who have had contact with the criminal justice system. Our intent is to synthesize available knowledge regarding evidence-based programming when providing housing for this population. While many programs exist and are currently being developed to serve this population, there is little evaluation research from which to draw conclusions about what works. Hence, we focus on extracting best practices across the various literatures—homeless assistance, mental health and criminal justice—that inform housing programs for persons with mental illness who have had contact with the justice system. In addition, we seek to develop a sound framework from which to draw conclusions about key components of promising programs and that future studies can utilize to measure outcomes.
In the past two decades a number of circumstances have coincided to result in the dramatic growth in the number of persons with mental illness who have had contact with the justice system and who are at risk of homelessness. These circumstances include the de-institutionalization of persons with mental illness from psychiatric facilities, large increases in the number of prisoners returning to their communities, cuts in public assistance, and declines in the availability of safe and affordable housing. The confluence of these and other factors has created serious challenges to public safety, public health and the overall quality of life of individuals and communities nationwide.
It has been estimated that thousands of persons with mental illness exit prisons and jails each year (Beck & Maruschak, 2001; Harlow, 1998). Correctional facilities are unlikely to provide optimal treatment for those persons with mental illness, let alone those persons with both substance use problems and mental illness. And many of these individuals do not receive any type of mental health services while incarcerated—possibly reducing the chances of successful community reintegration. A Bureau of Justice Statistics study found that only 60 percent of prisoners suffering from major mental illnesses receive mental health treatment while in prison (Ditton, 1999). Furthermore, persons with mental illness—and in general most individuals—are too often released without any type of aftercare or pre-release planning (Beck & Maruschak, 2001; Petersilia, 2003; Travis, 2005). Another Bureau of Justice Statistics study (Beck & Maruschak, 2001) found that only two-thirds of prisons help released prisoners obtain mental health services in the community, post-release.
Without attention to mental health needs, individuals who have been involved with the criminal justice system may be at high risk of continuing criminal behavior and/or substance abuse. A study tracking 261 persons with mental illness who were released from jail in Ohio found that 72 percent were re-arrested within one year (Ventura, Cassel, Jacoby, & Huang, 1998). Of those arrested, 15 percent were arrested for violent felonies and 23 percent were arrested for violent misdemeanors. Individuals who received case management in the community were less likely to be rearrested than subjects who received no case management. In a longitudinal study of prisoner reentry in Maryland known as Returning Home, mental health status had implications for measures of post-release success (Malik-Kane, 2005). Respondents with a mental health condition were significantly more likely to report drug use or alcohol intoxication post-release than those without such conditions. The study also found that only half of the respondents who had been taking medications regularly in prison were still taking medications four to eight months after release.
Similarly, research has shown that the provision of housing to persons who have mental illness and have been involved with the justice system can enhance an individual’s chances of becoming self-sufficient. A recent study of California’s AB2034 housing programs for persons with mental illness found that individuals provided with housing were much more likely to remain in programs designed to increase residential stability than individuals not provided with housing (Hunter, 2005). Overall, the importance of housing as a component of good mental health care has been documented (Mechanic, 2003). With regard to returning prisoners, research suggests that residential instability and incarceration are compounding factors influencing both later residential instability and reincarceration. A large study examining persons released from New York state prisons found that both having a history of shelter use and a history of incarceration increased the risk of subsequent reincarceration and shelter use (Metraux & Culhane, 2004). Furthermore, shelter usage post release increased the likelihood of reincarceration. Those individuals with links to the mental health system had considerably higher proportions of shelter stays and reincarcerations post release than those without links to the mental health system. A few other studies have found that persons with mental illness who experience housing instability are more likely to come in contact with the police and/or be charged with a criminal offense (Brekke, Prindle, Bae, & Long, 2001; Clark, Ricketts, & McHugo, 1999). Furthermore, there is new evidence that returning prisoners view housing as a key component—perhaps even the most important component—of successful community reintegration (Malli-Kane, 2005).
Regardless of mental health status, securing housing upon release from prison or jail is a challenge for many (Roman & Travis, 2004). The process of securing housing is often complicated by a host of factors: the scarcity of affordable housing and general lack of service-enhanced housing, formal and informal regulations and prejudices that restrict tenancy and the development of new housing for this population, and strict eligibility requirements for food stamps, veterans’ benefits and benefits through the Temporary Assistance for Needy Families program (TANF). For persons with mental illness the challenges are compounded. Mental health practitioners repeatedly voice concerns that community-based services for special needs populations only serve a very small fraction of those in need.
For those individuals searching for housing who do not live with family or friends the housing options include: the private market; federally-subsidized and administered housing; corrections-based housing facilities; transitional (service-enriched) housing (non-corrections based and non-HUD funded); homeless assistance supportive housing; other service-enhanced housing, and special needs housing supported through U.S. Department of Housing and Urban Development (HUD). This report discusses these housing options with the exception of the private market and federally-subsidized and administered housing.
As stated earlier, the goal of this paper is to collect evidence on successful housing program models that serve persons with mental illness who have been in contact with the justice system. We begin this paper with the assumption that persons with mental illness returning from prisons and jails or those with previous justice system contact who are at risk for homelessness would benefit from the provision of housing. Some research indicates that not all persons with mental illness at risk of homeless need to be provided with housing (see, for example, Clark & Rich, 2003). However, this paper focuses on “what works when housing is provided” as opposed to “for whom housing should be provided.”
We define housing models or programs as those programs that provide a place to sleep and provide at least some access to support services intended to increase residential stability and self-sufficiency. This definition excludes case management programs or other community-based programs designed to increase residential stability but that do not offer housing. We also exclude those programs that only offer vouchers. Our review includes discussion of some residential treatment programs designed to serve persons with mental illness who are also chemical abusers, but we only included those programs that had stated goals relating to decreasing homelessness and/or increasing residential stability. We also reviewed housing programs that had the explicit goal of reducing criminal behavior—although the overwhelming majority of these programs did not have associated outcome data. We define justice system contact in broad terms—contact may involve incarceration in jail or prison, or arrest with some type of diversionary program provided. The individuals in this population include both those returning directly from jails and prisons and those in the community who have had previous criminal justice system contact.
Despite the lack of evaluation data for housing programs that serve persons with mental illness who have been in contact with the justice system, these programs generally represent themselves as evidenced-based. Our review found few housing programs that link the program configuration and services delivered to demonstrated housing-related and criminal justice outcomes. However, many of these programs have recently begun to collect a range of outcome data. Furthermore, these programs are based on models that have been demonstrated to be effective in mental health and homelessness prevention practice. In current practice, evidence-based principles for successful housing programs that serve those who are mentally ill have been translated into principles hypothesized to work in developing and running housing programs for previously incarcerated people with mental illness.
In order to provide sufficient foundation to enable more progress toward evidence-based programming, we organized this paper into four sections:
There are numerous issues that complicate an evidence-based practice synthesis of the literature on housing programs for persons with mental illness. There is no pre-packaged model for providing housing to persons with mental illness. Similarly, the evidence establishing that certain programs improve outcomes is particular to specific models, approaches, or housing configurations operating in the studied jurisdiction; the numerous combinations of variables across studies renders useless any comparison of housing models, with the exception of comparisons conducted within studies. The evidence overwhelmingly suggests that varied types of housing are needed to meet the heterogeneous needs of such a diverse group (Lipton, Siegel, Hanningan, Samuels, & Baker, 2000). In addition, program evaluation research examining supportive housing programs for persons with mental illness is generally based on weak evaluation design—rigorous experimental designs such as random assignment are rarely possible and comparison groups are not often used. Furthermore, outcome variables examined across studies vary widely. Studies comparing types of supervision or service structure rarely control for specific attributes of the housing setting, such as structure type and number of units, or the level of integration of persons with mental illness with persons without psychiatric disability. These factors have been shown to influence outcomes, yet few studies exist that examine a comprehensive list of such variables. Finally, a common vocabulary has failed to emerge—practitioners and researchers utilize different terms (e.g., supported, supportive, etc.), and additionally, there seems to be little consensus with regard to the structure of service models (e.g., service configurations of community residences, group home, etc.). These issues have the potential to complicate future research examining housing models for persons with mental illness who have been previously incarcerated.
To mitigate some of these issues, we propose a general framework below to specify the various and often overlapping dimensions related to housing program models for persons with mental illness. The framework contains five broad dimensions or features of housing programs. Although the model displays the five dimensions in a vertical manner, placement does not necessarily signify levels or ordering of the dimensions. We believe the framework is an important step in moving the field toward evidence-based practice and programming. It is our intent that the draft framework will begin to elucidate the significant components that comprise evidence-based housing models. Undoubtedly, the complexity of the service needs of the targeted population and wide variation across housing programs hampers systematic assessment and rigorous evaluation. Conceptual frameworks can be useful tools in providing strong foundations for understanding these complexities and more systematically evaluating programs. Basically, we intend to use the framework to provide organization to this report, as well as to assist in laying out the key components that have been hypothesized to relate to the achievement of positive outcomes, regardless of whether the outcomes are related to residential stability, mental health status, or criminal behavior.
We hope that the panel discussion can focus on the refinement of the framework from which to examine key characteristics related to best practices of housing programs. The framework proposed here is a barebones framework that can serve to guide the panel discussion.
|
Figure 1. Draft Framework for Examining Dimensions of Housing Models |
|
Housing models for persons with mental illness generally exist on a continuum of approaches from “housing first” to “housing ready.” Within the continuum exists other approaches including “harm reduction.” These approaches are, in essence, underlying principles that guide the provision of housing and services to individuals who are homeless or have been deemed “hard to house.”
The housing first approach offers the direct placement from streets to housing with support services available, but not required. Often, the only requirements are that individuals not use substances on the premises and abide by the traditional lease obligations of paying rent and refraining from violence and destruction of property. Tenant stability is a central factor in the housing first approach, with the idea that mentally ill and chemically dependent tenants will be unable to fulfill the obligations of their lease without taking advantage of the support services available to them. Services are viewed as separate from housing and are usually off-site with little to no mandated participation. Staff often use motivational techniques to suggest services and services are made available when they are desired by consumers.
Some housing first approaches, however, institute some requirements for services, but the general philosophy tends towards non-integration of housing with services.
In summary, critical elements of this approach include (Clarke Institute of Psychiatry, 1998):
“Housing ready” housing starts with treatment and progresses through a series of increasingly less service-intensive options with the promise of permanent supported housing as people are “ready.” Housing is transitional; services are high demand (generally some form of residential treatment) where the receipt of some package of services is a condition of participation in the program. As residents progress through levels of readiness, they are often moved from apartment to apartment. However, it is important to note that many housing ready programs do not provide or have access to an adequate supply of permanent supported housing.
As stated above, approaches underlying housing models for persons with mental illness generally exist on a continuum from housing first to housing ready. However, program directors and developers may not always specify or name their approach. Our review of the literature also uncovered harm reduction as a general approach to housing. Harm reduction programs can fall anywhere along the continuum depending on the nature of the service integration and requirements. The harm reduction approach generally applies to substance abuse treatment services, but can be used as a strategy to guide the operations of either permanent or transitional supportive housing (Burt, Hedderson, Zweig, Ortiz, Aron-Thurnham, & Johnson, 2004). For substance abusing clients, the basic idea is to first promote safer use, then managed use and then abstinence if possible. Research indicates that a harm reduction approach has great potential for preventing and ending homelessness (Burt et al., 2004). In addition, the harm reduction approach is being applied to programs that serve persons with mental illness returning from prisons and jails.
“Service-related Approaches” is the term we use to capture the expectations and requirements for service participation and the ways in which programs sanction noncompliance with services. Although requirements and configurations of services vary tremendously across programs, service-related models cluster along a continuum from low demand to high demand. Although the outcome evaluations we reviewed often did not examine specific service-related models, let alone specify the service-related model, the literature describing housing programs seems to suggest that the service component is a key variable that will impact outcomes. Hence, we feel that it is necessary to create a framework that utilizes service requirements a key component of housing models, as well as to disaggregates service-related models from overall approaches.
The U.S. Department of Housing and Urban Development (HUD) defines low demand as:
“[T]he provision of health care, mental health, substance abuse, and other supportive services and referrals for services in a non-coercive manner, which may include medication management, education, counseling, job training, and assistance in obtaining entitlement benefits and in obtaining other supportive services including mental health treatment and substance abuse treatment.”
There is often great variation in the types of services provided across low demand programs. These flexible programs are designed to be conducive to engaging individuals who are distrustful of the service system. Case managers or house staff are expected to establish trusting relationships with residents and to broker services when requested.
Housing programs with high demand services are usually those programs designed to target special needs populations and provide services matched to particular needs. Intensive services are intended to facilitate the transition to permanent housing. The underlying premise of structured services is that structured, intensive services provide individuals with mental illness or chemical addiction the skills needed to function independently. Case managers are utilized as advocates, counselors, skills instructors and service brokers. Programs may use either team case management or individual case management approaches. Clinical treatment is usually a requirement of these programs, and residents often engage in other life skills training.
As with low demand programs, there is wide variation in the specific types of services provided, as well as combinations of services. Supervision is usually provided around the clock and residents’ sobriety, medication compliance, and attendance at services are closely monitored. Residents often have curfews, are required to submit to random drug testing, and to sign attendance forms for off-site programs (Barrow & Zimmer, 1999). Most often, these programs are transitional in nature, providing housing from three months to two years. As residents gain control and independence by demonstrating compliance and skills development, they are deemed ready to graduate.
With regard to our framework, we define two basic programmatic models in order to facilitate discussion of best practices in housing for persons who have had criminal justice system contact: (1) housing programs that have no specified eligibility criteria relating to the transition from prison or jail to the community (non-reentry), and (2) those programs that specifically target individuals returning from prisons and jails (“reentry housing programs”). Because there is a high likelihood that homeless persons with mental illness have been arrested or incarcerated, it is highly probable that many housing programs designed to serve persons with mental illness serve persons with criminal justice system contact. However, we did not find any evidence of programs specifically serving persons with criminal justice system contact unless they were designed as reentry programs.
More specifically, reentry housing programs can be defined as those programs that recruit clients directly from jails and prisons. Often, clients are screened for eligibility while they are still in prison or jail. Reentry programs may be funded by the corrections system, but many programs are funded through a mix of homeless assistance funding from federal and state agencies, as well as through private donations. As with non-reentry programs, innovation often comes in the form of using blended funding streams. Regardless of funding, reentry housing programs usually have close relationships with corrections and community corrections agencies.
Across the reentry and non-reentry program models, programs will vary on a vast number of components such as: eligibility criteria, types and configuration of services provided, length of stay, staffing and supervision, and building configuration. Our review of the descriptive and evaluation literature indicates that of these components, length of stay and configuration are useful ways to categorize program models.
Housing programs exist on a continuum from emergency shelter to permanent housing. The housing literature usually categorizes length of stay into four types: (1) emergency housing, (2) short-term transitional, (3) long-term transitional, and (4) permanent. Because shelters generally do not offer services designed to increase residential stability and reduce criminal behavior, we have not included shelters in our framework.
Short-term transitional housing programs have a finite length of stay, which may vary anywhere from one month to three months (or more depending on definitions). In most cases, consumers do not have occupancy agreements or leases. Social service provision is often the primary focus of short-term transitional programs, rather than housing. Hence, most often, services are structured and high demand. Configurations of transitional housing programs vary widely from barracks-type facilities, to shared living spaces, to individual apartments or houses. Programs most often will be site specific, but programs exist that are scattered site.
Corrections-based “halfway houses” are an example of short-term transitional housing, but few corrections-based programs exist that primarily serve persons with mental illness. Those that exist generally allow consumers to reside in the programs for longer than three months (e.g., Independence House, in Denver, Colorado).
Long-term transitional housing programs generally have a time limit spanning from three months up to two years. These programs offer an extensive range of services that include case management, mental health and medical services, counseling and general issues groups, life and social skills groups, anger management, vocational and educational training, advocacy, and assistance obtaining benefits and identification information.
For persons returning from prison and jails, site-specific long-term transitional housing facilities are relatively common (Corporation for Supportive Housing, 2004). Fewer programs exist for this population when mental illness is added as a program criterion. The configuration of these programs varies, but most often is site-specific, where a dedicated facility is provided where all clients served are persons with mental illness. Some clustered scattered site programs also exist.
One example of a long-term transitional housing program for persons with mental illness that have had justice system contact is the modified therapeutic community. Modified therapeutic communities (MTCs) are residential treatment programs modified specifically to treat person with mental illness who are also substance abusers. MTCs are based on the traditional therapeutic community (TC) model for treating substance abuse. TCs, introduced in the 1960s, provide highly structured, supportive, residential drug treatment. This model engages a structured daily routine, emphasizes personal responsibility and self-help, and utilizes peers as role models and the community as the healing agent (DeLeon, 1997; Sacks, Skinner, & Peck, 2002; Swan, 1997). Within this strategy, community-building activities and peer and mentor relationships enable and facilitate individual change among clients.
Modifications made to the TC model to serve persons with mental illness vary across programs, but typically include: (1) increased flexibility, allowing clients to graduate through the program at their own pace; (2) decreased intensity, removing the confrontation element common in traditional TCs; and, (3) increased individualization, taking into consideration each client’s needs and abilities in crafting an action plan for program participation. Although these models have flexibility in structure and service provision, MTCs are usually considered high demand models.
Overall, with the exception of HUD’s National Survey of Homeless Assistance Providers and Clients (NSHAPC) conducted by the Urban Institute, little research exists cataloging or describing the vast array of transitional housing models in operation across the country, let alone those that explicitly serve persons with mental illness. The NSHAPC study estimates that there are 40,000 programs serving homeless people in the United States that meet the NSHAPC definition of a homeless assistance program of which transitional housing programs account for 11 percent. Of the estimated 4,390 transitional housing programs, 43 percent have no special focus, 9 percent primarily focus on mental health and 5 percent have a mental health chemical dependency focus as their primary focus. When programs were asked which types of special needs populations they served, regardless of primary focus, of the 4,390 transitional programs, 24 percent served persons with mental illness. Appendix B provides a table displaying service characteristics of the transitional programs surveyed.
Permanent housing programs for homeless people provide long-term housing assistance with support services. For most programs, homelessness is a primary requirement for program eligibility (Burt, Aron, Douglas, Valente, Lee, & Iwen, 1999: p.B-2). Although there are few permanent housing programs across the country that specifically house persons who have had justice system contact, some providers have recognized the need for permanent housing units accessible to people that lack the necessary credentials that are typically necessary to secure permanent housing, such as rental histories, identification and employment histories. Unlike transitional programs, permanent housing programs typically require residents to sign a tenancy agreement and although programs may have social service assistance on-site, participation is not a requirement for eligibility.
Permanent housing programs that offer supports are generally known as supportive housing. Services are designed to maximize independence, be flexible and responsive to individual needs, available when needed, and accessible (Corporation for Supportive Housing, 1996). Support services may include case management, referrals, basic life skills training, substance abuse and mental health treatment, basic health care, income support, education, employment-related services, transportation, clothing, advocacy, and child care (Burt et al., 2004). Service configurations vary from community to community, as well as within community. This variation is not always deliberate, as the variety of stakeholders involved must sort through availability of services and capacity of stakeholders that include funders, housing developers, property managers, and on-site and off-site program staff. Housing configurations will vary across programs (see discussion below).
General examples of permanent supportive housing include the Shelter Plus Care Program, the Section 8 Moderate Rehabilitation Program for Single-Room Occupancy (SRO) Dwellings, and the Permanent Housing for the Handicapped Homeless Program administered by the U.S. Department of Housing and Urban Development (HUD) (Burt et al., 1999).
HUD’s NSHAPC found that of the 1,920 permanent housing programs surveyed, 65 percent have no special focus, 16 percent focus primarily on mental health and 5 percent have a combined mental health chemical dependency focus as their primary focus. When programs were asked which types of special needs populations they served, regardless of primary focus, 33 percent served persons with mental illness and 27 percent served persons with mental illness and an alcohol or drug problem. Appendix B provides a summary of programs and services associated with those permanent housing programs.
Although reentry programs offering permanent housing are rare, we see evidence that the number of these programs is increasing. Our review of housing programs for persons with mental illness identified at least three housing programs that offered permanent housing (Iyana House, The MIX program, and St. Leonard’s Ministries). As with St. Leonard’s Ministries and the MIX program, some programs are designed to have clients progress through a continuum of housing from short-term to permanent. Other permanent housing programs not reviewed in the appendix include the LAMP program in Los Angeles. LAMP uses a low demand philosophy by offering multiple levels of housing through a non-linear housing model with voluntary supportive services.
Building configurations relevant to housing programs for persons with mental illness include single site structures, clustered site and scattered site. Single site configurations are stand-alone structures where residents occupy entire buildings. The buildings can be single room occupancy (SROs) or multi-unit dwellings or private homes. Clustered apartment programs group residents with mental illness together in a separate wing or floor of a larger facility that houses people without mental illness. Scattered-site models are apartment buildings, housing projects or housing units spread throughout a city or county. The units are dispersed throughout regular apartment buildings or projects.
Configuration is important because it will impact the amount of individual privacy, the availability and nature of service provision, as well as the degree of integration residents have with the larger community (Barrow & Zimmer, 1999). Within clustered units, supportive services may be provided on-site, but some programs use an off-site model for providing services. For clustered and scattered configurations, staff most often do not live on site, but are available on-call to assist in crisis situations. For more structured programs, in particular transitional programs, case managers may visit daily to supervise residents. However, over time, visits are reduced.
This section provides a broad summary of key evaluation findings regarding housing programs that serve persons with mental illness. In general, the research evidence indicates that housing program models—both transitional and permanent housing—for persons with mental illness have been effective in producing positive outcomes related to residential stability (Ridgeway & Rapp, 1998; Shern, Tsemberis, Winarski, Cope, Choen, & Anthony, 1997). However, most studies were designed to examine whether community living was an effective alternative to institutional placement. As a result, it is difficult to draw conclusions from a large number of studies regarding the ingredients to successful programming. Furthermore, as discussed in the introduction, outcomes are specific to particular types of programs and specific clients. In addition, the outcomes examined vary to such great extent that it is almost impossible to draw conclusions about overall programs—much less their effectiveness compared to other community-based models. Below we briefly summarize a number of studies that yielded some important conclusions regarding effective model types and programmatic components such as integration of services and building configuration. Our discussion with program coordinators of reentry housing programs for the mentally ill revealed that these programs are utilizing evidence-based practice from these and other studies.
Demonstration programs utilizing housing first approaches have shown some success (Carling, 1993; Livingston, Gordon, King, & Srebnik, 1991). An evaluation of ten supportive living demonstration projects funded by the National Institute of Mental Health (Livingston et al., 1991) found that supportive housing can successfully serve persons with severe mental illness. Service integration varied widely across the demonstration project, but most demonstrations used scattered site housing where off-site teams provided support. Some projects utilized 24-hour availability of services and some relied on traditional office hour supports. The study found that use of hospitals and crisis services decreased after program entry. Housing stability outcomes were positive for those who chose their housing facility and for those who had fewer psychiatric disabilities.
A number of studies have found high retention rates in housing programs originating from the 1990 New York/New York (NY/NY) Agreement to House Homeless Mentally Ill Individuals, a joint city and state initiative that created roughly 4,000 units of affordable housing supported with clinical and social services (Lipton et al., 2000; Tsemberis, 1999; Tsemberis & Eichenberg, 2000). The housing agreement provides housing and social services in a variety of configurations. In one of the few supported housing outcome studies that examines criminal justice measures, Culhane, Metraux and Hadley (2002) not only found that individuals placed in NY/NY housing had markedly reduced shelter use and hospitalization compared to a matched control group of individuals not placed in supportive housing, but individuals in NY/NY housing experienced a greater average decrease (pre-placement to post-placement) in days incarcerated in jails. Furthermore, the decrease in the number of incarceration episodes was significantly larger for the NY/NY group than for the group of matched controls. The study was based on databases from eight government agencies and tracked 3,365 participants in New York/New York (NY/NY) housing.
An evaluation examining outcomes of 53 residents of Seattle’s Lyon Building, a permanent supportive housing program for person with disabilities, found that residents maintained consistent residence at the Lyon Building, compared to previous levels of housing stability (Northwest Resource Associates, 2002). Residents also perceived increased access to medical care, and an increase in overall quality of life compared to their past experiences. Housing was provided in a semi-structured environment, utilizing a harm reduction approach.
Although the evaluation literature provides evidence that supportive housing models offering permanent housing for persons with mental illness leads to good outcomes for persons with mental illness, some researchers and practitioners advise caution for those developing low-demand supportive housing approaches. Trainor, Morrell-Bellai, Ballantyne, and Boydell (1993) conducted a research review and came to the conclusion that it is not always advisable to rely on the supportive housing model in jurisdictions where community services are scarce, fragmented or not easily accessible. The authors stressed that those with severe mental illness who do not have intensive case management services may be at greater risk for a range of adverse outcomes because the needed supports are not available or accessible.
Similarly, the National Evaluation of the Shelter Plus Care Program found that the program’s intention to move persons with disabilities from the street directly into permanent housing proved impractical (Fosberg, Locke, Peck, & Finkel, 1997). The study concluded that most clients would have had better outcomes if they completed a transitional program using intensive case management, life skills training, and treatment before moving into the more permanent and independent S+C structures. Thirty-four percent of the study participants had severe mental illnesses, 33 percent were chronic substance abusers, 8 percent had AIDS, and 25 percent had multiple disabilities.
Goldfinger et al. (1997, 1999) used data from the Boston McKinney Research Demonstration Project to examine outcomes for clients randomly assigned to two different housing models. In one group, 63 clients were assigned to the evolving consumer choice model (ECH) and 55 were assigned to independent living. The ECH model was designed so that support services were gradually withdrawn over time. The residents did not have to move among various housing facilities. EC households began with 24-hour staffing but the residents were primarily responsible for operating the residence. Collectively, the tenants set household routine and determined the degree of services that was needed. Intensive clinical case managers were available to residents in both types of housing. The case managers acted as brokers of services, and assisted clients with obtaining benefits and coordinating service components. Outcome data collected at 18 months post entry showed that ECH subjects had significantly fewer days of homelessness (Goldfinger et al., 1999) compared to the group in independent living. Overall, only 19 percent of those placed in either housing program had returned to living in shelters or on the street by the end of the 18-month follow-up period.
Dickey, Gonzalez, Latimer, Powers, Schutt, and Goldfinger (1996) used the same data to examine whether the two housing settings in the Boston demonstration affected the number and type of services consumed 18 months after placement. The authors found that the staff and peer support in the ECH model had no significant effects on the type or intensity of service use. However, ECH residents did achieve somewhat greater residential stability measured by an index accounting for both the number of moves and the length of time between moves. Associations found between housing stability and fewer hospital days disappeared after controlling for age, sex, drug use, race, days homeless, and housing type. Interestingly, although living in independent apartments had no impact on the likelihood of being hospitalized, length of hospital stay was longer for the independent living group.
With regard to high demand transitional programming, Hawthorne, Fals-Steward, and Lohr (1994), using a single group, retrospective repeated measures design, tracked the outcomes of 104 eligible clients who had completed residential treatment at two publicly-funded treatment centers. The study found that clients had significantly fewer hospital and/or crisis residential admissions and fewer number of days homeless during the follow-up year than before the program. In addition, at the one-year follow-up, a significantly greater proportion of clients were employed and living independently.
Other studies showing positive outcomes for high demand treatment models include studies on modified therapeutic communities (De Leon et al., 1999; Shern, Tsemberis, Winarski, Cope, Cohen, & Anthony, 1997). De Leon et al. (1999) found that MTCs produce significantly more positive outcomes for drug use, criminal activity, and psychological depression than treatment-as-usual approaches.
In a randomized study comparing a housing first approach to a housing ready approach, Gulcur, Stefancic, Shinn, Tsemberis, and Fischer (2003) tracked 225 clients over two years. Of the total, 126 participants (56%) were assigned to the control group (using a “continuum of care” housing ready approach), while 99 participants were assigned to a program known as Pathways to Housing (housing first). In the continuum of care model, clients begin at a drop-in center and move to a series of congregate living arrangements with varying levels of on-site support before moving to permanent independent living arrangements. The Pathways to Housing program is a housing first, harm reduction, supported housing model that provides immediate access to independent apartments and supportive services. At Pathways to Housing, ACT (Assertive Community Treatment) teams are used to provide case management services, however clients in this program are given the opportunity to choose the frequency and types of services they receive.
The study found that the Pathways to Housing program was successful in reducing homelessness and psychiatric hospitalization of homeless individuals with mental illness. Participants who were randomly assigned to the Pathways to Housing program were housed earlier and spent more time stably housed than those in the continuum of care program and at a lower cost. The Pathways to Housing group also spent fewer days hospitalized compared to individuals assigned to the Continuum of Care program.
A recent study conducted by the Corporation for Supportive Housing that examined California’s AB2034 programs found that being housing ready was not a good predictor of housing outcomes (Hunter, 2005). Individuals provided with housing were much more likely to remain in programs designed to increase residential stability than individuals not provided with housing. In addition, the evaluation found that programs that enroll “more challenging” consumers (those that have had longer histories of homelessness or more barriers to housing stability) sometimes produced better outcomes than programs that served less challenging consumers.
The above studies generally indicate that housing first approaches may be more appropriate for the targeted population than programs that utilize housing ready approaches. In addition, overall, transitional housing—most often part of the housing ready models—remains controversial. Some view transitional housing for persons with mental illness as stigmatizing and destabilizing, while proponents of transitional housing believe specialized programs are needed to transition tenants into independent living. However, most agree that transitional housing needs to be implemented in conjunction with options for placement into permanent housing. For the high-demand transitional programs that target individuals and families with multiple problems (and particularly, substance abuse), the research shows that these programs can improve housing and clinical outcomes, but only for those who remain in the program until graduation. These programs often have very high attrition rates (Barrow & Zimmer, 2001). Furthermore, intensive residential treatment programs may not facilitate residential stability (ibid.). Researchers suggest that, for these programs, greater flexibility and tolerance about the management of relapse, including the use of harm reduction strategies may be necessary (Lipton et al, 2000).
Some housing providers have used combination housing or convertible housing to address some of the critics of transitional housing. Combination housing involves locations that consist of both transition and permanent apartments. The apartments are co-located in the same building. Individuals or families can move from one unit to another as they progress toward independence. Convertible housing involves the process of converting the terms of tenancy from temporary to permanent. The individual or family does not have to move from one unit to another. Many states have used this approach successfully (Barrow & Zimmer, 1999).
With regard to building configuration and size, there is some consensus from studies spanning two decades that tenants with mental illness are less residentially stable in buildings with more units (Harkness, Newman, Galster, & Reshcovsky, 2004; Harkness, Newman, & Salkever, 2004). A few studies indicate that smaller-scale developments may lead to better outcomes for persons with mental illness (Nagy, Fisher, & Tessler, 1988; Nelson, Hall, & Walsh-Bowers, 1998). It has been hypothesized that buildings with fewer units are more likely to foster a sense of community among residents. Regarding integration of persons with mental illness into normal housing, the research is mixed. Newman, Harkness, Galster, and Reschovsky (2001) found that residents with mental illness felt more comfortable and secure living among others with similar characteristics. Basically, their study examined the effects of living with others with mental illness on disruptive behavior and found that people with mental illness had better outcomes when they lived with others like themselves. The authors suggested that once tenants with mental illness are living in normal or regular apartments in the community, the presence of peers can be beneficial.
In contrast, Hodgins, Cyr, and Gaston (1990) found that levels of stress were higher among mentally ill persons living in groups of supervised apartments than for persons with mental illness living in integrated apartment settings. Similarly, Nagy et al. (1988) found that social adjustment outcomes improved for persons with mental illness when they lived among people who did not have any psychiatric disabilities. A 1993 review of 23 studies examining consumer preferences (Tanzman, 1993) found that, overall, residents prefer not to live with other consumers of mental health services.
As communities grapple with the large number of persons returning from prisons and jails, jurisdictions have begun to develop innovative programs designed to serve this population. In particular, communities that have demonstrated effectiveness in serving homeless persons with mental illness are also the communities developing promising housing programs targeted specifically to those individuals exiting prisons and jails.
Although rigorous outcome or impact evaluations for these programs are scarce, there are a few communities that have been building outcome data on their programs for a number of years. Maryland and California both have developed statewide housing programs that serve persons with mental illness being released from prisons and jails.
Maryland’s Shelter Plus Care program, operating in 21 counties, provides tenant and sponsor-based rental assistance to persons with serious mental illness coming from jails. Case management and supportive services are provided. Outcomes tracked by the State of Maryland demonstrate that recidivism to jails is less than seven percent. Only one percent entered hospitals and only percent were homeless (Substance Abuse Mental Health Services Administration [SAMHSA], 2003) during the evaluation period.
In 1999, California passed Assembly Bills 34/2034 to target mentally ill, criminally involved individuals for housing support. These acts appropriated $10 million to fund housing and treatment for homeless individuals with a diagnosed mental illness. Initial legislation (AB 34) provided for pilot programs in three counties that incorporated integrated service provision across a range of service providers and corrections officials. Based on positive early results of the pilot programs, in 2000 the state legislature expanded AB2034 services by providing an additional $55 million for implementation of 40 programs in 31 additional counties across the state. Today, AB2034 serves more than 4,500 individuals in California. Los Angeles specifically focuses their funds on those persons coming out of correctional facilities. Other counties include, but not require, persons served to be individuals returning from prisons and jails.
The programs are designed to provide comprehensive services to adults who have severe mental illness and who are homeless, at risk of becoming homeless, recently released from a county jail or the state prison, or others who are untreated, unstable, and at significant risk of incarceration or homelessness unless treatment is provided to them. State funds for this program provide for outreach programs and mental health services along with related medications, substance abuse services, supportive housing or other housing assistance, vocational rehabilitation, and other non-medical programs necessary to stabilize this population.
Housing strategies offered by AB2034 programs include advocacy to help clients secure housing, supportive services, landlord assistance, assistance with applications for housing subsidies as well as the provision of short-term assistance, long-term rent subsidies, emergency housing, transitional housing, and residential treatment programs. Some AB2034 programs also master-lease buildings or apartments and sub-let them to clients, and others secure units that are set-aside for clients. And nearly one-third of counties receiving AB2034 funds are developing or currently operating permanent supportive housing. The Corporation for Supportive Housing is currently conducting an evaluation of these housing strategies, determining whether there is a correlation between housing strategies and housing outcomes as well as the costs associated with each strategy (Hunter, 2005). Initial findings indicate that programs that serve the most challenging clients (those with longer histories of homelessness and incarceration) produce similar housing outcomes as those programs that serve less challenging clients. According to the researchers, this suggests that “housing readiness” is not necessarily a good predictor of housing outcomes.
In their analysis of housing outcomes for currently enrolled clients, who have been enrolled in an AB2034 program for 24 months, researchers clustered housing into two categories: (1) Semi-dependent/structured living situation with no lease; and (2) permanent housing where consumers hold a lease and supports are available. Of consumers who were homeless at the time of enrollment, 13 percent of clients who were in semi-dependent situations at enrollment had housing, compared with 35 percent in permanent housing at enrollment. At 12 months, 18 percent of consumers who started in semi-dependent situations had housing, compared with 56 percent in permanent housing. And at 24 months, 16 percent who started in semi-dependent situations had housing, compared with 66 percent in permanent housing. Overall, initial findings also show that the most successful AB2034 programs were utilizing multiple housing strategies (ranging from partnering with housing providers and landlords to securing housing units) (Hunter, 2005).
With the exception of the outcome and evaluation data described above for the California and Maryland programs, our review of the literature and conversations with criminal justice, housing, and mental health practitioners revealed no impact evaluations or rigorous outcome evaluations of housing programs. Given the dearth of evaluation research, but the wealth of small, new, promising programs designed to improve housing, mental health and criminal justice outcomes for persons with mental illness returning from prisons and jails, we conducted in-depth interviews with seven program directors of reentry housing programs. When evaluation data do not exist, understanding the nature and structure of these programs is paramount to moving the field towards evidence-based programming and practice. Furthermore, some of these programs have been operating for a number of years and as a result, are developing their own evidence base—program staff are re-assessing what works and modifying program design as lessons have been learned.
The program descriptions can be found in Appendix A. Here, we describe a number of common themes or program aspects that emerged across the seven programs. These aspects include the reliance on housing ready approaches, the integration of housing and services, daily structure, community location, the use of peers, and coordination with the criminal justice system.
With the exception of one program (Iyana House), the reentry programs reviewed utilize housing ready approaches. The programs are designed as transitional programs with a treatment focus. Some of the programs reviewed offer combination housing, where consumers can progress through different housing options. Related to the housing ready model, reentry populations usually have little service or housing choice in the beginning of their continuum. Tenant rights are usually program-based (but the program may transfer basic rights if participants move into more permanent housing within the program). There is often 24-hour supervision and surveillance and on-site service teams present during the day for mandated sessions and activities.
With regard to service provision, support services are usually integrated in the housing facility rather than just simply available on or off-site. These reentry programs usually have a structured daily environment where support services are required to assist participants in their transition to independent community living. Because these programs are working with presumably high risk persons who also may still be under correctional supervision, reentry programs offer housing that is coupled with a set of services designed to address the needs and risks of this population. In addition, because most of these programs are based on a housing ready model, the programs usually require at least some utilization of services initially. For example, St. Leonard’s Ministries requires an initial counseling and information session for all residents and continues to make them available as long as needed. While residents may not have a choice in whether to take advantage of services, they usually have some choice in which services they want to participate, as long as they maintain a certain level of productivity. This type of service choice is consistent with the idea that individualized plans are an important component of addressing the needs of a mentally ill and substance-abusing population. With regard to service availability, case management and counseling or therapy services are almost always available on-site. More extensive or specialized health services are usually referred out. For consumers that have more severe mental illness, most mental health services are typically on-site (e.g., Berkey House, FIR-ST).
Reentry housing programs often operate within a larger social service agency, and therefore incorporate their own services into the housing facilities (e.g., The Bridge, Project Renewal, St. Leonard’s Ministries, Gaudenzia, Inc., Pioneer Human Services). For persons who have been incarcerated, particularly for long periods of time, access to community resources is essential as they attempt to rebuild their lives. The housing programs are usually conveniently located near community resources and social service providers (e.g., Iyana House, St. Leonard’s House and St. Andrews Court, MIX program).
In general, reentry housing programs for persons with mental illness seem to be designed around the concept of a supportive peer community that acts as a change agent. In virtually every reentry housing program described, group counseling sessions and social activities with peers are an integral and generally required part of the service delivery model. Because the peer community is often an integral part of reentry housing programs, the housing is usually in single-site, congregate settings where most, if not all, residents are program participants. The housing configuration may also be clustered scattered-site or mixed-use, but there is still a significant portion of program participants living under the same roof.
With the exception of Project Renewal’s PSTP, all reentry programs we reviewed provide private rooms, and in many cases private studio apartments, for their residents. PSTP residents are given an opportunity to graduate into a studio. While most reentry housing programs offer participants private living quarters (a space that it not shared), this privacy is limited because most of these programs have staff on-site during most of the day and may provide some kind of overnight supervision or surveillance.
A few studies have attempted to quantify the costs of transitional programming and supportive permanent housing. The small body of research tying costs to outcomes suggests that increasing housing stability can lead to a reduction in service utilization among the homeless mentally ill population, mostly by saving public resources in providing emergency or fragmented services. This cost savings is purported to offset the upfront investment in providing supportive housing.
The most widely-cited study on costs of providing housing to the mentally ill is the research conducted by Culhane, Metraux and Bradley (2002). The authors used databases from eight government agencies to track 3,365 participants in New York/New York (NY/NY) housing, a program providing housing and social services in a variety of configurations (discussed earlier). The researchers quantified costs for each participant in government services two years before and two years after being placed in NY/NY housing. The study, which also tracked a control group of mentally ill homeless individuals who were not participating in the NY/NY program, found that while NY/NY participants cost the government roughly the same per year as those in the control group, there were significant cost reductions in service utilization (an average reduction of $16,282 per housing unit, per year) after NY/NY participants moved into supportive housing. The most significant reductions in service use were among shelter, health, and corrections services. The study found that NY/NY supportive housing resulted in a $12,145 net reduction in shelter, health, and corrections service use annually per person, over each of the first two years of the intervention.
Dickey and colleagues (1997) conducted a cost-effectiveness study using the Boston data described earlier. The study, which compared an evolving consumer household program to those in independent living, found that the annual costs for the evolving consumer group came to $56,434, compared to $29,838 for the independent-living group.
Studies quantifying the costs associated with modified therapeutic communities find that modified TCs cost no more than treatment-as-usual approaches; and modified TC treatment produces $13 of benefit for every dollar spent (French, Sacks, De Leon, Staines, & McKendrick, 1999; French, McCollister, Sacks, McKendrick, & DeLeon, 2002; McGeary, French, Sacks, McKendrick, & De Leon, 2000).
Finally, a large study estimating the cost of serving the homeless population in six settings in nine cities found the median daily costs of the settings to be (The Lewin Group, 2004):
The preceding sections and a review of the reentry housing program models in Appendix A make it clear that a one-size-fits-all approach to housing for persons with mental illness who have had justice system contact will not work. Taking a step back from housing programs that directly serve those with criminal justice system contact, even within the mental health and homelessness prevention literature, researchers and practitioners recognize that there is no body of compelling evidence regarding the most effective program structure, methods of choosing patients for particular programs, and measures of success (Thompson & Breakey, 1997). Below we outline a number of overlapping questions designed to assist the panel discussion.
Umbrella Approach: Housing ready
Service-related Model: High demand
Program Model: Reentry
Length of Stay: Long-term transitional
Configuration: Single-site
The Mentally Ill Offender Community Transition Program (MIOCTP) began as a five-year pilot program in July 1998 following legislation aimed at addressing the social and financial cost associated with severely mentally ill persons coming out of prison without adequate treatment and housing. The King County Regional Support Network contracted Seattle Mental Health (SMH) to provide the statutorily defined components of the pilot. SMH subcontracted with Pioneer Human Services (PHS) to provide the housing component for the 25 participants enrolled in the MIOCTP pilot at any given time. PHS’ Berkey House reserves 16 of its 30 units for MIOCTP participants.
Berkey House follows a modified therapeutic community model based on regular contact between the house manager, SMH service providers and justice system officials.
Eligibility criteria for MIOCTP participation is determined by the Washington State Legislature. To be eligible, the Department of Corrections must determine that incarcerated men and women have a serious mental illness that influenced their criminal activity, are in need of continued mental health treatment, are unable to obtain housing, and have at least a year remaining of their sentence but are within six months of being released to the community. Additionally, SMH and PHS look for potential Berkey House residents who are motivated to remain clean and sober. While persons with sex offense and arson histories are not excluded from MIOCTP eligibility, they are not able to stay at Berkey House.
Berkey House is a long-term transitional housing facility where residents can stay for up to two years until they are ready to move on to more independent permanent housing. Often, Berkey House residents move into other Pioneers Human Services’ clean and sober housing.
Berkey House is a four-story building with 30 units, almost all of which are reserved for individuals with involvement in the criminal justice system. In addition to the 16 units reserved for MIOCTP participants, there are nine shelter plus care units that are reserved for people who have had contact with the criminal justice system. The remaining five units are clean and sober housing for individuals with co-occurring chemical dependence and mental illness. Residents have their own room and bathroom and share a kitchen with other residents on the floor.
Berkey House is a highly supervised and structured setting that provides integrated behavioral, mental health, and substance abuse services to the presumably high-risk MIOCTP participants. MIOCTP participants have different degrees of functioning and treatment needs and each has an individualized service and supervision plan that combines requirements from Seattle Mental Health, the Department of Corrections, and Pioneer Human Services. Mental and behavioral health services and chemical dependency treatment are provided on-site by SMH service teams. These service teams consist of a case manager, project manager, house manager, psychiatrist, nurse, and a substance abuse counselor. At least one housing manager, jointly employed by PHS and SMH, is always on-site and has daily contact with the SMH service teams. Berkey House requirements include chores, weekly house meetings, and use of case management services. Berkey House also requires that residents sign in and out every time they leave or enter the facility. In this supervised and structured setting, Berkey House seeks to create a therapeutic environment that encourages peer support. Because MIOCTP participants have varying levels of need and ability, Berkey House establishes buddy systems to involve residents in the transition process of their peers.
The Department of Corrections first screens potential MIOCTP participants in prison and then refers them for interviews with SMH program case managers who determine eligibility and motivation. Ideally, participants are identified at least three months prior to release to begin reentry planning. Once participants are released to Berkey House, they often remain under justice system supervision and are subject to home inspections and random urinalysis by their community corrections officer. The Berkey House program director reports a good relationship between PHS, SMH, and the Department of Corrections in which all three agencies reinforce each other’s specific requirements for MIOCTP residents.
It appears that funding for the MIOCTP program comes through the Washington State Department of Social and Health Services as mandated by statute.
In the seven years that MIOCTP has been in operation, SMH has found permanent housing for every MIOCTP participant leaving Berkey House.
Umbrella Approach: Not specified
Service-related Model: High demand
Program Model: Reentry
Length of Stay: Long-term transitional
Configuration: Single-site, mixed-use
Gaudenzia, Inc. is a nonprofit substance abuse and mental health treatment organization that owns and operates a series of specialized treatment and residential programs in Pennsylvania. Gaudenzia’s Forensic Intensive Recovery – State Program (FIR-ST) is a 22-bed transitional housing program, in its seventh year of operation, under contract with the Pennsylvania Department of Corrections to serve mentally ill parolees returning to Philadelphia.
FIR-ST operates under a structured Modified Therapeutic Community model that includes a level system of privileges and responsibilities as residents progress through treatment. Residents are expected to take part in the progress of their peers in formal and informal group settings and learn how to address the behavior of others in a non-confrontational way.
The FIR-ST program accepts only adult parole-eligible men with a diagnosed mental illness or personality disorder. Eligible participants must be medically stabilized, willing to accept treatment, and misconduct-free for at least a year prior to entering the program. FIR-ST does not accept those with mental retardation.
As a transitional housing program, FIR-ST had planned to house residents for 6-9 months, however, the length of stay can be as much as 18 months depending on the individual’s progress.
FIR-ST residents are housed in a congregate mixed-use building in Philadelphia owned and operated by Gaudenzia. The 22 beds reserved for the FIR-ST program are located on one floor. Each room typically has two to three beds; some rooms have four beds. While no resident has his own private room, each is guaranteed his own chest of drawers, a closet, and a medicine cabinet. All FIR-ST participants share two bathrooms, each with two shower stalls, two toilets and two sinks.
FIR-ST provides on-site case management and support services that include mandatory morning meetings, individual and group counseling, mental health and substance abuse education, life skills and stress management, medication management, and support groups. One social worker is assigned to all FIR-ST participants. Each participant also has a counselor, who is generally responsible for a FIR-ST caseload of seven to eight. The PA DOC requires FIR-ST residents to have at least two contacts a week with Gaudenzia counselors and support staff. These contacts are broken up into one group session and one individual session. Residents are encouraged to attend more than two sessions a week.
Potential FIR-ST participants are first identified and referred by PA DOC counselors in prison and are then further screened by Gaudenzia counselors to determine eligibility. FIR-ST does not accept those convicted of a sex offense or arson or those who have histories of misconduct in prison. Parole officers are in regular contact with FIR-ST staff and counselors to check on the progress of their parolees. Once participants complete their parole sentence, the DOC no longer pays for them to remain in the FIR-ST program; however, in order to successfully complete the conditions of their parole, participants must have an approved home plan. According to the director of the FIR-ST program, it is very rare that participants max out their parole sentence before a housing plan is set up for them.
The Pennsylvania Department of Corrections contracts with Gaudenzia, Inc. to provide support services and housing.
Currently, there is no system in place for measuring outcomes. Neither the DOC nor FIRST follows participants after release from the program.
Umbrella Approach: Housing ready
Service-related Model: High demand
Program Model: Reentry
Length of Stay: Long-term transitional
Configuration: Single site
In 1997, with assistance from the National Development and Research Institutes (NDRI), the Colorado Department of Corrections (CDOC) developed a model modified therapeutic community (MTC) treatment program for inmates suffering from mental illness and chemical abuse (MICA). Independence House is a 20-bed aftercare housing program for people coming out of the MTC in the San Carlos Correctional Facility.
Independence House follows a Modified Therapeutic Community model.
MICA men who are enrolled in the (CDOC) Modified Therapeutic Community treatment program in San Carlos Correctional Facility have the opportunity, upon release, to enter Independence House, the post-release community aftercare portion of the program.
Independence House is a six-month transitional housing program.
The Independence House program reserves 20 beds in five apartments of a 15-apartment community correctional facility.
Three aftercare staff, trained in both mental health and substance abuse treatment, conduct programs and provide services on-site every day of the week from eight in the morning to eight in the evening. On average, residents participate in programs and services three to seven days a week for three to five hours of the day. Skills learned in on-site activities include cooking, money-management and banking, taking advantage of community resources and examining and managing emotions and behavior related to substance abuse, mental health, and criminal activity. More intensive mental health and psychiatric treatment is available off-site at local centers, and continues to be available once participants leave the program and establish permanent residency in the community.
As a MTC, Independence House seeks to create an environment where individuals learn to live in a community and where their recovery and growth process benefits from connections made with peers. Residents gather daily as a community to discuss problems or successes they are having with treatment, work, independence, etc. During their six-month stay, residents progress through different levels of independence and responsibility and take part in the progress of others. Many basic activities, such as cooking, are done together to foster the growth of the community as a peer support group and self-improvement tool.
The CDOC administration and prison staff are part of a working partnership with service providers and researchers from NDRI. As part of the MTC program, Independence House residents are still under correctional supervision during their stay.
The program was developed in 1997 as a four-year pilot with funding from the Department of Justice Edward Byrne Drug Control and System Improvement Formula Grant Program. Regular funding is currently provided by the Colorado Department of Public Safety, Division of Criminal Justice. The cost of medication is funded through the CDOC Parole and community budget.
Evaluation data available for Independence House come from a larger multi-year evaluation of the MTC program within the San Carlos Correctional Facility, where Independence House is viewed as aftercare for those leaving the San Carlos prison and who had participated in the in-prison MTC. The evaluation studied the MTC program (within San Carlos) impact on criminal activity and incarceration rates (new crimes only) for three groups by comparing individuals who participated in the MTC Program in prison and those who participated in a mental health program (MH). The MTC participants were grouped by those who did not enter Independence House upon release (MTC only) and those who did enter Independence House upon release (MTC + aftercare). The study traced three outcome measures for all groups: (1) reincarceration, (2) criminal activity, and (3) criminal activity related to alcohol and drug use. The MTC + aftercare group (Independence House) showed significantly lower rates of reincarceration and new criminal activity at follow-up than the MH group or the MTC only group.
Umbrella Approach: Housing first
Service-related Model: Low demand
Program Model: Reentry
Length of Stay: Permanent
Configuration: Single site
Iyana House is a new 16-bed reentry housing program operated by The Bridge, Inc. a nonprofit based in New York City that provides counseling, psychotherapy and residential services to persons with mental illness who struggle to lead independent and productive lives. Recently opened in October 2004, Iyana House provides permanent supportive housing in a congregate setting to women with co-occurring mental illness and substance abuse coming out of the Bedford Hills Correctional Facility. In addition to their own apartment, Iyana House offers residents comprehensive support services that promote independent living, self-sufficiency and recovery. Located in a renovated apartment building in East Harlem, Iyana House seeks to create a rehabilitative peer community to facilitate post-release transition into the community.
The Bridge has created its own model for Iyana House to address HUD’s three established goals: residential stability, increased income and skills, and self-determination. The Iyana House model combines the use of a peer community as a change agent and comprehensive on and off-site support services that strengthen the community.
Iyana House serves homeless women with mental illness, many of whom also have co-occurring substance abuse histories, who have been incarcerated at the Bedford Hills Correctional Facility for at least six months.
Each resident entering Iyana House has her own studio apartment for as long as she wants or needs it.
Iyana House is a single-site congregate apartment building in East Harlem. In addition to the 16 studio apartments, Iyana House includes a two-bedroom apartment that serves as a program office and a common space with a kitchen for women to socialize with each other and visit with guests. Iyana House is located in the middle of a lively community with convenient access to several resources including a health center, public transportation, grocery stores, and a recreational center.
Residents are offered a continuum of care through a variety of on and off-site individualized support services provided by The Bridge. Although women are highly encouraged to take advantage of support services, Iyana House does not require participation. The Bridge forensic Assertive Community Treatment (ACT) Team is available to provide on-site individualized case management and direct services 24 hours a day, 7 days a week. These on-site services include mental health assessment and medication monitoring, referrals to community resources, benefits counseling and a weekly trauma group, career group, and Dialectical Behavior Therapy (DBT) group. During the day, women can take part in an integrated mental health and substance abuse treatment program, vocational training and job placement off-site at The Bridge’s rehabilitation center. There are several other recreational and therapeutic programs available at The Bridge, including art therapy classes. In the evening, women return home are free to attend weekly group sessions, socialize, cook a group meal, or retreat to their private apartment.
The Bridge has a long-term lease on the apartment building, but transfers rights of tenure to the residents. This is the third time The Bridge has leased from the same owner.
Along with the State Office of Mental Health Forensic Bureau, The Bridge has a working relationship with the State Division of Parole and the Bedford Hills Correctional Facility. Iyana House staff conducts a weekly prison in-reach program 6 months prior to release to identify potential Iyana House residents, establish relationships and prepare women for their placement in Iyana House. One dedicated parole officer works with all women entering Iyana House and has regular contact with The Bridge staff and service teams to check on the progress of each woman.
Funding comes from HUD SHP grants, the State Office of Mental Health, and client fees. Women pay a third of their income, primarily from SSI, towards rent.
Because the program is so new, and not all enrolled women have been released yet, there is no client tracking data available. Iyana House collects data that is relevant to its yearly progress reports to HUD and are looking at two primary outcomes, (1) comparing the functionality of women upon entering the program and their functionality one and two years after, and (2) re-hospitalization rates, either for mental or medical reasons. There is no client tracking data related to recidivism.
Umbrella Approach: Housing ready
Service-related Model: High demand to low demand
Program Model: Reentry
Length of Stay: Short-term transitional and permanent
Configuration: Single-site and scattered-site
Heritage Health and Housing is a social service nonprofit that provides residential and treatment services to populations with special needs. In addition to owning and operating three housing facilities, Heritage leases scattered-site apartments throughout New York where staff provide on-site support services. Heritage’s MIX program is a 23-bed housing program that has provided both transitional and permanent supportive housing to homeless, mentally ill, and previously incarcerated men since 1997. The Federated Employment and Guidance Services (F.E.G.S) NYC LINK team identifies and refers potential residents, designated homeless by the New York City Department of Homeless Services, who are living in city shelters or are awaiting release from a correctional facility. In supported and safe environments, the MIX program seeks to provide the needed rehabilitative and supportive services to assure a successful transition to community life.
The MIX program follows the housing continuum model designed by the New York Office of Mental Health that moves residents from highly structured and supervised short-term housing to permanent independent supported housing.
The MIX program provides three levels of housing care as defined by the New York State Office of Mental Health: supervised, supportive, and supported. Supervised housing is highly structured short-term transitional housing in a congregate setting that provides 24-hour supervision, seven days a week for individuals with greater needs who are less prepared for independent living. Supportive housing is the next stage in the continuum and provides long-term transitional housing in a single building with support staff on-site during regular business hours. Supported housing is permanent independent housing for higher functioning individuals where support services are primarily off-site, and case managers visit once or twice a month. MIX program residents have access to all three OMH levels of housing. MIX provides six supervised beds, four supportive beds, and 13 supported beds.
The MIX program accepts formerly incarcerated mentally ill men between the ages of 18 and 60 who are determined to be homeless by the New York Department of Homeless Services. Participants must be officially released from confinement, without absconding, and may or may not be on parole.
Originally, the MIX program intended for residents to stay in short-term supervised housing for four weeks before moving them on to supportive and eventually supported housing. However, after the first year, MIX staff realized that most residents need more than four weeks to secure their benefits and become mentally prepared for less supervised housing. Supportive housing is long term transitional housing that further prepares residents for independent living in permanent supported housing.
MIX program supervised housing is located in Safe Space, a congregate building in South Bronx. Supportive housing is located in a single-site mixed used building, and supported housing is scattered throughout Northern Manhattan and South Bronx in apartments that Heritage leases.
Throughout all levels of housing, the MIX program requires residents to participate in support and treatment services. Residents living in supervised housing have access to support services on-site 24 hours a day. At this stage, staff assess residents’ needs, determine a rehabilitation and preparation plan and begin working with residents on their self-sufficiency skills. In supportive housing, staff is available on-site to continue case management services and skills training. On-site services include group and individual therapy sessions and various workshops on daily living skills, employment, social interaction, parenting, and nutrition. Mental health and substance abuse treatment are referred out. The agency referred to most often is the Upper Manhattan Mental Health Center that conveniently shares building space with Heritage’s headquarters. In supported housing, residents continue to receive support services, but in an off-site setting. Case managers conduct home visits once or twice a month.
Heritage rents supported apartments off the private market and the MIX program transfers tenant rights to program residents.
Funding is provided by HUD McKinney SHP grants and the NY State Office of Mental health.
According to the Director of Development at Heritage Health and Housing, since the MIX program began operation, only 11% of the residents have been reincarcerated. Internal evaluations and outcome data are unavailable.
Umbrella Approach: Housing Ready
Service-related Model: High Demand
Program Model: Reentry
Length of Stay: Long-term transitional
Configuration: Cluster scattered-site
Project Renewal, Inc. is a nonprofit agency that provides a holistic and comprehensive approach to address the needs of homeless men and women in New York, beginning with outreach and housing and including primary healthcare, substance abuse and mental health treatment, and employment services. Project Renewal is a large agency serving over 13,000 homeless men and women each year. Project Renewal’s Parole Support and Treatment Program (PSTP) was created in 2002 to provide transitional housing for 50 mentally ill and substance abusing men and women coming out of state prison on parole. Project Renewal’s Assertive Community Treatment (ACT)-like teams provide comprehensive wrap around services and mental health treatment to assist the successful transition into the community.
The Parole Support and Treatment Program follows a housing ready approach to independent living for the mentally ill, substance abusing reentry population. PSTP does not place participants directly in permanent housing until they have received support services and demonstrated readiness.
PSTP does not exclude based on the nature or history of criminal backgrounds.
The Parole Support and Treatment Program provides long-term transitional housing for up to 18 months. Once participants are ready, or have completed their parole sentence, they are moved into permanent independent housing settings that are often other Project Renewal facilities.
PSTP serves 50 men and women in 28 cluster scattered-site apartments in five buildings throughout Manhattan and the Bronx. PSTP participants live in full “supported” apartments that are typically used for permanent independent housing. Most participants share apartments with their own room, and some have an opportunity to graduate into a single studio.
Project Renewal’s ACT-like team of professionals consisting of social workers, a psychiatrist and nurse, and other substance abuse and mental health counselors work with PSTP participants to determine an individualized plan. This team helps participants secure benefits, access community resources, live independently, set and reach individual educational and employment goals, and access permanent housing. Mental and primary health services and substance abuse treatment is provided through Project Renewal’s continuum of care. In addition to complying with the conditions of their parole, participants are required to meet with their team of professionals regularly and follow their treatment recommendations. Participants are also required to attend peer group sessions. Although it is a highly structured environment with 24-hour emergency response, staff members do not stay overnight.
PSTP staff begin working with future participants in weekly group sessions when they are in the Community Orientation and Reentry Program (CORP) unit at Sing Sing Correctional Facility, the state prison in Ossining, New York. Inmates identified as having severe and persistent mental illness are transferred to the CORP unit 90 to 120 days prior to their release. PSTP participants can also be referred from the State Division of Parole once they are already released. PSTP staff work closely with parole officers to help PSTP participants transition into the community in a productive and law-abiding way.
Funding comes from the New York State Office of Mental Health “Supported Housing” dollars and New York State Division of Parole Federal “Byrne Grant” dollars. PSTP participants are also expected to pay 30% of their income toward rent.
Currently, there are no available or formal outcome data. Project Renewal, OMH and Parole are in the process of determining how they want to measure participant success.
Umbrella Approach: Housing Ready
Service-related Model: High demand to low demand
Program Model: Reentry
Length of Stay: Long-term transitional and permanent
Configuration: Single site
St. Leonard’s Ministries (SLM) is a faith-based nonprofit organization that provides case management services and both transitional and long-term housing for men and women coming out of prison or on parole in Chicago. St. Leonard’s House (SLH) is a 40-bed short-term transitional housing facility for men. St. Andrews Court (SAC) is a 42-bed long-term housing program for men who have completed the SLH program. SLM also provides transitional and permanent housing for women. Grace House is the 16-bed short-term transitional facility. Recently, SLM reserved 10 beds in Sanctuary Place, a 60-bed, single-site complex operated by the Interfaith Housing Development Corporation of Chicago (IHDCC), for women who are ready to move from Grace House.
St. Leonard’s Ministries follows a housing ready approach to permanent supportive housing for previously incarcerated individuals in need. St. Leonard’s also incorporates peer and community support in their program model.
St. Leonard’s Ministries serves men and women who are coming out of prison without the housing or skills necessary to lead productive lives. The program does not directly target those with mental illness. However, about one-third of SLM residents suffer from a mental illness. St. Leonard’s Ministries does not exclude sex offenders from its housing programs.
At SLH, men typically stay for six months, while the average stay for women at Grace House is nine months. St. Andrews Court and Sanctuary Place are permanent supportive housing options for men and women completing the program at SLH and Grace House. At least six current residents in SAC have lived there since the programs inception in 1998.
All of SLM residences are located in congregate single-site buildings. With the exception of Sanctuary Place, all housing facilities are fully occupied by SLM clients. St. Leonard’s House and SAC are located next to each and have established themselves as welcome neighbors to the larger community. When SLM was in the process of creating SAC, they involved the community throughout the planning and implementation stages. Community members helped furnish and stock each unit in SAC, and donated artwork and landscaping and legal services. St. Leonard’s Ministries made a huge effort to create an aesthetically charming facility from the inside and out to benefit and respect residents and encourage community support. Residents at SLH and SAC have close and convenient access to a health and mental health center, public transportation, a public library, and Chicago’s business and financial district.
Residents at St. Leonard’s Ministries receive an array of individualized case management and support services on-site from 45 staff members and 15 volunteer. Many support staff are former residents, which strengthens the program’s ability to reach out to clients and engage them in treatment participation. According to the director of SLM, residential staff make every effort to address substance abuse and other mental health issues on-site. However, physical and serious mental health issues are referred out. Employment counseling and placement is a major service component at SLM.
Initial counseling and information sessions are mandatory for SAC residents and free of charge. Ongoing sessions are available on an as-needed basis and continue to be free of charge. To instill a sense of ownership and responsibility for their homes, SAC initiated a residents’ council. Each of the four floors of SAC elects a Residents’ Council representative to attend the regular meetings and act as a liaison with the director of social services. A professional facilitator is brought in to instruct new residents about the responsibility of the Residents’ council. St. Leonard’s Ministries has recently completed the Michael Barlow employment center, a job development and training center for SLM participants at all residential levels.
St. Leonard’s Ministries receives partial funding from the Illinois Department of Corrections and has a close working relationship. Residents at SLM are referred either from parole officers or directly from the Department of Corrections prior to release. About 99% of the men and 80% of the women participating in the program are on parole.
St. Leonard’s Ministries receives private donations and a mix of grant funding and contracts from federal and state agencies, including the Illinois DOC.
Currently, there is no development of outcome measures and performance indicators; however, the program does track recidivism rates three years after completion of the program. These rates have consistently been at 20%, significantly lower than Illinois’ average rate, which exceeds 50%.
Programs and Services Attached to Housing Programs |
||||||
|
|
Transitional |
Permanent |
|||
|
Total |
MH Focus |
MH/CD |
Total |
MH |
MH/CD |
Total |
4,390 |
400 |
|
1,920 |
300 |
100 |
Located With: |
|
|
|
|
|
|
Emergency Shelter |
40% |
24% |
11% |
30% |
16% |
12% |
Permanent Housing |
20 |
30 |
17 |
-- |
-- |
-- |
Transitional Housing |
-- |
-- |
-- |
44 |
44 |
55 |
Outreach |
23 |
35 |
23 |
35 |
42 |
27 |
Drop-In Center |
14 |
12 |
10 |
13 |
14 |
24 |
Voucher Distribution |
11 |
7 |
3 |
18 |
5 |
7 |
Housing w/Vouchers |
9 |
12 |
8 |
13 |
14 |
12 |
|
|
|
|
|
|
|
Services Provided on Site: |
|
|
|
|
|
|
Food |
76% |
74% |
74% |
61% |
65% |
58% |
Clothing |
69 |
61 |
64 |
52 |
54 |
45 |
Life Skills |
73 |
68 |
65 |
65 |
66 |
57 |
Housing |
81 |
88 |
68 |
84 |
83 |
72 |
Employment |
63 |
52 |
74 |
59 |
47 |
53 |
General Health Care |
55 |
61 |
63 |
47 |
52 |
43 |
Substance Abuse |
56 |
55 |
67 |
|||
