Toward Understanding Homelessness: The 2007 National Symposium on Homelessness Research. Incarceration and Homelessness. What Can We Learn from Emerging or Promising Practices in the Field?


Given the tendency to re-adapt existing housing models and to use patchwork financing schemes to address post-incarceration homelessness, most practice in this arena remains ahead of, and therefore uninformed by, research that evaluates effectiveness. While many of these models are considered best practices or national models, empirical evidence that these models reduce homelessness and recidivism is often lacking. For example, studies have demonstrated that assertive community treatment (ACT), which is an intensive community-based case management regimen, targeted at criminal justiceinvolved persons with mental illness (often referred to as Forensic ACT or FACT) and other intensive case management models significantly reduce rates of recidivism and criminal justice involvement (Lurigio, Fallon, & Dincon, 2000; Ventura et. al, 1998; Hartwell & Orr, 1999; Lamberti, Weisman, & Faden, 2004). However, they have seldom been evaluated for their impact on homelessness and housing stability. An exception is a FACT program in Atlanta that showed substantial reductions in both criminal justice involvement and homelessness among its caseload of persons diagnosed with mental illness (Georgia Rehabilitation Outreach, 2005).

Only been recently have any programs made deliberate attempts to draw upon this limited body of research to shape their approaches. For example, Culhane and colleagues (2002) evaluation of the New York/New York housing initiative was a basis of the New York City Frequent Users of Jail and Shelter Initiative in their providing supportive housing to individuals identified as frequent recidivists to emergency shelters and city jails (Anderson, 2006). Moreover, New Jerseys Program to Return Offenders with Mental Illness Safely and Effectively (PROMISE) Initiative also draws upon the New York/New York study as well as research on the effectiveness of FACT at reducing recidivism in its combination of supportive housing and ACT-like intensive case management to parolees with mental illness who are found to be homeless at the time of their release. And in Philadelphia, the State of Pennsylvania and the Council of State Governments has funded a study to assess the impact of Gaudenzia FIR-St., a residential program for persons with mental illness leaving prison, upon community services use, homelessness, and reincarceration. While these studies show promise in providing insights into the performance of housing programs targeting persons released from incarceration, they are still too early in their implementation to provide evidence of program effectiveness.

Several trends are apparent among emerging practices. First and foremost, there is assumed to be an interaction effect between homelessness and incarceration. Just as incarceration, with its clinical and social consequences, leads individuals to homelessness, the lack of stable housing is considered to be a criminogenic factor. Therefore, assisting individuals to achieve housing stability can itself reduce returns to criminal justice involvement. Housing stability stands as a prerequisite to service delivery as well as a strategy for preventing returns to criminal justice involvement and incarceration. Promising programs tend to integrate services and treatment with either permanent or transitional housing rather than offer either services (e.g., case management) or affordable housing alone.

A second trend is the use of case management models for service delivery. ACT and intensive case management models are common, particularly for individuals with mental illness leaving incarceration, and these services are usually provided to this population at a higher intensity level than for other populations. This means that such case management regimens involve higher numbers of staff, often with more specialized expertise, and more frequent and intensive client contact. This case management is often supplemented with services from specialized personnel such as community supervision officers (as in FACT teams) or certified substance abuse counselors. Similarly, many housing projects targeted towards persons with histories of both homelessness and incarceration tend to use higher service worker-to-client staffing ratios (Hals, 2003). This larger and higher credentialed service staffing makes possible the incorporation of a range of clinical approaches such as cognitive or dialectical behavioral therapy, motivational interviewing, group counseling, and substance use counseling, as well as more frequent contact with tenants than is found in programs for non-justice involved homeless populations. Many programs make weekly or even daily contact with clients or tenants.

A third and related trend is the practice of front-loading services. Here a more intense level of services provision is utilized during a critical time period, usually defined as the first 30 to 90 days immediately following release from incarceration. During this period, persons are believed to be at particularly high risk for reincarceration and homelessness (Nelson, Deess, & Allen, 1999). Front-loaded service models attempt to assist individuals during this high-risk period by either first placing them into a housing setting with more intensive services and then transitioning them to another, less intensive setting, or by simply varying the number and type of staff assigned to each resident as he or she becomes acclimated to community living (Hals, 2003; Roman, McBride, & Osborne, 2006). In the latter approach, the resident agrees to participate in a formal bonding period, in which the resident is accompanied to most or all outside engagements by a service worker who assists the resident with navigating public systems, evaluates the residents cognitive patterns, and troubleshoots and mediates potential problems and conflicts.

Fourth, many interventions serving persons who are homeless leaving incarceration have working relationships with the criminal justice system, particularly those programs that serve people under community correctional or court supervision. This capacity may take the form of a formal agreement with a community supervision agency (Iyana House, Parole Support and Treatment Program), an informal arrangement with community supervision personnel (the Fortune Academy), or even dedicated community supervision personnel who serve as staff to the program (PROMISE Initiative). In programs that have incorporated this function, service personnel assist community supervision authorities to assess clients progress in achieving service goals and lower the risk of reoffending. Because service personnel are able to interact with clients and observe their progress without a law enforcement obligation, they are able to assist supervising authorities to take into account individual service needs and issues in their monitoring. In doing so, they provide community supervision authorities with a means of distinguishing between behaviors that lead to true offenses and those that constitute a part of a process of recovery and stabilization.

A fifth critical trend emerging from promising programs for persons who are homeless upon release from incarceration is the emphasis on client engagement. Whereas many programs serving homeless individuals typically use a first-come first-serve approach to client intake and recruitment, many effective supportive and transitional housing programs serving persons who are homeless leaving the criminal justice system tend to be more proactive in their client recruitment. For example, programs will perform in-reach into correctional institutions to identify and engage potential clients, some even spending several months meeting with and preparing potential clients prior to their release. Even programs serving highly transient populations such as frequent jail users place a heavy emphasis on up-front client engagement, performing outreach in various settings where potential clients might be found in order to establish a relationship with clients prior to their housing placement. Such active and sometimes aggressive engagement helps to build trust with members of a target population who tend to be distrustful and avoidant of service providers (Hals, 2003).

Defining the Target Population

One promising emerging trend in practice is not a service delivery strategy but rather an approach that uses greater specificity in selecting a target population. In these emerging initiatives, data systems are used to identify and target housing and services to individuals who are frequent users of correctional institutions as well as other costly emergency public services such as homeless shelters, emergency rooms, and psychiatric hospitals. By targeting these frequent users of multiple systems, these programs not only seek to extend assistance to those corrections-involved individuals who represent the highest levels of housing instability, but also to achieve disproportionate reductions in the demand for public services, thus justifying continued public investment. For example, in the New York City Frequent Users of Jail and Shelter Initiative, individuals who cycle in and out of local corrections and homeless shelters are engaged and placed into permanent supportive housing with enhanced services with the intention of breaking their cycle of incarceration and homelessness (Anderson, 2006). Similarly, a Portland, Oregon program operated by Central City Concern called Housing Rapid Response provides permanent supportive housing to individuals who are identified as disproportionate users of both uncompensated healthcare and incarceration. Similar initiatives are being tried in Rhode Island, Chicago, and Atlantic County, New Jersey.

View full report


"report.pdf" (pdf, 242.57Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®