by
Susan Barrow, Ph.D.
Rita Zimmer, M.P.H.
Despite HUD’s endorsement of transitional housing as an essential part of a comprehensive continuum of care, consumers, providers, and advocates have frequently disagreed on its value, on the best ways of linking services to housing, and on appropriate mechanisms and sources for funding transitional housing programs. Critics have emphasized the stigma associated with transitional programs as well as the diversion of resources that might otherwise serve to expand the supply of and access to affordable permanent housing; proponents counter that homeless families and individuals with multiple problems need help with more than housing alone if they are to achieve residential stability. To sort through the conflicting claims about transitional housing requires some consideration of the diversity of the programs thus labeled and what we do and don’t know about their impact on homelessness. We begin by clarifying what the concept encompasses; review the evolution of transitional housing; and describe variations in the major approaches developed for homeless families and individuals in terms of differences in target populations, physical structures, service intensity, and other program characteristics that cluster along a continuum with "high demand" service-intensive facilities at one end and "low demand" programs with flexible requirements and optional services at the other. Available research assessing the major models indicates that scattered-site transitional housing programs that convert to permanent housing constitute one effective (and cost effective) approach to helping families and possibly individuals exist from homelessness. Future research should not only test the relative effectiveness of different transitional program models but should compare transitional housing approaches to alternative strategies for ending homelessness for individuals and families.
Rapidly rising homelessness in the late 1970s and early 1980s initially evoked crisis responses from federal and local agencies—responses focused on expanding the capacity for emergency shelter. Only later in the 1980s did the emphasis shift to developing housing and service combinations that would address longer-term needs of the homeless population. Federal support for both transitional and permanent housing has been provided since 1987 with McKinney Act funds; and since 1994, when the U.S. Department of Housing and Urban Development (HUD) began to require that applicants for federal funds create an integrated "continuum of care," transitional housing has been deemed one of the necessary components of a comprehensive response to homelessness (HUD Report to Congress, 1995; Barnard-Columbia Center for Urban Policy, 1996).
Despite HUD’s endorsement of transitional housing as an essential part of the continuum of care, consumers, providers, and advocates have frequently disagreed on its value, on the best ways of linking services to housing, and on appropriate mechanisms and sources for funding transitional housing programs. Critics have emphasized the stigma associated with transitional programs as well as the diversion of resources that might otherwise serve to expand the supply of and access to affordable permanent housing; proponents of transitional housing counter that homeless families and individuals with multiple problems need help with more than housing alone if they are to achieve residential stability (HomeBase, 1998).
To sort through the conflicting claims about transitional housing requires some consideration of the diversity of the programs thus labeled and what we do and do not know about their impact on homelessness. We begin by clarifying what the concept encompasses and where the boundaries between transitional housing and related concepts—emergency shelter, residential treatment programs, permanent supportive housing—can most usefully be drawn. We then review the evolution of transitional housing for homeless families and individuals and the major approaches that have been developed for each of these groups. Finally, we examine the relatively limited research literature on the effectiveness of the major models, and conclude with a discussion of unresolved issues requiring further study.
Concepts and Definitions
In the context of a continuum of responses to homelessness, transitional housing occupies an intermediate position. It consists of relatively private accommodations provided on a temporary basis along with intensive services intended to facilitate the transition to permanent housing. The distinctions between transitional housing and other types of temporary and/or service-enriched accommodations for homeless people are not hard and fast: what one locality labels as "transitional" may look a lot like the "shelters" in another setting. Despite this overlap, however, we can clarify core features by examining how transitional housing contrasts with emergency shelter, residential treatment programs, and permanent supportive housing.
Transitional Housing vs. Emergency Shelter
Transitional housing usually differs from emergency shelter in offering smaller facilities, more privacy, and more intensive services with greater expectations for participation. While shelter services address basic needs (food, clothing, a place to sleep), the services in transitional programs almost invariably extend beyond meeting survival needs. They tend to be coordinated by case managers and are geared toward helping residents define goals and achieve greater independence. Finally, transitional housing is almost always time limited, with lengths of stay usually capped somewhere between three months and two years. Emergency shelter stays, in contrast, tend to be either more limited (e.g., less than thirty days) or, in places where the courts have mandated the provision of shelter, unlimited.
Transitional Housing vs. Residential Treatment Programs
While it is in the context of widespread homelessness that transitional housing programs have proliferated, transitional accommodations play a role in other types of continua as well. As the locus of mental health care shifted from institutions to community settings, a graduated series of transitional residence programs emerged (Arce et al., 1982) to facilitate ex-patients’ adaptation to community living. The substance abuse treatment system, likewise, has produced a continuum that usually begins with detoxification, followed by short-term rehabilitation and long-term residential treatment—all conceived of as transitional steps en route to recovery. As homelessness has increased in the mentally ill and substance abusing populations served by residential treatment programs, they increasingly double as a transitional stage in both rehabilitation or recovery and in the process of exiting from homelessness. There has in fact been little consideration of how these two processes are related, whether the transition out of homelessness entails distinctive service and housing issues, and how these are related to those of recovery and rehabilitation.
Transitional Housing vs. Permanent Supportive Housing
The boundaries between transitional and permanent housing are clearer, although here, too, there are ambiguities. In general, transitional housing is time-limited; permanent housing is not. And, when residents reach the time limits built into transitional housing, they are expected to "graduate" to more independent, "normal" housing settings. Thus transitional housing is a stage in a progression, while permanent housing entails no assumptions about personal growth and development. An additional factor that sometimes distinguishes transitional from permanent housing is tenancy rights. For transitional residents, tenure is usually contingent on participation in services and compliance with program rules, whereas permanent tenants usually hold leases and have full tenancy rights.
Why Transitional Housing? Evolution of Concepts and Practice
During the 1980s, there were major changes in both the population affected by homelessness and in government and community responses. These shifts were reflected in the 1987 Stewart B. McKinney Act, which supported a varied set of housing and service programs to assist an increasingly diverse homeless population that included growing numbers of homeless women with children and homeless youth, as well as individual adults, many with disabling health, mental health and substance abuse conditions. McKinney funds also spurred growth in the numbers of not-for-profit agencies serving the homeless, as new grassroots organizations, critical of government approaches to homelessness, emerged to develop alternatives that were more responsive to needs of homeless individuals and families as well as the communities affected by homelessness.
In 1992, HUD was charged with coordinating plans and consolidating programs involving homelessness. It’s 1994 report, "Priority: Home! The Federal Plan to Break the Cycle of Homelessness" (Interagency Council on the Homeless, 1994), recommended implementing the "continuum of care" concept, intended to bring together all parts of a community interested in addressing homelessness; providing a coordinated system of care for the homeless; building partnerships among localities, states, nonprofits, and the federal government; and encouraging localities to seek long-term solutions.
Development of Transitional Housing for Homeless Families
The increasing numbers of families joining the homeless population in the 1980s quickly overburdened existing emergency facilities in many locales. In New York City, for example, where needs vastly outstripped both emergency and permanent housing resources, homeless families could spend as much as ten days in overcrowded municipal intake centers where the only bed was a chair, the only meal served on paper plates with plastic utensils, infectious and communicable diseases spread rampantly, and fights broke out among clients and staff. Shelters were also overloaded, and as New York continued to lose affordable housing throughout the 1980s, lengths of shelter stays increased.
As homeless families in New York and elsewhere stayed longer in temporary accommodations, providers and families alike recognized that many would benefit from (and others desperately needed) special help and services. The community-based not-for-profit agencies that took an increasing role in sheltering homeless families created facilities that were not only smaller in size and better managed, but also provided more services than shelters operated by government or disaster relief agencies. By incorporating parenting classes, counseling, substance abuse treatment, basic education and job training into the accommodations provided for families, community organizations thus transformed crisis shelter facilities into full-service transitional housing while shifting focus from the systemic problem of not enough permanent housing to the more tractable problems of individual families that could be addressed through services. Many transitional programs began requiring clients to sign contracts stating they will work on personal goals; others mandate classes and workshops; and some contain a research component, assessing each client upon intake and evaluating her progress while in transitional housing. The Continuum of Care encouraged the development of such services and their integration with other resources in each locality.
Development of Transitional Housing for Homeless Individuals
Transitional housing for homeless individuals has also been developed by service providers to address unmet needs and to fill gaps in the existing continuum of services. Initially these were often small-scale efforts—for example, renting a few rooms at a YMCA to temporarily accommodate individuals receiving case management services from street outreach or drop-in center programs. When agencies serving homeless individuals with mental illness and substance abuse problems found that providers of permanent housing for people with these disabilities were reluctant to accept homeless clients, they were able to use HUD McKinney monies to develop more service intensive transitional housing programs designed to enhance "housing readiness" by fostering the skills needed to get and keep housing: budgeting and money management; household management skills including shopping, food preparation, nutrition; and, where relevant, self-management of medication and achieving and managing sobriety or recovery. Some programs also offer vocational and employment services. At the same time they build skills, transitional programs also help residents meet stringent entrance criteria used by many providers of permanent supportive housing: a specified period of time clean and sober, compliance with and management of a regime of psychiatric medications, willingness to participate in structured services or activities. In such contexts, transitional housing can provide the requisite credentials for admission to permanent housing.
The Knowledge Base for Transitional Housing Practice and Research
Providers have accumulated considerable experience with transitional housing over the last two decades. Some of this is represented in program descriptions (Fogel, 1997; Lowery, 1992; Proscio, 1998), conference presentations (Lipton, 1993; National Resource Center on Homelessness and Mental Illness 1995; O’Hara 1997; Porterfield 1997), and technical assistance manuals (HomeBase 1998; Sprague 1991; Meister 1993), and taken together, these sources document a rich variety of program approaches and practices. However, description and documentation cannot reconcile the sharply contrasting perspectives on the value of transitional housing as a means to self-sufficiency and stable housing. This requires more systematic examination of the various transitional housing models as well as assessments of their outcomes as compared to other approaches to these goals. These kinds of studies are few in number. The most broadly based reviews of transitional housing have been done as part of the evaluations of HUD’s supportive housing programs, which provide an overview of the programs the agency has funded, the populations they serve, and their short-term outcomes (Matulef et al., 1995; More Than Housing 1995; National Resource Center, 1995; 1997). These reviews are complemented by the results of local evaluations of HUD-funded programs (e.g., Wilder Research Center, 1998).
To assess the effectiveness of transitional housing requires research designs that control for other factors that may influence outcomes while comparing transitional housing programs to policy-relevant alternatives. Evaluations conducted as part of the McKinney AFDC Transitional Housing Demonstrations supported by the US Department of Health and Human Services (HHS) provide almost the only controlled studies of transitional housing for homeless families (Nathan et al., 1995; Roman and Zhu 1996). For homeless individuals, research demonstration efforts have been supported by the agencies that serve mentally ill and substance abusing populations—CMHS, NIDA, NIAAA, NIMH, VA. While most of these projects have focused on services or permanent housing, a few have included transitional housing models (e.g., Grella, 1993; Rahav et al., 1997; Ware et al., 1992). A small number of studies have focused on outcomes for particular types of transitional housing (Barrow and Soto, 1998) or for individual programs (e.g., Baier et al., 1996; Prabuki et al., 1995). In general, however, the accumulation of experience in providing transitional housing has not been matched by research on its effectiveness. Thus while we offer a synthesis of descriptive writings on transitional housing, our assessments of practices and models that are most effective in helping homeless individuals and families successfully exit from homelessness are necessarily provisional. More definitive evaluation of particular approaches or of transitional housing more generally must await the development of more rigorous outcome research that will complement the elaboration of practice in this area.
The programs encompassed by the transitional housing label are remarkably diverse. They draw on a range of philosophical and disciplinary traditions, are targeted at different subgroups of the homeless population, utilize a broad array of physical structures, and offer differing configurations and intensities of services, as well as contrasting levels of demand for service participation. Their admission criteria, lengths of stay, and strategies for facilitating moves to permanent housing also differ, and they are embedded in geographical and resource environments with enormous contrasts in available services and housing. These factors do not exhaust the types of variation that characterizes transitional housing, but they define some of the most important dimensions of difference among the programs. We review here how program characteristics are distributed along several of these dimensions and describe some of the more typical program configurations that result.
Target Populations
Transitional housing is aimed at subgroups of the homeless population thought to need special assistance in transitioning to permanent housing. These include homeless families, battered women, youth, mentally ill individuals, persons with substance abuse problems, people living with AIDS or HIV, developmentally disabled and physically disabled persons. Because there is considerable heterogeneity within these categories ("homeless families," for example, may consist of single parent households with children; pregnant women with and without children; couples with or without children; and women whose children are in the foster care system or are living with relatives), as well as overlap among them (mentally ill individuals, for example, may also be members of homeless families, victims of domestic violence, people living with AIDS, or substance abusers), the targeted groups do not neatly coincide with program types. Thus while most transitional housing programs focus their efforts on either homeless families or homeless mentally ill individuals, the form a program takes will be influenced by "secondary" issues the residents are also dealing with. In addition, some programs target their efforts at groups defined by differences in their homeless experiences—"street" homeless or long-term shelter residents. Variations in the target population have significant implications for many other elements of the transitional housing program—including but not limited to its needs for and use of space, the services provided and those available in the community, the funding available, and the availability of resources for permanent housing.
Type of Physical Structure and Privacy of Living Space
Transitional housing programs run the gamut of available building structures—from former convents, schools, or SROs to multi-unit dwellings and private homes. "Stand alone" transitional programs occupy entire buildings, while "clustered" apartment programs group transitional residents together in a separate wing or floor of a larger structure, and "scattered-site" apartments are dispersed through "regular" apartment buildings or housing projects. Variations in the physical structures have implications for the amount of privacy individuals and families have; the availability of space for on-site programming; and the degree to which transitional tenants are integrated into the building and broader community.
Stand Alone Programs
Stand alone programs include congregate facilities and multi-unit apartment or single room occupancy (SRO) buildings. Some congregate programs for families have been created by converting SRO buildings, former schools, or convents into housing for pregnant women or small families. These are often older buildings that require significant capital improvements to make them usable by families. Congregate facilities are also frequently used as residential treatment settings for homeless individuals or families with problems of mental illness and/or substance abuse. In programs for individuals, single or double rooms (or in some cases dormitories) comprise the sleeping areas; there may be offices, storage rooms, reception areas, kitchen and dining rooms, and staffed program space on-site. The congregate setting allows individuals to learn or hone skills for independent living while participating regularly in treatment and rehabilitation services on- or off-site. Transitional residential treatment programs for women with children usually use a rehabilitation treatment design, based on the development of a "community". Child care is the most critical design feature that supports women coming into and staying in a 10-12 month recovery program, and play, study, and day care spaces are designed to meet the needs of children.
Individual apartments in small, medium and large multi-unit apartment buildings constitute an alternative variant of "stand alone" transitional housing. These are the most frequently used accommodations for homeless families in medium to large cities (HUD Report to Congress, 1995:17). Most of these facilities are owned by not-for-profits and operated through contracts with the local government unit responsible for housing homeless individuals and families. A combination of public and private resources support the building purchase and renovations, including building out office, childcare and services space. Located in communities where homeless people previously lived, most multi-unit residences became vacant because of fire, abandonment by the landlord or disrepair. Rebuilding and turning them into homeless residences aids in the restoration of neighborhoods while providing families who have never held a lease the opportunity to prepare for moving out on their own.
Shared apartments may be created by subdividing larger apartments or by reconstructing small units in a building into larger apartments. In rural communities, shared single family homes may be used to accommodate two to three homeless families. The families are expected to contribute to the upkeep and pay some rent based on a sliding scale. These settings tend to have little space available for program activities, childcare and services, and only limited services are provided on-site. Shared apartment programs may, like congregate settings, occupy entire buildings; or the apartments may be either clustered or scattered within buildings housing non-homeless individuals and families.
Clustered or Scattered-Site Apartments
Clustered or scattered-site apartments differ from stand-alone programs in that they disperse homeless individuals and families through "normal" housing settings. Scattered-site apartments, in particular, are popular with families and may be cost effective for providers. Other than security deposits, costs for minor alterations, and the purchase of furniture and furnishings, the major costs are rent and utilities plus some staff. The apartments are already part of the community, minimizing problems of siting and stigmatizing homeless families. Families requiring limited on-site services make good candidates for scattered-site housing, as services are primarily offered at a central location off-site, although staff may make regular or unscheduled visits to the apartments to monitor activities. For families seeking to be reunited with other family members, this type of "residence" provides both the flexibility families desire and the structure required for monitoring by caseworkers or family court staff.
Scattered or clustered units of transitional housing for individuals are often sited within privately owned apartment buildings, YMCAs, or commercial SROs that also house other transient and/or low-income groups. When the transitional program leases a small number of units, services are typically provided off-site or by mobile teams that visit residents. The scattered rooms or apartments provide a more normal living environment for those who reject living in a "program" setting, but the delivery of services requires greater effort on the part of both the residents and service providers. Some providers are able to lease a wing or floor of a building, providing a cluster of transitional units within a larger structure. Supportive services—case management, group activities, various clinical services—may be available on site, although some clustered programs use a visiting or off-site service model.
Clustered or scattered apartments shared by two or more individuals also provide "next step" housing for mentally ill or substance abusing individuals who have graduated from congregate facilities but can benefit from continued support. The apartments, which are leased by the provider agency, usually have single or double bedrooms but common living, kitchen and bathroom space. Individuals are expected to attend off-site treatment, rehabilitation or vocational programs. Staff do not live on site but are on-call to provide crisis services. In the most structured programs, case managers visit frequently, even daily, in the first months, to supervise and monitor residents’ ability to handle living in an unstaffed setting. Visits taper to perhaps twice monthly over time. Graduates are candidates for either minimally-serviced "graduate housing" or fully independent housing in the broader community.
Services: Location, Type, Intensity and Level of Demand
It is the services provided in conjunction with transitional housing programs that are expected to provide a "bridge to self-sufficiency and permanent housing" (Sprague, 1991). As with every other dimension, there is broad variation in where services are provided (on-site, off-site), the types of services provided (case management, concrete assistance, child care, supportive services, living skills training, employment training and job placement, and medical, mental health and substance abuse treatment), and intensity or level of service participation required (ranging from low barrier, low demand services to high intensity, high demand programs providing specialized treatment services). Whether and how services are "sequenced" also varies.
Location of Services
Location of services on-site or off-site depends on both the physical space available at the site, the availability of the needed services in the local community, and the programs’ expectations about residents’ ability to access community-based services. Highly structured service-intensive programs using congregate housing settings are most likely to provide services on-site, where participation and compliance can be encouraged and monitored. Safe Havens-type programs using low barrier, low demand approaches may also provide some services on site as a way of minimizing service barriers. However, programs for individuals and families with less extensive service needs often expect residents to access mainstream services in the larger community. This approach also reduces the institutional quality often associated with serviced residences and the stigma associated with "facility"-type living.
Level of Demand
Level of demand refers to the expectations and requirements for service participation that transitional programs impose on residents and the ways in which they enforce these expectations. At the "low demand" end of the continuum, programs for individuals and families operate somewhat differently. Low demand programs for homeless families tend to target those who do not need intensive, specialized programming but who might nonetheless benefit from supportive services, including child care, education and training, and other "soft" services like counseling and life planning and housing relocation assistance, which includes identification of public and private housing resources, preparation for interviewing with landlords or tenant groups, transportation, childcare, understanding the financing and lease arrangements, and help with budgeting, securing household furniture, security deposits and moving expenses.
Low demand transitional programs for individuals, in contrast, are intended to minimize barriers that prevent people from seeking or accepting assistance. They are primarily directed at those who are distrustful of housing and services programs and easily deterred from seeking or accepting assistance—usually mentally ill individuals who cannot or will not comply with the contingencies that accompany most offers of help. These programs for include "Transitional Living Communities" (TLCs), Safe Havens, and a variety of interim accommodations that serve those described as the "hard to reach" homeless (Lipton, 1995; National Resource Center, 1997). Some low demand programs, such as Safe Havens, are exclusively directed at individuals with severe mental illnesses and were developed to "support and assist them in overcoming specific problems that impede access to permanent housing, and develop the integrated supports needed for successful residential tenure" (Interagency Council 1994, p. 39).
Low demand programs for individuals are predicated on the assumption that a supportive, low demand environment with flexible admission criteria and few initial requirements—a low threshold, low barrier approach—is conducive to engaging those who are most estranged from the service system (National Resource Center, 1995; 1997) and will allow staff to build trust and prove their responsiveness before they "contract" with residents for more intensive service involvement. Such programs pose challenges for both staff and residents. Staff may find the flexible, individualized approach is at odds with their training and experience and may have difficulties defining and enforcing rules in ways that are consistent with a low demand philosophy (Erickson & Page, 1997; Proscio, 1998). Residents may find the increase in expectations and requirements over time to be confusing, and the individualized application of rules and expectations may appear unfair or inequitable. Administrators of one low demand transitional residence have noted the "paradox of an effective Safe Haven is that its greatest strength—that of creating a warm, accepting, engaging environment where residents feel respected and safe can also be its downfall—that residents will not want to leave" (White et al., 1997:13). They suggest, however, that "safe havens" foster a sense of possibility and entitlement that residents carry with them into more independent settings. In locales where permanent housing options are limited, inappropriate, inaccessible or unappealing to residents, providers often need to become advocates for the expansion of permanent housing (White et al., 1997).
At the "high demand" end, transitional housing programs for both individuals and families tend to entail specialized treatment and rehabilitation services and are configured in ways that reflect the special needs of those they are designed to serve. Homeless individuals and families troubled by serious mental health and drug abuse problems, domestic violence or HIV/AIDS are among the subgroups whose needs have spurred the development of specialized "high demand" residences. More recently, welfare reform and increased federal funding for vocational and job training and placement have resulted in the development of specialized transitional housing/employment projects oriented toward moving homeless families off welfare and into work. Unlike specialized programs based on disability, these projects target families with members closer to being "job ready".
Specialized "high demand" programs are service intensive. They are based on the assumption that homelessness and residential instability are prompted or exacerbated by problems and disabilities such as mental illness, addictions, and serious health conditions. To achieve residential stability and a satisfying quality of life, individuals with these problems need interventions that address their treatment and rehabilitation needs along with their housing needs. A structured environment is viewed as necessary for individuals with psychiatric disabilities and substance abuse problems to develop the skills needed to avoid relapse and function more independently. Residential treatment programs offer a context for learning and practicing life skills such as managing money, medication, and sobriety, vocational training, job placement, etc. They usually also require that residents engage in clinical treatment, and most also provide services directed at self-management of medications and sobriety—as a means of sustaining clinical gains and to help residents qualify for admission to supportive housing. Residential treatment programs oriented toward rehabilitation and growth encourage residents to meet high expectations and to continuously improve their levels of functioning. Time frames range from three months to two years.
Specialized programs usually involve substantial structure and supervision to inculcate skills and behavior seen as necessary for success in long-term housing, although the particular mix of services offered varies considerably. They tend to impose the most stringent requirements on newly entering residents, whose sobriety, medication compliance, and program attendance are closely monitored through random urine testing, directly observed medication administration, and signed attendance forms for off-site programs. Curfews, participation in group activities, required room clean-ups and other chores may be strictly enforced, with penalties (e.g., restricting off-site travel, requiring an escort) imposed in response to rule infractions. Over time, residents earn greater privileges and independence. When they have demonstrated sustained compliance with program requirements and adequate skills managing money, medication, treatment participation, and household maintenance, they are deemed ready to graduate to more independent settings.
"High demand" programs often do double duty as treatment settings and as preparation for permanent housing. While these two aspects of program mission are usually viewed as mutually reinforcing, they can create conflicting demands and present residents with dilemmas regarding treatment, education, training, work, and permanent housing. The extended duration of residential substance abuse treatment programs, for example, often conflicts with the time it takes to become eligible for permanent housing or with court deadlines for decisions regarding custody of children or time limits on welfare eligibility. In these situations, residents who remain in treatment may forfeit custody of their children or lose an apartment or exceed the time limits on receiving benefits. Additional issues emerge from the "high demand" structure itself, which imposes rules and requirements that many residents fail to adhere to, leading to high rates of attrition and early discharge in the most structured programs. Specialized transitional housing programs need to develop ways of handling these setbacks with clients who return to alcohol or drug use, refuse to take prescribed medication, or engage in other behavior considered detrimental to their safety and the safety of their children. Questions of how to serve those who do not remain in these programs, and the impact of "failure" on subsequent service efforts remain unaddressed and unsolved issues.
Types of Services
Types of services provided in transitional housing programs vary with the populations targeted. Low demand approaches such as the "safe havens" programs for homeless mentally ill individuals initially focus on providing basic services and necessities (e.g., telephones, a mailing address, and a place to securely store belongings) but over time seek to engage residents in goal-oriented services that will address their mental illness and other barriers to housing.
Programs for homeless families usually offer a range of supportive services—job training/placement, child care, substance abuse treatment, mental health services, and instruction in independent living skills—though specific combinations and approaches differ. The service components of these programs have been designed to respond to the emotional distress brought about by dislocation as well as to address longer term or ongoing problems. In addition to parenting classes, budgeting, and small home repair classes, transitional programs increasingly offer training and employment services geared toward preparing adult family members to meet local welfare-to-work requirements or to enter the job market. These services will likely increase with the impact of welfare reform. Housing relocation assistance is also considered a necessary service for success in permanent housing. This assistance includes: identification of public and private housing resources, preparation for interviewing with landlords or tenant groups, transportation, childcare, understanding the financing and lease arrangements, along with budgeting, securing household furniture, security deposits and moving expenses. In urban areas homeless families, notably those with a history of domestic violence, are often relocated away from familiar resources, shopping, family members and friends. Identifying resources in the community for childcare, education and training, medical and other social service needs can be critical for a family’s successful transition into a new community.
Twenty-three percent of homeless mothers have been found to have needs that span the areas of "human capital" (education and employment), mental health and substance abuse (Rog et al., 1995). These families and those headed by parents with psychological problems, victims of domestic violence, teenage mothers with no prior independent living experience, as well as individuals with dual problems of mental illness and substance abuse may need more extensive services in transitional residences.
Specialized treatment services are often provided in transitional housing programs designed for individuals and family members with substance abuse problems. Since drug and alcohol abuse have an enormous impact on all family members, programs targeting homeless families may need to address a variety of issues related to substance abuse, and treatment and intensive services can effectively address many of the most serious consequences for infants, older children, and adults. Treatment opportunities make it possible for pregnant substance abusing women to deliver babies without a positive toxicology thereby reducing or eliminating physical problems associated with alcohol and substance abuse. The impact of alcohol and substance abuse on pre-school and older children may necessitate a variety of specialized services, including therapeutic day care and professional staff (family therapists, pediatric social workers) that can recognize and treat children’s stress related illnesses and behaviors. Structured service-intensive programs allow adult residents to proceed with recovery gradually gaining more control and independence. Women find that it is possible to be reunited with older children in foster care and with family members from whom they have been estranged. Specialized transitional residences provide a context for the women not only to remain sober and drug free but to learn and practice daily living skills needed for independent living, vocational training, and eventual work. Success in treatment has been shown to directly impact on a resident’s ability to remain in permanent housing over time
Across widely differing programs, case managers are the linchpins of the service effort, working with residents to define individualized goals, coordinate resources and services, and monitor progress. In low demand "safe havens" programs for homeless mentally ill individuals, case managers have the task of establishing a trusting relationship with persons who may reject service offers and are willing to return to the streets if efforts at service engagement are too aggressive. Even in programs that impose demands for service involvement, however, case managers’ effectiveness depends on the relationship established with the homeless individual or family. Regardless of whether programs use team or individual, clinician or broker case management approaches, case managers in most transitional housing programs combine advocacy work, counseling, skill development and service coordination functions.
Service Sequencing
Service sequencing, which specifies the order in which services are offered, is practiced by many programs, but they vary in the order they specify. Low demand programs like Safe Havens usually begin with "low threat" services such as those addressing basic needs and those which entail little risk by consumers. They offer more intensive, potentially threatening services—substance abuse or mental health treatment, educational programs, others that carry the risk of failure—only after consumers are engaged, unless specifically requested at the outset. High demand programs are more likely to require that consumers achieving some substance abuse or mental health treatment goals before they are offered employment training, housing search assistance, or educational programs. There is variation across programs in the flexibility with which these sequencing approaches are used.
Making the Transition: Program Admission, Tenure, and Moving On
Transitional housing programs gear the services they provide to changing both the behavior and life circumstances of those whom they accommodate. Across diverse program contexts, the process of making these changes is described as becoming "housing ready". However, the specific ingredients of housing readiness are contingent on what is required to obtain permanent housing for a given population in a given locale, as well as the program’s judgment about what experience and skills will be necessary for maintaining housing, once acquired. Thus there is some variation in how transitional housing programs define and approach housing readiness at various points in the process of moving from homelessness to housing.
Admission
Funding sources and local political priorities have a primary influence on what population groups are targeted for transitional housing services. Programs may be mandated to serve specific segments of the homeless population (literally homeless people living on the streets; long term residents of shelters; people with mental illnesses or substance abuse problems; "multiproblem" families, etc.). However, within a targeted population, thresholds for admission to transitional housing programs are often related to the program's level of demand. "Low demand" settings such as safe havens seek to minimize barriers to entry and impose few requirements. Over time, as residents and staff build relationships of trust, expectations are increased. Motivation is expected to result from rather than precede the service process. Service intensive programs at the "high demand" end of the continuum, in contrast, screen potential residents to identify those who are motivated to make changes in their lives and are thus more likely to comply with the program structure and rules. Those who fail to do so will lose privileges and may be asked to leave the program.
Tenure and Tenancy Rights
Transitional programs vary significantly in the kinds of rights residents have over the terms and duration of tenure. Programs that provide intensive services, particularly those addressing mental illness and/or substance abuse, usually retain the right to determine not only who enters the program, but how long they stay and when they should leave. These programs often emphasize their "residential treatment" nature, with the provision of housing seen as a factor that facilitates the treatment rather than being the program's primary purpose. In these clinically administered settings, tenure is conditional on participating in services; residents may sign occupancy agreements but do not hold standard leases and do not have the "tenant protections" that are available to renters under local landlord-tenant laws. In some low demand interim or safe havens programs, tenure is less conditional on adherence to strict rules, and the length of stay may be "indefinite" for those who are unable to move to more independent settings. However, here too, it is the transitional housing provider who ultimately determines whether a resident stays or goes. In contrast, in some low demand housing programs—particularly scattered-site housing for families—residents have leases and thus whatever protections are offered under local landlord-tenant law.
The nature of tenancy protections for residents in transitional housing is closely tied to local or state ordinances and the extent to which these support tenancy rights for transitional residents. Ironically, transitional housing providers in some localities with strong legal protections for residents, such as New York City, may seek to limit lengths of stay for transitional housing residents because of the difficulties they face evicting residents who decide to claim their rights as tenants, even when the rent is paid by the program. Thus some programs in New York City restrict stays to 28 days or less or require residents to temporarily leave after 28 days when residents acquire tenancy rights in the premises where they are staying.
While local laws exercise some effect on tenure and tenancy in transitional housing, the duration of stay is strongly related to the availability of permanent housing. Where housing is in short supply or difficult to access because of cost or high admission barriers, residents may need to remain longer in transitional programs in order to locate permanent accommodations, to qualify for admission (e.g., logging clean and sober time or months of medication compliance), and to move through waiting lists. In many locales, transitional housing has become the only or major means through which homeless families can gain access to public housing and other subsidized accommodations. A limited number of vacancies results in longer stays in transitional housing, which becomes a device for titrating the flow of families or individuals to the level that can be absorbed by available housing resources.
Moving On: Links from Transitional to Permanent Housing
Within-agency and between-agency referral linkages, co-location approaches, and convertible housing are all efforts to smooth and ensure a viable transition into permanent housing. These approaches have been developed in response to a number of challenges that transitional housing poses for those seeking to end homelessness. Transitional programs sometimes seek to establish informal or formal referral agreements with particular supportive housing providers in order to enhance their residents’ access. This can be particularly effective when the same agency operates both the transitional and permanent supportive housing programs, making movement between the two relatively seamless. "Co-location" (also called "combination" housing) of transitional and permanent accommodations within the same building; and "convertible" housing, in which staff and or services are phased out after a transitional period and the housing itself "converts" to permanent are two newer approaches to facilitating transitions to permanent housing.
Combination Housing
Combination Housing consists of transitional and permanent apartments "co-located" in the same building. Usually these are new construction, owned by a not-for-profit or community development corporation. The financing is complicated and involves using multiple resources for construction such as tax credits, construction loans, grants; debt service is paid out of contracts for the homeless families, rent and section 8 and other types of housing subsidies. The permanent housing is developed as an out-take for homeless families.
In a number of communities "co-location" (combination) projects provide transitional and permanent housing as next steps for families that have completed substance abuse treatment programs or as a spectrum of options for people who are mentally ill. Some of these also combine units for single individuals and for families within the same building. Co-location approaches assume that movement along a service-housing continuum is facilitated by locating the various programs under one roof (Proscio, 1998). One variant of "co-located" housing is particularly targeted at those who are reluctant to leave a known environment with familiar supports for a more independent setting because it will require severing valued ties and reconstructing a support system from the ground up. This model is also seen as well-suited to those whose progress towards independence may not be linear—e.g., severely mentally ill or MICA individuals. The on-site availability of a less demanding setting permits individuals to weather a period of relapse without being fully dislocated.
Convertible Housing
Convertible Housing offers an alternative way to link transitional and permanent housing. Unlike co-location, which entails the movement of individuals or families within the same building, this approach permits residents to convert the terms of their tenancy from temporary/transitional to permanent. In particular, families with minimal needs for services often find that scattered-site housing is exactly what they need, and they would like it to be permanent. This approach has been successfully implemented in a variety of locales, including Minnesota, Massachusetts, and parts of New York State. Denver offers an example of transitional housing that was successfully converted from McKinney-funded housing for homeless individuals to permanent stable housing by shifting the funding source. Originally funded with McKinney monies a decade ago, the individuals residing in the program had achieved a decade of housing stability and could hardly at this point be considered homeless. Rather than continuing to use McKinney homeless monies for what was clearly stable permanent housing, providers were able to shift the source of funding to regular section 8 certificates, freeing up McKinney monies to develop housing for currently homeless individuals.
A small number of providers are currently experimenting with similar approaches for people with substance abuse problems or severe mental illness. These programs gradually withdraw services and staff as residents’ rehabilitation proceeds. "Sober housing" provides for the continuation of services for families while allowing for the gradual withdrawal of services as recovery progresses. Two types of sober housing have evolved: single family homes and apartments. In single family homes, two to five women in recovery take on the responsibility of managing a home. The lease may initially be in the name of the not-for-profit but the goal is eventually for the group of women to take over the lease. An alternative approach involves using rental subsidies to obtain scattered site apartments or an apartment building. The HUD Shelter+Care program pays fair market rents thus making it possible for residents to move into the next stage of housing before moving into permanent housing. Agencies apply for the subsidy which they must match with supportive services.
A small number of agencies have developed convertible housing programs for mentally ill individuals that after a period of time shift from transitional to permanent housing. One impetus for this model comes from a recognition that "graduation" from a supportive transitional residence to an unknown setting where a support system must be constructed anew is a dubious reward for mentally ill and other vulnerable individuals who have successfully completed transitional housing programs. The convertible model assumes from the start that as residents acquire skills for living more independently, less intensive staff involvement will be required. However, rather than expecting residents to move on to minimally- or un-staffed settings, the program itself reduces staffing and services and converts to long-term "graduate" housing. There are few examples of fully implemented versions of this model. However, experiments in New York and Boston bear watching.
While transitional housing has been developed as a means of addressing the barriers that homeless families and individuals face in their efforts to exit homelessness, several challenges remain to be addressed if transitional housing is to be an effective means of accomplishing this exit. First, transitional programs generally reward those who do well by requiring them to move on. This has been criticized by proponents of "supported housing" (normalized housing with flexible off-site supports) for mental health service consumers as well as by others who note that such moves can be destabilizing, removing valued supports for individuals not easily able to construct new ties and support systems. A second issue is that transitional housing programs can only be effective if adequate permanent housing is available. Many of the residents of these programs have difficulty accessing and supporting fully independent housing. Helping them move on requires adequate permanent housing and available subsidies plus the supports needed to sustain their tenancy. Moreover, an expanded supply of permanent housing and ready availability of flexible supportive services for those who need them might obviate the need for a transitional phase for some individuals who could do well in permanent housing with less extensive preparation. At present, however, competition for limited permanent housing and limited knowledge of how to identify those who most need the special help provided by transitional housing.
Sources and Types of Funding
McKinney funding administered by HUD has been the largest single source of funding for transitional housing programs. A variety of HUD programs—most notably the Transitional Housing component of the Supportive Housing Demonstration Program, but also the Supplemental Assistance for Facilities to Assist the Homeless (SAFAH) program—have supported the development of transitional housing. However, most providers draw on diverse funding streams and cooperative relationships in assembling funding for their projects. In addition to HUD funds, providers of transitional housing receive support from McKinney funds administered by other federal agencies—particularly HHS—through demonstration initiatives; by state-level social service agencies; by city or county-level government; and donations from foundations and private contributors. The Corporation for Supportive Housing has also supported the development of transitional housing models with funding from the Conrad Hilton Foundation. Funding streams affect everything from the scale of the project to the population targeted, the amount and focus of services provided, lengths of stay, and follow-up services. Restrictions on how funds can be used have also been one of the difficulties in implementing convertible approaches, although Denver's success in replacing McKinney funding with Section 8 Certificates illustrates one approach to conversion.
While HUD Supportive Housing funds have been underwriting virtually every type of transitional housing program, one issue of contention is the high proportion of HUD dollars going to support intensive treatment and other services rather than housing. Both providers and advocates have recommended that HHS and state or local mental health and substance abuse (as well as labor and education) agencies take on more of the responsibility for residential treatment and other service intensive efforts, freeing HUD money for housing.
Transition to What? Defining and Measuring Expected Outcomes
Transitional housing programs are designed to assist people in moving from homelessness to stable housing. Additional program goals vary with the subgroups targeted for services, with the way the barriers to stable housing are conceived and approached, and with the operant philosophy about how to overcome those barriers. To assess effectiveness of transitional housing, then, it is necessary to consider outcomes related to the specific barriers a given program is designed to address. In addition to housing outcomes, which reflect central goals in all transitional housing programs, these programs have mainly been designed to have an impact on four domains—services outcomes; behavioral or clinical outcomes; self-sufficiency; and cost effectiveness. Specific measures used to assess transitional housing's effects on these domains may reflect either short-term outcomes assessed at the point individuals or families leave the transitional setting or longer-term program goals that involve stability and self-sufficiency after leaving the program. Although conducting longer term follow-up entails complex logistics and can be expensive, only a long-range view of program effects can test key assumptions underlying transitional housing—e.g., that clinical and life skills services will enable individuals and families to weather the kinds of events and crises that previously resulted in homelessness and thus will contribute to residential stability.
Housing Outcomes
Housing outcomes are central to any assessment of the effectiveness of transitional housing programs. They include short-term measures of housing status at the conclusion of the transitional program, such as whether permanent housing is obtained; length of time required to obtain permanent housing; type and quality of housing; and residents' satisfaction. Longer term housing measures can consist of "snapshots" of housing status (housed/not housed) at a specified follow-up point or may reflect the stability of permanent housing over time—e.g., number of days homeless and number of days housed within a specified follow-up period. Operationalizing the relevant concepts requires clear definition of terms such as "housed" and "homeless", and conventions for categorizing ambiguous conditions (hospitalizations, incarceration in jail or prison, doubled up or living with relatives, etc.), and more complex assessments of residential patterns such as those developed by Hopper and colleagues (1997) and by Hurlburt and colleagues (1997) may be needed to assess whether transitional housing programs have successfully interrupted cyclical patterns of homelessness. It is also important to note that housing stability need not entail remaining in a single "permanent" housing setting. The critical distinction involves remaining housed as opposed to returning to homelessness.
Services Outcomes
Services outcomes include the extent to which a program successfully engages transitional housing residents in the services provided; the amount and range of services utilized by residents; the retention of residents in the program or in particular components; and linkages to community-based services that will be available following the transitional period. Transitional housing programs have been developed on the assumption that the services provided during the transitional period will equip homeless individuals and families to maintain residential stability after they move on. Service involvement is thus a proximal outcome that is expected to contribute to the longer range goals of residential stability and self sufficiency. While all transitional housing programs involve services, the specific service offerings are very different in programs targeting different segment of the homeless population, with corresponding variation in how service outcomes are measured.
Engagement is of particular concern to low demand programs such as Safe Havens that offer low-barrier access to a place to stay as an enticement for clients with psychiatric disabilities to enter the program and establish relationships with staff, seen as prerequisites to providing skill-building, clinical, and housing-focused services. Measures of engagement are not well developed and engagement is often inferred from participation in other services, but more direct indicators of engagement include talking with staff, disclosing information, agreeing to consider/accept more demanding services as well as actual use of those services. To determine how well transitional programs accomplish engagement goals will require development of reliable ways to detect significant changes in these domains.
Service utilization measures focus on how many services an individual or family uses, how frequently, and—for particular types of services—whether they complete the particular service program. In programs that require residents to complete a specific range of services to graduate, rates of graduation also measure service utilization. Where service plans are more individualized, service use must be measured within specific service domains. In interpreting use of services data, however, it is important to be aware that some programs practice service sequencing, i.e., requiring participation and/or completion of certain services (often substance abuse and psychiatric treatment, if relevant) before others—education, job training—are made available. In these situations, low rates of utilization of "higher level" services may be an artifact of the sequencing process.
Retention in transitional housing until "graduation" from services is often a program goal in its own right, but it is also viewed as a means of enhancing housing stability by equipping families and individuals to address problems that might otherwise result in loss of housing. However, relationships between housing outcomes and service retention can be complex and need to be assessed and interpreted cautiously. In localities where eligibility for subsidized permanent housing requires families to spend time in transitional facilities, families that fail to complete skill-building workshops can be discharged and thereby lose access to the major source of affordable permanent housing. In these contexts, analysis of relationships between housing outcomes and retention in services must take into account the likely confounding effect of the discharge policies.
Service linkages usually consist of specific referrals of aftercare arrangements that will ensure ongoing access to supports or treatment once an individual or family has moved to permanent housing. Most transitional programs also provide a period of follow-up case management services to enhance the likelihood that families and individuals will remain connected to community services such as child care, job training or mental health treatment. Measures of service linkage should capture the nature of the links established for various types of services.
Behavioral and Clinical Outcomes
Behavioral and clinical outcomes are often the major focus for transitional housing that doubles as residential treatment. Mental health and addiction treatment services are most likely to serve in this dual capacity. Residential treatment settings usually assume positive treatment outcomes are prerequisites for housing stability. Success is evaluated in terms of clinical outcomes and recovery/rehabilitation goals, as measured by, for example, decrease in psychiatric symptoms, adherence to medication regimens, abstinence or decreased use of drugs and alcohol, amount of time clean and sober, and level of functioning.
Self-Sufficiency
Self-sufficiency is viewed by many administrators and providers as the overarching aim of transitional housing programs, and the one toward which most of the service efforts are addressed. For families and some sub-groups of individuals, attaining self-sufficiency is usually conceptualized as a movement from welfare to work, and measures focus on income and employment. For severely mentally ill individuals, there is less expectation of total self-sufficiency. Measures of income and employment are more often used within this group as indicators of improved levels of functioning, but most are expected to continue to rely on disability benefits, perhaps in combination with part-time employment in supportive employment programs.
Cost Effectiveness
Cost effectiveness refers to the relative costs of achieving different levels of outcome, and thus its measurement is partially dependent on the measures of outcome discussed above. Cost effectiveness studies usually rely on measures of service utilization as reported by residents or documented from MIS systems or program records, in order to assess the service costs associated with successful outcomes—whether these are defined in terms of housing, services, behavioral measures, or self-sufficiency. At least one transitional housing demonstration program has focused on "cost neutrality" in developing innovative transitional housing services—i.e., the new service programs should be no more costly than the system of motels and shelters that was in use before.
Consumer Perspectives
Consumer Perspectives have not figured prominently in most research on transitional housing. Critiques of transitional housing programs coming from organized segments of the consumer population—notably mental health consumers and advocates for "supported housing"—emphasize that transitional housing is stigmatizing and note that graduation disrupts housing stability rather than fostering it. They argue that non-compliance and high attrition from transitional programs reflect strong consumer preferences for de-coupling services needed to consolidate a transition out of homelessness from housing, which should not be contingent on service participation. Research on transitional housing will benefit from expanding the focus beyond program-defined outcomes to prominently incorporate consumer-defined conceptualizations of outcomes as well. This will not only clarify factors that may contribute to attrition in transitional programs but will also increase the likelihood that efforts to resolve homelessness are responsive to perceived needs.
Relationships Among Outcomes in the Short Run and Over Time
In addition to measuring how well transitional programs meet their self-defined goals, it is critically important to consider how non-housing and housing outcomes are related. As noted above, programs have been developed on the basis of the assumption that housing outcomes are related to other program goals pertaining to service engagement, treatment, and rehabilitation or recovery. Research provides mixed support for this contention, as described below, and thus the relationships among outcome domains warrants closer scrutiny.
What Works? Program Experience and Research Results
In the last two decades, providers have acquired considerable experience in developing and operating transitional housing for homeless families and individuals, and individual agencies have used this experience to inform their own ongoing program development. In addition to these direct renderings of the experience of providing transitional housing, several programs have been the focus of individual case studies that provide detailed documentation of how transitional housing has been implemented in particular contexts (Hannigan & White, 1990; Lowery 1992; Blankertz et al., 1992; Proscio, 1998). National or local conferences of providers and policy makers and providers’ manuals that offer technical assistance to others developing transitional housing have been additional vehicles for communicating lessons learned from doing (Meister, 1993; Sprague, 1991; National Resource Center 1995; 1997). While these accounts are primarily descriptive and focus on issues of process and implementation, both public agencies that administer homeless services and the foundations and corporations that have traditionally provided nonprofits with supplementary resources are increasingly requiring outcome information and using performance indicators in monitoring nonprofit organizations operating transitional housing. Agencies have also implemented innovative program components that enhance their own monitoring and assessment of program outcomes and facilitate more rigorous evaluations.
Accounts of how specific programs work are complemented by more broadly-based descriptive research on federally funded transitional programs. The largest studies have focused on transitional housing funded under the Supportive Housing Demonstration Program and its successor, the Supportive Housing Program. In 1990, the General Accounting Office (GAO) reviewed HUD’s Transitional Housing Program to determine whether the program was serving the targeted population with a wide range of services, whether it was helping homeless people move to independent living, and what factors influenced successful transitions. The GAO conducted a telephone survey of program directors of 360 (94%) of funded projects and visited 32 of the project sites. The study found that the programs used a variety of facilities, ranging from converted warehouses or hospitals to renovated hotels, apartment buildings and newly constructed buildings. Lengths of stay varied enormously, with maximum duration ranging from one month to 24 months, the limit set by HUD. The programs provided—either directly or by referral—an array of supportive services: case management, housing placement, benefits or entitlements assistance, psychological counseling, job training, medical care, child care, and guidance in life skill, as well as specialized mental health and substance abuse services.
The GAO concluded that the program was reaching its intended targeted population of homeless families and mentally ill individuals; that about 40 percent of the individuals served by transitional programs succeeded in obtaining housing and a source of income upon leaving the program; and that the clients most likely to succeed were those who remained in the program longer and those who used more supportive services. Families and couples without mental health or substance abuse problems were most likely to succeed, and people whose homelessness was assessed by the GAO investigators as resulting from domestic violence, eviction, or money-related matters were more likely to succeed than those whose homelessness was attributed to mental illness. The GAO report noted that transitional housing programs tended to screen out those with mental illness or substance abuse, except in programs specifically targeted at those groups. While they were not always successful in excluding such persons, the selection process may have screened in the most motivated portion of the population and those most willing to accept the structure and rules of the transitional programs. Follow-up data were not available to assess longer term outcomes.
In 1995, the Final Report on Westat, Inc.’s National Evaluation of the Supportive Housing Demonstration Program (Matulef et al., 1995) updated the GAO findings. The proportion of participants who remained in transitional housing programs until they graduated was 57 percent; another 24 percent withdrew voluntarily, and 19 percent were dismissed. The mean length of stay for project residents was 9 months, with slightly shorter stays for battered women (8 months) and slightly longer for severely mentally ill residents (11 months). Overall, 56 percent of all residents (70% of graduates; 30% of those who withdrew or were dismissed) went on to stable housing—mostly in unsubsidized housing without services. Income and employment both increased. Although follow-up data were not available, program directors reported that there was great stability in the housing of those who had graduated from their programs. Project sponsors attributed successful outcomes to several factors: the availability of a safe, secure, private place to live; case management; screening for those who were most motivated to succeed; and a range of specific services. Impediments to success included pre-existing problems like mental illness and substance abuse; and community level factors such as lack of affordable housing and lack of employment or vocational opportunities.
Reports on implementation of state and local-level transitional housing programs offer additional evidence that substantial numbers of families and individuals move on to permanent housing, most express satisfaction with the transitional services and with the new housing they have obtained, and modest gains are reported in income and employment. (See, for example, Wilder Research Center's 1998 report on five years of the SAFAH project in Minnesota.) While both national and local descriptive evaluations of the major federal initiatives offer a general overview of transitional housing and the issues that have emerged in implementing it, few studies of transitional housing programs have attempted more rigorous assessments of the effectiveness of particular types of transitional programs on clearly defined subgroups of the population. Yet in the absence of studies that use experimental or at least comparison group designs, it is impossible to tease out the extent to which the positive outcomes reported in the descriptive studies can be attributed to the transitional housing programs or whether these outcomes are better or worse than the same population would achieve using alternative approaches. For homeless families, in particular, research of this sort has been extremely limited.
Studies conducted as part of the McKinney Transitional Housing Demonstration Projects for AFDC-eligible homeless families are therefore particularly notable, in that they raise and begin to answer some of the key questions about transitional housing as an approach to ending family homelessness as well as about the relative effectiveness of particular models. The AFDC projects were carried out in Massachusetts, New Jersey, and Westchester County in New York State, and were designed to develop service intensive transitional housing alternatives for homeless families. Evaluation reports from these projects describe local implementation of a variety of models of transitional housing for families and compare their costs and their effectiveness to "usual" services—i.e., shelter and hotels—in their locale. The models evaluated differ in a number of ways, with notable contrasts between the different projects. In Massachusetts, for example, all transitional housing programs targeted "multi-problem" homeless families, and all used scattered-site approaches, with three different models that, respectively, provided families with apartments in public housing projects, the private rental market, and newly refurbished subsidized units. All of the public units and a portion of both the private and refurbished "converted" to permanent housing as the families graduated from services. In Westchester County, in contrast, the models examined were all based in single-site facilities—two of them newly constructed as transitional housing for families, the third a notorious welfare hotel that underwent extensive rehabilitation to become a service-intensive transitional facility, although many of the families this program served had been residing there for years before the renovations.
While both of these studies used the existing system of shelters and motels as the comparison condition against which the transitional programs were assessed, their findings on effectiveness diverge. In Massachusetts, the families in transitional housing made greater gains in every area of outcome examined than the shelter/hotel families. The outcome analyses focused on behavioral changes in such areas as paying bills on time, children skipping school, using illegal drugs, or socializing with drug users. In Westchester, where the primary outcomes of interest were service utilization, reducing lengths of stay, and attaining stable housing, the differences between the demonstration sites and others were less striking: the participants in the McKinney programs were offered and used more services, but their lengths of stay in transitional facilities were slightly longer than those of families in non-McKinney programs, and there was no difference in the likelihood of obtaining permanent housing. However, the demonstration project coincided with a major reorganization of services for homeless families throughout the county, and across all types of facilities there were reductions in lengths of stay during the course of the study period. Moreover, the McKinney programs had less access than others to scattered-site Emergency Housing Units (EHUs) which were administered like the single-site programs but used regular apartments which, upon the family’s graduation, converted to permanent housing. This option—which closely resembles the demonstration condition in the Massachusetts study—proved a particularly effective—and cost effective—means of transitioning families out of homelessness. The Westchester investigators estimated that up to 61 percent of the families served in the McKinney programs could have been sent directly to EHUs instead, suggesting that the single-site intensive service model may have actually prolonged homelessness for some families. Moreover, analysis of cost data indicated that the longer stays in the demonstration models resulted in higher costs, whereas in Massachusetts the demonstration models actually reduced costs compared to the shelter/hotel alternatives.
The Massachusetts and Westchester studies also had both similar and contrasting findings on service utilization and service satisfaction. On the one hand, in both studies, families in the demonstration sites expressed the greatest satisfaction with their transitional housing experience and with the specific services they received. However, the Massachusetts project found satisfaction unrelated to initially perceived need, while the Westchester study found that those who felt coerced into using services they did not feel they needed were less likely to express satisfaction with them.
The evaluations of the McKinney AFDC projects makes it clear that a variety of transitional housing models can be used to enhance the services provided to homeless families. They also suggest, however, that even "multiproblem" families can be effectively served in the scattered site convertible models used in different ways in Massachusetts and Westchester County…if the local public and private housing resources are available. In Massachusetts, one of the three demonstration models actually refurbished and brought dilapidated housing back into circulation, thus expanding the available supply of units. In Westchester, the demonstration program augmented the stock of transitional units but did not address the county’s severe shortage of affordable permanent housing, and limited availability was a major reason that the successful EHU approach was not more widely used. Thus these studies serve as important reminders that in assessing transitional approaches, we must carefully attend to the context in which they are implemented, particularly in terms of the availability of permanent housing stock and mechanisms for making it affordable.
Going beyond the suggestive results from descriptive studies, these demonstration evaluations provide the groundwork for an empirically-grounded knowledge base of policy-relevant research on transitional housing for homeless families. Moreover, they underscore the importance of making the relevant comparisons—not only between differently configured combinations of transitional housing and services, but between transitional housing and alternative approaches to achieving housing stability that do not as a matter of routine subject families to a prolonged transitional limbo.
Research on transitional housing programs for homeless individuals is only slightly more developed than that on families. Few evaluations of low demand programs have been carried out, and most have been descriptive studies with weak study designs and limited follow-up data. A retrospective record study of 160 residents of low demand transitional residence programs in Philadelphia that offered psychosocial rehabilitation services (Blankertz et al., 1992) found that many residents improved their housing status (only 10% returned to the streets) and were linked to mental health services (39% entered treatment), but high rates of missing follow-up data and the absence of control groups make these results difficult to evaluate.
An evaluation of a program in Chicago that "co-located" transitional housing within the same structure as a shelter and permanent housing found that residents of the transitional program did indeed move on to permanent housing and in that sense, the program was a success. However, expectations that most transitional residents would enter the program from the co-located shelter were not borne out. At least in the early phases of the project, referrals to the transitional program from other agencies vastly exceeded the number of "in-house" moves. Similarly, most transitional residents went on to permanent housing elsewhere, with only a small number remaining within the co-located system. The evaluation report suggests that the pressures resulting from scarcity of all kinds of housing ensures that any available resource will be quickly filled, often with those who could do well in more independent settings if they were available. To ensure the units were fully utilized, the program was under pressure to accept these individuals rather than work on persuading more reluctant shelter residents to try the transitional setting (Proscio, 1998).
Another recent study of six "interim housing" programs in New York City (Barrow & Soto, 1996) compared outcomes of individuals entering interim housing operated in conjunction with outreach and drop-in center services with a matched sample drawn from outreach/drop-in programs that provided comparable supportive services but did not offer interim or transitional housing. The interim residents were significantly more likely to be housed three months after exit than the controls, although there were no significant differences in their involvement in service and treatment programs at follow-up. The study also found that within the interim housing group, longer stays and more intensive services were associated with better housing outcomes. These results suggest that low demand transitional programs can enhance the housing prospects for homeless individuals who are among the most estranged from the service system.
In another small group of studies, researchers have looked at the effectiveness of "high demand" transitional housing programs that offer specialized services for individuals with severe mental illness, substance abuse problems or both. These range from retrospective record reviews to experimental studies that randomly assigned subjects to alternative residential settings; and they consider a variety of outcomes, including retention in the program, housing status at exit and/or follow-up, employment and income, and in some studies, clinical outcomes such as changes in psychiatric symptoms and use of drugs and alcohol.
Studies of a transitional residence for severely mentally ill individuals in St. Louis, MO (Murray et al., 1995; Murray et al., 1997; Murray & Baier, 1995) based on a retrospective record review found that 48 percent of 228 individuals who participated in the program over a five and a half year period completed the program as planned—i.e., obtained housing and a source of income, while 18 percent were discharged without "graduating" and 22 percent left "against medical advice". Those who completed the program spent a longer time in residence, participated in more program activities and had a greater number of prior psychiatric hospitalizations. Two studies have been conducted in transitional Domiciliary Care for Homeless Veterans (DCHV) facilities (Prabuki et al., 1995; Leda & Rosenheck, 1992). One of these found that program participants showed improvements in clinical status (symptoms of mental illness and substance abuse), social functioning (employment and vocational) and housing status. However, they also noted that these domains of outcome were only weakly related to one another (Leda & Rosenheck, 1992). In the other study, completing the program was associated with improvements in housing, income, and vocational stability, but was not associated with improvements in psychiatric symptoms (Prabuki et al., 1995). Transitional programs thus may not have the same impact on the varied issues that homeless individuals grapple with in their efforts to improve their quality of life.
Retention in high demand programs, particularly those tailored to the needs of substance abusers, is a focus of several reports. In a Los Angeles study that compared a long-term residential program for homeless alcoholics with a briefer version of the program, the experimental group had higher rates of retention, but within both groups, retention in the initial program and housing placements at discharge were associated with race, gender and associated differences in employment histories, economic status, and subsistence patterns while homeless (Grella, 1993). The theme of low retention rates reappears most dramatically in the one controlled, experimentally designed study to assess transitional housing programs for homeless individuals. This study compared two approaches (modified community residence, derived from a mental health services tradition and modified therapeutic community, derived from a substance abuse treatment tradition) to transitional residential treatment for homeless mentally ill chemical abusers in New York City. While the findings showed that for those who completed the program, the modified TC approach was more effective, the most striking result was that only 13 percent of over 600 men who were referred and found eligible in fact completed either program. Moreover, significant attrition occurred even before admission to the programs, with additional attrition in the first sixty days after admission (Rahav et al., 1997).
Descriptive accounts of particular transitional programs and the findings that 40-60 percent of residents of HUD’s Transitional Housing Programs move on to permanent housing settings suggest that transitional housing can be effective in helping individuals and families accomplish an exit from homelessness. However, individuals and family members with substance abuse and mental health problems were less successful than other groups. While the limited research on safe havens and other low demand programs for homeless individuals suggests these can be effective alternatives to the streets and drop-in centers for mentally ill and other particularly vulnerable groups, the small number of studies that have examined these programs and the limited duration of follow-up indicate a need for more research in this area. Since low demand programs are often developed to fill gaps in a given agency’s continuum of services, evaluations need to avoid confounding the effects of the safe havens component and the more comprehensive service program in which it is embedded.
Research on transitional housing programs that offer specialized services such as mental health or substance abuse treatment to homeless individuals is also extremely limited. Most studies indicate that transitional housing residents improve their housing situations and often experience gains in clinical status as well. However, the weak relationship between clinical and housing outcomes challenges one of the basic premises of transitional housing programs that focus on providing treatment and services for mental illness or substance abuse problems and raises questions about the relevance of using clinical criteria to gauge "readiness" for permanent housing. In addition, existing studies either lack comparison groups or use control conditions that differ in treatment approach but not in housing condition. These studies also usually report only short-term follow-up and/or low follow-up rates. Thus it is difficult to assess whether, in the long run, individuals who go through transitional housing programs achieve more stable and/or independent housing than those who do not. Finally, and perhaps more critically, low rates of retention in transitional residential programs—particularly those for substance abusers and dually diagnosed mentally ill substance abusers—make it very difficult to determine how widely findings of positive outcomes can be generalized.
As with studies of transitional housing for families, because of limitations of method and design, the research on programs for homeless individuals offers only partial answers to key questions about transitional housing. Thus while studies of low demand programs suggest that adding transitional housing to the services offered in drop-in settings has an impact on housing outcomes, until we have studies that test transitional approaches against permanent housing approaches for comparable populations, it will remain unclear whether transitional housing offers a useful way to enhance housing access and stability or primarily serves as a substitute for permanent housing when the latter is inaccessible, undesirable, or in short supply.
Conclusions: Recommendations for Research and Policy
Descriptions of transitional housing programs can be found in annual reports, program brochures, and proposals submitted to funding agencies as well as in conference presentations, case studies written for practice journals or popular media, and technical manuals for those planning similar programs. When these sources are combined with the broad-based descriptive surveys of transitional housing programs undertaken by HUD, smaller descriptive studies of single programs, and research demonstration projects mainly supported by agencies within HHS, transitional housing emerges as an amazingly diverse set of programs directed at a varied population of homeless individuals and families.
Given the diversity encompassed under the umbrella of transitional housing, the limited research on these programs precludes easy generalization. Descriptive surveys of transitional housing programs tell us that a substantial number—usually around half—of the individuals and families that enter these programs go on to permanent housing—and among those who remain in the transitional programs until graduation, a much higher proportion obtain housing. They also confirm that individuals with psychiatric disabilities and/or substance abuse issues have greater difficulty obtaining permanent housing than families without these problems.
Although descriptive accounts of individual programs are plentiful and succeed in conveying many of the variations in target populations and program types, they remain largely outside the published literature. Moreover, they have focused overwhelmingly on three subgroups—homeless families, mentally ill individuals, and substance abusing individuals; they have mainly been carried out in urban and suburban settings; and though they document high proportions of their populations drawn from racial and ethnic minority groups, they tend to shy away from addressing issues of cultural competence. In addition, to the extent that transitional housing programs operated by or primarily staffed with consumers exist, they are notably absent from both the "fugitive" (i.e., unpublished) literature and from published research.
While descriptions of particularly notable or innovative programs sometimes do enter the published literature, well-designed studies of their effectiveness are rare. It is thus noteworthy that in the limited work on effectiveness of transitional housing for homeless families, there are convergent findings in favor of a scattered site model that eventually converts units from transitional to permanent housing. Moreover, in one variant of this model, dilapidated units are renovated and brought back into use, thus increasing the total stock of permanent housing. This approach has been shown effective even with multi-problem, hard-to-serve families, and thus may be widely generalizable—given availability of subsidies to maintain the affordability of the permanent units. The model not only seems to work; it does so at lower cost than single-site alternatives while also addressing critics’ concerns that transitional housing is stigmatizing, disrupts stability by requiring multiple moves, and siphons resources away from permanent housing development.
There is no comparable consistency in the findings from studies of homeless individuals. A similar model of convertible transitional units scattered through a permanent housing settings was one of several included in a study that found that adding transitional housing to drop-in or outreach program services increases movement into permanent housing. However, small sample sizes precluded testing the relative effectiveness of the various models.
Other controlled studies of transitional housing for homeless individuals are few. Moreover, studies of transitional housing for individuals have often focused on residential treatment programs and have been concerned only incidentally with their role as housing. Such studies indicate that attrition is a major issue for "high demand" programs providing intensive substance abuse and/or mental health services, and whatever their merits as treatment programs, they do not offer a route out of homelessness for most who enter them. And, despite the assumption that services addressing underlying clinical issues and skill deficits will enhance housing stability, research support for this is equivocal and raises questions about the relationship between housing outcomes and the clinical and skill-building goals of many transitional residence programs.
To date, then, the research on transitional housing programs offers only a few consistent findings and many major gaps. Thus we conclude with the few recommendations for policy that are suggested by recent research as well as several recommendations for the kinds of studies that are needed to develop more informed approaches to aiding the transition from homelessness to housing.
Expanding the supply of affordable housing should be the highest HUD priority. Inadequate availability of affordable permanent housing limits the effectiveness of all efforts to support the transition from homelessness to housing.
Scattered-site transitional housing units that convert to subsidized permanent housing reduce time families spend homeless, facilitate their transition to permanent housing, and avoid the stigma associated with single site programs, while using case management and community-based services to provide the support needed to maintain housing. Policies facilitating conversion should be supported and providers should be encouraged to experiment with developing convertible models for homeless individuals as well.
Research should focus on comparison of transitional housing with non-transitional alternatives: The most critical need is for studies that compare major variants of transitional housing with alternative approaches to helping people exit from homelessness (see Burt, 1997). Such comparisons can help resolve current controversies over the value of transitional housing and may provide a basis for determining what approaches work best for which subgroups of the homeless population.
For homeless families, research should compare transitional housing programs with several major alternatives—short-term housing assistance (rent arrearages, security deposits, negotiation with landlords), supportive/rehabilitative services, and long-term housing vouchers or certificates, as well as key combinations (transitional housing plus vouchers; supportive services plus vouchers).
Long-term follow-up studies are needed. We know almost nothing about what kind of service/housing combinations are needed to sustain housing stability over time for the various homeless subgroups. Studies of transitional housing and the alternatives described above should be longitudinal, with follow-up extending at least one year, ideally two, from the point of entry to permanent housing.
Consumer perspectives must be brought to bear on research questions, designs, and conceptualization of outcomes. Consumers’ voices have not been heard in most research on transitional housing. Researchers need to go beyond consumer satisfaction surveys and incorporate consumer perspectives and concerns in conceptualizing research questions, designs, and outcomes for transitional housing studies.
The importance of context. Evaluations of transitional housing programs cannot take as given the broader context within which these programs operate. Studies must, for example, take into account how contextual factors like shortages of affordable permanent housing may serve to inflate the importance of transitional housing and services. They must also contend with significant recent changes in that broader context—particularly the array of policies enacted under the rubric of "welfare reform"—that are altering and restricting the terms under which homeless assistance can be provided. As states enact workfare requirements and time limits that make no allowance for the "work" involved in the struggle to exit from homelessness, the once unthinkable disappearance of even minimal social welfare provisions may come to seem as normal and routine as the once unthinkable presence of widespread homelessness already has. Research that focuses on effective strategies for helping families and individuals exit from homelessness must centrally consider the effects of the already advanced erosion of the social contract.
Arce, A., Vergare, M., Adams, R. & Lazarus, L. (1982). A Typology of Community Residential Services. Report of the Task Force on Community Residential Services. Washington, D.C.: American Psychiatric Association.
Baier, M., Murray, R., North, C., Lato, M. & Eskew, C. (1996). Comparison of Completers and Noncompleters in a Transitional Residential Program for Homeless Mentally Ill. Issues in Mental Health Nursing, 17: 337-352.
Barnard-Columbia Center for Urban Policy. (1996). The Continuum of Care: A Report on the New Federal Policy to Address Homelessness. Washington, DC: U. S. Department of Housing and Urban Development.
Barrow, S. M. & Soto, G. (1996). Closer to Home: An Evaluation of Interim Housing for Homeless Adults. New York: Corporation for Supportive Housing, 1-103.
Bassuk, E. (1992). Homeless Families with Children: Research Perspectives.
Blankertz, L. E., Cnaan, R. A. & Saunders, M. (1992). Assessing the Impact of Serving the Long-term Mentally Disabled Homeless. Journal of Sociology and Social Welfare, 19(4), 199-220.
Burt, M. (1997). Future Directions for Programs Serving the Homeless. In Culhane, D. P. & Hornburg, S. P. Understanding Homelessness: New Policy and Research Perspectives. Washington D.C.: Fannie Mae Foundation, 361.
Erickson, S., & Page, J. (1997). Safe Havens: Program Rules and Expectations. Paper presented at Developing and Operating Safe Havens Programs, Washington, DC.
Fogel, S. J. (1997). Moving Along: An Exploratory Study of Homeless Women with Children Using a Transitional Housing Program. Journal of Sociology and Social Welfare, 24(3): 113-133.
Garrett, G. R. (1989). Alcohol Problems and Homelessness: History and Research. Contemporary Drug Problems 16: 301-332.
General Accounting Office. (1991). Homelessness: Transitional Housing Shows Initial Success but Long-term Effects Unknown. Report to the Chairman, Committee on Governmental Affairs, U. S. Senate GAO/RCED-91-200. United States General Accounting Office.
Grella, C. E. (1993). A Residential Recovery Program for Homeless Alcoholics: Differences in Program Recruitment and Retention. The Journal of Mental Health Administration, 20(2): 90-99.
Hannigan, T., & White, A. (1990). Housing Hard-to-Place Homeless Mentally Ill Women, The 350 LaFayette Transitional Living Community: A Status Report. New York: Columbia University Community Services.
HomeBase. (1998). Transitional Housing: A Bridge to Stability and Self-Sufficiency. San Francisco, CA: Center for Common Concerns.
Hopper, K., Jost, J., Hay, T, Welber, S., & Haugland, G. (1997). Homelessness, Severe Mental Illness, and the Institutional Circuit. Psychiatric Services 48: 659-665.
Hurlburt, M. S., Wood, P. A., & Hough, R. L. (1996). Providing Independent Housing for the Homeless Mentally Ill: A Novel Approach to Evaluating Long-Term Longitudinal Housing Patterns. Journal of Community Psychology, 24(3): 291-310.
Interagency Council on the Homeless. (1994). Priority: Home! The Federal Plan to Break the Cycle of Homelessness. Washington, DC: U. S. Department of Housing and Urban Development HUD-1454-CPD.
Leda C., & Rosenheck, R. (1992). Mental Health Status and Community Adjustment After Treatment in a Residential Treatment Program for Homeless Veterans. American Journal of Psychiatry, 149(9): 1219-1224.
Lipton, F. R. (1995). Housing Homeless People With Severe Mental Illness: Why Safe Havens? Paper presented at the Creating Safe Havens for Homeless Persons With Severe Mental Illness, Washington, D.C.
London, S. (1997). Developing and Operating Safe Havens: Staffing Issues. Paper presented at Developing and Operating Safe Havens Programs, Washington, DC.
Lowery, M. (1992). LAMP in L.A.'s Skid Row: A Model for Community-Based Support Services. In L. I. Stein (Ed.), Innovative Community Mental Health Programs, 56: 89-98. San Francisco, CA: Jossey-Bass Publishers.
Manhattan Borough President David N. Dinkins’ Task Force on Housing for Homeless Families (1987) A Shelter is Not a Home. New York, 1987.
Matulef, M. L., Crosse, S. B., Dietz, S. K., Van Ryszin, G., Kiser, M. L., Puhl, L. M., & Ficke, R. C. (1995). National Evaluation of the Supportive Housing Demonstration Program: Final Report. Washington, D.C.: U. S. Department of Housing and Urban Development and the Office of Policy Development and Research.
Meister, J. (1993). On Our Way Home: Transitional Housing Inc. Program Director's Manual. Cleveland, OH: Transitional Housing Inc., 1-221.
More Than Housing: A Guide to Developing and Managing Supportive Housing for Homeless People (1995). Unpublished draft report prepared for Interagency Council by Westat, 1995.
Murray, R. L. E., & Baier, M. (1995). Evaluation of a Transitional Residential Programme for Homeless Chronically Mentally Ill. Journal of Psychiatry and Mental Health Nursing, 2: 3-8.
Murray, R., Baier, M., North, C., Lato, M., & Eskew, C. (1995). Components of an Effective Transitional Residential Program for Homeless Mentally Ill Clients. Archives of Psychiatric Nursing, 9(3): 152-157.
Murray, R., Baier, M., North, C., Lato, M., & Eskew, C. (1997). One-Year Status of Homeless Mentally Ill Clients Who Completed a Transitional Residential Program. Community Mental Health Journal, 33(1): 43-50.
Nathan, R. P., Pulice R. T., Nakamura, R. T., & Schwartz, M. (1995). The New York State McKinney Transitional Housing Demonstration Project. Final Report. Prepared for the New York State Department of Social Services.
National Resource Center on Homelessness and Mental Illness, Policy Research Associates Inc. (1995). Creating Safe Havens for Homeless Persons with Severe Mental Illness: A Workshop Summary. Center for Mental Health Services and U. S. Department of Housing and Urban Development.
National Resource Center on Homelessness and Mental Illness, Policy Research Associates Inc. (1997). Developing and Operating Safe Havens Programs. Washington, DC: The Center for Mental Health Services, Substance Abuse and Mental Health Services Administration, U. S. Department of Health and Human Services, Office of Special Needs Assistance Programs, Office of Community Planning and Development and U. S. Department of Housing and Urban Development.
New York City Commission on the Homeless. (1992). The Way Home: A New Direction in Social Policy. Report of the New York City Commission on the Homeless.
O'Hara, A. (1997). The Safe Haven in the Continuum of Care. Paper presented at the Developing and Operating Safe Havens Programs, Washington, DC.
Porterfield, D. (1997). Safe Havens: Development and Operations. Planning, Designing, Siting and Financing Safe Haven Housing. Paper presented at the Developing and Operating Safe Havens Programs, Washington, DC.
Prabuki, K., Wootton, E., McCormick, R., & Washam, T. (1995). Evaluating the Effectiveness of a Residential Rehabilitation Program for Homeless Veterans. Psychiatric Services 46(4): 372-375.
Proscio, T. (1998). Under One Roof: Lessons Learned from Co-locating Overnight, Transitional and Permanent Housing at Deborah's Place II, A Case Study Commissioned by the Corporation for Supportive Housing: Corporation for Supportive Housing, 1-19.
Rahav, M., Rivera, J., Nuttbrock, L., Ng-Mak, D., Sturz, E. L., Link, B. G., Struening, E. L., Pepper, B., & Gross, B. (1997). Homeless, Mentally Ill, Chemical Abusing Men in Different, Community-based Treatment Programs. In: F. M. Tims, J. A. Inciardi, B. W. Fletcher, A. M. Horton Jr. (Eds.). The Effectiveness of Innovative Approaches in the Treatment of Drug Abuse.. Westport, CT: Greenwood Press, 84-97.
Rog, D. J., McCombs-Thornton, K. L., Gilbert-Mongelli, A. M., Brito, M. C., & Holupka, C. S. (1995). Implementation of the Homeless Families Program: 2. Characteristics, Strengths, and Needs of Participant Families. American Journal of Orthopsychiatry, 65(4), 514-528.
Roman, A. M. & Zhu, X. D. (1996). An Evaluation of the Massachusetts AFDC Transitional Housing Demonstration Project. Prepared for the Massachusetts Department of Public Welfare and the U. S. Department of Health and Human Services.
Sprague, J. (1991). More Than Shelter: A Manual on Transitional Housing. Women’s Institute for Housing and Economic Development.
U. S. Department of Housing and Urban Development. (1995). Review of Stewart B. McKinney Homeless Programs Administered by HUD. Report to Congress : U. S. Department of Housing and Urban Development.
Walachy, M. E., & Ray, J. (1997). Sounds Really Good, But Will It Work? Paper presented at the Developing and Operating Safe Havens Programs, Washington, DC.
Ware, N., Desjarlais, R., AvRuskin, T., Breslau, J., Good, B. & Goldfinger, S. (1992). Empowerment and the Transition to Housing for Homeless Mentally Ill People: An Athropological Perspective. New England Journal of Public Policy 8(1): 297-315.
White, A., Kirk, C., & Wagner, S. (1997). Transitions: From Safe Havens to Other Supportive Housing Settings. Paper presented at Developing and Operating Safe Havens Programs, Washington, DC.
Wilder Research Center. (1998). Minnesota’s Fifth Year Experience with Supplemental Assistance for Facilities to Assist the Homeless (SAFAH): A Report to the U. S. Department of Housing and Urban Development.
Wittman, F. D. (1989). Housing Models for Alcohol Programs Serving Homeless People. Contemporary Drug Problems 16: 483-504.
Wittman, F. D. (1991). Current Status of Housing Initiatives for Homeless People With Alcohol and Other Drug Problems. Housing Initiatives for Homeless People with Alcohol and Other Drug Problems: Proceedings of a National Conference. National Institute on Alcohol Abuse and Alcoholism and Interagency Council on the Homeless, San Diego, 112-125.
Return to the Table of Contents, ASPE Home Page, or HHS Home Page.