Service and Housing Priorities and Preferences
Since the early 1990s, research and other literature have consistently and unambiguously documented discrepancies between provider and consumer perspectives on how to prioritize housing and services, and on the kind of housing needed. A comprehensive array of outreach, transitional, and permanent housing programs is now well established at city and state levels through provider participation in the annual continuum of care application to HUD, which funds and shapes the direction of local services. While this broad offering of services provides some people who are homeless with options for exits from homelessness, point of entry into housing is usually not determined by consumers themselves, but by service providers. Housing providers often worry about the potential for disruptive and dangerous behavior, as well as the possibility of recurring homelessness, if people with mental illness and addiction problems are given independent housing before demonstrating that they are ready for it. Similarly, clinician beliefs in the limited ability of persons with mental illness to maintain independent housing can restrict the paths that consumers take into housing. Indeed, when surveyed about housing needs, providers more often recommend staffed settings with on-site treatment or supports (Bebout & Harris, 1992; Goering, Paduchak, & Durham, 1990; Goldfinger & Schutt, 1996). Therefore, in most programs across the country, approaches to housing and services for people who are homeless have reflected provider perspectives, emphasizing outreach and shelter as first responses. They offer emergency supplies (food, clothing), support, and referrals to transitional settings such as drop-in centers and safe havens that offer support and treatment services to help consumers become housing ready. In contrast, researchers and advocates studying consumer preferences in housing and consumers themselves all report that consumers who are homeless want housing first housing on their own terms: independent, integrated into the community, with the support services an individual chooses available off site but not required (Carling, 1993; Howie the Harp, 1993; Tanzman, 1993).
While there are numerous examples of very successful housing readiness programs, the visibility of a highly vulnerable group of consumers who have been homeless for years has recently prompted researchers and other observers to focus attention on chronic homelessness and to join consumers in challenging the treatment-as-usual service approaches. Kuhn and Culhane (1998) found that among all who stay in shelters over the course of a year, a small group (about 10 percent ) with long and repeated stays use half of all system resources (days of shelter use). Their work suggests that extensive engagement and contact with outreach, shelter, transitional housing, and clinical services have been very costly but ineffective for this group. Gladwells (2006) poignant portrait of Million Dollar Murray highlights the experience of thousands of people who are chronically homeless. Gladwell provides a litany of well-intentioned but failed service efforts by outreach workers, police, detox and psychiatric emergency room staff, and many others who tried to help Murray leave the streets of downtown Reno, Nevada. Murray wanted a place of his own but no one believed that he was ready for it (in fact, he did well when he once briefly had a place, but without support services he soon lost it and there were no more chances after that). A tally of the cost of providing 12 years of failed services to Murray gave Gladwell a nickname for his anti-hero, and highlighted the need for new strategies.
In the last decade, new approaches emphasizing consumer choice and recovery (Deegan, 1988; Mead & Copeland, 2000; Ridgway & Press, 2004) have challenged traditional models requiring treatment and sobriety as preconditions for housing. What is shared by well-documented, consumer-driven approaches, such as wellness recovery action plans (WRAP) (Copeland, 1997), motivational interviewing (Miller & Rollnick, 2003), shared decision making (Deegan, 2007), and housing first (Tsemberis, 1999), is that programs must allow consumers to be equal partners in choosing providers and determining the type and sequence of services they receive, including the right to refuse services altogether (New Freedom Commission on Mental Health, 2003).
Approaches to housing and services cannot be neatly dichotomized between treatment and sobriety first providers on the one hand and housing first providers on the other. Although treatment and sobriety are common preconditions for accessing permanent housing, some homeless assistance programs make participation in treatment and support services optional once individuals meet initial treatment requirements for admission. Since the 1980s, community development activists and homeless advocates, who have viewed an expanded supply of permanent affordable housing as essential to any efforts to end homelessness, have collaborated to reclaim or replace disappearing housing stock by creating integrated housing developments (Hopper & Barrow, 2003). These programs offer SRO accommodations or efficiency apartments for diverse groups with housing needs, including people with mental illness, people living with HIV/AIDS, and other low-income members of local communities. Subsidies ensure affordability; standard leases ensure tenancy rights; and nonprofit service agencies make optional support services available on site, usually in inconspicuous locations, which tenants may use or not, as they wish. However, integrated housing development programs, like most congregate housing settings, often face challenges, juggling as they do to support individual tenants through periods of symptom or substance use relapse, maintain a viable environment for the tenant community as a whole, and keep admission requirements reasonable (Barrow, Soto Rodriguez, & Cordova, 2004; Hopper & Barrow, 2003).
Housing First: A Paradigm Shift in Homeless Services
The inclusion of consumer voices in the planning and implementation of housing and services has fostered a paradigm shift in homelessness interventions and moved outreach, engagement, treatment, and housing services to be more in line with consumer preferences. In the last decade, housing first, developed in partnership with consumers, has emerged as a well-documented practice that effectively provides permanent housing that is not contingent on prior services or readiness criteria. The housing first approach was pioneered by Pathways to Housing. By offering consumers an apartment of their own as a direct exit from homelessness, Pathways fused several programmatic steps outreach, engagement, and housing into a single powerful and desirable invitation (Tsemberis et al., 2003, p. 310). This approach is consistent with consumer priorities for an independent apartment of their own without requirements for psychiatric treatment or sobriety as a condition for housing entry or retention.
The pathways to housing model. The pathways to housing model emerged from an ongoing dialog among consumers, staff, and researchers (Shern et al., 2000; Lovell & Cohen, 1998; Tsemberis et al., 2003) who had developed an outreach and drop-in center program as an NIMH research demonstration project. An ethos of respect for consumers and their wishes was fostered by training staff in consumer-centric clinical approaches such as psychiatric rehabilitation (Anthony et al., 2002). In addition, several staff members were consumers themselves, and consumers shared responsibility for policy and program decisions. Howie the Harp, an early consultant, brought to the program a commitment to social justice and revolutionary fervor to change the mental health system (Tsemberis & Asmussen, 1999; Tsemberis & Eisenberg, 2000; Tsemberis et al., 2003).
In the drop-in center, neither status nor salary distinguished consumer staff from non-consumer staff, an approach that also blurred boundaries between staff and participants and fostered collaboration on the critical problem of access to housing. Staff and consumers witnessed how existing housing providers used the need for housing to leverage consumer acquiescence to unwanted treatment and abstinence requirements (Allen, 1996). After repeated failures to secure housing for consumers, the group began a trial and error process to design a housing program that would be desirable to consumers and manageable to staff. They determined that the scattered-site supported housing model met consumers requirements for normal housing, tenancy rights, privacy, and an affordable rent contribution (30 percent). Consumers and staff collaboratively worked out operational details, occupancy policies, and program and consumer fiscal responsibilities, including a program account that required both staff and consumer signatures for checks to be cashed.
This collaborative plan evolved into the pathways to housing model, which focuses on ending homelessness by offering permanent, independent housing and comprehensive, consumer-driven supports without contingencies for treatment or sobriety. The pathways model, now called housing first, prompted several programs and agencies around the country to adopt this approach in their efforts to end chronic homelessness. However, the Pathways program and the numerous replication sites are not the only programs that are described as housing first. Locke and colleagues (2007) identify several variations of housing first models with different housing and service approaches. Other examples of successful programs called housing first include Direct Access to Housing (DAH), operated by San Franciscos Department of Public Health, which is aimed at housing people who are homeless and frequent users of medical emergency rooms, and a housing program operated by Seattles Downtown Emergency Service Center that provides immediate access to an SRO room for people who are living on the streets and suffer from addiction disorders. These programs vary in type of housing provided (single site buildings vs. scattered site) and how services are provided (on site vs. off site). This variability of housing first approaches offers communities seeking to serve individuals or families who are chronically homeless a variety of program models: from a few scattered-site units with off-site case management support services to single site options of various sizes with services on site. Despite their differences, housing first programs share many dimensions: immediate access to housing with no treatment preconditions; services are optional; housing is permanent and affordable. Thus more communities are beginning to include a housing first option in their menu of programs options in continuum of care plans (Locke et al., 2007).
In the discussion of choice-based housing and services that follows, we focus on the Pathways model for several reasons. The Pathways program is well researched and well documented, while also offering housing in the most integrated settings. In addition, the Pathways model not only centers on consumer choice, but the program is consumer driven. Consumers are integrated into every dimension of the program: planning, operation, and policy. Pathways has sustained its consumer-centered character by attending to the importance of consumer presence and voice at every level of the organization. All services promote individual self-determination and social integration. Thus upon admission, consumers choose their apartment and neighborhood of residence, restricted only by the availability of affordable housing. Because the independent scatter site apartments are leased from existing units in the community and comprise less than 20 percent of any building, while all program services are off site, both housing and services are offered in the least restrictive and most socially integrated settings. Finally, consumers are encouraged to fashion their own path towards greater social inclusion, whether through employment, returning to school, or reuniting with children.
In addition, several practices promote consumer voice more broadly within the agency: (1) as tenants, consumers participate on an advisory committee that meets with agency heads to express tenant concerns and to provide programmatic input; (2) every officer of the organization has an open door policy to all tenants; (3) consumers are hired as service providers and managers and are elected as members of the agency board of directors; (4) in meetings, accommodations ensure full participation appointing a moderator, taking turns speaking, and having an active, moderated, question and answer period; and (5) social and recreational events provide opportunities for staff and consumers to meet informally and expand the repertoire of their dialogue. Both the organizational environment and service approach foster empowerment for greater community, civic, and political participation.
Research on housing first. A series of increasingly rigorous studies, conducted over several years, in different settings and with different subsets of the homeless population have focused mainly on the pathways to housing model and its replications. The studies summarized here have begun to establish an evidence base for housing first as an effective approach to ending homelessness and achieving positive outcomes in mental health and other domains.
Early findings. Early work focused on Choices Unlimited, a drop-in center and forerunner of Pathways to Housing. The program was funded as one of six multisite NIMH-McKinney research demonstration projects. The goal was to test a psychiatric rehabilitation approach to engaging consumers. The program did not impose treatment requirements, predetermine lengths of stay, or require that services be used in a given sequence as conditions for drop-in center use. To test the effectiveness of the approach, potential participants were randomly assigned to either the experimental program or to traditional outreach and drop-in center programs focused on housing readiness. Over two years of follow-up, time spent on the street declined for both groups, but the 55 percent decrease for the experimental group was almost twice the decrease (28 percent) for the control group (Tsemberis et al., 2003). Those in the experimental group also found it easier to obtain food, find a place to sleep, and remain sober than did the control group; they participated in more services, including day programs and self-help groups; and they received more help with alcohol and drug problems, financial entitlements, and health insurance (Shern et al., 2000). However, after 24 months only 38 percent of participants had moved to permanent housing. This finding guided program developers to focus on reducing barriers to housing access more directly by providing immediate access to independent scattered-site apartments and services based on consumer choice.
Housing retention. The housing first approach, which evolved from the Pathways to Housing drop-in center work, was initially investigated in a study that used administrative data to compare housing retention of housing first tenants to housing retention of tenants in supportive housing programs that required treatment and sobriety as preconditions to housing. Controlling for the effects of client characteristics, this study showed that participants in housing first achieved better housing tenure than did the comparison group (Tsemberis & Eisenberg, 2000). After five years, 88 percent of housing first participants remained housed, compared to 47 percent of those in more traditional housing programs. However, in the absence of random assignment, it is impossible to be sure that the better outcomes were due to program effects rather than unmeasured participant characteristics.
A subsequent experimental and longitudinal study of the effectiveness of the housing first approach in New York City was conducted as part of a SAMHSA multisite study of homelessness prevention. In this four-year study, 225 participants with severe mental illness, who were literally homeless and many diagnosed with co-occurring substance use disorders, were randomly assigned to receive either housing first or services as usual. Participants were interviewed every six months to examine changes across a range of outcomes, including residential status, substance use, and psychiatric symptoms. After six months, 79 percent of housing first participants were living in stable housing compared to 27 percent of participants in the control group (Tsemberis et al., 2003); and throughout two years of follow-up, housing first participants spent more time in stable housing and showed far greater reductions in homelessness than the control group (Tsemberis, Gulcur & Nakae, 2004).
Another experimental study examined housing retention among long-term shelter dwellers with psychiatric disabilities and often co-occurring addictions in a suburban county. Participants were randomly assigned to one of two programs using housing first approaches (Pathways to Housing or a local consortium) or a services-as-usual control group. Replicating and extending previous findings, this study demonstrated that participants assigned to housing first obtained permanent, independent housing at higher rates than the control group. A majority of consumers housed by both housing first programs retained their housing over four years, with78 percent of participants in Pathways to Housing remaining housed over that period (Stefancic & Tsemberis, in press). This suggests that housing first services can be successfully replicated in non-urban environments and with a population of homeless persons with extensive shelter histories. These findings have been replicated with shelter populations in Salt Lake City (Flynn, 2006) and Hartford (White, 2005).
In 2003, a Collaborative Initiative to Help End Chronic Homeless was coordinated by the Interagency Council on Homelessness (ICH) and funded by HUD, HHS (SAMHSA and HRSA), and the VA to provide housing and services to chronically homeless populations. Seven of the eleven programs funded used the Pathways housing first model to provide scattered-site housing and off-site support and treatment services. They achieved similar housing retention results, with approximately 80 percent stably housed after 12 months (Rosenheck, 2006), successfully replicating both the model and the findings on retention across diverse contexts.
In domains other than retention, the results of several of these studies, including two controlled trials, are building a case for housing first as an evidence-based practice for addressing homelessness: Participants in housing first obtained housing earlier and remained stably housed at higher rates than control groups receiving services as usual through more traditional housing and treatment programs (Tsemberis, Gulcur, & Nakae, 2004; Gulcur et al., 2003). Further, housing first participants spent significantly less time in psychiatric hospitals and incurred fewer residential costs than controls (Gulcur et al., 2003), though findings from another study showed modest increases in societal costs for housing first (Rosenheck et al., 2003), suggesting a need for further examination in this domain.
Psychiatric symptoms and consumer choice. Further analyses of the housing first experimental data focused on psychiatric symptoms. Although housing first participants used fewer psychiatric treatment services than control group participants at every time point, there were no significant differences in self-reported symptomatology (Padgett, Gulcur, & Tsemberis, 2006), suggesting that optional participation in treatment services is as effective as mandatory participation in services. An examination of the relationship between stable housing and psychiatric symptoms shows a reduction in psychiatric symptoms if the person has been stably housed for the preceding six months (Tsemberis & Fischer, under review). Separate analyses compared the impact of consumer choice, a principal component of housing first, on the mental health of housing first and control participants. Ratings of perceived choice were significantly higher for participants in housing first compared to those in the control group; and perceived choice significantly accounted for a decrease in psychiatric symptoms, a relationship that was partially mediated by mastery (perceptions of personal control) (Greenwood et al., 2005). This strong and inverse relationship between perceived choice and psychiatric symptoms supports expansion of all housing models that increase consumer choice, thereby enhancing mastery and decreasing psychiatric symptoms.
Substance use and consumer choice. In a recent study Milby and colleagues (2005) examined the effectiveness of providing direct access to housing to individuals who were homeless and experiencing cocaine dependence disorder. A total of 196 participants were assigned to receive: a) abstinence-contingent housing; b) non-abstinence contingent housing; or c) no housing. Participants were followed for 24 weeks. While the abstinence housing group showed significantly higher rates of abstinence (as required by the housing program) than the non-abstinence contingent group, results for days housed showed that the groups did not differ significantly from each other at any time point. The investigators conclusion favors abstinence-contingent housing, and the studys findings have been described as contradicting the positive findings on housing first. However, the focus on abstinence as the critical outcome reflects the studys focus on treating addiction rather than ending homelessness and obscures the important findings on housing tenure. The data clearly show also that for individuals with addictive disorders, housing without abstinence contingency is as effective as abstinence-contingent approaches in addressing homelessness. In New York City, Project Renewal, one of the HUD Chronic Inebriates Initiative sites, obtained similar results. Using a housing first approach with people who were frequent users of their detox services, this program achieved an 80 percent housing retention rate for non-abstinence contingent housing (Ed Geffner, personal communication, October 31, 2006). In a later program intervention, as one of the HUDs chronic inebriate grantees, Project Renewal decided to vary their approach and required 90 days of abstinence and treatment prior to providing housing for their second HUD program (calling the approach housing-second). Results indicated that the number of people who lost housing due to relapse into alcoholism was higher in the housing-second program (Cowles, 2007).
At another HUD Chronic Inebriates site, Pathways to Housing DC (a Pathways replication site), 35 of 36 people who were chronically inebriated were still housed after the first six months, and there was a reduction in (average) expenditures on alcohol from $87.06 a month prior to entering the program to $17.90 per month after entering into housing. Furthermore, results indicated that consumers who chose to participate in drug treatment had significantly reduced their consumption (Kent, 2007).
The experimental study of 225 housing first participants also examined substance use outcomes. Analyses show that although the services-as-usual group utilized more substance use treatment, there were no significant differences in self-reported substance use between the control and Housing First groups. Moreover, though the control groups greater service use continued over the four-year follow-up period, absence of group differences in alcohol and drug use persisted over the four-year period as well (Padgett, Gulcur, & Tsemberis, 2006).
Consumer satisfaction and challenges. While evidence for housing first has been accumulating, many providers continue to view the approach as appropriate only for high functioning individuals. This issue was addressed in a quasi-experimental study that compared differences in housing, psychiatric outcome, and satisfaction of formerly homeless participants in housing first and another supported housing model versus those in structured, service-intensive community residences (Siegel et al., 2006). Because participants were not randomly assigned to housing groups, a statistical procedure called propensity scoring was used to categorize participants into three strata depending on how they ranked as candidates for supported housing. Regardless of stratum, individuals in supported housing remained stably housed, and housing type had no effect on tenure. However, at every follow-up point, participants in supported housing reported greater satisfaction in terms of autonomy and economic viability than those in community residences.
Some participants in this study who were in supported housing, but who were ranked as more likely candidates for community residences, reported greater isolation, a finding also supported by a qualitative study of community integration of housing first consumers. The qualitative study found that, for most consumers, entering housing after a long period of homelessness was associated with improvements in several psychological aspects of integration (e.g., a sense of fitting in and belonging) as well as feelings of being normal or part of the mainstream human experience. However, the study also uncovered challenges faced by housing first participants, including difficulties in coping with loneliness, adjusting to living independently, feeling safe without any monitoring presence, and fitting in in the community (Yanos, Barrow, & Tsemberis, 2004).
The weight of the evidence. Participants in housing first obtain and maintain independent housing with consumer-chosen supports without negative effects on psychiatric or substance use symptoms. Housing retention rates remain around 80 percent for periods of four to five years. And consumers in housing first report higher levels of choice and residential satisfaction compared with participants of more traditional programs. Further, consumer choice a key aspect of housing first services positively affects psychiatric symptoms, a relationship that is mediated by mastery (perceived choice leads to increased mastery, which is associated with reduced psychiatric symptoms). Finally, providing housing first has been shown to be less costly than traditional residential treatment, though it may be associated with modest increases in societal costs. Further research is now underway on fidelity measures of housing first that define the models key features and assess how closely they are approximated as this rapidly disseminating model is implemented in diverse contexts and with various consumer subgroups.
Harm Reduction, Choice, and Homelessness
The predominant approach of housing programs for people who are homeless and dually diagnosed requires psychiatric treatment and a period of sobriety as preconditions for permanent housing. Specific variants include therapeutic communities, modified therapeutic communities, residences for mentally ill chemical abusers (MICAs), and other abstinence-based housing programs. (For a recent review see Center for Substance Abuse Treatment, 2005.) This approach has a documented history of clinical success for consumers who choose or are able to complete the programs and is favored by providers because it limits liability risks and management problems, and is consistent with the widely held view that people with dual disorders are unlikely to maintain housing without first developing housing readiness.
The programs presented here are focused on how substance use affects the efforts of persons who are homeless to obtain housing as opposed to treatment. As noted previously, abstinence-contingent housing can serve to exclude subgroups of people with dual diagnoses, thus leaving a significant proportion of individuals chronically homeless. Abstinence-contingent housing and treatment models usually emphasize, if not mandate, participation in 12-step mutual support groups (AA or NA). Despite the overall strength of their peer-based approach, abstinence programs tend to be stringently and hierarchically structured, with consumer choice and input significantly circumscribed beyond the initial choice to participate. For these reasons, and because these models are not a new development in the last 10 years and thus beyond the scope of this report, we have not reviewed them as choice-based approaches. By eliminating sobriety and psychiatric treatment as a preconditions for housing, housing first programs have proven highly effective in housing and keeping housed people with addiction disorders and dual disorders who had repeatedly failed in or been rejected by other programs.
Consumer choice is the foundation of the harm reduction approach (Inciardi & Harrison, 2000), in which consumers define their needs and goals as well as the pace and sequence of services. Harm reduction has emerged as an alternative to the ubiquitous 12-step abstinence/sobriety models of drug use and addiction (Marlatt, 1998). As applied to homelessness, this gradual approach of encouraging consumers to reduce substance use and related risks replaces the pervasive sobriety and psychiatric treatment requirements that prevent individuals who are homeless from attaining and retaining housing (Rowe, Hogue, & Fisk, 1996). It offers an individualized approach to assisting consumers progress towards recovery but starting at their stage of readiness for change rather than insisting on abstinence as a prerequisite for housing (Tsemberis & OCallaghan, 2004). Harm reduction is consumer driven and seeks to minimize personal harm and adverse societal effects of substance abuse while the consumer strives towards recovery.
Although sobriety and abstinence are considered ideal outcomes of harm reduction, the model allows alternative paths to sobriety as long as they serve to contain or reduce the many risks or risk behaviors associated with addiction, such as drug overdose, incarceration, impoverishment, prostitution, malnourishment, chronic homelessness, and ill health. Consequently, consumers are offered a range of treatment alternatives, which can include AA/NA, and are supported in making positive steps towards recovery, whether it means striving for abstinence or making use less risky. Harm reduction approaches often incorporate DiClemente and Prochaskas transtheoretical model of addiction and recovery, often referred to as the stages of change model, to help consumers and clinicians define and track gradual goals on the path to attaining the ideal of sobriety (DiClemente, 2003; Prochaska & DiClemente, 1992; Marlatt, 1998).
In treatment for dually diagnosed populations, harm reduction works particularly well with other evidence-based practices, such as integrated dual-diagnosis treatment (IDDT) (Mueser et al., 2003) and illness management recovery (IMR). When housing programs for individuals who are dually diagnosed require abstinence as a condition for obtaining housing or for remaining housed, consumers with histories of substance abuse who are at risk of relapse are also at risk for housing loss and continued homelessness. This fear of eviction inhibits consumers who begin to use drugs or to experience psychiatric symptoms from discussing their emerging problems with housing counselors. Housing first programs separate the terms and conditions for continued tenancy from treatment concerns and provide integrated services that encourage honest feedback from consumers, including disclosure that they are using drugs or alcohol or no longer taking their psychiatric medication without fear that this will lead to eviction (Tsemberis & Asmussen, 1999).
Harm reduction approaches to homelessness for consumers with dual diagnoses. In some locales, harm reduction is codified as public policy, which requires its application to housing approaches that address homelessness. Thus, for example, harm reduction principles are central to San Franciscos Direct Access to Housing program, which creates integrated SRO housing developments to address homelessness for people with dual diagnoses. A descriptive study found that in these contexts, tenants with addictions were not at elevated risk of housing loss, and 70 percent retained housing over two years (Barrow et al., 2004).
The housing first approach implemented by Pathways to Housing is the most extensively described and researched homelessness intervention based on harm reduction principles. As described above, Pathways imposes neither abstinence nor treatment conditions for accessing or remaining in housing. The program offers consumers access to an interdisciplinary ACT team of skilled clinicians who offer integrated dual diagnosis treatment (IDDT) and other assistance on the persons own terms. Those who continue to use addictive substances or remain symptomatic may go in and out of treatment but they are not at risk of housing loss. Some newly housed consumers are motivated to reduce substance use or seek psychiatric treatment because they desperately want to avoid jeopardizing their newly-obtained housing (which, in some cases, is the first housing they have ever been able to call their own). Once survival is assured through housing, others may seek treatment for psychiatric symptoms or become ready to address other needs such as employment or family reunification (Tsemberis & OCallaghan, 2004).
Previously cited evidence from experimental studies shows housing first participants have dramatically better homelessness outcomes and show equivalent levels of substance use compared to participants in services-as-usual programs that do not employ harm reduction (Padgett, Gulcur, & Tsemberis, 2006). While a specific test of abstinence-contingent housing versus non-abstinence contingent housing reported the unsurprising result of higher rates of abstinence in the housing program that required it, investigators report no difference in housing retention (Milby et al., 2006). Given that the most important outcomes for dually-diagnosed consumers who are homeless are either neutral or improved when harm reduction is employed, the study supports a harm reduction approach to homelessness and substance addictions among people with co-occurring addiction and mental illness.
Harm reduction approaches for consumers with long-term substance addictions. In 2006, a Health Care for the Homeless Council position statement endorsed including harm reduction models of addiction treatment in SAMHSAs best practices. The statement notes that programs that incorporate harm reduction strategies are more likely to attract active users (and hence those most in need of resources for reducing drug-related harm), enhance motivation for positive change, improve retention in treatment, and reduce attrition and premature termination of services (National Health Care for the Homeless Council, 2006). Canada Mortgage and Housing Corporations brief review (2005) of homeless assistance programs that use harm reduction approaches describes a dozen programs in the United States, Canada, and the United Kingdom. Most provide permanent housing in dedicated units or scattered sites. To minimize harms associated with high-risk behaviors, programs emphasize helping residents reduce their usage, focusing on the strengths and capacities of the person rather than on the substances they consume, and encouraging changes in consumption habits (e.g., a move to less harmful substances, safe disposal of used syringes) or ensuring that there is not an increase in use. All emphasize the consumer-determined pace and content of services. Descriptions of harm reduction programs specifically addressing alcohol addiction among people with long-term homelessness include a Canadian report on a managed alcohol intervention (Podymow et al., 2006) and Anishinabe Wakiagun, a Minnesota program that provides permanent housing and case management services addressing health and support needs of men and women with late-stage chronic alcoholism and extensive homelessness (Canada Mortgage and Housing Corporation, 2005).
Research on programs that use harm reduction principles in housing for people with chronic homelessness and long-term alcohol dependence remains rudimentary (Hwang, 2006). A managed alcohol intervention was studied by Podymow and colleagues (2006), who used pre- and post- measures to assess changes in 17 program participants. They documented significant decreases in both emergency room visits and encounters with police. Participants also spent fewer days in the hospital and self-reported decreases in alcohol use. Such initial reports indicate that this harm reduction approach deserves serious attention in future research with this group.