The Architecture of Recovery: Two Kinds of Housing Assistance for Chronic Homeless Persons with Substance Use Disorders

Abstract

Purpose

Roughly half a million persons in the United States are homeless on any given night and over a third of those individuals have significant alcohol/other drug (AOD) problems. Many are chronically homeless and in need of assistance for a variety of problems. However, the literature on housing services for this population has paid limited attention to comparative analyses contrasting different approaches.

Approach

We examined the literature on housing models for homeless persons with AOD problems and critically analyzed how service settings and operations aligned with service goals.

Findings

We found two predominant housing models that reflect different service goals: Sober Living Houses (SLHs) and Housing First (HF). SLHs are communally based living arrangements that draw on the principles of Alcoholics Anonymous. They emphasize a living environment that promotes abstinence and peer support for recovery. HF is based on the premise that many homeless persons with substance abuse problems will reject abstinence as a goal. Therefore, the HF focus is providing subsidized or free housing and optional professional services for substance abuse, psychiatric disorders and other problems.

Practical Implications

If homeless service providers are to develop comprehensive systems for homeless persons with AOD problems, they need to consider important contrasts in housing models, including definitions of “recovery,” roles of peer support, facility management, roles for professional service, and the architectural designs that support the mission of each type of housing.

Originality

This paper is the first to consider distinct consumer choices within homeless service systems and provide recommendations to improve each based upon an integrated analysis that considers how architecture and operations align with service goals.

Introduction

The human and financial toll of homelessness in the U.S. is staggering. During 2013 on any given night there were over half a million people with no place to live (National Alliance to End Homelessness, 2014). In a review of problems associated with homelessness Polcin (2015) cited research showing homelessness is associated with substance abuse, mental health symptoms, HIV risk, and other health problems. Persons who are homeless have a mortality rate that is triple that of persons who are stably housed (O’Connell, 2005). Petersilia (2003; 2006) and Petteruti and Walsh (2008) implicated homelessness as a cause and consequence of criminal justice incarceration and massive overcrowding of jails and prisons. The U.S. Secretary of Housing and Urban Development estimated that healthcare, criminal justice, and other costs result in taxpayers spending $40,000 for each homeless person each year (Moorhead, 2012).

For decades there has been and urgent yet neglected need to address homelessness and the concurrent problems that accompany it. A number of studies indicate over a third of individuals who are homeless experience alcohol and drug problems (e.g., Gillis, et al., 2010) and up to two-thirds have a lifetime history of an alcohol or drug disorder (Roberston, et al., 1997). The U.S. National Coalition for the Homeless (2009) cited studies indicating 20 to 25% of the homeless suffer from some type of severe mental illness. Many of these individuals do not receive treatment for their substance abuse and mental health problems and even when they do there is typically no provision of permanent housing during or after treatment. Lack of stable housing leaves them vulnerable to substance abuse relapse, exacerbation of mental health problems, and a return to homelessness. Efforts to address homelessness among persons with alcohol and drug problems include distinct models: abstinence oriented recovery homes, called sober living homes (SLHs) in California, and housing first (HF), which is designed to provide immediate access to free or subsidized housing with few conditions (i.e., low demand) and assist residents to access services they need and want (e.g., substance abuse, mental health and medical treatment).

Purpose

The goal of this paper is to provide an analysis of HF and SLH models in the U.S. and make recommendations for practice that are based on an approach to architectural planning that emphasizes the confluence of settings and operations. How settings and operations interact to influence resident experiences has largely ignored in the discussions about services for homeless persons, particularly subgroups, such as persons with AOD problems. In addition to standard ways of understanding these models, such as explicit philosophies, services offered, and explicit goals, we examine these programs in terms of key architectural considerations that we believe influence operations and outcomes (Wittman, 1993; Wittman, Jee, Polcin & Henderson, 2014). Considerations include characteristics of settings, such as location (neighborhood context), appearance, design for sociability, design for personal space, facility oversight and security, and practices for upkeep. Considerations also include ways that setting characteristics interact with operations, such as house rules, mobilization of peer support, role of professional services, and cultivation of a shared approach to recovery. The paper ends with recommendations for the design and operation of HF and SLH facilities, issues in need of more research, and development of more comprehensive approaches to homeless services.

Sober Living Houses

Sober living houses (SLHs) are alcohol- and drug-free living environments for persons with substance use problems who wish to abstain from substances and embark upon a program of recovery. The housing stock used for SLHs is typically ordinary housing that fits in with single-family neighborhoods, multi-family neighborhoods, and mixed-use residential-commercial areas. Wittman and Polcin (2014) pointed out they originated in California in the 1940’s as “12^th step” houses modeled on the principles of Alcoholics Anonymous. The houses were a peer-based, grassroots service for persons with alcohol and drug disorders rather than residential treatment facilities managed by professionals. If professional services were required (e.g., medical care, medications for addiction or psychiatric disorders) SLH residents sought them out in the community.

In recent years there have been increased efforts to use SLHs within existing continuum of care services. For example, some treatment programs have designed SLHs specifically as places where clients can live after completing residential treatment or while they attended outpatient treatment (Polcin, 2015). Whether affiliated with treatment or not, houses are primarily financed through resident fees and individuals are free to stay as long as they wish provided they abide by basic house rules such as abstinence from alcohol and drugs, payment of rent, and participation in household maintenance (e.g., chores and attendance at house meetings). Typically, a manager or operator oversees house operations such as payment of rent and bills and monitors residents in terms of their sobriety. Individuals who return to substance use are asked to leave at least for some minimum period of time.

Housing models with some similarities to SLHs include Oxford Houses (Jason, Olson, Ferrari, & Lo Sasso, 2006), which are located primarily in the U.S., and Psychological Informed Environments (PIE) (Breedvelt, 2016), which ae located in the U.K. Like SLHs, Oxford Houses emphasize abstinence and peer support and residents can stay as long as they wish. However, rather than having a house manger oversee operations, Oxford Houses engage residents in rotaing leadership positions to oversee operations. In addition, unlike some SLHs, they do not have affiliations with treatment programs. PIE’s are similar to both SLHs and Oxford Houses in their emphasis on involving peers to shape a supportive or “enabling” environment. However, PIE is a broadly conceived approach designed to help the homeless in a range of settings. The model lacks specific recommendatins about issues such as abstinence from substances.

Research on SLHs has documented favorable outcomes. In an evaluation of 245 individuals over an 18 month time period Polcin, Korcha, Bond and Galloway (2010a) found significant improvements on measures of substance use, alcohol and drug problems, employment, psychiatric symptoms and arrests. Importantly, the improvements were maintained at 18 months. Similar 18 month outcomes were found for 55 persons living in SLHs affiliated with a treatment program (Polcin, Korcha, Bond & Galloway, 2010b). That sample included a significant number of persons with recent homelessness (35%). In a study of the community context of SLHs Polcin, Henderson, Trocki, Evans, & Wittman (2012) found neighbor perceptions and community stakeholder views were supportive. SLHs were generally viewed as an asset and residents were generally viewed as good neighbors. Neighbors expressed appreciation that the houses mandated abstinence from substances and stressed the importance of the houses to practice a “good neighbor” policy toward others.

One of the limitations of SLHs is that they require a commitment to abstinence, which some individuals are unable or unwilling to pursue. Thus, there is a need for housing services that can accommodate substance use. SLHs also require a level of independence and interpersonal functioning that can be difficult for some persons, such as those suffering from severe psychiatric disorders (Polcin, 2015). Thus, there is a need for additional homeless services that can accommodate these groups.

Housing First

Housing First evolved during the 1990’s as a reaction to the large numbers of persons who remained chronically homeless even if they received professional substance abuse and mental health services. Wittman and Polcin (2014) pointed out after 1980 federal and state support for long-term housing was greatly reduced in the planning and funding of services for persons with substance abuse and mental health problems. As a result, low-income persons with serious problems often had difficulty finding stable housing and were vulnerable to becoming homeless. HF was initiated by service providers and public health advocates seeking to engage homeless persons with co-occurring mental health and substance abuse disorders who were ready to accept help in a supportive living situation but not necessarily ready to stop drinking/using (Tsemberis et al, 2004).

Rather than promoting an abstinence-oriented recovery from substance abuse, HF uses a harm reduction approach which focuses on reducing harm caused by the individual’s substance use and mental health problems. Foremost in the HF approach is immediate access to free or at least subsidized housing. Some programs use a “scattered-site” approach where individuals are provided apartments within the general housing market. Other programs provide housing within a centralized setting where multiple homeless persons are housed together. HF uses a “low demand” approach that does not require abstinence or participation in a supportive milieu. In some programs, on-site staff consisting of case-managers assist residents to access to health, mental health, educational/vocational, and legal services. In programs that do not have onsite staff, case managers often visit residents on a daily basis to monitor problems and provide referrals to off-site services if desired by the resident.

Reviews of the research on HF indicate when homeless persons are provided free or subsidized apartments they tend to stay in those locations for extended periods of time (Kertesz, et al., 2009; 2015; Waegemakers, et al., 2014). While some studies have documented other favorable outcomes (e.g., substance abuse and mental health), reviews of the current literature have described a variety of research design limitations. Concerns have been raised about the measures used, study procedures, sampling, and descriptions of comparison groups (Kertesz, et al., 2009; Waegemakers, et al., 2014). One factor shown to be associated with favorable outcomes is the provision of case management services (Hwang et al., 2005). However, lacking in the research on HF are studies examining neighborhood and community stakeholder views about HF residences. Although research has not yet addressed the issue, there could be concerns from neighbors and other stakeholder groups about ongoing substance use among some of the residents and problematic behaviors related to substance use.

Conceptualizing Housing Approaches

Our conceptualization and analysis of HF and SLHs draws from a number of sources:

• The existing literature for both housing models.

• Our years of researching and providing residential substance abuse and mental health services.

• Considerations from architectural planning papers that emphasize the interaction of settings and operations to achieve service goals (e.g., Wittman, Jee, Polcin & Henderson 2014).

• Our involvement in forums designed to discuss housing models for homeless persons (e.g., the National Substance Use Treatment and Housing Leadership Forum sponsored by the Corporation for Supportive Housing & National Council for Behavioral Health, 2014).

One way to conceptualize differences between HF and SLHs is to consider their central focus. HF represents a “personal autonomy” approach where the need and desires of the individual take precedence. The approach uses a “low demand” philosophy that minimizes expectations of the individual and requirements to live in the facility. The personal autonomy approach can also be seen in the emphasis on providing only the types of services requested by the resident. The residential setting is primarily a way to be off the street (i.e., sheltered) and a way to access services desired.

SLHs represent a view that the setting itself is the essence of the service offered (Wittman, et al., 2014). Residents derive benefit from participation in a recovery environment (i.e., setting) that emphasizes peer support, abstinence from substances, and the practice of recovery principles during day to day interactions. Management of SLHs is typically conducted by persons in recovery themselves and frequently the residents of the houses have input into decision making and management. The setting, including the physical characteristics, the social environment, the operational procedures, and the surrounding neighborhood, invite individuals to engage in a peer focused recovery community that establishes abstinence as the common foundation for achievement of other personal goals. The focus is a “place- centered” socio-physical setting dedicated to recovery of the residents who choose to live there.

Increasingly, SLHs are being used as places for residents to live after they complete residential treatment or while they attend outpatient treatment programs (Polcin et al, 2010b). As noted in pervious papers (e.g., Kertesz, et al., 2009; Polcin, 2015), lack of attention to housing in the treatment process has been an enormous problem for decades. Without an alcohol- and drug-free living environment persons receiving treatment services have been vulnerable to relapse and homelessness. The focus of SLHs on maintaining an abstinence oriented living environment over the long run complements the acute care offered by many treatment programs. In addition, the focus on maintaining a recovery lifestyle, typically through involvement in 12-step and other recovery groups, is consistent with most treatment programs.

Complementary Roles of Housing Models

The HF and SLH approaches present different areas of strength and weakness that could potentially complement one another. While HF can accommodate a wide variety of individuals because it adapts to their needs and desires, it sacrifices the power of the social environment and the influence of peer support that can enhance the functioning of residents (e.g., Polcin, et al., 2010a, Polcin & Korcha, 2015). For residents of SLHs, peer support and social influences within the household environment are the mediums through which residents “work their personal program” for recovery on a daily basis. Sobriety permeates the home environment through daily living among peers – other recovering housemates.

While the experience of being part of a community of recovering persons is at the heart of what is thought to be helpful in SLHs, there can be downsides to this type of community for some individuals. SLHs and similar types of residences have been shown to be helpful for a wide variety of persons (Jason, et al., 2014), but some residences offer limited adaptation to individual needs. Some papers have suggested that some persons with severe mental illness may not be an appropriate match for residence in a SLH (e.g., Polcin, 2015, Polcin & Korcha, 2015). The requirements of abstinence and active participation in a community of recovering persons may be overwhelming and may require modifications, such as smaller households and staff who are trained to monitor mental health symptoms. Some SLHs have made adjustments to adapt to these types of residents (Polcin & Korcha, 2015), but the HF approach may be more appropriate for some of these individuals, particularly those who are not interested in abstinence. The U.S. Department of Housing and Urban Development currently only funds HF residences despite the potential for both housing models to contribute to decreasing homelessness (Polcin, 2015).

Recovery as Abstinence, Aspiration, and Harm Reduction

HF and SLHs present two fundamentally different approaches to “recovery” housing based on very different definitions of the term “recovery.” The origins and operations of SLHs are largely based on the principles of Alcoholics Anonymous (Wittman & Polcin, 2014) and the vast majority of SLHs have some level of participation in 12-step recovery groups. Within this traditional definition SLHs define “recovery” as the daily practice of abstinence within the fellowship of other recovering persons. “Being in recovery” means abstaining from alcohol/drugs daily and indefinitely (one day at a time). Life in the SLH consists of living in a recovering community of peers working toward the same end of remaining sober on a day-to-day basis. There is no end-point to this recovery, it is an ongoing lifelong process that becomes deeper and more meaningful with the passing years.

HF proponents advocate a different definition of recovery. In their discussions about what constitutes recovery they draw upon the Substance Abuse and Mental Health Services Administration (SAMHSA) definition of “Recovery Oriented Systems of Care” (ROSC). SAMHSA (2010) defines ROSC as “a coordinated network of community-based services and supports that is person-centered and builds on the strengths and resilience of individuals, families, and communities to achieve abstinence and improved health, wellness, and quality of life for those with or at risk of alcohol and drug problems.” The ROSC model advocates for an extremely broad-based approach that includes prevention, education, substance abuse treatment, alternative therapies, linkages to medical and mental health services, and services designed to support long-term well-being, such as housing and employment assistance. The definition emphasizes that building upon person-centered strengths is important to “achieve abstinence,” but it also emphasizes health, wellness, and quality of life. HF advocates therefore adopt a very broad definition of recovery that includes any effort toward positive change, even while individuals continue their substance use. While recovery for some individuals includes abstinence, for others it may be a long-term goal to which they aspire. For still others, abstinence may not be visualized as a goal, even a possible future goal.

We suggest the use of new terms to differentiate types of recovery: the first is “abstinence recovery,” with its emphasis on not using substances in the present and for the foreseeable future. The second is “aspirational recovery,” where abstinence is not a current goal, but the individual can visualize that as a goal at some future point. Some individuals may have no desire to change substance use currently or in the future. However they still wish to reduce harm associated with use, such as health, psychiatric, legal, financial, and social problems and improving other aspects of their lives. For these individuals we suggest the term “harm reduction recovery.”

For the aspirationalist and harm-reductionist, “being in recovery” means making qualitative progress toward resolving one’s life-problems. “Abstinence recovery” views abstinence as an essential means to those ends. At the core of these differences lie competing values placed on the centrality of abstinence and the importance of sobriety in the personal recovery process. These differences are reflected in the design and use of housing to support each approach as well as the role of peers and professionals in the recovery process. However, it must be emphasized that persons often transition back and forth among the different recovery goals. Thus, they need different types of services at different time points to respond to different recovery objectives.

Peer and Professional Influences

While peer support and peer involvement are at the heart of the recovery process in SLHs, they are emphasized minimally by HF advocates (Polcin, 2015). In contrast, the involvement of case managers is standard for HF programs (Stergiopoulos, et al., 2015) but not for SLHs. For scattered HF programs that house a limited number of persons within the general housing market, a variety of professional services are offered that may be provided onsite at the resident’s apartment or offsite through referrals to community services. However, not all HF programs use a scattered housing approach (Pearson, 2007). Some consist of large buildings with multiple units owned and operated by HF programs. In these programs case managers monitor the residents and the facility and on-site professional services (e.g., medical and mental health) are typically offered by not required. Although multiple individuals may be housed in the same building, there appears to be very little in the way of mobilizing peer interaction and peer support as therapeutic influences. Descriptions of HF programs rarely mention any role for peers in terms of having input into management of the residential environments. Absent in the literature are descriptions of ways that residents come together to support achievement of goals, operation of the facility, or participation in recreational activities.

One point often put forward by HF proponents is these residences are good options for persons with chronic mental illness because of the low demands on the resident (Tsemberis, Gulcur & Nakae, 2004). The implication is that persons with severe mental illness are thought to be unwilling or unable to practice abstinence and actively participate in a community oriented milieu. However, engagement in an appropriately designed therapeutic milieu that is not overly demanding has been shown to be helpful for persons with severe mental illness (e.g., Whitley, et al, 2008). We contend the problem of homelessness among persons with severe mental illness is not the result of inappropriate treatment or that treatment requirements have been too demanding. Rather, it is a problem of limited suitable housing that supports recovery from mental illness during and after treatment. Thus, an important question for HF programs is how might operations of these facilities be modified to maximize peer support and resident involvement among the tenants in ways that enhance the mission?

SLHs use a “social model” approach to recovery that emphasizes peer support and empowerment of the peer group in managing daily operations of SLHs (Polcin, et al., 2014). While SLH managers typically have the authority to make final decisions about admission, eviction, and consequences for rule violations, that authority should be exercised cautiously. For SLH residents, interactions with professional service providers typically occur outside the house in the surrounding community. As described elsewhere (Polcin & Korcha, 2015), SLH residents with similar issues can help one another access and successfully use community services. Although historically there was a clear policy of professional services being delivered offsite, some SLHs have made modifications in recent years. A limited number of houses allow for outside providers to visit the houses and present workshops on various topics, such as finding employment, managing psychiatric problems, communication skills, and parenting. It needs to be emphasized that the house itself typically does not provide these services. In addition, the outside professionals who deliver them do not have influence over house policies or operations.

Building Designs and Locations

Missing in most of the literature on housing models for homeless persons is recognition of how the architecture of the buildings used to house individuals interacts with the goals and purposes of the programs. Wittman et al. (2014) suggested that the architectural characteristics of SLHs need to be viewed as a vital partner in service delivery rather than a neutral container that simply provides shelter. Residences and the premises where they are located can be designed to facilitate the goals and purposes of housing for homeless persons and designs can also detract from them. Building designs and locations influence a variety of factors, including the quality of the social environment in the houses, available recreational activities the community, and factors in the community supporting recovery (e.g., 12-step recovery groups) and hindering recovery (e.g., alcohol outlets.

Persons entering a living environment react to the overall stimulus (gestalt) of the physical and social environment combined. The physical space and its furnishings shape interactions among residents and affect their feelings and self-image. These experiences can be welcoming or dispiriting. For example, persons entering scattered-site HF programs typically live alone in their own apartment within general community housing. In this scenario there may be few interactions with persons who have a history of homelessness. The result for some can be an initial experience of social isolation and alienation. Persons entering SLHs located in areas with a high concentration of alcohol outlets can be discouraging because maintaining abstinence can be difficult when access to alcohol is easy. On the other hand, SLH locations with easy access to 12-step groups and other services can be a welcome relief that supports the aims of the house. In addition, the spatial layout where residents are housed is critical for facilitating social interaction (socio-petal design) while other spaces provide appropriate privacy for personal time, sleeping and hygiene.

One way to consider architecture is to assess how the confluence of settings and operations fit with the purposes of the program. For example, the primary purpose of HF is to provide a safe and accepting place where the resident can manage his or her own life accordingly while continuing to drink/use under conditions where residents are not compelled to participate in therapeutic activities and services. This design-use plan poses challenges that must be addressed: How are HF facilities that are not scattered site designed to balance individual resident prerogatives against common concerns and minimum behaviors expected from all residents – that is, what are “house limits” and how are these enforced? Importantly, how do troublesome behaviors influence other residents, neighbors, and the reputation of the program? What arrangements are made to address neighbor complaints and concerns raised by local officials? The dilemma faced by planners of HF services is that architectural designs that facilitate social interaction among residents and interaction with the surrounding community can facilitate social support but also exacerbate the effects of troublesome behaviors.

The architectural needs of SLHs are quite different. When SLH architecture is well-designed it facilitates peer support for the maintenance of sobriety. For example, “socio-petal” designs are emphasized that include spatial layouts that naturally lead to residents congregating in common living areas (Wittman, 1993). Typically such areas are large enough to accommodate the entire household so all residents can attend house meetings and participate in community social activities. The goal is to engage residents with other members of the household in an open environment that invites participation and peer-monitoring.

Although there should be a modicum of privacy, such as a place to store valuable items, most bedrooms should be large enough to accommodate more than one person. Rooms are typically shared for several reasons. First, rent and other costs are covered by resident fees and sharing rooms reduces those costs. Second, SLHs discourage residents from isolating alone in their rooms because it can lead to relapse. The goal is to engage residents with other household members in an open environment that is easy to monitor (Wittman, et al., 2014).

Ideal neighborhoods where individuals are housed differ as well. For HF residences, neighborhoods with some degree of tolerance for substance use may be helpful. For example, if the local area takes a highly punitive stance to any drug use and readily involves police if they have suspicions about substance use it could put residents who use drugs at risk for legal problems. In a study of neighbors and other stakeholder groups located near SLHs Polcin et al (2014) found a high level of support for the homes. However, there was overwhelming support among all stakeholders for the policy of requiring abstinence. Thus, HF residences may not have been a good match for those localities. There is a serious need for research examining community perceptions of HF residences, both centralized facilities housing many persons together as well as scattered housing approached where individuals are placed in apartments within the general housing. There is also a need for research on strategies to help residents minimize the destructive influences of substance use on neighbors and the surrounding community.

The characteristics of neighborhoods that are a good match for SLHs are markedly different from those that are a good match for HF residences. Unlike HF, services are usually not delivered onsite at the houses. It is therefore helpful to locate houses where residents can easily access community services they need (e.g. medical care, mental health treatment, job training, etc.). Easy access to public transportation is important, particularly for the large number of residents who do not possess a driver’s license or have access to a vehicle. Given the focus on abstinence, it is best to locate houses away from neighborhoods that have large numbers of alcohol outlets or are known to be areas for drug distribution. These locations contain obvious triggers for persons in recovery that can precipitate relapse. Because involvement in 12-step recovery groups is required or strongly encouraged, it can also be helpful to locate houses in areas where large numbers of 12-step meetings and other recovery-oriented activities are held.

Zoning Issues

There are considerations for zoning regulations of both types of housing. HF advocates attempt to straddle the two basic housing and zoning categories of “private residence” and “residential care facility.” On the one hand, HF provides private, regular housing for homeless persons as a right. On the other, in some of their facilities they acknowledge they provide specialized on-site services not normally considered part of ordinary housing for independent living. Local codes typically make clear that specialized care residential facilities are a separate type of housing permitted only in certain designated areas. Reconciling these different classifications may require modification of existing zoning regulations.

SLHs are considered to be ordinary private housing that is permitted in all areas where housing is permitted, including single-family residential zones. SLHs are protected against discriminatory zoning by the federal Fair Housing Amendments Act. Additionally, SLH as well as HF housing both come under federal legislation that protects access to housing for people with disabilities, including alcoholism, drug dependence, and mental illness Wittman & Polcin, 2014).

Despite the legal protections offered to HF and SLH residences, both been subject to NIMBY complaints and prejudicial treatment in land-use planning and zoning matters. Although research to date indicates that SLH residences are perceived as good neighbors (e.g., Polcin, et al., 2014), some localities have attempted to modify zoning laws to discourage their existence. Examples include efforts to limit the size and density of SLHs in neighborhoods (Wittman, Jee, Polcin, & Henderson, 2014. For both types of housing to survive, there is a serious need for residents, their families, friends, and providers to engage in advocacy efforts that protect housing for homeless persons. Polcin (2014) has suggested that providers of all types of recovery services should emphasize citizenship activities (e.g., voting and political activism) as an integral component of the recovery process.

Conclusions

Our analyses of housing approaches for homeless persons in the U.S. with AOD problems leads us to a number of suggestions for provision of services and research:

1. Planners of services for homeless persons should develop policies and programs that triage homeless persons to various housing options based on needs, preferences, and considerations of the person-housing match. HF and SLHs should be developed to run robustly in parallel ways that allow homeless people to move freely back and forth between them as the individual chooses. For additional analysis and suggestions for ways service systems can better respond to the needs and preferences of persons who are homeless see Paquette and Pannella Win (2016). Considerations for ways service systems can address obstacles and mobilize community support are described by Pannella Winn and Paquette (2016).

2. Funding from the U.S. Department of Housing and Urban Development should support a range of housing options for homeless persons with substance abuse problems, including SLHs and HF residences. Funding for permanent housing after completion of substance abuse and mental health treatment is essential.

3. Architectural designs for different types of housing need to be based on resident needs and preferences. Considerations should include how the spatial layout of the facility, level of privacy, degree to which the household environment can be monitored, availability of professional services, and characteristics of the neighborhood fit with the purpose and mission of the houses.

4. Current housing models should consider ways residences can be modified and enhanced to be more responsive to resident needs. For SLHs, examples include modifying operations to address issues such as mental illness. For HF, examples include ways that operations and physical designs could mobilize peer support to address substance use and mental illness.

5. Social stigma and NIMBY (not in my back yard) attitudes toward substance abuse, homelessness and mental illness need to be addressed. Housing providers and consumers need to see community activism (i.e., citizenship) (Polcin, 2014) as an integral part of their work and recovery, regardless of how one defines recovery.

6. Research is needed in a number of areas. First, there is a need to measure architectural design characteristics and correlate them with outcomes that address the core missions of the housing programs. For HF this might be characteristics associated with housing retention and minimizing negative interactions with neighbors. For SLHs it might be characteristics associated with reduction in substance use and employment. Second, there is a need for studies to address operational characteristics and their associations with outcomes. Important questions include: What is the differential impact of housing that is affiliated with professional services, such as treatment or case management, versus housing that is freestanding? What is the differential impact of houses that are primarily staff run versus peer run? What is the differential impact of houses that are communally organized (shared space) versus individually organized (private room or apartment. Third, there is a need to assess population characteristics of homeless persons in different localities and assess how well housing and other services respond to their needs.