With the advent of deinstitutionalization in the 1960s, people with serious mental illness (SMI) have become increasingly at risk of becoming homeless, living in poor quality housing, and languishing in the community (Rochefort, 1993). Since this time, a variety of housing approaches for this population have been developed, implemented, and evaluated. In the past 10 years alone (since 1999), there have been at least 13 reviews of this research literature (see Leff et al., 2009; Nelson, Aubry, & Lafrance, 2007; Nelson & Saegert, in press). The purpose of this chapter is to describe approaches to housing for people with SMI and to review research related to those approaches. The chapter is divided into three main sections:
- housing without rehabilitation
- housing with rehabilitation, which includes an identification of best practices in housing for people with SMI, and
- change strategies for shifting to best practices in housing people with SMI
Housing Without Rehabilitation
From roughly 1850 to 1950, people with SMI in North America were warehoused in psychiatric institutions, away from mainstream society, as they were deemed to be incurable. The prevailing attitude was one of "out of sight, out of mind," and people with SMI were left in these institutions, in many cases, for their entire lives. Deinstitutionalization of individuals with SMI began in the 1950s, at which time many institutions were downsized or closed altogether (Nelson, 2006). While this was a period of major change in how society treated people with SMI, there was poor planning in terms of where to house the people who were released from the hospitals. As a result, many people were forced to live with their families or in substandard housing, shelters or on the streets with little or no support.
While there have been great strides over the past 50 years in terms of housing for people with SMI, there are still a significant number of individuals who live in housing in which no support or rehabilitation services are available. This lack of support is associated with an increased risk of becoming involved in the criminal justice system, being hospitalized, not having access to treatment, and increased risk of being victimized. Types of housing options that offer little or no support include living with family members, custodial housing, normal rental housing market, and on the streets.
Living with Families
Families represent the largest group of community caregivers for individuals with SMI (Canadian Mental Health Association, 2004). While many individuals with SMI live with family members, this type of living arrangement often does not include support or rehabilitation. Many family members are well-intentioned and caring, but the lack of support can cause great strain on the caregivers and the individuals themselves (Solomon & Draine, 1995). Many family caregivers work outside of the home and are forced to juggle their paid job with that of being an informal caregiver (Canadian Mental Health Association, 2004). This juggling act is associated with increased stress and emotional difficulties for the caregivers. Family caregivers are often forced to play many roles, such as advocate, nurse, counsellor, and crisis worker (Canadian Mental Health Association, 2004) and may have a difficult time navigating the complex mental health system or may not have access to certain services because they are not "professionals." While male caregivers are becoming more common, the majority of informal caregivers are women, mostly wives and daughters (Canadian Mental Health Association, 2004). Furthermore, eventually, caregivers will no longer be able to care for their loved one due to aging. This can result in the individual with SMI not receiving the appropriate supports and treatment for her or his illness which may cause an increase in psychiatric symptoms and decreased quality of life.
While there has been quite a bit of research on the challenges that face families who live with a family member with SMI (Loukissa, 1995), there has been very little research on the outcomes of living with family on people with SMI. In a study in Germany, Kallert, Leisse, and Winiecki (2007) found that people with chronic schizophrenia who lived with their families were less symptomatic and disabled than people with chronic schizophrenia who lived in more sheltered or therapeutic housing settings. However, over a two-year follow-up period, those who lived with their families did not show significant changes in psychiatric symptoms, social disabilities, or quality of life. In a study of people with schizophrenia in five Scandinavian countries, Hansson et al. (2002) found little difference between those living with their families and those not living with their families on measures of housing, quality of life, and social networks.
Custodial housing usually consists of large, congregate facilities operated by non-professional, private landlords for profit. These settings, which include lodging homes, foster families, and single-room occupancy hotels, are primarily located in inner-city core areas and support is oriented toward care and dependency (Parkinson, Nelson & Horgan, 1999). This type of setting represented the dominant form of housing following deinstitutionalization in the 1960s and provided little more than meal preparation and house cleaning by non-professional caregivers (Sylvestre, George, Aubry, Durbin, Nelson, & Trainor, 2007). Staff typically has little training in mental health rehabilitation and do not place emphasis on skills training or independence (Parkinson et al., 1999). This type of housing focuses on consumer deficits and is based on the assumption that individuals with SMI are unable to care for themselves or make their own decisions, which has a negative impact on individuals' mental health and quality of life (Nelson & Saegert, in press).
The previously cited study by Hansson et al. (2002) found that those living independently (in their own place) scored significantly higher on housing independence, influence, and privacy and higher on availability and adequacy of emotional relations than those living in a sheltered setting or on the streets. In an Australian study, Browne and Courtney (2004) found that while people with schizophrenia who were living on their own did not differ in terms of psychiatric symptoms from those living in boarding homes, the boarding home residents had signficiantly lower levels of social support, meaningful activities, work, and global functioning. Segal and Kotler (1993) followed up residents of sheltered care settings in California and found that residents became more dependent over time. This research and first-person accounts (Capponi, 1992) demonstrate that custodial housing is not a viable option for promoting the quality of life and recovery of people with SMI.
Normal Rental Housing Market
Many individuals with SMI are forced to rely on social assistance as their primary source of income. In Ontario, Canada, this type of social assistance called Ontario Support Disability Program (ODSP) does not provide enough money for people to live above the poverty line. In fact, ODSP is 34% below the poverty line (Canadian Mental Health Association Ontario, 2007). Furthermore, in Canada, there is a severe shortage of affordable housing units and the waiting lists for subsidized housing are often years long (Hulchanski & Shapcott, 2004). This limits individuals' choices in terms of housing options and often results in people living in substandard housing in disadvantaged neighbourhoods in which prostitution, substance abuse, violence, and other social problems are prevalent (Arai, Pedlar, & Shaw, 2006). Living in low-income neighbourhoods exposes individuals to criminal behaviour and substance abuse which increases the chances of coming into contact with the criminal justice system (Fisher et al, 2003) and/or being victimized (Hiday, Swartz, Swanson, Borum & Wagner, 1997) and has a negative impact on mental health and quality of life.
Many individuals with SMI live in chronic poverty and do not have adequate resources to obtain decent housing. This often results in individuals being forced to utilize the shelter system or live on the streets. Many shelters have limits on the number of days an individual can stay which results in a large proportion of individuals becoming homeless after reaching this maximum number of days. Being homeless has numerous negative consequences, including an increased risk of being victimized, becoming involved in the criminal justice system, contracting diseases and death. Homeless individuals also face numerous barriers in accessing healthcare (Frankish, Hwang & Quantz, 2005). Many homeless individuals do not have a health card, have difficulty keeping appointments, and lack a continuity of care due to transience. This lack of access to appropriate healthcare can result in an increase in psychiatric symptoms and a decrease in quality of life.
In summary, individuals with SMI who are living in the community as a result of deinstitutionalization often have limited housing options. These limited options often result in a lack of support from community agencies, which can result in an exacerbation of psychiatric symptoms and have a negative impact on the individual's quality of life.
Housing with Rehabilitation
Beginning in the early 1970s, mental health systems began to develop housing that focused on rehabilitation (Sylvestre, Nelson, Sabloff, & Peddle, 2007). Influenced by housing created in other sectors for people with disabilities, a variety of housing programs were developed including quarterway and halfway houses, group homes, and supervised apartments (Carling, 1993). These programs combined housing and services in a single setting and were typically segregated, professionally staffed, and congregate in nature.
These initial efforts were intended to offer rehabilitation services based on a residential continuum with housing offering varying levels of support that would correspond with the level of severity of an individual's mental health difficulties (Ridgway & Zipple, 1990). As individuals made progress in achieving their rehabilitation goals, they would navigate down the continuum taking on an increasing amount of autonomy until eventually they would be ready for independent living in the community.
Few mental health systems in North America were actually successful in developing a comprehensive continuum of residential programming options (Goering et al., 1997). Moreover, there was a lack of clarity and no consensus in the community mental health field on the elements of an ideal continuum (Ridgway & Zipple, 1990). The "continuum" model was also criticized for confusing an individual's need for housing in the community with the need for treatment (Carling, 1993). Essentially, it placed them in a position where they had to accept treatment in order to access housing. Other criticisms of the continuum approach included the temporary nature of the housing provided requiring individuals to experience disruptive moves, the frequent lack of correspondence between consumer needs and their placement on the continuum, the exclusionary and segregated nature of specialized housing in communities, and the artificiality of congregate facilities for enabling individuals to achieve rehabilitation goals related to living in the community (Ridgway & Zipple, 1990). Finally, consumer preference surveys conducted in regions across North America have consistently shown that the large majority of consumers have a strong preference for living in regular housing (Nelson, Hall, & Forchuk, 1993; Piat et al., 2008; Tanzman, 1993).
In response to the criticisms of the residential continuum model of housing, Paul Carling argued for the development of a "supported housing" approach in which individuals live in normal housing in the community and receive separately individualized, flexible, and portable supports (Carling, 1993, 1995). The development of community mental health programs, such as Assertive Community Treatment and Intensive Case Management, which offer these kinds of supports, has facilitated the creation of supported housing programs over the past 15 years. Indeed, there is evidence at least in the U.S. that supported housing has now become the predominant housing model for people with SMI (Yanos, 2007).
In light of this history of the development of housing programs with rehabilitation, the research literature now differentiates it into the two broad categories of supportive housing and supported housing (Dorvil, Morin, Beaulieu, & Robert, 2008; Goering et al., 1997; Nelson & Saegert, in press; Parkinson et al., 1999; Sylvestre, Nelson, et al., 2007). These will described next along with the research on their demonstrated effectiveness.
In line with the residential continuum approach from which it developed, supportive housing links housing and support in specialized settings such as group homes or supervised apartments located in one building (Sylvestre, Nelson, et al., 2007). The housing is congregate in nature with groups homes having 6-12 residents and supervised apartments having 2-3 residents living together (Parkinson et al., 1999). The focus of the rehabilitation in supportive housing is on building skills and the intensity of the support varies depending on the setting (i.e., high support group homes, intermediate support groups homes, low support supervised apartments) (Ridgway & Zipple, 1990).
Staff in supportive housing typically provides standardized support that can include supportive counseling, case management, and social and life skills training (Parkinson et al., 1999). A deficits perspective is favored but consumers are usually also viewed by staff as having strengths and a capacity to change (Sylvestre, Nelson, et al., 2007). Support is usually provided in the setting but in the case of supervised apartments staff may be on call and not present on site (Parkinson et al., 1999). Consumers have a limited say over their choice of housing, housemates, or the services they receive.
There is usually an emphasis in this type of housing on developing a supportive community among those living together and professional support focuses both on rehabilitation and housing issues (e.g., group decision-making, conflict resolution) (Sylvestre, Nelson, et al., 2007). Although an underlying assumption of programs may be to ultimately help consumers transition into an independent living situation, many supportive housing programs are open-ended in terms of length of stay and residents can stay long-term (Carling, 1993; Parkinson et al., 1999).
Overall, there remains a paucity of rigorous research evaluating the effectiveness of supportive housing and a wide range of housing programs are subsumed under this category, complicating the interpretation of the research that has been conducted. Research conducted on supportive housing has shown that its residents experience a reduction in homelessness, hospitalizations, psychiatric symptoms, and drug abuse and an improvement in housing and financial stability, quality of life, and satisfaction with their living situation (Leff et al., 2009; Nelson et al., 2007; Nelson & Saegert, in press; Parkinson et al., 1999).
In contrast to supportive housing, supported housing is an approach that separates housing and support, assisting consumers to access regular housing in the community while offering them individualized and flexible support services (Sylvestre, Nelson, et al., 2007). The approach emphasizes consumer choice in terms of both housing (i.e., location, type, roommates) and supports (i.e., intensity, location, focus) (Nelson & Saegert, in press). In order to maximize the choice available to consumers, supported housing programs usually offer rent supplements. The support available is portable so that it follows consumers if and when they move or even if they are hospitalized. Values guiding supported housing include consumer empowerment and community integration (Parkinson et al., 1999).
In this approach, housing is defined as a place for consumers to live and not a treatment setting (Goering et al., 1997). The type of housing that consumers access with the assistance of rent supplements are private market apartments, housing co-ops, and social housing (Parkinson et al., 1999). Because they are given choice, consumers can choose the location so that they are close to friends, family, or community resources. Important by-products of living in regular housing facilitated by this approach are the enabling of consumers to take on regular roles as tenants, to come into more frequent contact with non-disabled individuals, and to integrate into the community.
Services provided to consumers are oriented to strengths (Sylvestre, Nelson, et al., 2007) and the intensity is regulated based on need and interest (Parkinson et al., 1999). A particular form of supported housing known as "Housing First" has become the intervention of choice in many cities for assisting people with SMI and a history of homelessness to become stably housed (NREPP, 2007). The original "Housing First" program was developed and evaluated by Pathways to Housing Inc. in New York City (Tsemberis, 1999). The Pathways program combines subsidized housing with Assertive Community Treatment, providing services that include psychiatric and substance use treatment, supported employment, illness management, and recovery services (NREPP, 2007). Spinoffs to the Pathways program have been developed that includes the provision of less intensive support such as through intensive case management (e.g., City of Toronto Shelter, Support & Housing Administration, 2007).
Like supportive housing, there has been only a small amount of research evaluating the effectiveness of supported housing (Leff et al., 2009; Nelson et al., 2007; Nelson & Saegert, in press). This research has shown supported housing, when compared to treatment as usual in the community that does not include housing, to achieve greater improvements in housing stability, housing choice and control, employment, social networks, and subjective quality of life, as well as decreases in hospitalization, psychiatric symptoms, and substance use for individuals with SMI (Leff et al., 2009; Nelson et al., 2007; Nelson & Saegert, in press).
To date, results of research on supportive and supported housing show the two approaches to yield very similar results, although the effects of supported housing are generally more pronounced (Leff et al., 2009). In the only study comparing supported and supportive housing, residents of supported housing reported greater housing choice and control relative to residents in supportive housing (Tsemberis, Gulcur, & Nakae, 2004). The use of a supported housing approach with people with SMI who have also experienced chronic homelessness has yielded impressive findings that include the achievement of housing stability and a reduction in the use of institutional services (i.e., hospital and corrections-based) (Greenwood, Schaefer-McDaniel, Winkel, & Tsemberis, 2005; Tsemberis et al., 2004).
Change: Strategies for Shifting to Best Practices
While much is known from extant research about what constitutes best practices in housing for people with SMI, best practice approaches and policies to support them are the exception rather than the rule in Canada and the U.S. (Nelson & Saegert, in press). For example, between 1993 and 2000, fewer than 1,000 units of social housing were created in all of Canada, compared with more than 25,000 units that were created in the year 1980 alone, which has contributed to a growing homeless population (Hulchanski & Shapcott, 2004). Moreover, an ideology of "blaming the victim," which holds that people are homeless because they choose to become homeless or because of some personal defects (bad behaviours, choices, or lifestyle habits), has guided Canadian housing policy since the 1980s. In 2002, Canada's current Minister of Finance called for jailing homeless people. Also, board-and-care homes and other custodial facilities that do not reflect best practices remain the norm for housing people with SMI in many North American communities. A multi-dimensional approach to change that employs several different strategies is needed to shift governments and communities towards best practice approaches.
Challenging Assumptions and Building a Vision and Values
Fundamental to all change strategies is the need to challenge assumptions and beliefs that people with SMI are not capable of making their own decisions, need to be taken care of, are too sick to know what they want, have reached their highest levels of functioning, and cannot grow further. Such beliefs can be used to rationalize warehousing people with SMI in custodial types of housing that offer no rehabilitation services. As such, these assumptions, beliefs, and values are the deep structures of systems that resist change and maintain the status quo of poor quality housing for this population (Foster-Fishman, Nowell, & Yang, 2007). A paradigm shift towards best practices in housing must be based on an alternative vision of recovery and a set of values (citizenship, holistic health, power, social inclusion, and social justice) that guides the journey towards that vision (Foster-Fishman et al., 2007). Moreover, this change in deep structures must be accompanied by innovative and creative strategies for change at all levels with multiple stakeholders (consumers, family members, staff, managers, planners, and policy-makers). The Pathways program in New York City is an excellent example of Housing First that embodies these recovery principles (Tsemberis & Eisenberg, 2000). Pathways is:
"... founded on the belief that housing is a basic human right for all individuals, regardless of disability, the program provides clients with housing first - before other services are offered. All clients are offered immediate access to permanent independent apartments of their own." (p. 488)
Moreover, Pathways practices "... radical acceptance of the consumer's point of view" (p. 489) and hires consumers as staff members.
Education and Advocacy
To counteract major strides backwards in federal housing policy in Canada, social activists, including those with a history of SMI, formed coalitions in the 1990s (e.g., the Dream Team [a group of mental health consumers], the National Network for Housing and Homelessness, the Toronto Disaster Relief Committee, Community Action on Homelessness – Halifax, Raising the Roof) to address the issues of housing and homelessness. They have argued that if all levels of government (federal, provincial, municipal) contribute 1% of their budgets to housing, that the homelessness problem in Canada would be significantly reduced, if not eliminated. As a result of education and advocacy activities by these activists, the federal government developed a National Homelessness Initiative in 1999, which has put some funding back into social housing.
Earlier supportive housing coalitions created in the Toronto (Trainor, Lurie, Ballantyne, & Long, 1987) and the Waterloo Region (Nelson, 1994) of Ontario, Canada in the 1980s successfully lobbied the provincial government for more supportive housing for people with SMI. More recently, local-level mental health coalitions have successfully renewed efforts to expand Housing First options for people with SMI. For example, the city of Hamilton, Ontario had 674 people with SMI living in second-level lodging homes and 65 in Homes for Special Care (for a total of 739 living in custodial housing), but only 49 in supportive housing in the 1990s. Advocates formed the Supported Housing Network in Hamilton and were able to develop 158 new units of housing based on the Housing First approach. Similarly, in London, Ontario, a Community-University Research Alliance (CURA) entitled "Partnerships in Capacity Building: Housing, Community Economic Development, and Psychiatric Survivors" held annual community conferences, developed fact sheets and summary bulletins, and held all candidates meetings during provincial and municipal elections. Through the CURA, this group successfully advocated for 111 new units of housing also based on the Housing First approach. Ottawa's Alliance to End Homelessness developed a Community Action Plan on Homelessness to create "A City Without Homelessness" and has issued an annual report card on the state homelessness in Canada's capital city (Fuller, Browne, Beaulac, & Aubry, 2006). The overall thrust of these education and advocacy activities is to increase the number and quality of low-income housing units for low-income Canadians, including those with SMI.
While increasing the stock of new housing is important, a parallel strategy that is also very important is to change existing housing so that it reflects best practice principles. Some mental health organizations have used internally-driven initiatives to make change. In Winnipeg, Canada, for example, Options for Supported Housing sold its group homes and created independent apartments to shift towards a Housing First approach (Parkinson, 1999). Other organizations have used external consultants to assist with the change process. With the help of a consultant, Waterloo Regional Homes for Mental Health, Inc. in the Waterloo Region of Ontario decided not to sell its group homes, but it did de-link housing and support and created more individualized, consumer-directed services as it shifted to a more independent housing approach (Lord, Ochocka, Czarny, & MacGillivary, 1998). Moreover, all of the more than 100 new units of housing that this organization has created are independent apartments that are consistent with the Housing First approach. Finally, consultation with planning and policy-making bodies has been undertaken to create change at the regional (Parkinson, Nelson, & Horgan, 1998) and provincial levels (Sylvestre, George, et al., 2007) towards a Housing First approach.
In this chapter, we provided a brief introduction to types of housing for people with SMI. We made a distinction between housing without rehabilitation and housing with rehabilitation. While housing that offers no hope for rehabilitation and recovery of people with SMI is clearly an unacceptable social policy, it is all too prevalent in North America. Commenting on this sorry state of affairs, Drake and Wallach (1999, p. 589) made the following statement 10 years ago:
"Homelessness reflects our societal values. People with psychiatric disabilities are generally poor and disadvantaged, and safe and decent housing is often beyond their means. Access to affordable housing and appropriate supports for people with disabilities might easily undergird the social structure of the wealthiest nation on earth; indeed, deinstitutionalization could make sense only in the context of providing affordable housing and supports."
Unfortunately, housing without rehabilitation remains the norm in many North American communities, leaving people with SMI to languish rather to flourish. This is a major blemish on our societies, as well as a waste of important human assets.
We made a distinction in this chapter between supportive and supported housing. While supportive housing is clearly an improvement over custodial approaches, this approach has important limitations. Housing First, based on the supported housing model introduced by Paul Carling in the 1990s, bears great promise in solving the problems of poor housing for people with SMI. By promoting choice and providing rent supplements, Housing First enables people with SMI to access the types of housing that they want. Significantly, research has shown the positive impacts of this approach for people with SMI and co-occurring problems, such as substance abuse and homelessness.
In the final part of the chapter, we underlined strategies for creating change to shift towards a more supported housing approach. This not only requires the translation of research into practice, but the articulation of an alternative vision and values for people with SMI and strategies of consultation, education, and advocacy to make change. Changes in social policy and practice are desperately needed to address the problems of housing and mental health for people with SMI. This chapter provides directions in the types of changes that are needed and how they can be achieved.