Enhr Conference 2011 – 5-8 July, Toulouse Changes in Irish homeless policy: What next for homeless people with a high level of need? Niamh Murphy Respond! Housing Association e-mail: niamh.murphy@respond.ie Abstract With the aim of ending long-term homelessness and the need to sleep rough in Ireland, a range of homeless policy changes have been made since 2008, most notably in the Dublin area. This paper sets out to analyse these changes in the context of homelessness models and attempts to place Ireland within this spectrum of services. The new policy will be critiqued in relation to research on these models and their positive and/or negative effects on shortening the homeless ‘careers’ of the most vulnerable homeless people. As the restructuring of services in Dublin is still underway, the full effects will not be evident for a number of years. However, it is possible to make some conclusions and recommendations in relation to the approach currently being implemented, in order to ensure that the most vulnerable benefit from the new services and can obtain and maintain independent homes. Introduction Homeless services in Ireland, and more significantly in the Dublin region, have undergone many changes since 2008. The national homeless strategy, The Way Home, was published by the Department of Environment, Heritage and Local Government in 2008 with the aim of ending adult homelessness in Ireland by 2010. This department is primarily responsible for the formulation and implementation of policy and legislation in relation to housing in Ireland. Most housing services, however, are delivered through the local authorities (municipalities). Each local authority area has a Homeless Forum, comprising statutory and voluntary representatives, who are responsible for the creation of local homeless action plans. Due to the higher levels of homelessness in Dublin than the rest of the country, the Homeless Agency was set up in 2001 and replaced the Dublin homeless forum. The Homeless Agency is responsible for the planning, co-ordination and administration of funding for the provision of homeless services in the Dublin area. They work in partnership with a range of voluntary and statutory agencies to implement their agreed plan and work towards achieving its vision. This group is known as the Homeless Agency Partnership. A localised strategy – Pathways to Home – was published for the Dublin area by the Homeless Agency in 2009. It outlined a restructuring of services in the city; moving away from traditional transitional services towards short-term emergency accommodation and supported housing services (independent and semi-independent housing with supports). This paper will outline the recent changes to homelessness policy, detailing the Homeless Agency’s Pathway to Home model and the new ‘Mixité’: an urban and housing issue? 2 services and programmes that have been introduced since the publication of this and the national strategy. Research on the different models of homeless service provision will be discussed in order to determine whether any one model of services works best. The models described include: 1. the linear model of service provision (also known as staircase or continuum of care models) which involves the traditional transitional approach to homeless services with a ‘treatment first’ philosophy and the movement towards independent housing in stages; and 2. the model of normalisation consisting of homeless services such as Housing First and supportive housing, involving the provision of independent or semi-independent housing with support services. From the available research, it appears that services based on a strictly adhered to Housing First model (independent housing with supports and an emphasis on consumer choice and harm reduction), would have the greatest impact in terms of ending homelessness for those with a high level of need and who prove difficult to rehouse through traditional transitional approaches to service delivery. The paper illustrates how the changes to services in Dublin mark a clear move away from the traditional transitional approach, towards independent or semi-independent housing with supports. This restructuring of services has only happened in Dublin. This has resulted in what can be described as a two-tiered system of service delivery for homeless people in Ireland; the normalisation model dominating in Dublin with the linear model continuing to dominate outside the capital and the larger urban centres. Homelessness in Ireland The MakeRoom campaign to end homelessness (an initiative of a number of voluntary housing groups) has estimated the number of people who are homeless to be significantly higher than those included in the Assessment of Housing Need (numbers on social housing waiting list). The campaign’s figures estimate that at least 3,305 households or 4,143 persons are homeless (2008 and 2009 figures). The estimate, however, is very conservative in relation to the number of children and is lacking in up-to-date comparable national data. In relation to people under the age of 18, Focus Ireland has stated that 1,500 are homeless in Ireland each year. The number of young people using the “out-of-hours service” in Dublin has increased from 492 young people in 2006 to 705 in 2009 (FEANTSA, 2011: 6). Table 1: Estimated number homeless in Ireland 2008 and 2009 figures Data Source Counted In Dublin region 2008 Counted In Galway City 2008 Counted In Limerick City 2008 Counted In Cork City 2008 Housing Needs Assessment 2008 (Areas where no Counted In took place HSE, 2009 (unattached under 18s) Total Homeless Total Population (2009) Households 2,144 157 214 396 314 Persons 2,942 163 223 421 314 Adults 2,366 160 220 411 314 Children 576 3 3 10 0 80 3,305 -- 80 4,143 4,459,300 0 3,471 -- 80 672 -- Source: MakeRoom http://www.makeroom.ie/download/pdf/2010_homeless_numbers_memo_final.pdf (accessed 04/05/2011) 3 Workshop 14: Welfare Policy, Homelessness and Social Exclusion Homelessness policy in Ireland: Recent developments Strategies, action plans and evaluations The Irish government launched their new national homeless strategy in August 2008. The strategy set out three core objects for the following five years: • • • the elimination of long-term occupation of emergency homeless facilities; the elimination of the need to sleep rough; and the preventing of the occurrence of homelessness as far as possible. The government set a target date of ending long-term homelessness and the need for people to sleep rough by 2010 (Department of Environment, Heritage and Local Government, 2008: 7). To keep up the momentum in meeting this shared vision, the Homeless Agency commissioned a comprehensive review of homelessness services in Dublin (2008). The reviews sought to measure the extent of homelessness in the city and to evaluate current measures in response to homelessness. Expenditure was reviewed in order to determine whether the resources were being used in a manner which met the strategic aims of the partnership, as well as examining how overall value for money is achieved (The Homeless Agency Partnership, 2009: 5). An Implementation Steering Group was set up in 2009 by the Board of the Homeless Agency Partnership. Their primary task was to consider and propose – to the partnership and central government – an implementation plan in response to these reviews. Their implementation plan, Pathway to Home, sets out what the Homeless Agency Partnership considers the most appropriate way to realise the target of ending long-term homelessness and the need for people to sleep rough (The Homeless Agency Partnership, 2009: 6). The Pathway to Home model The Pathway to Home document sets out a new model for the delivery of homelessness services in Dublin. The model is based on a shifting of resources away from the provision of temporary accommodation to long-term supported housing solutions. The plan seeks to reconfigure current homeless and housing services into a ‘Pathway’ model of provision; prevent homelessness; provide effective services in each of the four Dublin local authority areas; and provide sufficient long term housing with appropriate supports as required (The Homeless Agency Partnership, 2010: 3). Box 1: Use of the term ‘Pathways’ The use of the term ‘Pathways’ in relation to homeless service provision is sometimes confusing and needs clarification. The organisation accredited with the development of the Housing First model is called the Pathways to Housing organisation. The new model for homelessness services in Dublin is called Pathway to Home, yet it is different from the Housing First Model in a number of ways (although a Housing First project has recently been set up by the Homeless Agency as part of the overall services). Also, some local authorities in the UK, for example, the London Borough of Camden use what they call a ‘pathways’ approach through supported accommodation schemes from ‘assessment beds’, progress beds’, through to a penultimate stage similar to a ‘trial tenancy’ in a selfcontained unit (Johnsen and Teixeira, 2010: 15). The following sections outline the new accommodation and support services for Dublin. The new services include supported temporary accommodation, temporary emergency accommodation, housing support services, long-term supported residential accommodation, the Support to Live Independently Scheme (SLI) and the Homeless Agency Housing First project. ‘Mixité’: an urban and housing issue? 4 Supported Temporary Accommodation Supported Temporary Accommodation (STA) is specifically for persons who have specialised health, care and support needs, or complex needs in addition to housing needs. It is a ‘key-working’ service in which a holistic needs assessment will be undertaken for each resident. All residents are allocated a project key-worker for the duration of their stay to help link them to mainstream services. In addition to this, STA will deliver programmes that address a range of requirements of their residents such as training, education, life-skills and well-being. The Pathway to Home plan states that STA will require the provision of dedicated intervention programmes that can stabilise, for example, a person’s drug and/or alcohol related lifestyle and address behaviour associated with addiction. It proposes that a range of visiting mental health, addiction and Primary Care Teams provide services in situ for persons resident in STA. Residency in STA is limited to a maximum six month period before residents progress to more permanent housing with supports (The Homeless Agency Partnership, 2009: 44). Temporary Emergency Accommodation For those with general accommodation need and low or no support needs, emergency provision will be in the form of Temporary Emergency Accommodation (TEA). During a person’s time in TEA, a holistic assessment of housing need will be completed as appropriate. The service provider will ensure, where a person can be diverted into private-rented accommodation in a short period of time that this is undertaken as a priority. As in STA, the key-worker will establish and maintain links with the necessary agencies for the delivery of healthcare, welfare, education, training and other services. Again, residency in TEA is limited to a maximum of six months. Housing support services Housing support services are a key component of the Pathway to Home model in order to secure and maintain progress towards independent living. The Local Authority Housing Service is to work in collaboration with the housing support team, key workers, case managers, and Health Services Executive (HSE) staff to identify the level and duration of supports required for every person who is assessed. The person’s support requirements are to be adjusted as their needs change. The proposed support will be outlined in their housing support plan, and will be agreed and put in place, prior to their move on from temporary accommodation into housing (The Homeless Agency Partnership, 2009: 46). Figure 1. Type of housing support required by homeless households, 2008 Source: The Homeless Agency Partnership, 2009: 61. 5 Workshop 14: Welfare Policy, Homelessness and Social Exclusion Figure 1 outlines the estimated housing support requirements of households in 2008 based on data from the Evaluation of Homeless Services 2008 Series. The chart shows the importance of housing support services, with only 17 per cent of the 1531 homeless households included in the assessment considered as needing no support (service managers assessments of approximate housing support needs of homeless service users). The level of support required, however, varies greatly between clients, with 26 per cent in need of short-term visiting support and 26 per cent in need of long term visiting support. A total of 31 per cent of the service users were assessed as needing on-site or nursing home services. It is important to note that these figures only include people who were engaged with homeless services. For other people who are homeless and are not engaged with services, their need may just be one of housing rather than supported housing. Long-term supported residential accommodation The Pathway to Home model states that a small cohort of formerly homeless people, who are presenting with complex multiple health/social support needs, may not be able to live independently. For this group, they suggest, there will need to be intensive on-site healthcare and housing supports made available as appropriate. The plan does not include much information on this cohort of people. Rather, it states that additional scoping and clarification is necessary in order to further identify the level and range of competencies required, so that this provision can be put in place (The Homeless Agency Partnership, 2009: 46). Support to Live Independently scheme Support to Live Independently (SLI) is a scheme of mainstream housing and support which involves the use of accommodation procured through the Social Housing Leasing Scheme (Circular N3/09), or accommodation which is available to local authorities in the form of unsold affordable housing which is considered unlikely to sell in the current market. As well as accommodation, the scheme provides for low to moderate level visiting supports on a reducing basis for a period of time in order to help homeless households address challenges that they may face in making the transition to independent living. People with a high level of need are not eligible for SLI as the level of support they require is not available under the terms of the scheme. Funding for SLI is to be secured through the cost savings on emergency and transitional housing accommodation and will only be provided for people who move from homeless accommodation currently funded under section 10 of the Housing Act 1988. There should be no overlap or duplication of services which would result in an overall increase in the cost of homelessness services. High level or specialist support will not be provided through the scheme; rather tenants will be helped by their support worker to access other mainstream support services. Homeless Agency Housing First project The Homeless Agency recently set up a Housing First project in Dublin. Housing First is a type of homeless service developed in the US by an organisation called Pathways to Housing. People with a high level of need, such as drug addicts or people with mental health difficulties, are housed straight from the street into independent housing with supports available if they chose to use them. Unlike many transitional projects, abstinence or engagement with treatment services is not a requirement of housing. The Homeless Agency project is currently running on a small scale. Twenty-three participants are proposed for the first year, with 5-6 people expected to be taken in to the project in each quarter, allowing for a phased increase in the numbers. The project is currently (May 2011) engaging with the first group of people to be offered housing. The project was set up with guidance from the Pathways to Housing organisation in the United States (US) and has been adapted to a Dublin context. To this end, there are two deviations from the original Housing First programme. Firstly, all the participants in the New York projects have an axis 1 mental ‘Mixité’: an urban and housing issue? 6 health diagnosis (Axis 1 consists of clinical disorders, including major mental disorders, and learning disorders). However, the Homeless Agency found that the numbers were too small to use the same target group in Dublin. Instead they liaised with Pathways to Housing who agreed that different criteria could be used for participation as they were still working with the most entrenched rough sleepers. Secondly, the American projects have core staff that provide the support services. However, the Dublin project is still in very early stages and much of the support team is voluntary at present. According to the Homeless Agency, the Dublin Housing First project seeks to end the need to sleep rough for an identified number (23 in year one) of entrenched rough sleepers with significant support requirements. The target group have remained sleeping rough over a number of years and have generally not engaged with other accommodation, social care and primary care services. According to the new government, homeless services in Ireland which are aimed at ending long-term homelessness are to be based on a Housing First approach as stated in the Programme for Government 2011. It is important that consideration is given to the fact that many of the emergency and transitional housing services have now been closed or reconfigured to fit with the Pathway to Home model of temporary emergency accommodation and housing with supports, rather than a Housing First model. Whether the government plans to introduce a national Housing First programme, like the pilot Homeless Agency project, is not yet evident. Research on models of homelessness service provision Homeless services are often described as falling within a particular model of service provision such as the linear model or the model of normalisation. The linear model refers to services such as transitional housing units, where a number of people live in a hostel setting or in self-contained units with some communal facilities for a period of around two years before they move on to more independent housing. Traditionally this model has dominated homeless service delivery. More recently, services based on the model of normalisation have been developed, such as Housing First and supported housing. Housing First involves the provision of housing to homeless people with a high level of need straight from the street, without the requirement for transitional housing. Support services are available to those housed through Housing First, although it is not a requirement that they engage with these services. For those with addiction or mental health issues, treatment is optional and is not related to entitlement to their home. Supported housing services are sometimes similar to Housing First projects, in that elements of Housing First are included in the programme. Supported housing projects generally include the provision of independent or semi-independent permanent housing with supports. Unlike Housing First, however, some supported housing projects make it a requirement that tenants engage with the support services offered. Linear models To date, linear models such as the ‘staircase of transition’ and the ‘continuum of care’, have dominated homeless service provision in many countries. There are a number of different forms that linear models can take. However, the feature they all have in common is the fact that they require some form of progression of homeless people through a number of homelessness services towards independent living. The staircase of transition and continuum of care In Sweden, and many other European countries, ‘the “staircase” metaphor is used to describe shelter/housing systems where an individual’s housing becomes progressively more normal’ (Johnsen and Teixeira, 2010: 4). The higher people climb from the streets, the better their condition in terms of physical standard and space, freedom and security of tenure. During their time in the various forms of housing provision, the service users are expected to solve what are seen as ‘underlying problems’, for 7 Workshop 14: Welfare Policy, Homelessness and Social Exclusion example paying off debts, dealing with substance abuse issues and gaining employment (Sahlin, 1996, 1998 in Sahlin, 2003: 4). Figure 2. The ‘Staircase of Transition’ Source Sahlin, 2003: 4. Within ‘staircase’ provision, independent housing is only considered appropriate when the service user exhibits sufficient ‘housing readiness’ – that is, the service provider deems them rehabilitated enough to be able to maintain a home. This is founded on a ‘treatment first’ philosophy, which requires detoxification and sobriety before enabling access to independent housing (Johnsen and Teixeira, 2010: 4). There are considerable differences in the demands and expectations of transitional housing programmes internationally. Some have strict rules that residents must abide by such as in Sweden, while others are more flexible in their approach, for example, as in the UK where horizontal, as well as backward or forward moves between services may be possible if the placement is in danger of breaking down (Johnsen and Teixeira, 2010: 15). Are linear models of homelessness service provision effective? In terms of getting people from rough sleeping to the indoors and preparing them for independent living, it has been acknowledged that linear approaches do ‘work’ (Johnsen and Teixeira, 2010: 5). However, transitional housing programmes have been developed in the belief that they will increase the likelihood of successful retention of an independent tenancy for homeless people with complex needs, as some people are believed to need more assistance than is available through emergency shelters before they will be able to sustain housing on their own (Burt, 2006: 5). These programmes act as a place where people can work towards achieving ‘housing readiness’, a term which implies a change in behaviour or circumstances. This is the essence of transitional housing. The model assumes that the skills a person needs to live independently can be learned in transitional congregated living. However, for those with mental health issues or psychiatric disabilities, research in psychiatric rehabilitation indicates that the most effective place to teach a person the skills required for a particular environment is within that setting itself (Tsemberis et al., 2004: 651). As well as this, Tsemberis et al. argue that consumers’ perceptions of linear forms of care differ from service providers in that they experience the continuum or staircase as a series of hurdles which many of them ‘Mixité’: an urban and housing issue? 8 are unable or unwilling to overcome. For these consumers, housing is an immediate need; yet access to housing is not available to them unless they first complete treatment or are in some way deemed rehabilitated (2004: 651). Many of the critiques of linear models focus on the high attrition rate of services, in other words, the loss of service users between stages. According to Johnsen and Teixeira (2010: 5), this is commonly attributed to a number of factors: • • • • • the stress of constant change as clients move between projects; the reduction in support at each stage which may not suit people with multiple needs; use of standardised (‘one size fits all’) support programmes; lack of service user choice/control; and the ineligibility/rejection of potentially problematic clients. Another aspect of linear models which has received considerable criticism is the fact that recovery from addiction or mental health problems is a non-linear process. However, a strict linear approach to homeless services does not recognise this. In many cases relapse may be inevitable and this can result in ejection from a transitional housing project, for example, one which forbids the use of alcohol, leading to the person either ending up in emergency accommodation again or back on the street. Linear approaches can leave little room for this relapse stage when strictly applied and can, therefore, lead to a failure to prove ‘housing readiness’ and the resulting move on to independent housing. Figure 3. Prochaska and DiClemente’s Cycle of Change model Source: http://studentlife.tamu.edu/adep/addiction_cycle, accessed 27/04/2010. Of those people that do successfully pass through transitional housing, Novac et al. argue that no single characteristic assessed so far will distinguish an individual’s chances of success. Housing outcomes were not found to have a relationship to characteristics such as gender, age, psychiatric 9 Workshop 14: Welfare Policy, Homelessness and Social Exclusion disability or addiction, ethnicity, length of time homeless, main means of support, sleeping place, and pre-baseline services according to Barrow and Soto (1996, 2000 quoted in Novac et al., 2009: 16). However, they did find that a particular constellation of characteristics was associated with negative outcomes where a person left or was discharged from a programme without a placement. These people were more likely to be women in their forties with the most severe psychiatric conditions, who were actively abusing substances when first admitted to the programme. The findings of Hawthorne et al. (1994, quoted in Novac et al., 2009: 16) support those of Barrow and Soto as they found that various socio-demographic and clinical factors such as diagnosis, age, gender number of previous hospital or crisis centre admissions, employment and living situation, and length of stay, were not related to successful treatment outcomes. Models of normalisation ‘Housing First’ The Housing First model is generally attributed to Dr. Sam Tsembris; a psychologist based in the Pathways to Housing organisation in New York. The model was originally developed as a means of addressing homelessness among people with psychiatric conditions. This model of provision sees housing as a fundamental need and a human right (Padgett et al., 2006: 75). Therefore it bypasses the transitional stages of homeless service provision and places the most vulnerable homeless people directly from the street or emergency shelters into permanent independent housing. Consumer choice is a key part of the Housing First philosophy so support services are offered to tenants yet it is not a requirement that they engage with these services. There are only two requirements of tenants for the programme. They must meet with a staff member at least twice a month and they must participate in a money management programme. The Housing First model is a clear move away from the ‘treatment first’ philosophy of linear types of service provision. It does not attempt to ‘fix’ homeless people in order to make them ‘housing ready’, but ‘rather is premised on the assumption that the best place to prepare for independent living is in independent accommodation’ (Johnsen and Teixeira, 2010: 6). The key elements of the Housing First model are outlined in Box 2. Box 2: Key elements of the Housing First model – as endorsed by Pathways to Housing in the US 1. Immediate provision of independent accommodation in ‘normal’ private rented scatter site housing leased by the provider. No more than 15 per cent of housing units in any single building are used to accommodate clients so as to promote community integration. 2. No requirement regarding ‘housing readiness’; that is, an absence of high threshold admission criteria regarding sobriety, basic living skills, or motivation to change. Consumers may refuse clinical services, such as taking psychiatric medication, seeing a psychiatrist, or working with a substance use specialist. Housing is regarded as a basic human right, not something that should be earned or used as an enticement into treatment or sobriety. 3. Deployment of a harm reduction, rather than abstinence, approach to substance misuse. This separates clinical issues from housing issues, such that a clinical crisis (e.g. relapse) does not result in eviction. There is no expectation that users enter treatment for either mental health or substance abuse problems; they may refuse both without compromising their housing. 4. Provision of permanent housing and support. Apartments are permanent and kept open for service users if they are temporarily incarcerated or hospitalised, and access to support is not time-limited. Consumers only risk eviction for the same reasons as other building tenants such as non-payment of rent, creating unacceptable disturbances to neighbours, or other violations of a standard lease. ‘Mixité’: an urban and housing issue? 10 5. Respect for consumer choice regarding a) choice of apartment and furnishings; b) level of engagement with support; and c) the location and times of contact with support workers. Clients must express an interest in receiving housing, but may refuse services (such as seeing mental health or substance use specialists). The only requirement is that they meet with staff a minimum of two times per month in the first year, and participate in a ‘money management’ programme. 6. Provision of integrated and comprehensive community-based support to consumers through multi-disciplinary Assertive Community Treatment (ACT) teams. ACTs comprise social workers, nurses, psychiatrists, peer counsellors (former homeless persons with similar experiences) and employment workers. The teams are located off-site, but are on-call 24 hours a day, seven days a week, on a time-unlimited basis and provide most services in a client’s home or neighbourhood. 7. Targeting of the most vulnerable consumers – that is, those who have difficulty coping with or succeeding in traditional services and/or are resistant to service interventions – and then accept these consumers on a first-come, first-served basis. Source: Johnsen and Teixeira, 2010: 7. Consumer choice, mastery and psychological outcomes Consumer choice is a core element of the Housing First model. Despite the fact that the consumer choice movement has claimed for a long time that treatment choice is a fundamental right that should be restored to consumers of mental health and homelessness services, researchers have ‘only recently begun to systematically investigate the legitimacy of these claims and the effect of such programmatic changes on psychosocial and service utilization outcomes’ (Greenwood et al., 2005: 225). ‘Mastery’, or the extent to which people see themselves as being in control of forces that affect their lives in significant ways, is believed to be an important personal coping mechanism. Low rates of mastery have been linked to both a decrease in overall well-being and to hopelessness and passivity. On the other hand, high rates of mastery have been shown to boost mental health and functioning (Greenwood et al., 2005: 225-226). The purpose of the study by Greenwood et al. is to demonstrate a link between mastery or personal control – that is consumer choice in treatment services as offered through Housing First programmes – and psychiatric symptoms. The study consisted of 197 participants who were randomly selected for either Housing First or traditional linear care. The researchers state that the results demonstrate that Housing First is a: …critical distal link in a chain of associations between structural and psychological outcome. These links form a chain of relationships between structural and psychological factors that, taken together, suggest that consumer choice-driven models of service delivery may not only have a direct effect on reductions in homelessness and increases in perceptions of choice, but may also have a distal effect on important psychological outcomes such as reductions in psychiatric symptoms (2005: 234). Linear models are based on a relinquishment of control as there are particular requirements for service users in order to remain part of the service, for example sobriety or engagement with psychiatric services. This model, in effect, erodes ‘an important tool for coping with the very circumstances they are intended to redress’ (Greenwood et al., 2005: 234). 11 Workshop 14: Welfare Policy, Homelessness and Social Exclusion Housing retention A number of studies on Housing First interventions have found high rates of housing retention after a number of years amongst programme participants. A study carried out by Einbinder and Tull (2005: 4) of the Beyond Shelter organisation in Los Angeles looked at the situation for 200 previously homeless families seven years after they were moved into permanent shelter by the organisation’s Housing First programme. The study found that the vast majority (89.5 %) of families who were housed through the programme maintained long-term residential stability. Just over half of these families (55%) lived in the same house that they first moved into with help from Beyond Shelter. Those that did move generally did so due to the owner selling the building or a desire for a larger home. Tsemberis et al. (2004) examined the longitudinal effects of a Housing First programme for homeless mentally ill individuals. A total of 225 participants were randomly assigned to receive housing conditional to treatment and sobriety (control group) or to receive immediate housing without treatment prerequisites (experimental group). According to the researchers, their results confirm the effectiveness of using a Housing First approach to engage, house and keep housed individuals who are chronically homeless and dually diagnosed. Those housed immediately without the prerequisite for treatment sustained around an 80 per cent housing retention rate; a rate that the researchers argue presents ‘a profound challenge to clinical assumptions held by many continuum of care supportive housing providers who regard the chronically homeless as “not housing ready”’ (Tsemberis et al., 2004: 654). The results also challenge the long-held clinical assumption that people with mental illness have a lesser ability to maintain a home of their own. Tsemberis et al (2004: 655) argue their findings indicate that: …ACT programmes that combine a consumer-driven philosophy with integrated dual diagnosis treatment based on a harm-reduction approach positively affect residential stability and do not increase substance use or psychiatric symptoms. In another study, Tsemberis and Eisenberg (2000) reported that a sample of people in linear services had a four times greater risk of discontinuous housing than those who were availing of the Pathways to Housing organisation’s Housing First programme. (Quoted in Johnsen and Teixeira, 2010: 8). Clinical outcomes Larimer et al. (2009) studied the level of health care and public service use of chronically homeless persons with severe alcohol problems housed through a Housing First scheme. They found that the length of time in housing was significantly related to reductions in the use and cost of services, with those housed for the longest periods of time experiencing the greatest reductions. They also found that the Housing First participants involved in the study experienced reductions in their alcohol use and the likelihood of drinking to intoxication over time. The researchers found that the ‘HF intervention was associated with substantial declines in drinking despite no requirement to abstain from or reduce drinking to remain housed’ (pg. 1355). Different results in relation to levels of alcohol and drug use were found in a study by Padgett et al. (2006: 79-80). This study looked at data from an experimental longitudinal study (48 months) of 225 adults availing of Housing First or treatment first services. No significant differences were found in the level of drug and alcohol use after the 48 months, despite the fact that the treatment first participants were significantly more likely to use treatment services. Housing First participants have an absence of restrictions on substance use. However, this did not result in a link to increased use of drugs and alcohol. The researchers argue that the results of the study show that: …individuals with severe mental illness and substance use problems do not have to undergo mandatory treatment to be able to live independently in the community. Moreover, consumer ‘Mixité’: an urban and housing issue? 12 driven programmes that practice housing first and harm reduction are not linked to increased substance use despite the absence of restrictions. The current study has also revealed consistent (and probably underreported) use of illicit substances by individuals enrolled in treatment first programmes despite abstinence requirements (2006: 81). However, it is important to note that academics have argued over the effectiveness of Housing First for people with severe and active addiction. Kertesz et al. (2009) ‘assert that the addiction severity of people entering Housing First programmes is in fact “relatively modest”’ (quoted in Johnsen and Teixeira, 2010: 12). Despite this, the studies undertaken to date show positive outcomes for a large majority of the Housing First participants in terms of housing retention and some positive clinical outcomes. This kind of service provision has only been introduced in Ireland on a very small scale to date with the Homeless Agency Housing First project currently engaging with potential participants in Dublin. The project evaluation will hopefully provide some more evidence to add to the US research. The new government has stated that Housing First will become a core element in homeless service delivery. Whether the use of the term Housing First relates to a model with a high level of fidelity to the Pathways to Housing programme as developed in the US is unclear. It could be that the Government is referring to a scheme such as SLI, which is actually quite different from Housing First in that it excludes those with a high level of need and provides reducing levels of time-limited support. Or they could mean quasi-independent supported housing as is outlined the Pathway to Home, with elements of a Housing First approach. Departures from the Housing First model Studies comparing the outcomes of different Housing First programmes indicate that those most closely aligned with the model developed by the Pathways to Housing organisation tend to report the best housing retention rates (Johnsen and Teixeira, 2010: 10). Stefancic and Tsemberis (2007) studied the outcomes of two different Housing First projects and a traditional homeless project (control group). The researchers found that the outcomes in terms of housing retention rates were better for the Housing First project run by an experienced provider (Pathways) than for the one run by a new consortium with no experience. After 47 months, Pathways had a retention rate of 78.3 per cent, while the consortium had a rate of 57 per cent. This may be attributed to the challenges that face new Housing First providers in shifting their perspective from an emphasis on mental health and substance use needs towards a greater appreciation of the housing needs of consumers. According to Stefancic and Tsemberis (2007), the lower retention rates of the consortium suggest that their discharge policies may differ from Pathways in that they may not reflect the practice of separating housing from treatment. It is important that services are provided to Housing First participants even if they lose their home and they should be assisted in finding new housing if they experience difficulties in one building or neighbourhood. This separation of clinical matters from housing matters is core to the success of the service. The results of the study also suggest that the agencies used different approaches to enrolling participants to the programmes and placing them in housing. There was a disparity between the programmes in terms of the ratio of clients housed to those who were initially engaged or outreached. The large number of participants engaged by the consortium may suggest that they were ‘extremely rigorous in their efforts to screen-out ineligible applicants’. It is also possible that clinicians turned down participants who were eligible but whom they did not consider appropriate for placement straight into permanent housing. Despite these rigorous efforts at screening, the fact that the consortium maintained lower retention rates suggests that housing providers and clinicians are not able to successfully predict which potential consumers will be able to successfully maintain a home (Stefancic and Tsemberis, 2007). 13 Workshop 14: Welfare Policy, Homelessness and Social Exclusion Supportive housing model Supportive housing is not one specific programme model, rather it describes a number of programme types and housing arrangements, which often vary according to target group and location and whether the accommodation is permanent or not (Culhane and Byrne, 2010: 9). Many supportive housing models are underpinned by a Housing First approach but not all of them and some may in fact be contingent on acceptance of support services or conditions such as drug and alcohol free housing or psychiatric disability housing (Gordon, 2008 quoted in Johnsen and Teixeira, 2010: 13). There are typical defining features of supportive housing however: …firstly, the provision of safe and secure (typically self-contained and usually permanent) rental housing that is affordable to people on very low incomes; and secondly, the provision of support by staff with appropriate skills and expertise on-site or nearby (Johnsen and Teixeira, 2010: 13). The Family Housing Fund (Minnesota, US) outline a good example of the elements that would be included in a comprehensive supportive housing system for families. The Fund recommended the creation of a system of supportive housing, for the replacement of the transitional housing system in place at the time (1999), combining affordable housing and services, based on the following principles: • • • • • • • • Supportive housing provides affordable housing as the environment in which families receive services. Supportive housing providers have flexibility to determine the families’ length of stay in the programmes. Supportive housing assists families in making the transition to independent living. The supportive housing system includes housing for families who have difficulty complying with the requirements of transitional housing programmes. Supportive housing builds the capacity of parents to nurture and care for their children. Supportive housing encourages productive participation in community and society. Supportive housing includes an array of mental health, academic, social, recreational, and child care services to meet the needs of children. Supportive housing supports sobriety (Hart-Shegos, 1999: 2-3). The ‘Street to Home’ project run by Common Ground is an example of a well-known permanent supportive housing project (Johnsen and Teixeira, 2010: 13). This project, which began in New York and has since spread to many other parts of the US, focuses on the provision of housing for those who have been living on the street the longest and are considered to need housing the most. The programme replaces a random ‘first come, first served’ approach to homeless services with a targeted, strategic process in which the most vulnerable individuals on the street are identified and prioritised. Housing options are assessed and negotiated with these individuals who are then housed and supported in order to retain the housing. The Street to Home project reduced street homelessness in Times Square New York from an average of 55 people sleeping rough to seven over two years (2005-2007). Before the introduction of the programme, the area had one of the highest densities of street homelessness in the city (see www.commonground.org/?page_id=21, accessed 21/04/2011). The project is yet to be independently evaluated, ‘hence its effectiveness in terms of housing retention and other complex support needs has not been tested fully’ (Johnsen and Teixeira, 2010: 14). Does any model work best? In a number of countries, services are being developed for homeless people with a high level of need that would be categorised as ‘normalisation’ services. For example, Housing First projects (albeit ‘Mixité’: an urban and housing issue? 14 with some deviations) have been developed in France, Portugal, the US, and Ireland. Much of the research to date has come from the US rather than the European projects. However, the new Homeless Agency Housing First project will be evaluated from the outset and will provide some information on how the model works in an Irish context. Figure 4: Shift in homeless policy approaches Source: Adapted from Edgar in Frazer et al (2010) by Busch-Geertsema et al. (2011), pg. 45. A lack of rigorous research on outcomes makes it difficult to evaluate effectiveness of transitional housing as a means of addressing homelessness. According to Barrow and Zimmer, (1999: 4, quoted in Novac et al., 2009: 19-20), assessing the ‘effectiveness of transitional housing requires designs that control for other factors that may influence outcomes while comparing transitional housing programs to policy-relevant alternatives’. Research is also lacking on the long-term effects of transitional housing, that is, whether people maintain their housing over the long term and improve their financial situation or achieve financial independence. It is likely that linear approaches work best for people who are willing to engage with rehabilitation programmes and are well able to cope with shared housing arrangements (Tainio and Fredrikkson, 2009 quoted in Johnsen and Teixeira, 2010). However, shared accommodation as a transitional strategy to address homelessness should, according to Tremblay (2009: 29): …be re-examined and reconsidered since it may actually be counterproductive and may be alienating some individuals. Consideration should be given to more successful models of shared living, including development of small scale (i.e. 6-7 people) shelter/transitional housing, who may or may not work and learn together in order to create a shared sense of community which can then be transposed to a more independent living situation. The general evidence base for transitional housing effectiveness is actually quite weak, especially for people with a high level of need. Many of the critiques of linear approaches have focused on the high rate of attrition between the various stages of the ‘staircase’ (Johnsen and Teixeira, 2010: 5). 15 Workshop 14: Welfare Policy, Homelessness and Social Exclusion The traditional transitional approach to homeless services is quite linear in nature, unlike recovery from addiction which can be more cyclical and can include relapse. This means that the state of homelessness for those with a high level of need such as those addicted to drugs and/or alcohol, and the policies in place to alleviate it, do not always fit together well. A type of homelessness policy that acknowledges the cyclical nature of recovery is more likely to have a positive effect on service users. Unfortunately, many transitional services have prohibitory policies and practices which can result in a person being ejected from a service (and going back down a stage of service provision, for example, back to emergency accommodation or sleeping rough) if they experience the relapse stage of change. Housing First, on the other hand, practices harm reduction and therefore acknowledges relapse as a realistic stage in the recovery process. Although research is limited, there is some evidence that resource intensive linear models of service provision can have limited effectiveness, in comparison to lower intensity and lower cost Housing First models that use floating support services and emphasise service user choice, in providing sustainable exits from homelessness (Busch-Geertsema et al., 2011: 5). As stated previously, Housing First type provision is becoming more popular in Europe in recent years. However, the Pathways to Housing organisation in the US who developed the model, have argued that much of the policy being adopted is only including elements of Housing First, rather than adopting the whole model. The organisation argues that the services which are closest to the model which they have developed will get the best outcomes. They are currently devising a ‘Housing First Fidelity Scale’ against which policies can be measured to see how closely they align with the original model. Where do homeless services in Ireland fit in relation to the various models of provision? In Dublin there has been a move away from the linear model of service provision with the closure of many transitional services towards a model of normalisation, consisting of temporary emergency accommodation and supported housing. However, the requirement of the need for ‘housing readiness’ is still evident in some of the new services, for example, STA is to include dedicated intervention programmes that can ‘stabilise’ a person’s drug or alcohol lifestyle. This implies a need for treatment and stabilisation for ‘housing readiness’, which is associated with the linear model of service provision. However, the Pathway to Home plan does not state whether failure to comply with the programme will result in ejection from the accommodation or if it will affect whether the person can move on to more independent housing. Although the movement has been away from transitional service provision, the plan to replace the older services does not include a Housing First programme. Under the plan, those with a high level of need will have ‘quasi-independent’ housing with on-site support as opposed to independent housing with support. This is contrary to the recent announcement by the government in their Programme for Government 2011 that a Housing First approach will be used in order to tackle long term homelessness. Therefore it is likely that Housing First will have a more prominent role in homeless service provision in the future. No detail is available as to whether this Housing First approach will be an extension of the Homeless Agency Housing First project to other parts of the country or whether it will be in the form of a Housing First type project devised by the government. The Programme for Government does not include details on the Government’s understanding of the term Housing First and whether they are referring to the programme as devised by the Pathway to Housing organisation in the US or services that include elements of Housing First, without strictly adhering to the US model. The development of the Homeless Agency Housing First project is to be welcomed as it allows for the assessment of the model in the Irish context. However, the project is in the very early stages and it will take time before any outcomes can be evaluated. It is currently a small project (a maximum of 23 participants in the first year) so it will be difficult to robustly assess its impacts until it is extended further. The Housing First approach may provide an opportunity to achieve the goal of ending rough sleeping and long-term homelessness, as the project deals directly with the section of the homeless ‘Mixité’: an urban and housing issue? 16 population who are the hardest to engage in services. It is important to ensure that the cohort of people who will benefit from this project are catered for sufficiently in the time between the extension of the project and the closure of transitional housing projects, as many transitional projects have already closed before the Housing First project has been developed on a wide scale. In trying to place Ireland within a spectrum of homelessness service types, it is evident that homelessness policy is still in a process of transition; moving further from the linear model of service provision towards a model of normalisation. In Dublin, the closure or reconfiguration of transitional services and the development of housing with supports and a Housing First project, albeit on a small scale, is a clear indicator of this. This transition is likely to take place outside of the capital in the future with the government’s announcement on the adoption of a Housing First approach to homeless services. Until these changes are implemented outside of Dublin, there are two-tiers of homelessness services in place in Ireland – with the model of normalisation dominating in Dublin and a more linear model still dominating outside of the capital. 17 Figure 5. Workshop 14: Welfare Policy, Homelessness and Social Exclusion ‘Mixité’: an urban and housing issue? 18 Discussion and conclusions 1. The changes proposed in the Pathway to Home document describe a model of long-term supported residential accommodation for those with a high level of need, rather than the more independent Housing First model. This does not fit with the new government’s statement in their Programme for Government 2011 that a Housing First approach will be used to tackle long-term homeless. Housing First is not mentioned in the Pathway to Home document at all. The Homeless Agency has, however, recently launched a small scale pilot Housing First programme. This project will be evaluated from the outset and could, if successful, potentially work as a model for a service to be rolled out throughout the country for entrenched homeless people. 2. Changes to homelessness policy have not been limited to Ireland. Many countries have developed homelessness policies that move away from traditional transitional approaches towards policies more closely aligned with a Housing First approach, for example, in parts of France, Portugal and the US. Research from the US has demonstrated encouraging evidence in support of the Housing First model. However, the research shows that in order to have the highest level of impact in terms of housing retention, Housing First projects need to adhere to the principles of the programme as it has been developed by the Pathways to Housing organisation in the US. The higher the fidelity to the Pathways programme, the better the outcome. 3. It is important to note that in many instances the linear or normalisation models do not operate exclusively of each other within countries. For example, in the US, there are a number of cities, or pockets within the cities, that operate a Housing First policy. However, the linear model still dominates in many other parts of the country. The same applies for Ireland in that a normalisation model now dominates in the Dublin region, with a linear model playing a much larger role in the areas outside of Dublin. 4. It is important that organisations which provide Housing First services avoid risk aversion. Research has shown that professionals are not good at deciding who will succeed in independent housing, or in other words are ‘housing ready’. Rather, Housing First should be administered on a ‘first come, first served’ basis, irrelevant of whether the service provider considers the client ready for housing. 5. Much of the debate around models of homelessness service provision really only relates to people with medium to high level needs, or the most entrenched homeless and rough sleepers. Many people with a lower level of need are simply in need of housing, possibly with a low level of visiting support. The outlined services in the reconfiguration plan for Dublin should be sufficient to meet their needs once the required level of housing and support is provided. 6. With the employment freeze in Irish health services (HSE) and other budget cuts in health, an important consideration is whether the health care teams necessary to deliver a country-wide Housing First programme would be funded? It is crucial that the advocates of such a programme illustrate to the necessary bodies and Departments the returns in terms of cost savings through Housing First, for example, cost savings made on lower emergency hospital usage. The fact that this service has been shown to cost less in the immediate and long term, than traditional transitional services, means that new services should in fact save money for the exchequer, rather than demand more from it (for information on cost savings see Stefancic and Tsemberis, 2007, Gulcur et al., 2003 and Larimer et al., 2009). 7. SLI is to be funded through cost saving made on the closure of emergency and transitional housing units. However, many of the people who availed of these services may have had a higher level of need than is permissible for eligibility for the SLI scheme. Some of the cost 19 Workshop 14: Welfare Pol savings could be redirected towards a Housing First programme in order to cater for those with a high level of need who can no longer avail of transitional housing and are ineligible for SLI. 8. Many of the restructured services are based on time-limited support structures. However, homelessness can be a cycle rather than a state; from which people can enter and exit a number of different times throughout their homeless ‘career’. It is important that policy reflects this so that people can avail of support services again after their initial period of support if it is necessary to prevent a re-entry to homelessness. For example, SLI time-limited support can be reinstated if it is considered possible that a tenancy might breakdown without the support. However, this is only in ‘exceptional circumstance’. References Anderson, I. (2010) ‘Services for Homeless People in Europe: Supporting Pathways out of Homelessness?’ in O’Sullivan, E., V. Busch-Geertsema, D. Quilgars and N. Pleace, Homelessness Research in Europe, FEANTSA, Brussels. Benjaminsen, L., N. Wilhelm Benn, A. Holm (2008) ‘The Effects of Housing Interventions for the Homeless: Targeting of interventions and selection effects’, The Danish National Center for Social Research, Copenhagen. Burt, M. R. (2006) ‘Characteristics of Transitional Housing for Homeless Families’, Urban Institute, Washington. Busch-Geertsema, V., W. Edgar, E. O’Sullivan and N. Pleace (2011) Homelessness and Homeless Policies in Europe: Lessons from Research, FEANTSA, Brussels. Culhane, D. P. and T. Byrne (2010) Ending Chronic Homelessness: Cost-Effective Opportunities for Interagency Collaboration, University of Pensilvania, available from: http://works.bepress.com/denis_culhane/94, accessed 13/05/2011. The Department of Environment, Heritage and Local Government (15/02/2011) Minister Finneran Speech ‘Minister Finneran welcomes milestones reached for homeless people on “The Way Home”’, available from: http://www.environ.ie/en/DevelopmentandHousing/Housing/SpecialNeeds/HomelessPeople/News/Main Body,25412,en.htm, accessed 13/05/2011. The Department of Environment, Heritage and Local Government (2009) Housing Needs Assessment, available from: http://www.environ.ie/en/Publications/StatisticsandRegularPublications/HousingStatistics/, accessed 13/05/2011. The Department of Environment, Heritage and Local Government (31st July 2009) ‘Circular HU 03/2009 SLI (Support to Live Independently) – new scheme of housing and support to progress from homelessness’. The Department of the Environment, Heritage and Local Government (2009) Homeless Strategy: National Implementation Plan, Stationery Office, Dublin. The Department of the Environment, Heritage and Local Government (2008) The Way Home: A Strategy to End Adult Homelessness in Ireland 2008-2013, Stationery Office, Dublin. Einbinder, S. D. and T. Tull (2005) ‘The Housing First Program for Homeless Families: Empirical Evidence of Long-term Efficacy to End and Prevent Family Homelessness’, Beyond Shelter, California. ‘Mixité’: an urban and housing issue? 20 FEANTSA (2011) ‘FEANTSA Flash March 2011’, available from: http://feantsa.horus.be/code/EN/pg.asp?Page=1358, accessed 28/04/2011. FEANTSA (2008) ‘Homeless in Europe’, Magazine of FEANTSA Winter 2008, available from: http://www.feantsa.org/files/freshstart/Communications/Homeless%20in%20Europe%20EN/PDF_2009/ Homeless%20in%20Europe_Winter2008.pdf, accessed 28/04/2011. Focus Ireland (2nd March 2011) ‘Conference on Youth Homelessness: Breaking the Cycle’, Dublin Castle. Information available from: http://www.focusireland.ie/files/docs/advocacy/breaking%20the%20cycle.pdf, accessed 28/04/2011. Greenwood, R. M., N. J. Schaefer-McDaniel, G. Winkel and S. Tsemberis (2005) ‘Decreasing Psychiatric Symptoms by Increasing Choice in Services for Adults with Histories of Homelssness’ in American Journal of Community Psychology, Vol. 36, Nos. 3/4, pg. 223-238. Gulcur, L., A. Stefancic, M. Shinn, S. Tsemberis and S. Fischer (2003) ‘Housing, Hospitalization, and Cost Outcomes for Homeless Individuals with Psychiatric Disabilities Participating in Continuum of Care and Housing First Programmes’ in Journal of Community and Applied Social Psychology, Vol. 13, pg. 171186. Hart-Shegos, E. (1999) The Supportive Housing Continuum: A Model for Housing Homeless Families, Family Housing Fund, Minnesota. Harvey ,B. (1999) ‘Models of resettlement for the homeless in the European Union’ in Marsh, A. and P. Kennett, Homeless: Exploring the New Terrain, Policy Press, Bristol, pg. 267-292. The Homeless Agency Partnership (2010) Pathway to Home: New Configuration of Homeless Services in Dublin, The Homeless Agency Partnership, Dublin. The Homeless Agency Partnership (2009) Pathway to Home: On implementing the National Homeless Strategy. The Way Home 2008-2013 and realising the 2010 Vision of the Homeless Agency Partnership’s action plan on homelessness in Dublin, A key to the Door 2007-2010, The Homeless Agency Partnership, Dublin. The Homeless Agency (2008) Counted In, The Homeless Agency, Dublin. Johnsen, S. And L. Teixeira (2010) Staircase, Elevators and Cycles of Change: ‘Housing First’ and Other Housing Models for Homeless People with Complex Support Needs, The University of York and Crisis, London. Larimer, M, D. Malone, M. Garner, D. Atkins, B. Burlingham, H. Lonczack, K. Tanzer, J. Ginzler, S. L. Clifasefi, W. G. Hobson and G. A. Marlatt (2009) ‘Health Care and Public Service Use and Costs Before and After Provision of Housing for Chronically Homeless Persons with Severe Alcohol Problems’ in Journal of the American Medical Association, Vol. 301, No. 13, pg. 1350-1357. Novac, S., J. Brown and C. Bourbonnais (2009) ‘Transitional Housing Models in Canada: Options and Outcomes. In Hulchanski, J.D., P. Campsie, S. Chau, S. Hwang and E. Paradis (eds.), Finding Home: Policy Options for Addressing Homelessness in Canada, (e-book), Toronto Cities Centre, University of Toronto, available from: www.homelesshub.ca/FindingHome, accessed 13/05/2011. Padgett, D. K., L. Gulcur and S. Tsemberis (2006) ‘Housing First Services for People Who are Homeless with Co-occurring Serious Mental health Illness and Substance Abuse’ in Research on Social Work Practice, Vol. 16, No. 1, pg. 74-83. Sahlin, I. (2003) ‘The Staircase of Transition: Survival through Failure’, paper for presentation at the ICCR seminar on Housing and Social Inclusion, Brussels, 16-17 January 2003. 21 Workshop 14: Welfare Pol Stefancic, A. and S. Tsemberis (2007) ‘Housing First for long-term shelter dwellers with psychiatric disabilities in a suburban county: a four-year study of housing access and retention’ in Journal of Primary Prevention, Vol. 28 265-279. Tremblay, J (2009) Keeping the Homeless Housed: An exploratory study of determinants of Homelessness in the Toronto community, Wellesley Istitute, Toronto. Tsemberis, S., L. Gulcur and M. Nakae (2004) ‘Housing First, Consumer Choice, and Harm Reduction for Homeless Individuals with Dual Diagnosis’, in American Journal of Public Health, Vol. 94, No. 4, pg. 651-656. The Vancouver Sun (28th August 2008) ‘Does ‘Housing First’ model make sense?, available from www.canada.com, accessed 10/02/2011. Websites Common Ground: www.commonground.org/?page_id=21, accessed 27/04/2011. Word Press: www.datadesign.wordpress.com/2011/01/14/maslows-pyramid/, accessed 13/05/2011. Homeless Agency: www.homelessagency.ie/getdoc/2beb2e84-7fe5-42dd-930c-6638ef9c242a/EnhancedProgramme-for-Long-Term-Provision.aspx, accessed 13/05/2011. Irish Council for Social Housing: www.icsh.ie/eng/policy/homelessness/homelessness_statistics_in_ireland, accessed 04/05/2011. MakeRoom Campaign: www.makeroom.ie/download/pdf/2010_homeless_numbers_memo_final.pdf, accessed 04/05/2011. Pathways to Housing: see www.pathwaystohousing.org/content/research_library, accessed 13/04/2011. Texas A&M University: www.studentlife.tamu.edu/adep/addiction_cycle, accessed 27/04/2010.