‘It maybe doesn’t seem much, but to me it’s my kingdom’: staff and client experiences of Housing First in Scotland

Abstract

Housing First (HF) represents a significant shift in the way that the problem of homelessness and co-occuring challenges including problem substance use, is addressed. HF interventions have been the focus of much research. Quantitative studies have consistently shown positive findings regarding housing outcomes, with results regarding health and well-being outcomes more mixed. To date, limited attention has been paid to the experiences and perspectives of HF service providers, and few studies have explored the views of those HF recipients. In enabling providers and recipients to share their professional and personal experiences of HF, qualitative insights can help inform, and improve, service provision and practice. Semi-structured interviews were conducted with seven HF staff members and 11 clients in a single third sector service in Scotland. Overall, clients experienced HF positively and described how involvement in HF had enabled positive changes in their lives. Service providers reported positive views on HF alongside ways to maximize the effectiveness of the model. While our findings provide support for current efforts to promote HF as an approach to help end homelessness, a number of challenges exist. To address these, we propose a set of recommendations for those planning and implementing HF services.

Keywords:

• Housing First
• homelessness
• substance use
• qualitative research
• public health

Background

Homelessness is a complex term encompassing a range of housing situations including both sheltered (e.g. temporary accommodation) and unsheltered settings such as the streets (Baxter et al., 2019; Kertesz & Johnson, 2017). While homelessness does not have a uniform definition, a key criterion is the lack of ‘access to minimally adequate housing’ (Busch-Geertsema et al., 2016, p. 125). Recent estimates suggest that homelessness in the United Kingdom (UK) is increasing (Shelter, 2019). Figures indicate that 307,000 people in the UK (Shelter, 2017), 567,715 in the USA (National Alliance to End Homeless, 2018) and 235,000 in Canada (Gaetz et al., 2016) experience homelessness in a year, however it is likely that the true scale of the problem is under-reported due to variations in definitions.

Those who experience homelessness typically encounter some of the most complex and intersecting health and social challenges within society. These include being at greater risk of suffering from poorer physical health, including infectious diseases (Beijer et al., 2012), as well as using alcohol, drugs and tobacco (Fazel et al., 2008), and being at increased risk of premature fraility and ageing (Rogans-Watson et al., 2020) and death (Bean et al., 2013; Kerman et al., 2020). Mortality rates among people who are homeless have been estimated to be between three to four times higher than the general population (Fazel et al., 2014; O’Connell, 2005). Moreover, those who experience homelessness tend to report disproportionately high rates of co-occurring problem substance use, poor mental health and physical health (Hewett & Halligan, 2010; Levitt et al., 2009). In addition to being disproportionately affected by health inequalities (Stafford & Wood, 2017), people who are homeless tend to experience numerous social challenges, ‘deep social exclusion’, and are highly stigmatized (Johnstone et al., 2015). Such inequalities can result in social isolation, feelings of worthlessness, loneliness and depression (Sanders & Brown, 2015). Since the mid-1990s, concerns about the vulnerability of this group, alongside high levels of public service utilization, such as ambulance service callouts and Accident and Emergency attendances, have led to increased efforts to tackle homelessness (Stahl et al., 2016). Recognition that individuals are entitled to access to a safe home, and that adequate housing is vital to health and well-being, is growing internationally (Shelter, 2019), with a particular focus on meeting the needs of people who are ‘chronically’ homeless, and experience concurrent problem substance use and/or poor mental health (Burt et al., 2004; Larimer et al., 2009).

As a response to ending homelessness and improving housing stability, and because many services were not meeting the needs of those considered to be ‘difficult to engage’ (Making Every Adult Matter, 2009), Housing First (HF) emerged in the early 1990s with the introduction of the Pathways to Housing program in New York (Stergiopoulos et al., 2014). The ‘HF’ approach aims to assist clients, often with experience of serious mental health problems, and/or problem substance use, to access permanent housing as an initial step to addressing homelessness, with housing provision not contingent on compliance with health treatment or substance abstinence, and with the additional offer of ongoing support (Baxter et al., 2019). There are two types of housing typically offered by HF: the scattered-site approach offers a choice of individual housing units in the larger community; and a single-site/congregate approach provides individual units (e.g. private studio apartments) within a single housing project (Stahl et al., 2016). What accompanies the housing in HF programs varies, with some offering unfurnished housing with arrangements to secure furniture/household essentials; and others providing personal budgets to clients to buy these. A minority of HF programs offer fully furnished apartments (Breatherton & Pleace, 2015).

HF is built on four key principles: 1) immediate provision of housing and consumer-driven services; 2) separation of housing and clinical services; 3) providing supports and treatment with a recovery orientation; and 4) facilitation of community integration (Tsemberis, 2010). Consumer choice is the core value that drives provision of both housing and support services, with clients being encouraged to select the type of housing and neighborhood and the type, sequence, and intensity of services that best meets their needs, aiming to encourage active participation (Aubry et al., 2015). It is important to recognize, however, that in order to deliver the consumer choice, and succeed as an intervention, HF requires a sufficient supply of (adequate/desirable) housing, which can be difficult to achieve in practice due to a lack of housing, especially in the UK (e.g. Wilson & Barton, 2020; Wilson & Loft, 2021). It has therefore been suggested that tenancy provision may need to expand into private rental and not rely solely on the State (Parkinson & Parsell, 2018). The provision of HF represents a significant shift in the way that homelessness is addressed, providing a ‘low-barrier’ approach which differs from more traditional ‘continuum of care’ or Treatment First (TF) approaches that require individuals to meet certain criteria before qualifying for permanent supportive housing (Stergiopoulos et al., 2015). Instead, HF conceptualizes housing as a human right (Kertesz & Johnson, 2017), with support services combining both recovery orientation and harm reduction approaches (Gilmer et al., 2013; Raitakari & Juhila, 2015; Tsemberis, 2010), thus any treatment for problem substance use is optional (Somers et al., 2013). HF also differs from TF with its focus on individuals with ‘multiple and complex needs’ (Bramley et al., 2019) including mental ill health, co-occurring problem substance use, and chronic homelessness (Kennedy et al., 2016). These differences have led to HF being described as ‘innovative’ (Volk et al., 2014, p. 78).

Treatment engagement and retention for problem substance use and/or mental health problems whilst homeless can be problematic (Miler et al., 2021) and a lack of stability when living in temporary housing can negatively impact well-being (Aubry et al., 2019). The provision of safe and consistent housing is critical in enabling individuals experiencing both problem substance use (Martinelli et al., 2020; Pauly et al., 2013) and poor mental health (Kerman et al., 2018; Kirst et al., 2014), to recover from these challenges. HF aims to actively address such issues (Aubry et al., 2015). By providing immediate permanent housing, HF provides an opportunity to offer more consistent support services than TF, with evidence suggesting that individuals are less likely to remain in one area for an extended period of time when housed in temporary accommodations (Holmes et al., 2017). While individuals in TF programs may be receiving support, the HF concept of ‘permanent supportive housing’ specifically includes provision of support. This typically includes either Assertive Community Treatment (ACT), which uses multi-disciplinary staffed teams with shared caseloads to provide a full range of direct services to clients (McGrew & Bond, 1995), thus providing support directly through specialists (e.g. psychiatrists) on the team; or Intensive Case Management (ICM), which provides similar support using an individual case management model where participants are refered outside of the team (Tsemberis, 2010). ICM encompasses a range of service delivery practices that are less intensive and not as standardized as ACT (e.g. Schaedle et al., 2002), and involves assertive outreach, assessment of client need, and negotiation and coordination of care (but no direct care/treatment provision).

HF has become an internationally promoted community housing model for people who experience chronic homelessness and other complex needs (Greenwood et al., 2013; Johnsen & Teixeira, 2012; Johnson, 2012; Parsell et al., 2014; Tsemberis et al., 2004), receiving support from the United States Department of Housing and Urban Development (Pearson et al., 2009) and extensive attention in Canada with the At Home/Chez Soi HF program run as a randomized controlled trial in five major cities (Aubry et al., 2015). There are a number of established HF projects in North America and Scandinavia, and growing interest in other countries including the UK (Baxter et al., 2019), with a recent report from England urging the government to scale-up the program (The Centre for Social Justice, 2021). In Scotland, a HF pilot was developed in 2010 in response to high levels of repeat homelessness amongst people with problem substance use in Glasgow. Run by Turning Point Scotland, this was the first HF project to be developed in the UK, and one of the first internationally to explicitly focus on people who were homeless with problem substance use, and accommodate clients in independent self-contained apartments on a scatter-site basis (Johnsen, 2014). Since then, five areas across Scotland have established Scotland’s HF Pathfinders, working to expand HF provision on a local authority-wide basis (Homeless Network Scotland, 2020), with support from the Scottish Parliament and funding from Scottish Government and charities (Homeless Network Scotland, 2020). It is important to note that in Scotland only the ICM model is currently commissioned and utilized, and only scattered-site HF housing is available: congregate HF services are not presently an option.

Over the same period, HF interventions have become the focus of increasing (primarily quantitative) research interest, related to measuring outcomes, including: program participation, housing placement, residential stability, program retention, service use/cost, changes in substance use and psychiatric symptoms, and Quality of Life (QoL) (e.g. Desilva et al., 2011; Patterson et al., 2013; Tsemberis et al., 2004). Research consistently shows positive findings regarding housing outcomes but mixed results regarding health and well-being outcomes (Baxter et al., 2019). The recent systematic review by Baxter et al. (2019) found that HF resulted in unclear short-term impact on health and well-being outcomes. For substance use, mental health, and QoL, no clear differences were seen when comparing HF and treatment as usual (TAU) clients. However, the results indicated a clear reduction in non-routine use of healthcare services, for HF over TAU, which may be an indicator of improvements in health. Moreover, compared to TF approaches, individuals who use HF services appear to have higher rates of housing stability and retention (Desilva et al., 2011) which has been shown to improve QoL (Chung et al., 2018) and mental health/well-being (Hainstock & Masuda, 2019). Overall, many studies have shown that HF programs are useful for participants (e.g. Aubry et al., 2016; Nelson et al., 2015; Stergiopoulos et al., 2015; West et al., 2014), and highlighted that participants are generally more satisfied with HF housing conditions. It has also been reported that individuals housed in HF facilities, compared to those in TF approaches, are less likely to be hospitalized due to poor mental health (Kerman et al., 2018), or to use emergency services (Kerman et al., 2020; Mackelprang et al., 2014). HF has been demonstrated to decrease incidences of suicidal ideation (Collins et al., 2016) and improve medication adherence (Driscoll et al., 2018). Moreover, mortality rates of individuals in HF programs appear to be lower than those who are not using any service (Henwood et al., 2015). While we see these as being important health outcomes, and also focus on these when discussing findings from our study, it should be noted that housing is itself a social determinant of health (e.g. World Health Organization, 2018), with the physical benefits of housing (e.g. living in a safe environment), and other less tangible benefits including attachment to a ‘home’, well-recognised (Rolfe et al., 2020).

Qualitative research affords affected individuals the opportunity to contribute to program and policy discussions (Stahl et al., 2016) and provides nuanced information to improve programs (Weir et al., 2007). To date there has been limited attention given to the experiences/perspectives of HF service providers who are integral to implementation. There is also a substantive gap in relation to exploring the experience of those receiving HF, with noted recent exceptions (see Sandu et al., 2021). Qualitative research undertaken to explore service providers experiences of HF suggests that it is more effective in housing people than TF services (Kennedy et al., 2016), with access to stable housing, availability of ongoing support, community integration, and the opportunity to build sustained relationships with clients, highlighted as strengths. Despite these positive experiences, interviews with providers (Kennedy et al., 2016) also revealed implementation challenges including lack of housing stock and housing options, unwillingness of some landlords and property owners to participate in programs, as well as a range of challenges experienced by HF participants. Similarly, the mixed-methods evaluation of the At Home/Chez Soi project reported views concerning the significant challenges of finding good quality, affordable housing in areas in which participants desired to live, with many tenancies being explicitly smoke- and pet-free, further limiting clients’ options (Macnaughton et al., 2015).

Similarly, qualitative studies conducted with clients found that they are broadly satisfied with HF: most residents in a congregate HF program described their dwellings as ‘homes’, and their environment as a ‘community’ (Parsell et al., 2015). A community environment and the stability offered by another congregate HF program were praised by clients, but the need for adequate privacy was flagged (Stahl et al., 2016). In another qualitative study, clients described the benefits of simply having access to an independent apartment, stating that HF helped them to improve their health and praising the support from staff with whom they had developed trusting relationships (Kennedy et al., 2016). Although most qualitative studies have reported positive findings about HF programs, Brown, Malone and Jordan’s (Brown et al., 2015) study reported mixed results. Alongside positive themes, such as proximity to amenities and availability of support staff onsite the congregate HF program, negative themes including the presence of drugs, high levels of crime, and lack of privacy were noted. In particular, the presence of drugs in this single-site HF program and neighborhood emerged as a negative theme (e.g. ‘drug activity’, ‘too many drug dealers’) among over a third of intervewees (33.3%), suggesting this to be a notable problem. Brown et al. (2015) concluded that maintaining adherence to the harm reduction philosophy, while minimizing the consequences of drug activity in a setting with a high concentration of individuals with substance use disorders, is an inherent challenge in implementing single-site HF programs. They suggested that HF programs should develop strategies for effectively managing drug dealing and other criminal activity occurring near the building. However, there is some evidence suggesting that individuals in single-site HF support one another by sharing resources to ward off life-threatening alcohol withdrawal symptoms (Stahl et al., 2016), and that individuals struggling with problem substance use in scattered-site housing report feelings of isolation (Parsell et al., 2015). More research directly comparing the two approaches is needed.

Our study was designed to address this gap in understanding the experiences and views of those providing and receiving HF tenancies and aimed to generate information to improve the service offered to clients and support offered to staff. It focused on a scattered HF program in central Scotland and was designed by two researchers (JM, TP), in partnership with a service manager (HM) to address three research questions using a qualitative interpretive study design (Ames et al., 2011; Thorne et al., 1997):

1. how do clients experience the HF service, and how does it compare to previous support/services received?

2. how does HF compare with clients’ initial expectations?

3. how do staff experience providing HF?

At the time of data collection, the service had enrolled 72 clients, all of whom were defined as homeless, including living in emergency/temporary accommodations, a small proportion coming directly from prison release, sleeping rough or living in unstable living conditions, with 83% of clients reporting having support needs in the areas of mental health, physical health, problem substance use, and criminal justice, including being on probation. Typically in this relatively small sample, the criminal justice involvement preceded or was linked to pre-existing problem substance use and/or mental health issues, such as for example legal problems relating to drug dealing. Overall the clients were largely homogenous in terms of the complexity of need, with most of the clients requiring support with problem substance use and/or mental health. In accordance with the HF principles, substance use is permitted and not penalised among the HF clients, and there is no requirement of treatment enrollment. The ethos of the service is to provide safety, which predominantly ties in with the harm reduction approach. Clients have a choice regarding any engagement in treatment and in type of treatment, with a vast majority choosing harm reduction services, including opioid substitution therapy (OST), but some clients also have specialist addiction workers. The organization provides the HF service in one site only, and the program was initially commissioned for two years. Clients are signposted to other services during the two-year tenancy to prepare them to lead independent lives (Srebnik et al., 2013).

Finally, it is important to highlight that, in contrast to many homelessness services internationally, staff working in homelessness services in Scotland (including but not limited to HF) do not typically come from a social work background. While they are required to possess health and social care qualifications, they do not have the same formalized training social workers receive (Galbraith, 2020). This is likely to shape their experience of delivering a frontline service to individuals presenting with complex needs.

Methods

Study design, sampling and recruitment

A qualitative interpretive study design (Ames et al., 2011; Thorne et al., 1997) was utilized, focused on one HF service; this was the first HF service run by the provider organization and thus created opportunities for learning that could be implemented for the benefit of clients, staff and the wider organization. Semi-structured one-to-one interviews were conducted with seven HF staff and 11 service users (2 female, 9 male; age range: 32−60; all White British/Scottish) by one researcher (JM). Support and mentorship was provided by another researcher (TP). Purposive sampling was used to identify provider participants based on gender, role, and previous service provision/work experience. Gender balance was considered and a mixed gender sample achieved (details not provided to protect anonymity). Staff were recruited via email by JM once a list of all staff and contact details were provided by the service manager. Service users (described as ‘clients’) were recruited via a two-stage process. A purposive approach to sampling was used to include individuals with a range of past homelessness experiences (e.g. rough sleeping/living in hostels) prior to moving to HF, and to ensure that people who identified as both women and men were represented. All clients who had been moved to their HF tenancy at least three months prior to data collection were given a letter by a staff member to inform them about the study. While the aim was to represent diversity of background, age and gender, given the small participant pool it was collaboratively agreed to interview all clients who expressed interest. Prospective participants were then provided with a participant information sheet to confirm interest.

Written informed consent was obtained at the beginning of each interview. The researcher took care to establish rapport with participants and was mindful of power imbalances (Mason, 2002). All interviews took place in a third sector service between December 2019-January 2020. The interview schedules (developed by researchers) are included in Supplementary File 1. Interviews were recorded and transcribed verbatim.

Data analysis

Data were analysed using Framework (Ritchie & Lewis, 2003) in NVivo (Version 12). Transcripts were read closely and coded line-by-line (JM) in two separate NVivo ‘projects’, one for clients and one for staff. A coding framework was developed for each of the projects after coding an initial selection of transcripts (three and six transcripts for staff and client frameworks respectively), chosen for the richness and variety of data in participant responses. This coding framework was discussed in detail with a second researcher (TP) and refined. These codes were then applied to the remainder of the data but adapted and expanded as new themes were identified. Coding was both iterative and deductive, with initial themes developing from the data (in-vivo) and active reference to the research questions. The lead researcher (JM) undertook coding of the initial three staff transcripts and another member of the research team (KH) coded the remainder. The final coding framework (Supplementary File 2) was discussed among all authors (excluding HM) and finalized.

Ethics approval

The University of Stirling General University Ethics Panel (GUEP 749) and the ethics committee of the participating HF provider provided ethical approval. After each interview participants were provided with a debrief sheet signposting them to support organizations, and client participants were provided with a £10 shopping voucher to recognize their time and contribution.

Results

Findings from clients are presented first, followed by findings from staff. All illustrative quotations retain participants’ dialect; a glossary is provided to guide readers less familiar with the Scottish dialect and colloquializms used.

Client experiences

Awareness of Housing First and previous service engagement

Despite the growing presence of HF internationally, knowledge about HF varied among clients and was sometimes limited. Clients had typically heard about HF from support workers in other homelessness services, rather than from peers. Perhaps connected to the limited knowledge of HF and what it encompassed, some clients were unaware of the package of support that accompanied HF, and generally had limited expectations:

When they gae me all this I didn’t expect the help that they’ve gae me, but it’s actually mare than what I expected, and mare than what I can ask for. I thought… you’d get your hoose… and then you’d have to fend fir yoursel. But no, it’s all there as soon as you walk in. It’s there, and you don’t need tae ask fir nothin.

(HF Client 3, male)

Alternatively, some clients saw the potential of HF, or were simply desperate for any form of additional support. Some described wanting to escape the streets because of the discomfort of being continually surrounded by drug use. Some saw HF as an opportunity to be safe. Some had been unable able to receive help elsewhere.

Client accounts revealed different prior experiences of homelessness, and interactions with homelessness services, as well of experiences of compounding challenges. Some described ill health with problems accessing health care and receiving medication. Some spoke about being let down by individuals and services at various junctures in their lives, leading them to stop trusting professionals. This did not seem to impact on their overall experience of HF but did explain why, for some, building trust with staff took time:

Aye, aye, but it’s all about building trust. Because I don’t trust nobody and that’s what I’ve been taught on the street and learned, ‘never trust anybody’. And you don’t get used to people helping you and gaeing you stuff because you are like ‘what do they want?’, ‘What are they after?’ They don’t just gae you that for nothing.

(HF Client 9, male)

Although a small number of clients had not received any support from other services previously, the majority of clients had, and had experienced difficulties. Hostels, in particular, were criticized and even compared to prisons, with their staff described as unsupportive, unfriendly, aggressive, and treating residents as though they were children. There was a general sense of having negative experiences with temporary accommodation.

Views towards Housing First

Many clients highlighted that HF differed positively from other services. For example, they described staff as being more caring with more support available. Interviewees expressed that the receipt of housing through HF was accelerated when compared with other services, with better engagement and communication. Regarding the ‘HF package’, a number of clients commented that receiving a fully furnished apartment made a huge difference to their lives as they did not want to wait for deliveries of furniture, or to be put in an empty house. Some commented that receiving a fully furnished apartment was less stressful, and that it removed or minimized the temptation to spend money on other items, including drugs:

They brought the TV up that night, made sure it was plugged in. […] with Housing First, to me, they bend over backwards for you.

(HF Client 10, male)

I walked in and all the carpets and that were down, all my white goods brand spanking new, brand new cupboards, bed, drawer. It was unbelievable man, unbelievable, and a big smart telly an everythin. Oh, it makes some difference and you’ve got less stress that way ‘cause I know what it’s like to walk into an empty hoose.

(HF Client 1, male)

Clients described the breadth of support offered as part of the HF package including: help with house insurance, TV licences, and welfare benefit applications; help with shopping and cooking, being reminded about/taken to appointments, and support for reconnecting with family. Some commented on the importance of being able to make their own decisions, or being supported to do so, and expressed that this was unusual. Clients reported having developed positive, trusting relationships with their support workers, being given helpful advice, and feeling valued and listened to:

They are really good listeners, it’s like having partly friends… but they are still doin their job. They are still listening to everythin and, you know, we are talkin as if we are friends, ‘cause I have made friends out of them.

(HF Client 8, female)

Some clients valued the ability to express a preference for the area in which their new home was located, and saw this as an opportunity to live in a familiar area, often with family nearby or, alternatively, to move somewhere new and have a ‘fresh start’. Most clients described how involvement in HF had enabled positive changes in their lives, helping them to feel safe, more motivated, independent and determined to get their life ‘on track’. Some described receiving support to access healthcare and medical treatment and to address substance use, for example:

They keep in contact wi me. They’ve helped me wi my mental health, they’ve helped me wi my drugs issue situation cause at [previous accommodation] my mental health was deterioratin and I started usin drugs… I went through the drug program to get rehabilitated an all that, they helped me wi that.

(HF Client 3, male)

I wouldn’t have followed through with anything, full of good intentions but the drugs an all that came first for me, afore even appointments, but it’s good the way this operation works because they come to you all the time.

(HF Client 1, male)

Others described using the support to apply for employment and volunteering opportunities. Moreover, many described positive feelings towards their apartments, and were happy to have a tenancy:

It maybe doesn’t seem much, but to me, it’s my kingdom […] I need to pinch myself because from where I was to where I’m livin now, it’s that hoose that has done it fir us. With the power o' they keys, to just know you can lock your own door instead of lying in alleyways, oh I get the shivers sometimes even thinking of half the places I’ve slept.

(HF Client 1, male)

Four interviewees described less positive views related to feeling that they were not adequately supported, or had been let down. One highlighted the importance of each individual being treated equally, stating that HF needed to deliver on promises. In addition, given that the level of support required in HF is, to a degree, led and directed by those using the service, there was also recognition that some clients had become more stable and were therefore receiving less support. Lack of staff continuity was also noted, with consistency considered by some interviewees to be a central and necessary component of HF.

Challenges experienced by clients

A prominent feature in the interviews were reported challenges faced since moving into HF apartments, such as a desire to decorate or concern about temperamental elevators. Problems with neighbors were also highlighted and primarily related to shared landings within blocks of apartments where other residents were using drugs and disposing of syringes. Other challenges concerned difficulty getting used to a new apartment, and feelings of boredom and isolation. Some clients, particularly those who had experienced long periods of homelessness prior to engaging with the HF service, found it difficult to be alone in their tenancy, having been used to constant companionship:

Aye, it’s alright aye, I’ve got my hoose noo, but I’m findin it hard in the hoose, cause I’m no used to it. I never ever thought I’d be in a hoose. It’s weird being in a hoose, sitting there yourself with a telly and all that. It’s boring.

(HF Client 9, male)

Finally, some described the types of support that they believed they still needed but were no longer receiving, such as assistance with housekeeping, using appliances, cooking, time management and keeping appointments, and budgeting. One client believed s/he needed to be assertive to be heard in terms of getting their needs met. Interviewees made suggestions for improvements. They were honest about their own vulnerabilities, as well as those of other clients, and reported that consistent, reliable support was essential for people who were vulnerable, but not necessarily offered. The need for HF staff and partner agencies to communicate more effectively was suggested. Returning to the feelings of isolation and boredom, and problems with substances discussed earlier, one interviewee suggestioned monthly social outings and others discssed the importance of companionship, even from a pet, to mitigate such feelings.

Staff experiences of providing Housing First

Complex circumstances of clients

A considerable theme emerging from the staff interviews related to client support needs which ranged from personal vulnerability and experience of past violence, trauma or neglect, and having been previously failed by caregivers or agencies. Adverse experiences had, according to staff interviewees, led to (initial) difficulties with trust and subsequently to behaviors such as difficulties coping with change, unpredictability, lack of perceived cooperation, and ‘self-sabotaging’. Some staff had the view that the ‘homelessness lifestyle’ had become ingrained for some clients who either become very dependent upon particular forms of service provision or, paradoxically, so familiar with living on the streets that living in an apartment was alien to them. Some commented that clients who had experienced prolonged or repeated episodes of homelessness had, in their opinion, particular difficulty with accepting change with many becoming accustomed to receiving diverse support in one place via a specialist ‘hub’ model (primary healthcare, dentists, addiction workers) so, when moving into a new area with HF, they would also have to change health/addictions workers. Staff described some clients struggling with independent day-to-day living and their view was that more emphasis on the teaching of practical skills might help avoid the frequent need for crisis intervention. A significant theme that emerged from all but one of the interviews was the isolation faced by clients, with one interviewee recounting a client’s drug overdose believed to be caused in part by loneliness and isolation. Working with clients to find activities in local communities to help re-engage them had been attempted, but not successfully. Isolation was most keenly felt by those who had previously resided in residential services. One member of staff proposed setting up a small fund to be used as part of an individual’s support plan to enable accessing of activities in the community.

Staff believed that poor self-esteem could prevent clients progressing their lives, with some feeling that their lives had no direction or purpose, and others being scared or overwhelmed by their own progress because they were not used to things going well:

Everyone’s trauma is unique and everyone’s pain is their own, you shouldn’t measure it against other people’s. It will be alright for a wee while and it will go down…Sometimes they get scared when things are okay because they have further to fall. There is a huge, huge trust issue here, there and everywhere. A lot of the time there is maybe self-sabotage.

(HF Staff 4)

The majority of staff spoke about the profound stigma and prejudice that clients experienced which undermined confidence. Attention was drawn to clients wanting to change their lives but not really knowing how to, experience of long waiting lists for wider services, and a general lack of more holistic approaches, which would enable addressing clients’ multiple needs simultanously. Staff commented that access to services for mental health and problem substance use was ‘almost impossible’ due to the requirement that clients addressed their substance use first. This was evident in both high and low threshold services, with issues arising even when trying to get clients seen for an initial assessment appointment, for which they were rejected if they presented intoxicated.

The Housing First approach

Staff discussed having a focus on providing a ‘forever home’. The majority felt that, in reality, HF was not the best option for everyone. There was recognition of the continued need for supported accommodation due to the challenges some individuals had regarding maintaining a tenancy, even with support:

Housing First is aimed at the most chaotic and the most vulnerable people, with the most severe issues and all this trauma… sometimes I think… is this really the right service for the most chaotic and the most traumatized? Are we aiming it at the right people?

(HF Staff 2)

Many interviewees discussed the importance of following a person-centred approach, recognizing that clients had different needs that required tailored support plans and risk assessments: working flexibly was essential. The importance of a harm reduction approach was stressed and some staff felt that clients should have freedom to live their lives without regular ‘check ins’, if this was what they wanted. In this HF service clients received a furnished apartment and a number of other benefits including a mobile phone. Staff viewed this as a ‘life link’ and something that was not necessarily common practice in HF elsewhere:

I think back to the old days when somebody got offered a house and… they were lucky if it had a bed. Bare floorboards, nothing to cook, no cooker. No wonder people failed.

(HF Staff 3)

Views towards Housing First

While staff believed that HF was doing the best that it could, they highlighted that success was difficult to measure, and that external professionals sometimes only viewed HF success as tenancy retention. Staff stated that recognising other improvements as ‘success’ was particularly challenging for those who did not know the clients and were unable to see how much they had progressed:

Some professionals see success as a tenancy not failing – that is not success. Success is… how they feel about themselves, how they are engaging with services, how their health is, how their mental health is. Are they even thinking about addressing their issues? If they are thinking about it then… we are on the right path.

(HF Staff 2)

Staff also stated that, while HF had worked extremely well for some individuals, others were still experiencing difficulties. Some expressed feeling disappointed that lasting changes were not taking place: others described the need to adjust expectations of ‘success’. A number of comments related to the service primarily being a crisis intervention for individuals who were still experiencing profound challenges:

It does vary and I’m not saying that it isn’t working, or that this isn’t a good project, but I think with about fifty percent of the guys we need to be practical. They are not signing up to college courses, they are not taking part in cookery classes in the community, we need to… be practical… and accept that we are [a] crisis intervention.

(HF Staff 4)

The consensus was that the job was challenging. Half of the support staff talked about difficulties with workload. Many discussed the need for flexibility, emphasizing challenges with planning ahead due to rapidly changing circumstances:

You can plan to the best of your ability, unfortunately, because the needs of the client group are so high, a lot of your planning can go out of the window. When people are saying ‘you need to manage your diary properly’, the clients we work with don’t have diaries.

(HF Staff 4)

All staff mentioned positive aspects of partnership and/or team working, with half commenting that support workers supported each other and that everyone was working towards the same goal, both within the teams and wider sector. Some staff also commented that they could positively input into the service and had ‘a voice’.

All mentioned challenges in working with clients, ranging from maintaining professional client-support worker boundaries, feeling challenged by particular kinds of behavior, and feeling under-qualified to respond to client needs, recognizing that clients may require specialist support. Continued or increased provision of regular training, supervision, team meetings, reflective practice and well-being days were suggested to address this:

There needs to be certain things that all staff are doing. There needs to be really strong boundaries with people and be consistent, or people will… expect that level of non-boundaries with everybody, and if they come up against boundaries it shuts them down and closes them off from engaging.

(HF Staff 2)

Most of the staff related challenges regarding accepting client choices:

So you are putting someone back in control of their life. I will talk about risks involved in that choice you are going to make… but if you still make that choice, I am still going to support you. But, at the end of the day, the consequences are yours.

(HF Staff 3)

Half of the interviewees also discussed challenges with client engagement, particularly evident when managing the nuances of the HF principles of flexibility and assertive engagement, acknowledging some contradiction between them.

Staff pressures

While there was a sense that everyone was doing their best, it is important to highlight that some staff reported feeling ‘out of their depth’ and discussed vicarious trauma impacting on their mental health and well-being. Indeed, the majority of staff raised concerns about not feeling fully equipped to support clients with complex and often intersecting challenges, and worried about letting clients down as a consequence. Drug overdose was also a significant concern:

Sometimes I think to myself ‘am I helping this person?’ I am taking someone who has got really bad mental health, really serious addiction issues, who are very high risk of overdosing, suicidal and we are putting them in a tenancy on their own. Am I helping this person, or am I putting them at higher risk of overdosing, or if he overdoses there is nobody there? That is something I struggle with a wee bit. Although I know first-hand the good that Housing First is, sometimes I feel ‘am I putting this person in more danger?’

(HF Staff 2)

These challenges and pressures led the majority of interviewees to comment that that the job had taken its toll. There was acknowledgement, from both ‘frontline’ and managerial staff, that support workers had an overwhelming sense of responsibility compared to, for example, staff working in residential services where colleagues are available 24 h a day to ‘share the load’.

In terms of the future of HF, staff recognized that HF was not a panacea to the problem of homelessness, emphasising that a multi-agency approach was required to achieve the common goal of ending homelessness. Some commented on the challenge of funding, stating that additional resources were required to effect lasting change. Staff described the potential for expansion of HF beyond individual clients, to support for couples and families in order to bridge the gap between families supported in assessment centres (which provide accommodation and support options to meet the needs of vulnerable individuals affected by homelessness), and independent tenancies. Staff highlighted that, given that HF tenancies were intended to be ‘forever homes’, the type of areas/housing offered should be carefully considered so that the housing could continue to accommodate clients as their needs evolved. Uncertainty was evident in approximately half of the interviews regarding provision after the end of the initial commissioned two years. Although staff offered suggestions regarding how clients could be better prepared for the conclusion of the support, there remained a concern that some clients would not be ready for support to end at that point.

Discussion

Parsell et al. (2015) argue that supportive housing models need to be informed by the views of those using services. This ‘insider’s insight’ can help to avoid pitfalls associated with implementation, as Henwood et al. (2011) outline. However, few studies have explored the views of those receiving a HF service (Pleace, 2020), and limited attention has been paid to the experiences of providers. Our study addresses this gap by interviewing staff and clients from one HF program provided by a third sector organization in Scotland utilizing an ICM approach.

Client expectations about the service varied but some commented that HF had exceeded them. Many expressed that HF differed from other services by virtue of staff being more caring where they gained substantially more support. Most clients reported positive experiences, stating that HF enabled them to make a range of changes in their lives. Clients described warm and trusted relationships with support workers. Clients also greatly appreciated receiving a fully furnished apartment. This finding resonates with McCarthy’s (2020) work regarding material possessions in the home contributing to womens’ sense of being ‘at home’. Four interviewees described less positive views, describing difficulties with adjustment feelings of loneliness and isolation. Recalling our earlier discussion, the provision of housing in itself can have positive health impacts (Rolfe et al., 2020), as well as help facilitate other health improvements such as better engagement with healthcare services.

Staff described providing more crisis support than anticipated. Relatedly, staff members expressed the view that the ‘scattered’ approach to HF was not always suitable, with some individuals potentially benefitting from the extra support provided via 24 h supported accommodation (including the single-site/congregate HF programs where staff are available on site 24 h a day). This echoes findings from the Candian At Home/Chez Soi project (e.g. Aubry et al., 2015; Macnaughton et al., 2015) where providers believed that intended participants of the HF model tended to have more complex needs than an ‘average’ person experiencing homelessness, leading to doubts concerning the ‘housing readiness’ of such clients (Kennedy et al., 2017). Our findings suggest that there is a need for a range of housing support types to suit individual needs, and that the provision of congregate HF programs in Scotland should be re-thought as one of the available options. Staff described the importance of flexible person-centred approaches, where client choices were respected and accepted even if these caused them problems. This finding corresponds with the concept of ‘thin rationality’ (McNaughton-Nicholls, 2009) which recognizes that people experiencing homelessness continue to exert the agency they can, while being constrained.

To date, most research on HF focuses on housing outcomes, revealing consistently positive findings (Baxter et al., 2019). When comparisons have been made between HF and other forms of support, evidence suggests that individuals who use HF services appear to have higher rates of housing stability and retention (Desilva et al., 2011) and demonstrate feelings of attachment to their housing (Parsell et al., 2015). These studies involved clients living in HF congregate rather than scattered housing approaches. Insufficient evidence exists comparing the two. The limited existing research is equivocal, for example, the scattered-site HF approach has been shown to be associated with increased housing retention (Tsemberis & Eisenberg, 2000) yet, in contrast, more recent studies have shown that congregate HF housing can be associated with increased housing stability (for those experiencing homelessness and problem alcohol use) (Clifasefi et al., 2013; Collins et al., 2012). Our findings lend some cautious support to this developing evidence base regarding housing stability and retention. However, as noted in our interviews, success is more nuanced: staff believed that each individual’s day-to-day progression, and improvement with a variety of substantial challenges in their lives, were equally important markers of success.

People who are homeless experience a range of health inequalities including being at greater risk of experiencing co-occurring problem substance use, poor mental health, and poor physical health (Hewett & Halligan, 2010; Levitt et al., 2009). In our study all of these were widely reported, with clients making improvements to their health and addressing their substance use and describing how HF had helped them to make these. This echoes findings of Kennedy et al. (2016) which highlighted improvements in the health status of clients, with some believing that HF had saved their lives. Our findings also highlight the importance of relationships between clients and staff which facilitated proactive engagement with primary care and other health services. However, both groups in our study expressed the need for more effective partnership and multi-agency working external to the service, a factor of particular importance in the Scottish context which operates the ICM model where clients are referred to agencies for specialist support. Other HF studies have commented on the importance of effective multi-agency working (Gaboardi et al., 2019; Pleace, 2020). While current evidence regarding health and well-being outcomes for HF is mixed (Baxter et al., 2019), a range of positive outcomes have been reported (e.g. Driscoll et al., 2018; Hainstock & Masuda, 2019; Kerman et al., 2018, 2020; Mackelprang et al., 2014).

The evidence of positive impact of HF on substance use is mixed. Overall, service utilization is associated with stabilization of drug and alcohol issues, rather than significant reductions or increases in use (e.g. Pleace & Quilgars, 2013; Padgett et al., 2006). Similary, both Baxter et al. (2019) and Beaudoin (2016) found that HF resulted in no clear differences in substance use when compared with treatment as usual. Qualitative research from a Norwegian HF program suggested that trusting relationships can contribute to clients talking openly about substance use and/or mental health problems (Andvig et al., 2018). While our study involved a small sample which limits the conclusions that can be drawn, our data lends some support to the idea that substance use in HF programs tends to stabilize rather than increase or decrease. It is also important to consider that, even if the evidence on HF in addressing problem substance use is still emerging, wider literature continues to highlight how critical stable housing is to facilitate recovery from problem substance use (e.g. Martinelli et al., 2020).

Qualitative research with service providers in the US and Canada has indicated that some of the key strengths of HF programs are the availability of ongoing support and community integration (Kennedy et al., 2017). However, the issue of community (re)-integration was reported to be one of the most significant challenges in our study, corresponding with recent Canadian, European and UK scholarship on HF (Pleace & Bretherton, 2019). People who are homeless are typically socially excluded (Fitzpatrick et al., 2011). Our study reported a significant level of stigma and prejudice experienced by HF clients from other services, professionals, neighbors, and the wider public. This aligns with research documenting how stigma can prevent those receiving HF being able to create a new life (Kennedy et al., 2017).

Importantly, our findings underscore the significance of loneliness and isolation in reducing QoL, especially for individuals who have spent long periods of time surrounded by staff and other clients in temporary accommodation, or being with others whilst living on the streets. Isolation is a prominent theme that emerges in qualitative accounts of HF experiences (see Stergiopoulos et al., 2014) and this may particularly be the case for those who have intentionally distanced themselves from former drug- or alcohol-related peer networks, or whose family ties were weak or severed (Johnsen, 2014). This sense of isolation has also been reported in Scottish research with older people who use drugs (Matheson et al., 2019), an issue recently compounded by the COVID-19 pandemic (Roe et al., 2021). While we have not found evidence directly comparing the scattered and congregate HF approaches on experiences of isolation and loneliness, findings regarding isolation are predominantly reported in respect of scattered, HF programs. Indeed, early UK evidence from Pleace (1995) suggested that use of scattered housing for vulnerable people could produce negative effects, including isolation. However, other evidence suggests that scattered HF programs, compared to congregate, can lead to greater independence, occupational functioning and subjective sense of choice (Chambers et al., 2018), with less risk of neighborhood stigmatization (Chen, 2019). Taken together, these findings suggest that the benefit of camaraderie on the streets, and of companionship in supported accommodation, may be overlooked.

Study data demonstrated the number and severity of adverse life events and associated trauma experienced, as well as their legacy. While clients feared their circumstances worsening, they also feared their circumstances improving: positive experiences were less familiar, and progress also meant there was ‘further to fall’. These complex emotions could result in what staff described as ‘self-sabotaging’ behavior. Research has shown that complex trauma (exposure to multiple traumatic events) can affect people’s behavior, including in forming trusting relationships and emotional management (Keats et al., 2012). However, if properly supported, people can and do recover (Cockersell, 2012). The importance of being sensitized to people’s histories resonates with Psychologically Informed Environments (PIEs) (Keats et al., 2012), a psychological framework designed to ensure services respond nonpunitively to the needs of those experiencing homelessness, particularly highlighting the need for recognition of the underlying trauma that many clients have experienced and how this may shape behaviors (Breedvelt, 2016; Johnson & Haigh, 2011; Phipps et al., 2017). Given the adversity and trauma commonly experienced by those who enter HF programs, as well as the lingering effects of these, our study participants suggested that independent living without specialist support might not be realistic, something also highlighted by other commentators (see Arslan, 2013; Barker & Pistrang, 2002; Maguire, 2012; Mcgrath & Pistrang, 2007; Phipps et al., 2017).

Staff in our study reported experiences of vicarious trauma which can be defined as being continually exposed both to traumatic situations, and to those who are traumatized, with consequent negative impact (Baird & Kracen, 2006). The impact of helping individuals who have experienced homelessness and wider adverse events can lead to the development of elevated rates of a combination of burnout, compassion fatigue, vicarious traumatization, and post-traumatic stress in key workers (Arslan, 2013; Seager, 2013; Waegemakers Schiff & Lane, 2019). Burnout has also been noted in evaluations of the Canadian At Home/Chez Soi HF program (Macnaughton et al., 2015). While trauma-informed care has become a focus for some services, it has been aimed at providing appropriate interventions that workers can use in helping those who are homeless, rather than addressing trauma-related responses among staff (Waegemakers Schiff & Lane, 2019). Our findings suggest that vicarious trauma is a significant risk for staff in HF settings and should be proactively addressed via mechanisms such as reflective practice and PIEs. Reflective practice groups, derived from Schon (Schon, 1983), include an active process of reflection and learning, are central to PIEs, and aim to support staff with these challenges.

Recommendations for practice

Despite the range of identified benefits, HF will not be the best option for everyone: other supportive accommodation services should be provided. There are significant risks of isolation and loneliness, and the potential for clients to depend solely on staff for meeting their social and emotional needs. Meaningful activity, as defined by individuals themselves, is important to give due consideration to, to prevent boredom. We believe that this requires sustained attention, including a greater appreciation of what individuals may be losing when moving to their own tenancy. The exclusion, stigma and prejudice experienced by clients should be addressed as a matter of urgency by national as well as local organizations, and governments. It severely hampers the opportunities for meaningful integration into communities. For HF to be optimized it should be part of integrated local strategies which coordinate and provide the full range of services people experiencing homelessness need, rather than a standalone response. Furthermore, as flexible, inter-agency working is essential for this client group, we would recommend enhanced communication between HF and partnering agencies, particularly those HF services utilizing the ICM approach which relies on external referrals. Support staff working with this client group can be at risk of vicarious trauma so support and reflective forms of staff supervision should be available.

Strengths and limitations

The strength of this study is the rich description gained concerning the value of the service according to both clients and staff. We were able to gain unique insights into aspects of this HF service that make a difference to the lives of clients because the lead researcher managed to successfully build rapport and trust with interviewees. One limitation concerns the inclusion of a single and relatively small service, rather than multiple services, and the fact that we did not involve wider stakeholders to gain a broader perspective on the program. This was because the provider organization operates only one HF service in Scotland. The research team saw value in exploring this single service given the scale-up of HF across Scotland, and other countries, and the need to learn from a range of service models and providers. Another limitation is the relatively small participant sample size. Given the intersecting challenges experienced by the majority of HF clients (for example, problem substance use), HF clients could be considered a marginalized and ‘hard-to-reach’ population which in itself can present a recruitment challenge. Some strategies to mitigate this in future research could include the involvement of peer workers/peer researchers, who have been shown to be successful in enhancing recruitment of harder-to-reach populations (e.g. Kaida et al., 2019). It is however important to note that published guidelines regarding sample size for ‘small’ qualitative studies utilizing interviews and using thematic analysis recommend between six and 10 participants (Braun & Clarke, 2006), with our study falling within that category and adhering to these recommendations.

Conclusion

This study has illustrated the benefits of one HF program in Scotland, helping individuals to move into independent tenancies, providing ongoing and intensive support to access a variety of services to improve health and well-being. In addition, we highlight a number of challenges that remain to be addressed if HF is to succeed in creating wider and more sustainable benefits for clients. These relate most specifically to the exclusion and stigma experienced by people who are homeless, isolation, loneliness and boredom, and problems addressing health needs such as substance use and/or mental health. Finally, our study has highlighted the need for extra support for staff providing HF programs and for multi-agency partnership working to be strengthened.

Ethics approval and consent to participate

The University of Stirling General University Ethics Panel (Reference number: GUEP 749) and the ethics committee of the participating HF provider provided ethical approval. All participants provided informed consent and received debriefing.

Consent for publication

All participants consented to their pseudo-anonymized data being used for publication.

Author contributions

JM, TP and HM conceived and collaboratively discussed the idea for, and contributed to the design of this study. JM and HM were involved in participant recruitment. JM performed all data collection. JM and KH performed data analysis, with input from TP and RF. JM wrote the first draft of this manuscript. All authors contributed to the interpretation of the findings and the final version of this manuscript. The authors read and approved the final manuscript.

Glossary

a wee bit: a little bit; aye: yes; doin: doing; draggin: dragging; fir: for; gae: give/gave; gaeing: giving; goin: going; hoose: house; housin: housing; lettin: letting; livin: living; mare: more; na: now; noo: now; nothin: nothing; o: of; tae: to; tellin: telling; wee: small/little; workin: working; yoursel: yourself.

Acknowledgements

The authors wish to thank all of the participants for their time, service managers and other staff from the HF service for facilitating the interviews and welcoming the researcher (JM) on their premises. We would also like to thank Dr Peter Matthews for his constructive comments on an earlier draft of this manuscript.

Disclosure statement

HM occupies a managerial role at the HF service evaluated in this study, for which reason she was not involved in data collection or analysis and was only shown anonymized data. The other authors declare that they have no competing interests. Authors who are researchers and funded by The Salvation Army (JM, TP, RF) are independent researchers employed by the University of Stirling: the funder does not seek to influence, nor is able to influence, study results and there is a contract in place clarifying these arrangements.

Data availability statement

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.

Additional information

Funding

The study was conducted by JM, TP, RF and KH who are part of the Salvation Army Centre for Addiction Services and Research at the University of Stirling and are funded to conduct research and knowledge exchange activities in the field of problem substance use and homelessness; with input from HM, who is part of The Salvation Army. SACASR is funded by The Salvation Army. Additional funding for travel and transcription were obtained from The Salvation Army.