Exploring Transitions Within a Project-based Housing First Setting: Qualitative Evaluation and Practice Implications

Abstract

The Housing First (HF) approach is a model of housing that entails the provision of immediate, permanent, low-barrier, supportive housing to chronically homeless individuals either in separate apartments within a larger community (known as scattered-site HF) or in a single building (known as project-based HF). One recent innovation is the application of project-based HF with chronically homeless individuals with alcohol problems. Although initial studies have shown its effectiveness, there is currently no research on residents’ and staff's experiences living and working in a project-based HF program. The purpose of this article was to document these experiences and highlight strengths and challenges of project-based HF programs. Using data collected from naturalistic observations, agency documentation, one-on-one resident interviews, and a staff focus group, we delineated transitional periods, including moving into project-based HF, community-building, managing day-to-day, and transitions from project-based HF. Findings are interpreted to help health care policymakers and providers envision the role of project-based HF in comprehensive public health efforts and to integrate lessons learned into their own clinical practice.

Based on 2009 estimates, there are 643,067 homeless people in the United States on any given night, of whom 17.2% are chronically homeless.¹ Health problems and mortality rates among chronically homeless individuals resemble those in the general populations of developing countries.^2,3 A recent meta-analysis of the literature on homelessness and health care indicated that housing instability is consistently associated with being uninsured, postponing health care, postponing medication, and higher hospitalization rates.⁴ Further studies have shown that homeless people,^5,6 particularly chronically homeless people with greater levels of co-occurring psychiatric and/or substance-use disorders,⁷ account for disproportionately large amounts of publicly funded and emergent health care, which is associated with greater health care system burden and costs.^8–10

Unfortunately, the traditional service infrastructure for homeless populations, such as emergency shelters, transitional housing, health care clinics, and abstinence-based substance-abuse treatment, often fail to engage this population and comprehensively address their complex problems.^11,12 Among chronically homeless individuals with co-occurring substance-use disorders, repeated contact with the traditional service infrastructure may result in a revolving door of jail, medical detoxification, mandated abstinence-based treatment, and failed attempts to obtain stable housing.^13–15 Of many important health-related predictors, the failure to obtain stable housing appears to be a critical risk factor for health problems and mortality.¹⁶

One reason for the failure to obtain stable housing might be the number of perceived barriers imposed by housing agencies.¹⁷ These might include stipulations on length of stay, chores, sobriety attainment and maintenance, curfews, personal money savings, rent provision, mandatory treatment attendance, and bans on onsite alcohol use. Policymakers have therefore called for the development of low-barrier housing programs that might more effectively engage, house, and attend to the basic needs of chronically homeless individuals.^18,19

The Housing First approach

The Housing First (HF) model entails the provision of low-barrier, immediate and permanent supportive housing to chronically homeless individuals many of whom also have co-occurring substance use and/or psychiatric disorders.^20,21 It involves the use of harm-reduction to address substance use.²² Harm reduction adherents use pragmatic strategies to minimize substance-related, negative consequences,²³ while maintaining a nonjudgmental, empathetic stance and supporting the realization of client-driven versus provider-driven goals.²⁴ The HF approach therefore differs from the traditional continuum-of-care housing model, which conversely requires individuals to fulfill certain provider-driven requirements (e.g., treatment engagement, in-house or complete abstinence from alcohol and other drugs) before they are allowed to move from a shelter to transitional housing to permanent housing.

The HF approach has been implemented in two main ways, both of which have been supported in the research literature and policy implementation.²⁵ Scattered-site HF programs were originally developed in the early 1990s by Tsemberis and colleagues^20–22 in New York City, and have now been implemented in cities throughout the U.S. In the scattered-site HF approach, residents are offered a choice of individual housing units in various locations throughout the larger community. In addition to housing, residents can choose to access a variety of supportive services (i.e., outreach; intensive case management; psychiatric, medical and/or nursing care; connections to external service providers; and assistance with basic needs) that are delivered using a client-driven assertive community treatment model. In project-based programs, residents are offered units within a single housing project, where they can elect to receive centrally delivered supportive services. Perhaps the most well-researched and widely cited project-based HF approach is the one established in the 1990s by Seattle's Downtown Emergency Service Center (DESC).^26–29

Quantitative studies have shown that scattered-site and project-based HF programs are associated with significantly increased housing stability, reduced public costs, decreased perceptions of coercion, and improved quality-of-life and alcohol outcomes.^{20–22,26–28,30} Additionally, qualitative studies have described scattered-site HF programs based on the experiences of individuals who participate in them^31,32 and based on staff's approach and goals.³³ A recent literature search has, however, yielded no studies to date that have documented how the project-based HF approach works in practice or how it is perceived by residents and staff. Such an exploration might highlight points for future enhancement of project-based HF programs and may assist in the development of similar programs across the U.S. Further, it may help health care professionals across different areas of practice understand these new housing programs and their cumulative effect on health care provision and health outcomes in their communities.

Importance of understanding housing transitions

In documenting how the project-based HF approach is perceived by residents and staff, it is important to consider all sides of the housing experience, starting with residents’ initial transitions into housing. Housing transitions are stressful,^34,35 and may be particularly stressful for chronically homeless individuals who have experienced a revolving door of treatment and incarceration,^14,15,36 and often have co-occurring substance use, psychiatric, and physical health problems that can complicate housing transitions.³⁶ Staff, who are helping affected individuals navigate housing transitions, might not fully understand these associated challenges,³⁷ and a better understanding might be key to facilitating this transition and increasing long-term housing stability more successfully.

Relevance of project-based HF to the health-care field

Project-based HF is a housing approach and not a clinical intervention as such. However, many of the issues that face this population and are addressed in project-based HF programs are relevant to the public health and health care fields more broadly. First, chronically homeless people with alcohol problems often have multiple, co-occurring, health-related problems, and account for a disproportionate amount of publicly funded health care utilization and cost.^7,26,37 Thus, any intervention tailored to this severely affected and underserved population is a clinically significant topic with cross-disciplinary relevance. In fact, recent studies have shown that project-based HF is associated with improved health outcomes for residents as well as decreased publicly funded and emergent health care utilization and associated costs.^26,28 Furthermore, HF is being advanced by government and public health agencies as evidence-based best practice for addressing homelessness and its various co-occurring conditions.^18,19,38 This growing support for and interest in project-based HF suggests it will become increasingly important to foster understanding and collaboration among housing agencies, public health officials, policy makers, and health care service providers to maximize project-based HF effects and ultimately improve health outcomes for this severely affected and costly population.^7,26,67

Current study aims

Because HF approaches are increasingly being endorsed as evidence-based best practices in addressing chronic homelessness,³⁸ studies on this approach have great public health relevance. To address the dearth of literature on this topic, we sought to examine residents’ transitions into housing and the day-to-day experiences of residents and staff who live and work in a project-based HF program. We aimed to explore the strengths and challenges of this approach and to elucidate potential points for program enhancement.

Methods

Participants and setting

Both staff and residents in a project-based HF program (i.e., a single housing project using an HF approach) located in downtown Seattle, Washington provided data in the context of a larger program evaluation conducted for a Seattle-based housing agency during 2009–2010. At the time of this study, the project was staffed with 16 full-time equivalent staff members: one Master's-level (MPA) project manager; one registered nurse; four Bachelor's-level and one Master's-level clinical support specialists providing clinical case management; one Bachelor's-level residential counselor supervisor; eight residential counselors with high school diplomas or equivalent, five of whom also had Bachelor's degrees.

Residents were chronically homeless individuals with alcohol problems (N=75) who were identified using two main sources: (a) a rank-ordered list of individuals who had incurred the highest public costs for alcohol-related use of emergency services, hospital, sobering center (i.e., sleep-off facility for homeless people with alcohol problems), and county jail in 2004; and (b) a list of eligible individuals suggested by community providers familiar with the target population. Case managers and housing agency staff sought out and recruited the residents from community settings. See Table 1 for a description of the demographics and alcohol use observed within the initial wave of residents.

Table 1.

BASELINE DESCRIPTIVE STATISTICS FOR THE INITIAL PROJECT-BASED HF POPULATION^a

Variable	M(SD) / %
Sociodemographic variables
Age (years)	48.39 (9.39)
Race/Ethnicity
American Indian/Alaska Native	27.4%
Asian	1.1%
Black/African-American	7.4%
Hispanic/Latino/a	7.4%
Native Hawaiian/Pacific Islander	3.2%
White/Caucasian	40.0%
“More than one race”	10.5%
Self-reported “Other”	3.2%
Relationship status
Married	2.1%
Consider self married	1.1%
Widowed	4.3%
Separated	7.4%
Divorced	33.0%
Never married	52.1%
Highest education level
Some high school	37.2%
HS graduate/GED	29.8%
Vocational school	8.5%
Some college	18.1%
College graduate	4.3%
Some graduate school/advanced degree	2.2%
Alcohol-use variables
Typical alcohol quantity (standard drinks/day)	24.39 (21.87)
Peak alcohol quantity (standard drinks/day)	39.85 (39.28)
Alcohol-use frequency (days/past month)	23.75 (10.49)
Intoxication frequency (days/past month)	19.82 (12.26)
Alcohol-related problems (SIP-2R)	23.34 (1.37)
Experience of delirium tremens (% lifetime)	65.2%
Number of alcohol dependence symptoms currently endorsed	5.22 (1.80)
Number of alcohol treatment episodes (lifetime)	17.19 (59.13)

^aQuantity and frequency variables are based on the past 30 days. All other variables based on past 3 months.

M = Mean

SD = Standard Deviation

SIP-2R = Short Inventory of Problems summary score

Data collection procedures

The data collection was multi-phased to document various aspects of the day-to-day experience in a project-based HF program from multiple perspectives. In the first phase, we conducted naturalistic observations of staff-resident interactions in shared community spaces in the HF project. To capture a variety of daily experiences, we took near-verbatim notes on resident-staff interactions in two-hour shifts at different times of day. These observations yielded more than 20 hours of hand-recorded transcripts. In the next phase, we followed staff on two separate two-hour rounds and took field notes on the activities and conversations that ensued. The third phase was a review of official housing project documentation regarding rules and regulations. In the fourth phase, we conducted individual, one-time resident interviews (n=17). Resident interviews were semi-structured, lasted approximately 30 minutes, and explored residents’ experiences at the project, their relationships with staff and other residents, and their ideas for future project enhancement. Residents received $10 for their participation. In the fifth phase, we conducted a single two-hour, semi-structured focus group with program staff (n=8). We discussed staff's day-to-day experiences, their roles at the project, their relationships with other staff and residents, their perceptions of the housing project community, and their appraisal of its strengths and challenges. We audio-recorded and transcribed focus groups and interviews and removed all names and identifying information to maintain confidentiality. Planned procedures were reviewed by the Institutional review board (IRB) at the University of Washington (UW) prior to the project start date. As this study was considered by the UW IRB to be quality improvement/program evaluation, it was not subject to full IRB review. We adhered to the American Psychological Association and Helsinki ethical guidelines in the conduct of this work.

Qualitative data analysis plan

The goals of the analysis were to provide a conceptual/thematic description of key elements involved in residents’ and staff's experiences of the housing transition process.³⁹ Naturalistic observation data, field notes, and transcribed focus groups and interview sessions were managed and coded in Atlas.ti version 6.⁴⁰ Qualitative data were independently coded using a constant comparative process consistent with grounded theory methodology.^41,42 Initial coding was conducted incident-by-incident using the observational data, whereby coders narrated the actions occurring in staff's and residents’ verbal exchanges.⁴³ After an initial independent coding phase, we created a codebook in consensus meetings, pooling incident-by-incident codes and removing or collapsing idiosyncratic or redundant codes. All observational sessions were re-coded independently, and interrater coding consistency (80%) was established.^42,44 The focused coding phase consisted of the most frequently observed initial codes. After data collection was complete, we pooled our memos from the focused coding phase and explored potential thematic codes.⁴⁵ Residents, program staff, and agency management had various opportunities to review the findings. Their feedback served as a means of assessing usefulness, fit, and resonance and was integrated into this article.

Results

Our examination yielded key themes that fit into staff members’ and residents’ progressions throughout the housing transition. We discuss these themes in the context of the following phases: (a) moving into project-based HF, (b) community building, (c) managing the day-to-day, and (d) transitions from project-based HF.

Moving into project-based HF

Move-in day

Residents were recruited into project-based HF after years of chronic homelessness, which often involved sleeping on the streets as well as overnight stays in hospital ERs, shelters, and sobering centers. In fact, some residents reported purposely drinking to high levels of intoxication, particularly in winter, because the local sobering center would allow them to sleep inside if they had reached a certain, clinically indicated level of intoxication. Many residents recalled feeling relieved that this was no longer the case:

I heard about this place, man. I was in a sobering center for a year and a half before finally they said, “That's it, [resident's name], we've got a place over here for you,” . . . I remember it was great! They said I don't have to get drunk, and I can come over here and go to sleep, and I don't have to be drunk to sleep inside. That's what I liked.

Residents uniformly reported being relieved to move into project-based HF. One resident reflected, “Now I've got a clean place to live, a kitchen I can use, Food Stamps every month. I'm happy. Things are good.” Another resident was “just glad to get in . . . [after] sleeping outside, you know, in the rain, in the snow, in the ice, in the cold. ‘Cause you can't drink in a mission.’”

At the same time, some residents reported being moved directly from the local sobering center or hospitals into the project, often very early in the morning, after recovering from a night of acute intoxication. Residents reported feeling disoriented, sometimes wondering where they were when they woke up. The transition into the project was, therefore, a jolt from people's previous day-to-day routine on the streets and acclimation took time:

They take us from sobering center. They pick us up over there and bring us here. Five of us. Then they took us to the building. Brand new. Nobody was here . . . and it was December 23rd. They took us here, these people. Then after we return, they give us our key. And that day I left. I visit my cousin in [city]. Live with him for a week. And after that I told him to drop me here. Then I ended up staying here.

One manager pointed out that this initial disorientation seems nearly unavoidable given the multiple problems that can compromise these individuals’ cognitive functioning at the time of recruitment (e.g., substance-related intoxication, physical/psychiatric disorders). She referred to this issue as a Catch-22 because “it is [difficult] to catch these residents sober enough for a more lucid move-in, [but] moving inside is the first step toward enhancing lucidity overall.”

Housing unit assignment

Staff indicated that housing project and outside agency staff, who are familiar with the recruited individuals and their needs, typically make collaborative decisions regarding resident unit assignment. Once recruited, residents are assigned to one of 49 private studio apartments or 26 semi-private cubicles (i.e., subdivided living areas without doors that contain drawers, a bed, and storage space). Residents do not have direct control over whether they receive a cubicle or private studio apartment.

Aware that housing unit assignment was not in their hands, residents discussed this topic with anxiety and speculation:

I've been in the same spot [in the cubicles] for almost two years. I don't have income so they haven't put me upstairs [in a studio apartment], that's the only reason why I can figure. Otherwise, other than that, there must be something in my paperwork that says I'm not very—I need to be watched or whatever. I don't know.

Originally, inclusion of cubicles in the housing design was to allow for greater monitoring and access to care for residents with chronic and severe medical conditions and/or greater need for assistance in activities of daily living. As residency has stabilized, however, assignment to cubicles versus studio apartments has become more dependent on immediate space shortages. This space-based means of assigning units was viewed by some residents as an obstacle to acquiring a more desirable private apartment. From the residents’ perspectives, moving into a private apartment from a semi-private cubicle was a clear priority for increasing privacy, security and quality of life:

I would like to move upstairs [from the cubicles] and get my own, private place, where I can lock my stuff up. I've been ripped off out of my cubicle more than once.

. . . [Staff] put some crazy person next to me [in the adjoining cubicle]. . . . There's been four or five of them and . . . there's been more than a couple instances with me, ’cause I—you don't wake me up at 2 in the morning. Especially if I haven't had enough booze, and I'm struggling to sleep as it is.

Community-building

Residents’ solidarity

Despite frequent physical altercations and arguments among some residents, we observed that residents also supported and helped one another (e.g., retrieving others’ wheelchairs, calling for help for an ill resident, asking staff to check up on other residents). Some residents described their community as being like a small town where people look after each other, “[My neighbor] brings over food. You know, I wake up to food in my room. . . . People really do look after each other.”

Staff members witnessed residents’ sense of solidarity, street-smarts, and pride in the uniqueness of the project community, and have integrated these characteristics into their own social network and individual identities. One staff member noted:

We are a part of this community. As staff you are a member of this community—in our role—but we are definitely integrated . . . I think we have learned that lesson from this population. That to be able to work with this population—part of that meeting them where they are at—is that they are really, really close and really familial, and as a result, amongst our team we are really familial, and we are really close.

HF is “an island”

The uniqueness of the community is also associated with a sense of isolation among staff—that the HF facility is an “island.” This sense was expressed in two ways. First, perhaps because of the isolated nature of a project-based HF setting, staff reported feeling physically separated from other agency and community services. Second, staff reported feeling figuratively isolated. One staff member said, “We're misunderstood. It's lonely. We're judged.” Staff often noted that health-care professionals at other local hospitals, clinics and programs have difficulty understanding and relating to the project-based HF model and its associated challenges. In the focus group, this became a main topic of discussion:

Staff 1: And I'd like to see there be more education at all of the area hospitals so that they understand what we do here. [So] they understand that it would be really helpful for the discharge planner to call us.

Staff 2: And to stop calling it the “wet house.”

Although staff indicated interest in more internal support and external collaboration to decrease their isolation and enhance their ability to better serve residents, they also reported that the physical isolation of this “island” allows for more independent decision-making, which was viewed positively by staff:

That is one of the things about our way of going through the process of the day around here is that our decision-making abilities—it's really done within the team . . . that processing and team decision-making . . . is a really amazing process because . . . collectively we have really great ideas.

Managing the day-to-day

Constant crisis management

Interpersonal, medical, and psychiatric crises occur regularly in this multiply affected population, and staff reported being on perpetual watch for the “next fire to put out.” This was mentioned by staff when asked to describe a typical day at the housing project:

Staff 1: There is no typical.

Staff 2: You've got to expect the unexpected here. I mean one day can be completely mellow and the next can be one crisis after another . . .

Constant crisis readiness appeared to complicate the implementation of a highly structured, day-to-day routine. The resulting open structure and strategies in approaching responsibilities allow greater staff flexibility to attend to crises, but this readiness also appears to contribute to high stress levels. One staff member commented on the chronic nature of this stress: “. . . it's respond, respond, and that response gets wearing, because it's such a high stress level.”

During this study, we were able to observe staff in various crisis management roles. The following situation reflected a successful resolution to an escalating situation during one late-night observation. A verbal exchange between two residents was escalating towards a physical altercation. Just before one resident made an attempt to physically hit another, two staff members stepped between the residents, and without physical contact, engaged them separately. One staff member retreated with one resident to another area to talk. The other staff member had the following exchange with the more agitated resident while remaining calm and speaking in a low tone of voice.

Staff: Give me five minutes, man. Let's go outside and talk.

Resident: No, I've been drinking.

Staff: I know you've been drinking, and that's ok. Let's go outside and talk. Give me five minutes.

Resident: That's bullsh*t, man. I'm not going. That's bullsh*t.

Staff: Yeah, but you are disturbing people. Come on, man, give me five minutes . . .

Resident: You are being a f*cking *sshole, man. Do you have a cigarette?

Staff: Let me see what I can do. Let's go outside, man. [Resident nodded, and the two go outside together.]

Among the various factors that might contribute to the outcomes of crisis situations, interpersonal style employed by staff appeared to have the greatest effect on the resolution or escalation of crises. The previous example shows the effectiveness of remaining calm, even in the face of resistance. On the other hand, some approaches meant to de-escalate potential crises, such as staff ignoring agitation or using a highly directive interpersonal style, were not able to diffuse crises and were instead associated with greater levels of agitation and externalizing behavior from residents.

Multiple roles

Staff faced the challenge of maintaining multiple roles in the context of their various and sometimes disparate duties within the housing project and the different ways in which they related to residents. The expectation that staff will perform multiple roles is central to day-to-day HF project management. There are strengths of the resulting diminished sense of hierarchy and the “everyone pitches in” spirit this structure affords: Staff have achieved a high sense of cohesion and teamwork. As one staff member noted, “Not only is our job kind of like walking a line between rigidity and flexibility . . . it's our job to look out after one another . . .”

The flexible job roles are, however, associated with less clearly defined delegation and distribution of tasks. Staff are often simultaneously faced with diverse and challenging situations and “things [that] are not going to come up for your ‘normal’ mental health case manager . . . It makes it really complicated. You're . . . wearing lots of different hats.”

Another challenge is the difficulty in knowing how to relate to residents while performing the varied and complex duties in the housing project. First, the HF, harm-reduction approach requires complex decision-making in clinically challenging situations: staff must maintain clinically appropriate boundaries while simultaneously meeting clients “where they are at” with a stance of unconditional positive regard.²⁴ Although staff undergo regular in-service trainings, most are entry-level human service workers and only a few have Master's-level clinical training to help them navigate these challenging situations. Most important, residents’ desire to connect with others can create a strong pull for staff into roles and relationships that are different from those encountered in other, more traditional housing and treatment settings. Some staff described residents’ pull to engage in close, familial relationships:

. . . I think we are filling a role no one has ever filled for them. Sometimes they don't get it, but when they do, it's kind of amazing. I think our roles are different. It's like mommy, daddy, brother . . .

We also observed some staff relating to residents via peer-type interactions, which ranged from innocuous rapport building to boundary blurring. For example, exchanges like the one below are discouraged by agency management:

Staff: You like that [brand of whiskey]? I like my whiskey smoother.

Resident: Yeah? What's your favorite?

Staff: My favorite is [brand name]. I like whiskey, I just like it smooth.

In this example, a less experienced staff member unwittingly engaged with a client around alcohol in a peer-type relationship, expressing a positive attitude toward alcohol use, which would be at odds with the value-neutral stance of the harm reduction approach towards alcohol use.

All in all, staff's ability to flexibly engage with clients in various ways throughout the day appeared to serve the residents’ needs and seemed satisfying to staff. Furthermore, staff's openness and ability to juggle multiple roles clearly arose from an adept and intuitive recognition of the need to empathize and build rapport with residents. Navigating these roles, however, appeared to be a delicate matter that contributes to workplace stress.

Power differential

It is clear that staff work hard to build strong relationships with residents. It is important, however, to acknowledge the power differential that lies at the heart of every staff-resident exchange. Residents appeared to be acutely aware of the power differential and expressed anxiety about navigating their relationships with staff. One resident explained how he negotiated a relationship with a particular staff member to maintain what he perceived to be a tenuous housing situation:

. . . don't get on [name]'s bad side. Stay on [name]'s good side. [name]'s willing to cooperate with you if you're willing to cooperate with [name]. . . . I don't know anybody who don't have fear, you know? What happens if I lose this place, you know? Am I gonna go back home to [name]? I don't wanna go to treatment. I did nothing bad . . .

Transitions from project-based HF

In project-based HF programs there are no formal plans for residents moving to other housing. That said, transitions out of the project were not uncommon, and the reasons were various. Some residents went to jail or were asked to leave due to violent incidents, although the majority of these individuals were able to move back into the project at a later time. Others reunited with their families or moved into other, independent housing. Still others elected to engage in inpatient or residential alcohol treatment programs. Some residents were moved to hospital, nursing, or hospice facilities when they needed a higher level of medical care. Many residents discussed their thoughts and feelings about potential transitions from the project.

Moving out

Some residents who had moved out and moved back in at a later time reported it was difficult to continue living at the project when their own goals were too disparate from the realities of living “inside” or when they had trouble following the rules. When reflecting on their experiences, these residents reported preferring to live at the project but were nevertheless prepared to return to the streets if necessary.

Interviewer: Do you ever worry about getting kicked out of here again?

Resident: No, it doesn't bother me. . . . They give me a 10-day notice a while ago ’cause I was swearing at some woman . . .

Interviewer: When you say it doesn't bother you, what do you mean?

Resident: I learn—know how to live on the streets. I don't care. I'll use some cardboards, and instead of sleepin’ on the cement, I'll just get some cardboard . . . I'll take my coat off, use it for a blanket. Hey, survival of the fittest!

Staff leaving

Staff members leaving also appeared to affect residents’ sense of stability and security in developing bonds. Residents sometimes perceived the staff turnover as a personal loss that reflected on them.

Resident: Everyone changes jobs [leaving the project].

Staff: A lot of turnover. You are right. It happens a lot in human services, had a lot of case managers in your time. That can be stressful.

Resident: Everyone sees me and leaves.

Staff: It's not about you, it's hard for providers to leave [too].

Resident: Hard for them, hard for us.

Fear of mortality

Premature mortality has been well-documented among chronically homeless populations,^46–48 and is an issue facing both staff and residents in project-based HF. Although the deaths that have occurred in this project-based HF program have been largely caused by long-term, chronic illness and none by acute alcohol toxicity, some residents were concerned about “drinking themselves to death.” This concern resembled a specter that is always in the back of residents’ minds. One resident noted, “Alcohol is serious. You got to talk to somebody, you got to have some friends, you got to have people who genuinely care for you. When you're on your own, you drink yourself to death. You're going to end up in trouble.”

Staff are also affected by frequent exposure to some residents’ experience of long-term, chronic illness, and in some cases, death. One staff member described how it feels to be privy to residents’ end-of-life experience:

I think the fact that we are in people's lives in such intimate moments . . . is really intense . . . the experience of sitting by people's sides as they are dying, is, is, I mean, I don't think there is anything closer. . . . That's hard. That is very hard to have sat through this much trauma, this much death, and this much of those hard intimate moments of life . . .

Discussion

Our analyses yielded key phases of staff and residents’ progressions throughout the project-based HF experience, including (a) moving into project-based HF, (b) community-building, (c) managing day-to-day, and (d) moving out of project-based HF. In contrast to another study documenting the transition from homelessness to housing,³⁷ our findings did not reveal unidimensional factors that represented challenges. Instead, we found that many traits of the project-based HF program could be seen as both strengths and challenges depending on the point-of-view and the situation. For example, the housing project was sometimes simultaneously viewed by residents as a warm, safe place to live and as a stressful environment in which they experienced interpersonal conflict. Furthermore, project-based HF was perceived by some residents as a community in which they felt accepted, but by some staff as “an island,” on which they felt isolated from other professional peers. Because strengths and challenges often appeared to represent two sides of the same coin, we found it important to consider housing transitions and their associated challenges on a case-by-case basis from both staff's and residents’ perspectives.

Main themes and implications for practice

Moving into project-based HF

During the transition into HF, residents reported feeling both relieved (because they were grateful to be inside from the elements) and disoriented (because many were recruited into the HF program when they were recovering from acute alcohol intoxication). Considering the challenges both staff and residents face when moving into the HF program, they might benefit from a more formalized orientation to the project-based HF approach. In addition to the training currently in place, staff members could benefit from more training addressing the physical, psychological, and environmental stress that housing transitions place on these particularly vulnerable individuals and how residents’ stress levels and needs may change throughout the transition into project-based HF. Residents could benefit from a more formalized, multisession orientation process that starts on but extends beyond the move-in day to provide staff with an accurate baseline and ensure residents’ adjustment to the new housing situation.

Most residents cited a perceived lack of control over housing unit assignment as a source of anxiety. This finding might have implications for residents’ sense of security and emotional well-being considering the positive association between perceived control over one's environment and quality-of-life outcomes.^49,50 Residents provided several suggestions to ease this perceived lack of control, including the use of waiting lists for more desirable units and allowing residents to exchange places. Because residents’ anxiety seemed to result from a lack of understanding of how units are assigned, it is also important to make this process transparent and explicit. Finally, although the intention of building the semi-private cubicle space was to ensure adequate attention for residents needing a higher level of care, more residents than anticipated were able to live independently. Therefore, to ensure residents’ sense of privacy and security, we recommend that future housing projects include fewer cubicles and more apartments.

Community-building

The sense of community in this HF program is a strength that appeared to increase residents’ sense of belonging, bolster their interest in building positive relationships with other residents and staff, and provide a safety net for and early detection of emergent physical or psychological problems. Residents’ sense of community seemed, at least on the surface, to be connected to their shared experiences with alcohol and chronic homelessness. Echoing observations in other community settings,⁵¹ these outward similarities also seemed to reflect deeper connections among residents that were sometimes not apparent to staff members. Such deeper, cultural connections seemed to be rooted in a mutual understanding of surviving on the streets. Thus, providers might capitalize on this existing sense of community that has brought these individuals together to support community-building within project-based HF programs.

Staff in this project-based HF program were immersed in the residents’ community, and over time, seemed to develop a greater understanding for and empathy with residents’ general sense of isolation and marginalization. This contrasted with observations made in single-room occupancy facilities, in which residents and staff seemed more clearly divided,³⁷ but was similar to findings in scattered-site HF programs.³³ Although the strong sense of community strengthened the staff's team cohesion, there was also a sense of professional isolation. This sense of isolation might occur because health care and housing professionals outside of the project-based HF program do not understand this population's specific needs or staff's need for more responsive professional collaboration and consultation.

To decrease this sense of professional isolation, project-based HF programs might consider seeking connections with other service providers more actively. Such community connections might increase awareness among local health care professionals of this population's needs as well as the project-based HF programs’ impact on the wider health care community. Fostering connections could also reduce the sense of professional isolation by increasing the flow of new information, services and technologies to project-based HF programs and vice versa. Team-building and increased information flow for higher quality decision-making are key points for successful teamwork in health care settings.⁵²

Managing day-to-day

The chronic stress staff reported experiencing in managing the day-to-day tasks and crises in this project-based HF setting echoes the experiences of other frontline human service and health care providers.^53–55 Despite the stressful nature of the daily workload, staff reported being prepared for, and even at times, appreciating the fast-paced environment. Residents also appeared to be noticeably resilient, even in the face of high-stress living situations. There might, however, be ways to further relieve staff's and residents’ stress and ease interpersonal interactions. Additional training on graduated crisis management and de-escalation techniques could be helpful in optimally managing crisis situations. The use of external clinical consultants could provide fresh clinical perspectives, additional links to the outside professional community, and help in coping with the challenges associated with the day-to-day crisis management.

Our findings indicated that it is a challenge to maintain clinically indicated boundaries while helping residents with intimate and personal issues, ranging from basic care and personal hygiene to physical and mental health issues to interpersonal relationships. Staff's openness and empathy with residents signaled an intuitive recognition of the need to meet residents “where they are at.”²⁴ Thus, staff appeared to be receptive to the client-driven characteristic of the harm reduction approach that is a key element of the low-barrier, non-abstinence-based project-based HF model. This harm reduction approach appeared to help staff build stronger, more empathetic relationships with residents. At the same time, even highly trained and experienced professionals find the tension between acceptance of clients where they are at and clinical, agency, and societal pressures to encourage behavior change to be a difficult dialectic to navigate.²⁴

Although no amount of training can fully resolve the challenges of working in a project-based HF program with severely affected populations, more intensive clinical training for staff might help guide prioritization and implementation of their many roles. Currently, staff attend monthly, didactic, agency-wide trainings, which build on staff members’ initial education and on-the-job experience. More intensive, site-specific and hands-on workshops might help staff put specific skills into practice. Such workshops could provide more personalized, small-group supervision, emphasizing role plays and rehearsal of clinical skills. Further, these workshops could introduce staff to other evidence-based interventions that might be relevant to this specific setting and the populations’ needs. For example, because most residents have been exposed to multiple severe traumas, training on posttraumatic stress disorder might enrich project-based HF program staff's understanding of trauma and coexisting psychiatric disorders as precipitants of residents’ externalizing behaviors and might give staff additional tools to avoid activation of maladaptive schemas (i.e., learned frameworks and images formed in the past that are used to interpret current situations).^56–58 Additional clinical training might also help staff learn to model new normative boundaries and behaviors to residents (e.g., nonjudgmental acceptance but not encouragement of residents’ drug use),⁵⁹ and build appropriate, empathetic, nonjudgmental relationships.⁶⁰ Finally, more training on client-centered approaches, such as harm reduction psychotherapy²⁴ and motivational interviewing⁶⁰ could teach staff new ways to help residents build and resolve discrepancy between their current and more adaptive behaviors.

Transitions from project-based HF

Residents perceived others’ and their own transitions from project-based HF as stressful. Life transitions have been shown to adversely affect psychological and physical health in the general population,^34,35 and might be particularly stressful for chronically homeless individuals,⁶¹ who have experienced multiple, severe stressors.^62,63 Our findings reflected those in the literature that have shown staff and residents perceive transitions differently, particularly when stressors unwittingly reactivate residents’ trauma histories.³⁷ Furthermore, studies conducted in longer-term health care facilities suggest that staff turnover can adversely affect staff-client relationship building and communication.^64,65 Staff could benefit from further training and clinical supervision on how to address these issues with residents. Residents might also benefit from support groups to cope with other difficult transitions, such as grief and termination issues.

Limitations

There are some important limitations to the current study. First, some of the defining features of the program—the housing approach, the setting and the population—might serve as limitations to the generalizability of the findings. Housing First approaches in general have different challenges and strengths from other housing models, such as continuum-of-care models. This research was also carried out in a very specific setting (i.e., project-based HF) and its larger social context (i.e., a politically progressive city in the Pacific Northwest). This HF project specifically recruited and intentionally serves chronically homeless individuals with alcohol problems. Our findings might therefore be less applicable to other types of housing projects, other locations where there may be more public resistance to the development of project-based HF, other segments of the homeless population, or in settings with a less uniform population. Finally, data were collected in the context of a larger program evaluation, which is a systematic method used to examine, describe and improve specific programs. Thus, by its nature, program evaluation may not have the same generalizability that is the hallmark of research evaluation.⁶⁶ For these reasons, the generalizability of the findings and our recommendations should be carefully considered in their interpretation and application within other populations, settings and approaches.

Relevance of the findings to public health and health care disciplines

Although these findings represent key themes within a specific housing setting, many of these issues are relevant to the public health and health care fields more broadly. First, the population described in this study presents with multiple, co-occurring, health-related problems and accounts for a disproportionate amount of publicly funded health care utilization and cost.^7,26,67 Fortunately, recent studies have shown that exposure to project-based HF is associated with improved health outcomes for residents as well as reduced publicly funded and emergent health care utilization and associated costs.^26,28 Because HF is being advanced by government and public health agencies as evidence-based best practice for addressing homelessness and its various co-occurring conditions,^18,19,38 it is increasingly important to foster understanding and collaboration among housing agencies, public health officials, policymakers, and health care service providers to improve this population's health outcomes. By providing a step-by-step description of how project-based HF addresses the “public health crisis of homelessness,”⁶⁸ the current study may help health care providers and policymakers envision its role in future, comprehensive public health efforts. Finally, because many clinically related elements discussed in this manuscript may be applicable across disciplines and settings, this study may help both health care and other service providers integrate common principles and lessons learned from this project-based HF implementation into their own clinical practices.

Conclusions and future directions

In this exploration, we sought to document residents’ and staff's experiences of living and working in a project-based HF program; to explore the strengths and challenges of this approach; and to thereby elucidate potential points for program enhancement. Key transitional phases within a project-based HF setting included: moving into project-based HF, community building, managing the day-to-day, and transitions from project-based HF. Within each of these phases, we observed clear strengths of the project-based HF approach, including a strong commitment to community building among both staff and residents, staff's increased understanding of the harm-reduction approach, and the provision of safe and stable housing for some of the most vulnerable and marginalized residents of our community. This unique application of the project-based HF approach and the population it serves face numerous challenges as well. Our findings suggest the need for further multidisciplinary efforts to cut across the divides in policy, health care, housing and treatment fields and to shed light on how to address most comprehensively the needs of chronically homeless individuals. Most important, these findings highlight the importance of residents’ and staff's input at every step of the development of project-based HF programs to facilitate residents’ transition from the streets to the “inside.”