Introduction

Employment is an important goal for people who are homeless and have a mental illness.1 , 2 It may facilitate successfully exiting homelessness3 , 4 and contributes to recovery from mental illness.5 – 7 However, homelessness is associated with numerous barriers to finding and maintaining employment.3 , 8 Preliminary studies suggest that without housing, barriers to employment are unlikely to be overcome.9 , 10 Because the relationship between housing and employment is likely bidirectional (housing facilitates maintaining employment, and employment facilitates maintaining housing), researchers suggest that programs intending to reduce homelessness should include both supported housing and supported employment services.11 , 12

Few studies have employed rigorous scientific methods to evaluate the combination of supported housing and supported employment.13 , 14 Those that have employed rigorous methods report modest findings that suggest housing can improve vocational outcomes,9 but not to the extent seen in high-fidelity evidence-based supported employment programs.

Individual placement and support (IPS) is one of the most effective supported employment models for helping people with mental illnesses return to and sustain employment.15 , 16 IPS offers a one-to-one personalized approach to finding jobs and follows eight principles: (1) focus on competitive employment,Footnote 1 (2) focus on individual preferences for job placement, (3) rapid job search, (4) integration into the mental health-care team, (5) zero exclusion: those who wish to work are eligible, (6) benefit counselling, (7) systematic job development, and (8) continued individualized support.18 A meta-analysis suggests that IPS may be effective in samples with experiences of homelessness and mental illness.19 A randomized controlled trial, however, suggests that IPS is less effective when offered to people experiencing housing instability than when it is offered to those who are stably housed.20 Evidently, the added complexities of experiences of homelessness can reduce the effect of IPS.

Service providers are developing methods for overcoming barriers experienced by stably and unstably housed people with mental illness, such as overcoming criminal records21 , 22 and substance use,23 but less is known about the influence of housing on these barriers.8

The primary objective of this paper is to explore which barriers to employment, directly resulting from having been homeless, persisted once stable housing was provided to individuals with a diagnosis of mental illness. This will allow us to elucidate the impact of housing on employment for this population. A secondary objective is to determine how IPS services helped participants overcome or manage these barriers.

Methods

Participants

The present study used a subsample of data from a randomized controlled trial of IPS nested in a larger randomized controlled trial of Housing First (HF).24 Group membership and study nesting are illustrated in Fig. 1. The two parent studies will be referred to as the IPS trial and the HF trial, respectively. The HF trial protocol has been published elsewhere.25 Briefly, inclusion criteria were the following: 18 years of age or older, the presence of a mental illness,Footnote 2 and current homelessness.Footnote 3 From this HF trial, 90 participants of the experimental group were recruited for the IPS trial, which required an expressed desire for assistance finding work and unemployment at time of recruitment.20 From this pool of 90, 14 participants were recruited from the experimental group and 13 from the control group for the qualitative sub-study. None of the participants for the present study dropped out. Follow-up for the HF trial was exemplary.27

FIG. 1
figure 1

Participant group membership of the 27 participants who contributed to the qualitative interviews.

Full size image

Recruitment

Recruitment for the present study was conditional on the participants’ willingness to complete an hour-long interview. Convenience sampling was used.28 Participants were approached shortly after their recruitment into the IPS trial and recruited sequentially. This ensured that desire to participate was not influenced by negative or positive opinions formed during the intervention. Participants were contacted by phone. Letters were mailed to reach those without telephones. Six of the eight mailed invitations were accepted. Of the 31 people contacted by phone, 21 participated. No new content related to the themes reported in this article emerged after the 11th interviews in the experimental arm and the 13th in the control group. This suggests that sufficient participants were included in our sample.29 A remuneration of $25 was provided. Ethics approval was obtained from the ethics review board at the Douglas Institute affiliated with McGill University, in Montreal, Canada.

Procedure

For the purposes of the present study, participants were interviewed by the first author over a period of 2 months before the end of the 2-year HF trial. Participants had received the Housing First intervention for 12–22 months and the IPS intervention for a minimum of 10 months at the time of their interview.20 A topic guide written by the first author and an independent qualitative researcher familiar with the project was used to conduct semi-structured interviews.30 The topic guide included questions about the respondent’s experiences with services, their experience with transitioning to stable housing, and their employment experiences. Interviews were in English or French depending on participant’s preference. Approximately a quarter of the interviews were conducted in participants’ homes, the remainder in an office familiar to them. The interviewer had no prior or enduring relationship with the participants beyond the recruitment and the interview. The interviews were transcribed by the first author and coded in ATLAS.ti (version 7.0).

Intervention

All participants in this study received supported housing services by virtue of their inclusion in the parent HF trial.24

People in the IPS group of the IPS trial received services from two employment specialists with Masters Degrees in health-related disciplines who were trained and supervised by the senior director of a local IPS program. Each employment specialist had a caseload of 22 participants. Contact with the employment specialists varied in frequency depending on participants’ needs and preferences. Periods between contacts ranged from a few days to 3 months. A fidelity scale was used to gauge the program’s implementation and adherence to these principles.31 Repeated fidelity assessments indicate that services were implemented with good, but not exemplary, fidelity for the last 8 months of the study, suggesting room for improvement. Services offered job search assistance, resume and cover letter assistance, mock interview exercises, on-the-job follow-up, and support to employers (depending on participant’s preference).

Participants in the control group of the IPS trial could choose from the vocational services available in the community. These included services linked with emergency shelters and welfare agencies. Jobs were usually temporary, offered daily contracts, and were limited in range of options. None of the employment agencies which assisted people in the control group would have had contact with the clinical teams providing support to the participants.

Analysis

The goal of the project was the identification of barriers to employment; therefore, we used thematic analysis to analyze the data.28 , 29 The first step after transcription was coding of each interview. To ensure methodological rigor, the coding of the first four interviews was done separately by the first author and an independent qualitative researcher familiar with the project. Coding lists were compared and discrepancies reconciled. The first author completed the coding of subsequent interviews with the reconciled coding list. Coding lists were different for the two groups, in that the group receiving IPS had IPS intervention-specific codes absent in the control group. However, the codes related to barriers reported below overlapped entirely in the two coding lists. Interviews were coded in their original language. Code frequency, intensity, and primacy were used to assign importance. Co-occurring codes were reassembled into themes based on the nature of their relationship with one another.32 Themes were generated inductively, and guiding questions did not target specific themes or barriers.30 Excerpts presented below were translated by the first author and an independent bilingual researcher to assure an unbiased translation.

Results

At the time of interview, four of the 14 participants receiving IPS were competitively employed, one was about to begin a new competitive job, and the remainder had not obtained stable employment. Of the 13 participants in the control group of the IPS trial, one was in competitive employment, two were working in jobs reserved for people receiving welfare, and one had several casual jobs. The others had not sought employment services. All were housed. Participant characteristics are reported in Table 1.

TABLE 1 Participant characteristics
Full size table

Our participants experienced several barriers to employment, including discrimination on the basis of race, gender, age and education, and interpersonal challenges with family members and coworkers. These barriers are not unique to the intersection of homelessness and mental illness and have been reviewed in the literature.33 Barriers related specifically to housing and homelessness are discussed below.

Experiences of homelessness resulted in several persisting barriers to employment, but some participants saw their experiences as sources of strength and resilience. While not immediately contributing to successful employment, these strengths and resiliencies were an important part of their conversation on their persisting barriers to employment.

Persisting Barriers to Employment

The barriers to finding employment that persisted once participants were stably housed included the following: (1) apprehension about disclosing sensitive information and explaining absence from work force; (2) wavering motivation; (3) continued substance use; and (4) being disturbed by fear, pain, and anxiety about re-experiencing homelessness-related trauma. Some of these barriers, such as substance use, existed to a certain degree prior to homelessness, while others, such as the fear of renewed victimization, were a direct result. Participants who received IPS spoke of the strategies co-developed with their employment specialist to deal with the barriers. These strategies are described following the details of each barrier.

Disclosing Sensitive Information

Apprehension about disclosing life events that explain an absence from work was a frequent barrier to searching for employment. Some participants were worried that their work history would appear suspicious if they did not have an explanation for their absence from the work force, but life events were difficult to discuss with potential employers. They felt that they could not tell employers about their mental illness or experiences of homelessness:

Well, if they are curious, you have to explain those 29 months! What can you say? The dissolution of the marriage? Living in shelters? No matter the supposed openness of the employer, let’s be realistic, these are not things that play in your favour when you are sitting in front of someone who is interviewing you. (translated, IPS participant)

Others had experienced discrimination when employers found out about their experience of mental illness or homelessness. This had an important impact on participant’s opinion about disclosing information to future employers, especially those in the control group who did not have support:

And when they found out I had a psychiatric history, I was laid off also. It has happened four times. So now with my employers and co-workers, I don’t usually let them know that I have a psychiatric history. (control participant)

This participant was symptom-free for several years but nonetheless experienced stigma because of prejudice and misconceptions about people with a history of mental illness and homelessness. Many other participants worried about disclosing sensitive information and were uncertain about what information should be disclosed. Participants receiving IPS reported being willing to disclose their experiences to employers. The employment specialist supported participants’ wishes and proposed approaching the subject in different ways. They spoke extensively about it and practiced interviews to prepare for difficult questions:

We are going to talk about it first. I want to be sure of myself to be open-minded about trying different approaches [to disclosing]. If I am not open-minded about it, it’s sure that nothing good will happen. Or I will already have the expectation of failure. (IPS participant)

Participants in the control group never spoke about discussing this obstacle or seeking advice. They dealt with it as best they could, which usually resulted in either avoiding discussing the issue or accepting the consequences (sometimes termination of their job) once employers found out.

Wavering Motivation

For approximately half of the participants in both groups, motivation to return to work fluctuated. This delayed their search for employment. They felt they needed more time to readjust to having a home before searching for employment:

Well it’s because when she proposed the project [IPS], I accepted, but deep-down, I had not taken care of myself. So I was not ready at all, I accepted that she would come to see me and all that, but in the two months that followed, I realized that I was not ready and that things were moving too quickly. I had too many downs still, and wasn’t motivated to find work. (translated, IPS participant)

Participants with wavering motivation sometimes pursued their goal of employment by submitting resumes and contacting potential employers. However, when motivation fell, they neglected to follow up with these initial steps. This meant that when motivation peaked afterwards, they had to start from the beginning, submitting new resumes to new employers.

Employment specialists helped strengthen motivation by actively engaging participants in the search process and following up with employers when participants were not. With the help of the employment specialist, some participants dedicated a certain period of time each day towards searching for a job as a strategy to maintain motivation:

I need to encourage myself regularly. So it’s good to see my worker because, let’s say otherwise you have the tendency to stay in your corner. (IPS participant)

However, other participants were ambivalent about employment, which influenced their engagement with employment specialists. For example, a younger participant accepted employment specialists’ help, but consequently neglected his job search. Participants in the control group noted that “there was no one to tell you, to motivate you to ‘go work, go work’, that voice has to come from within.” Participants that noted this internal dialog spoke about finding work as well, but none of the participants in the control group spoke about devising strategies to bolster motivation.

Continued Substance Use

Most of the participants consumed substances prior to homelessness, but their consumption was exacerbated by mental illness and homelessness. Once housed, however, participants indicated that substance use diminished. For some, it remained an issue, but for the others, addiction existed only in their past. What persisted was a fear of relapse. For participants receiving IPS, the progress they had made with the help of employment specialists motivated them to devote extra efforts to managing their sobriety. For those in employment, substance use diminished as a result:

I am currently reducing that [drinking], because last time you came, I was in the parties. But the partying, now, it’s over. I need to rest on Saturday and Sunday. Saturdays…like when I drink Friday nights, I wake up Saturday and I feel tired. I have no strength. […] Alcohol now, I have no choice, I have to limit it. I can’t get to the office smelling like beer. I don’t want to miss work. (translated, IPS participant)

This positive effect of employment and housing was not universally reported. Some participants reported opposing experiences where employment and housing provided the funds and location for excessive consumption. Boredom on weekends, lack of social networks from the marginalization of homelessness, and money from work precipitated substance use and contributed to long cycles of consumption and job loss:

Well, it’s that I don’t know what to do with it [money]. I should make use of it. I should find activities to do, I should have a social network, I should do something with my life. Instead of doing that I buy a case of beer and drink. And then drinking brings on other things. As soon as you drink you start smoking pot then you go on to cocaine, you know? It’s those cycles that I have a hard time breaking. I always find my way back into it. When I was working it was the same. I started to consume again. That’s probably why I got sick of working. […] You consume and you work. At one point you come to the realization that you no longer feel like working. You know…because consumption takes so much out of you… (translated, control participant)

For participants receiving supported employment, the support appeared to positively influence the way in which they managed their substance use. The collaboration between the clinical teams supporting the participants and the employment specialists was especially important. Support from employment specialists led participants to emphasize their goals of employment when discussing addictions with their clinicians, helping them better address both their substance use and their employment goals:

So I had given up drinking and my clinician told me “we will have to find you something to do before you replace [drinking] with something worse”. Exactly: my work. It’s healthy, it’s good, it’s productive, and it’s what I need. So that’s why I work really hard with [employment specialist] to rapidly find a job to focus on, rather than focus on searching for another dependence (translated, IPS participant).

Participants in the control group did not report such benefits, and substance use continued to have deleterious effects on employment.

Fear and Anxiety

Fear and anxiety about re-experiencing homelessness-related trauma were not the most frequent experiences, and participants were reluctant to share these sentiments. However, they were intensely discussed experiences with important impacts:

Interviewer: I am wondering if there are any experiences you lived while you were homeless that still bother you?

Participant: Ah! Certainly! When I go to bed, I barricade the door. I’ll tell you, I have a toolbox, it weighs pretty close to 150 lb. I put it in front of my door…you know, even with the chain and everything. I’ve got safety locks on all my patio doors and windows. If I hear a sound or something, I always have a knife with me next to my bed. I know it sounds crazy, but look, I’m left with these sequelae. (translated, control participant)

Fear and anxiety linked to experiences of homelessness played a big role in interfering with the lives of some participants. Symptoms kept them from functioning and seeking employment because they worried about experiencing the same anxieties at work. As a result, they doubted their ability to return to former occupations, like truck driving or construction, which required focus.

This barrier was never discussed with the employment specialists; therefore, they could not assist participants deal with this barrier.

Strength and Resilience

Participants emphasized the development of personal strengths and autonomy and a motivation to vanquish their past. Participants in the both control and IPS groups discussed the strengths that developed as a result of their experiences of homelessness.

It [homelessness] helped my development. I learned many things. I learned to socialize with strangers, to talk to people, to be more tolerant of people, to forgive people. And I think if I hadn’t…lived the life I have lived, I wouldn’t have become the person I am today. (control participant).

Other participants felt their experiences motivated them to seek ways of avoiding future homelessness. They recognized that their chance of returning to the streets was greater if they forgot their experiences:

I don’t want to forget the journey I have had, and everything that happened. Because it makes me stronger. If I forget, I could fall back into it. So I put aside a personal reserve of myself to remind myself that I am not recovered 100 %. (translated, IPS participant)

While these persisting effects did not have an immediate impact on participants’ search for work, they did frame participants’ employment goals during the interviews. Omitting this aspect of homelessness would be neglecting an important part of our participants’ experiences. Participants did not mention if the employment specialists made use of these strengths to motivate them.

Discussion

This study revealed that several barriers to employment persisted once people with a mental illness who were formerly homeless obtained stable housing, in addition to those known to affect people with mental illness.33 This is consistent with previous research suggesting that the impact of identifying as homeless has negative effects that outlast the actual period of homelessness.34 , 35 Unexpectedly, some of the participants also described positive experiences of homelessness and enduring strengths. This is in line with research demonstrating that people who are homeless adapt by developing attitudes and strengths.35 , 36 This study also highlighted participants’ perspectives of the role of employment specialists in addressing these barriers.

The way participants dealt with disclosing information exemplifies the differences between the IPS and control groups: participants receiving usual services had to devise their own strategies and withheld all information about their past. This could lead to negative consequences when employers discovered their history of mental illness, substance use, or homelessness. Participants receiving IPS devised different strategies, such as conducting mock interviews to prepare for difficult questions. Planning what to disclose is beneficial to job tenure, employer relations, and client satisfaction and is consistent with previous research.37 , 38 It is important to note that the source of this barrier is not the symptoms of mental illness, but the fear of stigma. Research related to stigma in the work place suggests that addressing the source of misunderstanding can have important impacts on reducing stigma between employers and coworkers39 and that withholding information may perpetuate stigma.40

It is not surprising that substance use persisted as an obstacle to employment. However, the contrast between the IPS and control groups is noteworthy: IPS participants managed to cut down on consumption because of the demands of their jobs, whereas those receiving usual services dealt with the stress of employment by consuming. The finding that boredom could lead to consumption has been explored in the literature.41 – 43 By helping people return to work, IPS may remove the boredom trigger, reducing the chance of relapse. However, employment alone may not be sufficient and the additional time-unlimited follow-along support from IPS may have contributed to management of substance use.

The trauma of homelessness is an established psychological stressor.44 This study provides qualitative data to support the link between the fear and anxiety about re-experiencing homelessness-related trauma and unemployment. A wide variety of trauma including injuries and motor vehicle accidents precipitate or perpetuate unemployment45 and subsequently homelessness.3 , 44 People experiencing this type of a barrier to employment did not discuss the issue with their employment specialist. It may therefore be beneficial for the employment specialist to broach the subject where appropriate, but further research is needed to explore the number of people who experience this barrier to employment.

Limitations

The relatively small sample was drawn from a homogeneous group of recently housed participants of a supported housing program who experienced mental illness and unemployment, but expressed desire to return to work at the time of enrolment. Despite the fact that a large number of people who are homeless experience mental illness,46 our findings may not be generalizable to other homeless subpopulations or supportive service contexts. Furthermore, the services were newly implemented and experienced difficulties achieving good fidelity.20 If the intervention had been offered by an established service, it is possible that employment specialists may have dealt differently with the barriers highlighted in this study. Finally, participants received IPS services for varying lengths of time, as recruitment for the IPS trial lasted 2 years. This is due to initial difficulties achieving good fidelity. Participants with longer involvement with IPS services may have more time to reflect upon their challenges and devise strategies to overcome them. However, this difference does not introduce a bias as the length of follow-up did not differ between groups.

Implications

The findings of this study have several implications. While returning to stable housing may be assumed to help people return to work, added supports are necessary. Supported employment helps address several persisting barriers to employment that result from having been homeless.

Furthermore, clear explanations of the service user’s and employment specialist’s roles and expectations can be helpful in strengthening partnerships. Discussing obstacles to returning to employment appears to prepare clients for the job search, as do discussions highlighting their strengths and accomplishments. Services may be improved by providing safe environments for successful clients to share their progress, an existing component of high-fidelity IPS programs.

Finally, certain sensitive topics concerning obstacles to employment may not be easy to discuss, especially if they have been associated with past experiences of stigma and discrimination. It is important for service providers to have heightened awareness of these difficulties and support service users in exploring them.