Introduction

Work is known to be a vital arena for recovery and social inclusion as well as a primary goal for the majority of people with severe mental illness (SMI). Nevertheless, employment rates for the population of people with SMI remain low, with typically only 10–20% having a paid job [1, 2]. Moreover, the results from two recent systematic reviews indicate that across countries, the portion of employed people decreases among those diagnosed within the schizophrenia spectrum [3] and that many people with SMI lack meaning, structure and daily activities in their lives and are at risk of poverty, victimization and homelessness [4].

During the past decade, systematic reviews [5,6,7], integrative reviews [8, 9] and meta-syntheses [10, 11] have explored predictors and barriers of competitive employment for people with SMI. Among the most reported individual barriers in the individual placement and support (IPS) literature are those related to demographic (e.g., age), cognitive (e.g., working memory), clinical (e.g., symptoms), and psychosocial (e.g., self-esteem) arenas and those related to the environment, such as service (e.g., employment specialists’ skills) and workplace features (e.g., work accommodation) [8, 12, p. 4]. Moreover, it has been revealed that job tenure is influenced by service users’ feelings of competence and enjoyment in performing job duties; their own strategies for problem-solving and managing symptoms at work; friendly, cooperative and inclusive coworkers; diverse support within and the workplace; and helpful work adjustments [9, 10]. Regarding job match, while previous IPS research has mainly focused on examining elements related to the person’s preferences, prior work history, illness-related difficulties, stressors, coping strategies and strengths, recent research has focused more on analyzing factors in the work environment [13].

Supported employment (SE) programs, mainly those using the IPS model, have been recognized as the most well-defined and efficient place-then-train vocational interventions to integrate people with SMI into the labor market [14]. Originally developed in the USA in the early 1990s, the IPS model spread to Canada, Europe, Australia and Hong Kong [14, 15]. However, although studies have shown limited success rates, with 60% of people with SMI obtaining and retaining employment [12, 16], IPS has been promoted as being two to three times more effective than traditional train-then-place models and has been indicated to result in “more rapid entry into competitive employment, more hours and weeks worked, and higher wages” [17, p. 3]. Due to the varied welfare systems across continents and contexts, implementation challenges of IPS have been reported [18,19,20]; however, according to a recent systematic review [21], the model can be generalized across diverse settings and economic conditions.

The IPS model is promoted as a pragmatic and person-centered approach that avoid extensive assessments, pre-employment training, and the demonstration of work readiness and may be available to anyone with an SMI who wants competitive employment [15, p. 1099]. The Employment Specialist (ES) is the key person in the IPS model. Working closely with mental health professionals, the ES coordinates every step in helping clients through the processes of finding, obtaining and keeping their preferred jobs. In addition, the ES provides counseling regarding benefits and outreach in the community to develop jobs and cultivate relationships with employers [15]. Research findings indicate that an ES typically “makes weekly contact with IPS clients immediately after a job start, within a few months reduce this to monthly, and maintain this frequency thereafter” [22, p. 150]. However, it has been noted that ESs’ contacts with clients tend to be more frequent during periods of employment than during periods of unemployment after a job loss [22]. As studies have shown 40–50% dropout rates of people with SMI prior to their obtaining employment, even with the assistance of SE programs, some programs have complemented IPS with other psychosocial interventions, such as cognitive remediation, motivational interviewing, social and job skills training, and with principles of the collaborative recovery model [17, 19, 20]. Nevertheless, such supplements have not yet proven to have significant effects over standard IPS interventions [4].

According to a scoping review [23, p. 197], “flexible scheduling/reduced hours, modified training and supervision, and modified job duties/descriptions, mainly provided by an ES” are the most commonly used work accommodations for employees with an SMI. Notably, contact of the ES with the job site has been found to be positively associated with job duration [22], including the ES actively helping their clients cope with cognitive impairments [24]. In a previous article [11], it was suggested that service users’ perspectives on the complex interplay between personal, workplace, material, and systemic factors [10] indicated that many people with SMI experience working to be similar to “ice-skating” [11]. Thus, workers with SMI often call for hands-on support in learning new skills and finding and maintaining the balance while being “on the edge” while working [11]. In other words, persistent follow-up support at the workplace, rather than the practice of “dropping” clients once they seem steady in their jobs, is known to be a predictor of job maintenance [22].

Overall, most people with SMI struggle to obtain and maintain work in the regular labor market, even with assistance from SE programs. Nearly half of IPS clients leave or lose their jobs within 6 months, and most of the jobs they obtain are part-time [12]. Moreover, studies have shown that ESs vary widely in their effectiveness (see, e.g., [25]), and it has been concluded that ESs need to possess a broad range of professional knowhow, including interpersonal, interviewing, assessment, job development, negotiation and problem-solving skills. Their relational skills and the resulting working alliance have been found to be essential components of these competencies [26,27,28,29]. To our knowledge, there is no systematic understanding of how ES competencies are experienced across diverse contexts and continents. Thus, the current review aims to explore and synthesize the views of IPS/SE clients, ESs themselves, and ESs’ supervisors on the core contributions of ESs to job support within the mental health field.

Methods

Design

We conducted a meta-ethnography [30] to perform an all-embracing synthesis of qualitative research findings on ES contributions. According to Noblit and Hare (1988), “synthesis refers to making a whole into something more than the parts only imply” (p. 29). Meta-ethnography involves systematic ways of comparing conceptual data from various qualitative studies about the same topic to develop new overarching concepts, theories, and models [30, 31]. In the processes of translation, the reviewers are encouraged to understand and transfer ideas, metaphors, concepts, and themes (second-order concepts) from one study to another and to integrate them by creating a common language for their interpretation, while emphasizing the preservation of meaning [30, 32]. Meta-ethnography includes the following seven steps: (1) getting started; (2) deciding what is relevant; (3) closely reading the selected studies and noting themes; (4) putting the studies together and determining how they are related, i.e., whether they are comparable (“reciprocal”) or in opposition to each other (“refutational”); (5) translating the studies into one another by grasping the meaning of each study account while being aware of the context; and (6) synthesizing, which involves two parts: To synthesize the translations; representing the reviewers’ interpretations of the translations (third-order constructs), and the line of argument synthesis, where the reviewers go further than translation and put any similarities into a new interpretive context by creating an overarching new storyline of the studied phenomenon. On the seventh step the synthesis is expressed [30, 31, 33, 34]. In this organized but artistic way of combining texts into a collage of new meaning, we engaged in hermeneutic processes of third-level interpretations [32, 34]. To validate our reinterpretations, we rigorously selected relevant participant quotes used in original studies and included them in our synthesized findings. As the included studies concerned similar topics, we primarily engaged in reciprocal translation analysis. In addition, to develop an overarching metaphor, a line of argument synthesis was used [30, 35].

Search Strategy and Processes

Initially, we planned to conduct an integrative review and include findings from both qualitative and quantitative research [36]. We performed a search to find everything published about the ES role, regardless of the setting or service. To plan the search, we used the Sample, Phenomenon of Interest, Design, Evaluation, Research type (SPIDER) tool [37], which is useful when searching for qualitative or mixed methods research. The search was kept very simple and consisted only of the following search phrase: job coach* OR employment specialist. We chose only databases indexing the health science literature so we would capture the relevant studies within the field of mental health. The first author (LGK) and fourth author (GA, librarian) developed and carried out the search strategy in the following databases: Amed (1985-present), Embase (1974-present), Medline (1946-present), and PsycINFO (1806-present) through the OVID SP interface and Cinahl (1981-present) through the Ebscohost interface. All databases were searched with the maximum time span to find as much literature as possible. We searched the title, abstract, and keyword fields of the databases. No limitations, such as language, type of article, time frame or geography, were used in the search. The search was conducted twice: the initial search on June 21st, 2017, and the final search on February 1st, 2019.

Selecting Primary Studies

Initially, we found 1071 hits in the databases, and after deduplication in EndNote, 558 remained. From the initial search, after the assessment of the titles and abstracts, 94 references were selected for full-text screening. Following the final search, the full texts of an additional 10 references were assessed. The screening process was conducted in two stages. The initial search was carried out in June 2017, and Rayyan in blinded mode was used to screen titles and abstracts. The title and abstract screening was conducted by the first author and a research collaborator who later withdrew from the project. A total of 454 references were excluded during the title and abstract screening. At this stage, we included both qualitative, mixed-methods and quantitative studies reported in original research papers from peer-reviewed journals published in English by employing well-documented research methodologies and selecting papers that were relevant to our research question. The second stage was conducted after the final search in February 2019 and included title and abstract screening performed independently by the first and third authors. Next, the first (LGK), third (IV) and fifth (RWA) authors screened all relevant studies by reading the full texts. Of the studies excluded during the full-text screening, the majority were theoretical publications, systematic reviews of all types, book reviews, program descriptions, opinion papers, and job support interventions targeting diagnostic groups other than people with SMI. However, as the inclusion process progressed, the research team found that there were too many relevant studies and decided to treat the qualitative and quantitative findings separately. Thus, the quantitative findings will be analyzed in a separate review, while the qualitative findings are synthesized in this meta-ethnography.

After the full-text screening, 23 qualitative studies were deemed relevant for this review. However, seven of these studies were later excluded (LGK, MC, and LD), some due to a lack of thick descriptions and some due to the samples being too heterogeneous. The final set of studies included in this review consisted of 16 articles. As rule-based judgments are contested in the qualitative research literature, the review team decided not to use checklists; instead, the included articles were evaluated through a reflexive hermeneutic dialogue [38]. Thus, quality alone was not used as an exclusion criterion. The following two screening questions suggested by Campbell (2003, p. 674) were used: “Does the article report findings of qualitative research involving qualitative methods of data collection and analysis and are the results supported by the participants’ quotes?” and “Is the focus in the article suited to the synthesis topic?” [39]

Analyzing and Synthesizing

The first author (LGK) and third author (IV) critically read each included article in its entirety but mostly focused on the results section. The first author developed mind maps of the articles’ results sections, which were brief summaries of the main findings or themes (second-order constructs) in each article and selected verbatim quotations representing informants’ perspectives (first-order constructs). The first author recorded second-order constructs along with some first-order constructs across the 16 studies in a preliminary matrix, which became the raw data for the synthesis. The tables, summaries, and matrix were discussed on two Skype meetings by the first, third and fourth authors (LGK, IV, and RWA). Next, the first author noted details from each article—authors, year of publication, country of study, aim, participant characteristics (gender, mean age, work experience), setting, design, data collection and analysis, relevant results, and identified ES contributions, which are presented in Table 1. To maintain the original meanings and concepts, the terminologies used in the included articles are preserved in the table.

Determining How the Studies are Related, Translating the Studies into each other and Synthesizing the Translations

The first author (LGK) and third author (IV) continued the processes of interpreting the data. Noticing common, recurring themes as well as differences, they compared how the second-order constructs across the studies, which were recorded in the matrix, were related to each other. They determined that the constructs were comparable (reciprocal analysis). When necessary, the original articles were used to recontextualize the constructs to enable reinterpretations. LGK and IV translated the comparable second-order constructs into each other by hand. Using a whiteboard, they categorized and coded concepts into a preliminary “mixture”, showing the relationships between the second-order constructs. Thereafter, they carefully performed a thematic analysis through interpretive reading and translation of the implicit meanings in the second-order constructs across the studies, involving an analytical development of preliminary third-order constructs [30, 31]. Their preliminary findings were noted systematically in a table, showing the relationships between the constructs and reinterpretations. For example, the first (LGK) and second (IV) authors translated the second-order constructs across studies that described ESs’ use of, e.g., motivating interviews or specific skills training and summarized them as “competences, approaches, and skills.” Next, in their processes of creating a collage of new meanings of the data [34], they synthesized this summary into “mapping the route,” which became one of the third-order constructs or final themes. Moreover, they wove the third-level constructs into a line of argument synthesis [30] by developing an overarching metaphor. The final third-order themes and the metaphor were developed and discussed first by a group of three of the authors (LGK, IV and RWA) and next by another group of three authors (LGK, MC, and LD). To validate the reinterpretations, two authors (LGK and LD) re-examined the findings examining Table 1 and the original full texts of the studies and reorganizing and recoding the data by using a whiteboard.

Results

Characteristics of the Included Studies

Key characteristics of the included studies are described in Table 1. Six of the 16 studies included were conducted in the USA, three were conducted in Australia, three were conducted in Sweden, two were conducted in Canada, one was conducted in the UK, and one was conducted in Denmark. Nine studies were published in mental health or psychiatric rehabilitation journals, four were published in occupational therapy or rehabilitation journals, and three were published in other journals. The methods used for qualitative analysis were qualitative content analysis (n = 9), ethnographic/grounded theory (n = 3), phenomenological analysis (n = 1), multiple case design with cross-case analysis (n = 1), and a combination of qualitative and quantitative approaches (n = 1). In one study [40], the qualitative research approach was not explicitly described. Five studies [40,41,42,43,44] were part of a larger IPS randomized controlled trial, and participants were selected from the IPS intervention group. One study was a part of a larger evaluation study of IPS using mixed methods [45]. The data collection methods employed were in-depth semi-structured interviews (n = 10), observation and field-notes (1), a combination of observation and interviews (2), semi-structured interviews combined with a semi-structured mail survey (1), and mixed data collection methods (2).

A total of 101 IPS clients living with SMI and 147 ESs (among whom six were IPS supervisors) were involved. The samples of IPS clients included from five to 27 adults; the samples of ESs ranged in size from three to 76. In eight studies, the samples included IPS clients living with SMI; in three studies, however, clients may also have experienced additional challenges. For instance, in two studies [40, 42], clients also had a history of homelessness or contact with the criminal justice system, and in one study [46] some clients also had physical comorbidities. The client samples included both females and males, with males outnumbering females. Moreover, samples in seven studies comprised IPS clients [40,41,42,43,44,45,46]. One study had a mixed sample of Individual Enabling and Support (IES) clients and ESs [47], and samples in eight studies comprised ESs/job coaches and their supervisors [48,49,50,51,52,53,54,55]. Given our research question, IPS was the predominant setting for investigation.

Table 1 Studies describing IPS clients’, employment specialists’(ES) and their supervisors’ experiences of the core contributions of ESs in job support included in the metasynthesis
Full size table

The analysis revealed seven themes, which are outlined below. To develop a comprehensive understanding of how the themes were related to each other, we created an overarching metaphor, which is intended to offer an overarching narrative of the studied phenomenon: ESs’ contributions in supporting people living with SMI in entering and remaining in the workforce, as experienced by IPS clients (n = 101), ESs (n = 141), and ES supervisors (n = 6) interviewed in the included studies. In the following section, we will outline the justification for the metaphor. Thereafter, the themes will be elaborated within the framework of this metaphor.

Navigating an Unpredictable Iceberg-Laden Sea (A Workplace)

To provide a fresh, third-order interpretation of how ESs’ employment support was experienced across studies, we elaborated on the well-known iceberg metaphor of culture [56] and put it into a new interpretive context. We found that this metaphor, which refers to a ship sailing in the icy North Atlantic Ocean, can portray the challenges that IPS clients and ESs often face in work integration processes. In considering this metaphor, one can imagine an ocean packed with icebergs. Since the risks of collision are prominent, the captain is on alert to steer the ship safely. However, to do so, the captain needs help from a trusted navigator who is in charge of planning the journey by continuously measuring the ship’s position, estimating the time to destinations and checking for potential threats in the sea. In addition, the navigator must ensure that the ship’s course-plotting maps and equipment are updated.

In this metaphor, IPS clients are like “captains” who are striving to steer their “ships” (job goals and work capacities) into “iceberg-laden seas” (unknown workplaces). To manage this vocational recovery journey, they need caring and well-informed navigators (ESs) nearby: professionals who hopefully communicate that a meaningful working life is in sight, who tell clients whether they are on the correct course for the trip, and who provide advice regarding the timing and pace of the journey. To act like “captains,” clients need help from accessible and trusted navigators to build the necessary self-confidence, drive and skills for a working life. They need support in learning how to build strength and trust their own abilities, even when they feel down; how to socialize; how to ask for help; and how to make their own decisions at work and learn from mistakes. When placed in unfamiliar jobs, it can be tricky for clients to identify the hidden and potentially harmful workplace “icebergs” and steer away to avoid crashing into them. In the workplace, where the demands tend to shift, it can be difficult for clients to understand what is happening because what can be seen in plain view at work is never the full story. Therefore, ESs must analyze not only the clients’ capacities for accomplishing the job but also the potential influence of the social dynamics and physical work environment on the client’s occupational performance.

Every workplace has various “icebergs.” The tip of the iceberg floats above the waterline and can easily be seen and recognized. However, as we know, this tip is the visible but smallest part of the iceberg, while most of the iceberg is hidden. Some layers of the iceberg appear and disappear with the tides, but the foundations are located deep under the surface. The same is true for workplace “icebergs.” The visible areas contain spoken and unspoken information: the strategies, procedures and routines that constitute people’s actions and appearances at work. In other words, these noticeable parts of workplace icebergs can be perceived by noticing the explicit and implicit cultural codes – the scripts for or expectations of, e.g., how to complete tasks, where to sit (and not sit) in meetings, how to communicate, and how to mingle and act at lunch. These invisible but major parts of workplace icebergs that embody unconsciously adopted practices are located below people’s levels of awareness. They are to a great extent concealed from the observer and to some degree are unarticulated; thus, they are difficult, but not impossible, to uncover.

Therefore, how can these partly concealed components of workplace icebergs be recognized? To capture these shifting levels of somewhat accessible information, the captain (IPS client) and the navigator (ES) together must be watchful of what is happening “at the waterline.” The parts of the icebergs that are exposed in these transitional zones are difficult to glimpse but can be observed by watching the transitional zones in the time when the sea tides are shifting (parallel to power fluctuations in workplace cultures). However, the fluctuating visibility of workplace icebergs is irrational and inconsistent, and one may think, “Now I can see it; oh no—now I cannot”! Furthermore, as these largest, unseen parts of workplace icebergs hover below the waterline (the workplace’s ethos and taken-for-granted norms), IPS clients need help to sense them. For example, employees know when their employer or colleagues treat them well or not. However, as the rationale behind such endeavors and interactions are hidden “below the waterline,” the actions are difficult to clarify and question. Hence, as many IPS clients tend to “crash their ships” into these hidden parts of workplace icebergs (i.e., leave or lose their jobs), they need helpful “navigators” who can prevent or remedy such occurrences.

Within the context of our meta-synthesis, we found that the IPS clients, the ESs and their supervisors interviewed focused on two major themes: (1) the nature of the relationship between the client and the ES and (2) the nature of the actual efforts entailed in the processes of seeking, obtaining, and keeping a job. Regarding the client-ES relationship, ESs were described as helpful in terms of the degree to which they were flexible, available, empathic, respectful, and positive motivators who were able to build encouraging and trustworthy relationships with clients, including by offering them personal and practical support. This helping relationship then served as a foundation for the various roles that ESs had to play from résumé preparation to the job search to the provision of ongoing job support. These efforts were described as competencies involving job development, an understanding of mental health problems and matters in the work environment, and thoughtful ways of bringing about work integration for this population. Successful ESs worked closely with clients to strengthen their job goals and skills, overcome barriers and negative beliefs, and made recommendations to employers to find useful work accommodations. With this framework now in place, we can focus more in depth on each of the following themes: (1)“It’s you and me looking”, (2) taking job seekers’ ambitions and needs seriously, (3) mapping the route, (4) exploring the hidden, (5) being on tap, (6) avoiding crashes, and (7) bridging.

It’s You and Me Looking

In various ways, the included studies described how ESs’ attitudes and ways of developing helpful relationships and collaborating with clients in the job-seeking phase influenced the quality of the job support. Of the eight studies [48,49,50,51,52,53,54,55] that reported ES and supervisor views, four [51,52,53,54] had an explicit focus on key qualities that differentiated high-performing from low-performing ESs. For example, it was revealed that while more successful ESs treated their clients “as equal partners,” less successful ESs were more “dominating” and not as aware of their clients’ preferences and needs [51, 52]. As one successful ES stated: “You know when you do a job search, it is you and me looking” [52]. In the same vein, two studies [53, 54] found that while higher-performing ESs built stronger working alliances with clients, lower-performing ESs described less confirming, reliable, and accepting attitudes and manners. As one ES said:

I could see that [the job seeker] didn’t really want to be there and so I was not going to waste my time with someone who was unmotivated and didn’t really want my help. I really just focus on those who are keen and ready to go [55, p. 52].

However, across the samples of IPS clients, ESs’ and supervisors’ various views and self-reported strengths of a typical high-performing ES, the following attributes were mentioned: a “close,” “sensitive,” open-minded, warm-hearted, and affirmative person [41 p. 592, 43, p. 403, 46, 48] who is “passionate,” “thick-skinned,” “persistent,” and able to understand his or her clients’ experiences, emotional and cognitive levels, and handle “rejection” [50,51,52]. One ES participant described his attitudes as follows:

“I wouldn’t call ‘em failures. Like the first job that they leave or get fired from. I wouldn’t call it a failure. Because it’s an experience they can grow from. And if you could point that out to them and find out what the good aspects of the job were, what the functioning aspect, aspects of the job were, that you could do successfully, that’s a skill. That you could build on” [51, p. 513].

Moreover, studies found that successful ESs were “optimistic” and “confident,” took “initiative,” were “hopeful,” “collaborative,” “team-oriented,” and “communicative,” and provided clients, current and potential employers, and mental health and SE teams with relevant information in qualified ways [51, p. 513, 514, 52, 53]. Less successful ESs were found to be more nervous and disconnected in their communication style [52]. The following collection of quotes illustrates how high-performing ESs often communicated with their clients: “What can we do to work on your confidence?”… “Tell me if I am being too pushy”… “When you decide to find a job, you have many things or qualities to offer an employer” [53, p. 213]. In the same vein, high-preforming ESs were found to put more energy into bonding with and supporting clients than low-performing ESs. As one high-performing ES said:

“When one of my clients stopped turning up to his appointments and seemed to be distancing himself from the whole idea of work, I visited him at home several times to convince him to get back into the program. I think going that extra mile made him feel valued and important for the first time in a long time….after that my clients seemed to have much more trust in me that he wasn’t just a number to me, but a real person and his attitudes changed” [55, p. 52].

This finding is echoed in most of the studies of IPS clients’ experiences [40,41,42,43,44,45,46,47]. Most of the IPS clients valued the job support they received; for some, the ES acted as an exemplar of the IPS model [41]. Notably, most of the studies [41, 43,44,45, 47] found that being treated as a person and as an equal were prominent elements of ES contributions. As one IPS client said: “Here, they see you as a human being. And that’s the difference. Things start to happen when you are treated like a human being” [47, p. 280]. Likewise, an ES said, “They [consumers] are just like us. They want a job. They want to do well, and we [ESs] are here” [52, p. 2014].

Moreover, several studies [41,42,43,44,45] found that ESs personalized ways of engaging clients in the job search made them feel like they were partners, creating “a sense of togetherness” [41, p. 592]. IPS clients appreciated that they were connected to only one ES [43, p. 403] and that they often met their ES in cafés, where they could “chat in a friendly and relaxed manner over a cup of coffee” [47, p. 280]. Many IPS clients described personal support as the most valuable feature of the program [44, 45]. They appreciated that ESs were empathically aware of their worries outside the work arena and that they sometimes chose to delay the job-seeking process because of such circumstances [41, 43, 45]. As one IPS client expressed: “The ES often puts her arm around me…it cheers me up” [41, p. 593]. Several participants contrasted the ESs’ professional behavior with how health and welfare service staff had treated them previously, experiencing treatment from the latter professionals as “stressful, impersonal and unproductive” [41, 43, p. 402, 47]. Moreover, two studies [43, 45] described IPS clients’ experiences of ESs as a means of boosting their confidence; even the simple presence of the ES in their meetings with authorities or in job interviews was felt as empowering.

However, several studies [40,41,42, 44, 46] reported some IPS/SE clients’ dissatisfaction with ES contributions. For example, some clients felt like they were “out of control” after the ES had pushed them into jobs they disliked [46, p. 655], while others described the ES as “too passive” or “too dominant and self-centered” [41, pp. 592, 593] and said that they wanted “more rigid structured job support or they could have done most of the work themselves” [44, p. 96]. Moreover, two articles [40, 42] noted that several clients with experiences of homelessness and criminal justice involvement spoke about “distrusting the advice of the ES” who “did not listen to their goals and opinions” and “did not possess appropriate job-seeking skills” [42, p. 23, 24].

Taking Job Seekers’ Ambitions and Needs Seriously

Unsurprisingly, consistent with IPS principles, studies across diverse samples found that ESs predominately used tailored and client-centered interventions. As one study noted: “job coaches tended to reject the notion of a typical client” [58, p. 413]. Similarly, a quote from one study of IPS clients’ experiences [41, p. 592] noted, “And that’s what so good… the ES doesn’t follow a schedule that tells you what you have to do and what not to do.” IPS clients felt that they were treated as partners (captains of their own ships) and that they were no longer as “a number in the system.” They valued the ES’s personalized job support and the focus on “suitable jobs” [40, p. 592, 41, 43, p. 403].

Similarly, many ESs interviewed in the studies underlined that clients’ personal barriers to employment, such as a lack of or fluctuating motivation or insufficient coping skills, could be overcome by tuning into their needs, negotiating their job wishes, and offering appropriate help [48, 57]. Consistent with the aforementioned ES explanations of client-centered approaches they used, one IPS client said: “He’s encouraging me to do, you know, apply for the jobs and he’s helping me when I don’t get the job. And also he’s a nice person to talk to, you just feel like you can talk to him about it” [44, p. 96]. Moreover, one study [43] noted that IPS clients talked about how their relationship with ESs had changed their minds and self-perception:

Before IPS, I didn’t expect that I would even attempt a job or school or anything. But that changed: I dared to start both a job and an education, and things changed even more, because now I believe I can do it, there are things I am good at. I’m not apologizing for myself anymore, I have something to offer (p. 403).

However, one study [48] highlighted that several ESs struggled to negotiate clients’ unrealistic job preferences while remaining client-centered. As one ES said:

Nobody wants the McDonald’s [job] and I don’t blame them. It can be kind of difficult sometimes because everyone wants that particular job, the office, and... we don’t say ‘you can’t do that,’ so we try and get them as close to that job that they want... so if someone said they want to be a pilot first, I’m not going to say ‘you can’t be a pilot,’ you know, but I will try and get them as close to the airport or the plane as [I] can [49, p. 526].

Moreover, it was reported that structural barriers, expressed as demands of “meeting numbers,” undermined the ESs’ client-centered practices, resulting in clients being placed in “old, not suitable jobs” [49, p. 527]. Four studies [51,52,53,54] found that while high-performing ESs actively advocated for clients’ abilities in meetings with potential employers, low-performing ESs tended to be more pessimistic and indifferent. One unsuccessful ES explained that she felt as if she were selling a “broken product” [53, p. 314]. Accordingly, the value of ES supervision has been highlighted in several studies [48].

Mapping the Route

This theme refers to how ESs’ knowledge of mental health and SE influenced their professional performance. Predominately, the studies [40,41,42,43,44,45,46,47] found that most IPS clients appreciated ESs’ ways of “navigating their ships.” For example, one study [44] reported that 28 of the 31 clients interviewed expressed being content with the support they received. Personal and practical support, such as help with résumé, interview preparations and job search, were the most useful components mentioned (p. 96). Most participants liked the planning of the job-seeking phase [41, 45]; as one IPS client said: “He understands where I am right now; if he has to do more or less, do I need more contact or less?” [43, p. 403]. Similarly, one study [49, p. 526] reported that client-centered ESs understood clients’ job requirements “by going beyond merely listening to what they said,” envisioning their reactions towards the particular workplace settings. Noticeably, across the various study samples [41, 44, 45, 51], the importance of the ES being accessible via mobile calls or SMSs and meetings in public places, cafés in the client’s neighborhood or the client’s workplace so that they could “map the job route” together was emphasized. As one IPS client said:

It is the best thing that ever happened to me. We meet once a week, or every other week. We sit down and have a cup of coffee and talk a little. It’s really pleasant, you feel you can relax and be yourself. No stress or pressure, but it allows you to concentrate and focus all your energy on what you yourself want [47, p. 280].

Regarding specific knowledge, across diverse study samples, it was found that ESs used various psychological interventions to help clients grow into the worker role, including empowerment models, cognitive behavioral therapy (CBT), motivational interviews (MIs), role-playing in job behavior, time management and communication skills training, and role modeling [47, 53,54,55]. These interventions were designed to align with clients’ physical, cognitive and interpersonal challenges [57]. Noticeably, both lower- and higher-performing ESs applied similar training techniques; however, it was noted that higher performers used additional, specific preparations and job coaching at the worksite more often than lower performers [54, 55, p. 53]. As two high performers explained:

I first showed the client how to perform the task and then got them to show me giving hints and suggestions as they went… I know people learn things differently—some are verbal people, and some are visual so I did the job for him chatting through why or how I was doing it, so he got both a verbal and visual explanation to help when he was learning the role. That was a trick I learnt from my supervisor [54, p. 25].

In particular, two studies [42, p. 284, 47] highlighted IPS clients’ experience of change and the relief of negative beliefs after their collaborations with ESs; as one participant with mental illness and a recent history of homelessness said:

[At] the beginning, I was very insecure because of my criminal record. I was telling myself “My God, they are going to ask me that question.” I was always afraid. For me, it’s been like three years that I flat-out refused to go to interviews because I was worried about being asked that question. And rejection and being refused by employers, me, I did not want to live that! You know? I would be too devastated. And then when I met my employment specialist she said “everything can be explained, and we can prepare for it, you will see...” she encouraged me… And when I saw how easy it was to pass interviews and that they were not pressing the issue, it encouraged me. I told my employment specialist that she was right [42, p. 23].

Exploring the Hidden

The studies reported the influence of ESs’ beliefs in vocational recovery and the IPS model on their ways of strengthening clients’ work motivations and advocating for their abilities in job development. The two following quotes illustrate how two ESs encouraged their clients: “When you decide to find a job, you have many things or qualities to offer an employer.” “The key is to keep you employed because that is the recovery” [53, p. 313, 314]. Notably, two studies [53, 54] highlighted that both higher- and lower-performing ESs recognized that clients’ mental illness and lack of work experience influenced their job motivation. However, it was found that higher performers spent more time addressing such issues. As one ES said:

The first thing I do is learn about client’s barriers to work, to employment, so they won’t get in the way of the client getting a job. So, with (this client) I ask her all about her previous work experiences and how she felt about getting a job. She was pretty forthcoming with all the stuff she was worried about… forgetfulness on the job, not getting along with a previous boss, not having worked for a long time… I spent much time talking about each of her worries and how we would get around them and ensure the same bad experience didn’t happen again [55, p. 52].

Likewise, several studies of IPS clients’ experiences [41, 43, 46, 47] explicitly noted the importance of ESs having expertise in mental health and relational skills in enabling vocational recovery. As one IPS participant said:

The ES realizes how I have been coping all these years and has knowledge about my difficulties. I do not have to worry about a lot of tricky questions… I don’t need to be afraid or show my claws as a way of defending myself [41, p. 592].

Being on Tap

This subtheme comprises descriptions of the ES being available and alert, ready to act at any time to navigate clients into the workforce. Several studies [48, 50,51,52,53,54] showed the influence of ESs’ ways of communicating with employers and mental health teams and their time management on the quality of job support. For example, it was found that successful ESs were more alert and proactive in their job development efforts and more efficient, but that they were flexible in their activities. Most of the time, they were out in the community to communicate with employers “face-to-face”; however, unsuccessful ESs acted in opposite ways [52, p. 205, 53]. Moreover, successful ESs networked during working and nonworking hours [51], focusing on “here and now actions,” as phrased in the IPS client studies [40,41,42,43, 47, p. 281, 48]. One IPS client said:

We were looking at a course that I wanted to go on and discovered that the deadline was the very same day. Lotta was on to it straight away: “Let’s go there at once!” She made some calls. “Where should the application be delivered? Where can we get hold of the necessary form?” … If she had not done all of this, it would never have got there in time, I can tell you! Then, they misplaced my application. She helped me to call round to those in charge and managed to find it, so I got the place. She accompanied me on the first day I was there, looked around a little and gave me support [47, p. 282].

Avoiding Crashes

IPS clients described a need to not feel cut off by the ES once they were employed [46, 47]. According to one study [57], more successful than unsuccessful IPS clients “received services involving on-site coaching, training (or skills acquisition prior to employment or outside of the job site), and participation in support groups” (p. 414). One ES study [53] revealed that both higher and lower performers described using a variety of job loss prevention strategies, emphasizing contact with employers and following newly enrolled employees to work on their first day and supporting them regularly during the first week. However, while higher performers maintained regular contact with employers during clients’ entire period of employment, this was not the case for lower performers [53]. As a high-performing ES said:

It is so much easier to save a job if you have a good relationship with the employer. That’s why I aim to regularly contact employers to have a quick chat about how things are going with the client. Firstly, to start with, this prevents small issues from becoming large issues that might warrant termination, but then if bigger problems do develop or a client is just not performing, often the relationship I’ve forged with the employer alone is enough to enable me to convince the employer to keep the client on or allow me to negotiate workplace changes. I’m sure that my good relationships with employers have saved numerous jobs over the years without even factoring in any change from the clients [54, p. 25]!

More precisely, it was found that higher performers made frequent and continuous workplace visits throughout the client’s job tenure, even when no problems arose [53]. As one high-performing ES participant explained:

“I see a lot of my colleagues just contact the clients in postplacement support and forget about the employers despite all the promises of support they gave them in the recruitment phase. Employers feel abandoned and if things start to go pear-shaped with the client such as poor performance, employers are more likely to just terminate them instead of working things out with the client and employment specialist. I think employer post-placement support is equally important to client support and saves jobs” [54, p. 24].

In contrast, the same study found that lower performers reported visiting the workplace primarily on the client’s first day or if problems arose on the employment path. They described that regular job visits could be viewed as a burden to employers; thus, they offered contact only when there were any problems. As one lower performer said: “Trust me, employers are far too busy to have to stop and spend time talking to me each week about one worker. Most don’t want me there and are happy to just call me if there’s a problem” [54, p. 24]. Importantly, the same study also reported that lower performers relied more on clients’ self-reported job performance than higher performers did, which is illustrated in the following two contrasting quotes:

Often, it’s only when a client calls that I find out they have been terminated. This really frustrates me as if the employer had have bothered to pick up the phone and let me know of the problems, something could have been done to save the job [54, p. 25] (lower-performing ES participant).

Often if you ask the client how they are going at work they’ll say ‘Great, no problems at all,' but then when you speak to the employer, it’s a totally different story and there are all manner of problems occurring and the client either has no insight or is too scared of telling me as they don’t want to let me down. That’s why it’s imperative to have frequent discussions with the employer too as that way I can really nip problems in the bud, before they become a bigger problem. That’s also why I like to visit the worksite so that I can observe the client perform their role and form my own opinion as well and see how I can help the client with strategies for improvement [54, p. 24] (higher-performing ES participant).

Moreover, regarding job retention, ES practices related to welfare benefits and disclosure were reported as important [53].

Bridging

Most studies found that flexible adjustments were associated with clients’ better prospects of obtaining and holding a job. In particular, one ES study [58, p. 413] reported that ESs put much effort into removing as many structural barriers as possible, such as offering clients help with transportation, childcare, affordable housing, and medical insurance. Moreover, ESs strived to ensure individualized accommodations for their clients at work: shorter shifts, frequent breaks, flexible schedule, frequent time off for medical appointments, or the presence of on-site coaching. Moreover, the studies noted that many ESs strove to “bridge” the gaps between workplace expectations, e.g., expectations regarding working hours, and the client’s capabilities. Notably, higher-performing ESs conducted more frequent initial on-the-job training and subsequent improvement coaching than lower-performing ESs, perhaps due to their higher presence at the worksite, even though the specific training strategies they employed seemed similar [53]. As explained by a high-performing ES:

I always try to get trained in a client’s job alongside them because that way, if there is a problem down the track, I don’t have to bother the employer with discussing techniques on how the client can improve…. I already know the job and how best to perform it and go straight into conducting additional training with the client to improve their performance. There have also been occasions where my clients have left unwell and I’ve stepped in for a few hours and performed their role for them so as the employer would not be left hanging! I’ve done car detailing, cleaning, cooking hamburgers—you name it [54, p. 25]!

In contrast, the same study reported that lower-performing ESs tended to drop such follow-up, explaining that employers wanted to provide the training themselves or that they “did not have time to provide this kind of intensive post-employment support” [55, p. 25].

Discussion

This meta-ethnography notes several aspects that can advance the understanding of ESs’ core contributions in helping people with SMI find and retain competitive employment and facilitate important discussions about useful job support activities. First, our overarching metaphor of “navigating an unpredictable iceberg-laden sea” suggests that successful ESs are perceived as being able to intuitively detect the employment barriers that IPS clients face on their paths to workplaces, embolden clients to overcome the perceived barriers and help them find ways of displaying and upholding their work goals and skills (“securing their ships”). Importantly, as noted in previous research [12], the ES role involves not only reducing such barriers but also bolstering clients’ abilities to address them. The same way a captain of a ship would rather not enter an unpredictable iceberg-laden sea without assistance from a well-informed navigator, many people with SMI will choose not to try to find a competitive job on their own. The stakes are very high. People with SMI must deal not only with challenges related to the illness and a lack of knowledge about working life but also the fears of being stigmatized and discriminated against at work; furthermore, they must do so in the absence of social and practical support. These environmental challenges are embodied in what we have labeled the invisible parts of workplace icebergs. Some people with SMI have difficulty identifying the range and size of the dangerous parts of workplace icebergs, which makes it almost impossible for them to navigate these waters alone.

In other words, our metaphor illustrates the importance of ESs who are competent in analyzing IPS clients’ work performance difficulties related not only to the individual’s illness, age, gender or cultural related challenges but also to psychosocial, behavioral, and environmental workplace factors [58]. As described in the person-environment-occupation model in occupational therapy theory [59], a person’s limitations in work performance may depend on their personal capacities, environmental components, and the activities involved in their occupation [60]. Consequently, as key persons in IPS practice, it is important that ESs analyze their clients’ work performance holistically in light of the dynamics between their abilities, work activities, and environmental demands [60]. As noted in the vocational rehabilitation literature (see, e.g., [12]), work accommodation may be crucial to helping people with SMI keep their jobs, but as such implementation requires changes in current social practices (which involves a risk of crashing into the invisible parts of the icebergs), the ES must examine existing social interactions and support in the workplace.

Second, in line with previous research [7, 26, 28], this meta-ethnography indicates that trusting bonds between ESs and their clients who are seeking competitive employment need to be forged. As has long been found in other helping relationships such as psychotherapy [61, 62] for many people with SMI who experience some of the barriers described above, competitive employment might become a possibility only in the presence of such a trusting and personally supportive relationship. More precisely, the theme “It is you and me looking” shows how a successful ES treats clients as equal partners when they go out together to find work. Such ways of collaborating with and empowering IPS clients as peers demonstrate that ESs in other ways depart from normal rules and procedures in traditional clinician–client relationships, and by doing so, ESs engage and motivate their clients, which is also known to be key for successful employment. Bearing in mind the important process of recovery through occupational engagement, such navigation skills are crucial. In other words, to help IPS clients obtain and keep jobs, ESs have to provide them with meaningful and exploratory routes into workplaces.

As noted in other mental health professions, the working alliance in this case includes an agreement between the ES and the client about goals, the tasks required to reach the goals, and the bond developed between the two parties [26, p. 280]. Specifically in the employment literature [12], ESs’ ability to build a strong working alliance with their clients is described also as grounded in their capacity to promote hope for a better vocational future, empower individuals’ vocational abilities, encourage individuals’ self-acceptance of their vocational strengths and histories, and help individuals generate a sense of worker identity—all of which are known to be keys for vocational recovery [28, 63]. Our findings indicate that, as noted in previous research, training programs to help ESs enhance their understanding of the working alliance with their clients and improve their communication skills to create more trusting relationships with mental health professionals and employers improve client engagement and reduce environmental stigma [8, p. 535]. Moreover, as ESs’ strategies for managing conversation about disclosure are less clear in the IPS literature, so are their strategies for preparing clients who lack work experience, relevant skills and self-confidence, ES training in countering such needs are recommended. Our subtheme “taking job seekers’ ambitions and needs seriously,” which is grounded in client-centered interventions, defines one possible direction to pursue.

Third, our findings indicate that communication with employers, illustrated by the subtheme “being on tap,” is another important component of ESs’ contributions to job support. Likewise, we found that IPS clients emphasized the need to continue to be linked to the program, as well as receive on-site coaching and training, after finding a job, which could allow them to potentially “avoid crashes,” thereby increasing the effectiveness of IPS. Not surprisingly, previous research has identified support from an ES as the most commonly used workplace accommodation by people with SMI [23]. To effectively fulfill this role, small caseloads for each ES are imperative. In working with clients at the jobsite, after clients are employed, the ES can help bridge gaps between their clients’ capabilities and workplace expectations while helping employers better understand and address the necessary accommodations. In addition, as it is known that contact is more frequent during periods of employment than during periods of unemployment [22], our theme also illustrates that ESs should be “on tap” for clients after a job loss.

Methodological Strength and Limitations of this Study

An important feature of this study is our multidisciplinary team approach and the fact that we as a research team come from different continents, so our synthesizing was influenced according to professional disciplines and cultural background. However, all reviewers have attempted to be trustworthy to the perspective and practice of reflexivity by critically reflecting on their own steps in the research processes [64]. There remain several limitations to be considered in interpreting the results of this meta-ethnographic study, such as comprehensibility of the search, the number of included studies, and their qualities. For example, due to poor indexing of qualitative studies, some important studies might have been missed. However, combining different search strategies in two different phases should have minimized this possible bias. A meta-ethnographic study includes processes of deep analysis; thus, we considered the included 16 studies to be a convenient size in the number, substance and scope of our sample. Moreover, limitations to be mentioned are that the studies included differed from each other regarding being conducted in seven different countries (Sweden, Denmark, UK, USA, Canada and Australia), on three different continents, in unlike mental health care and vocational services; aspects that all together may influence the contextual circumstances for the employments specialists’ contributions in supported employment. As well, this meta-ethnographic study involved synthesizing findings from a variety of epistemological positions and methods. Working as a team allowed us to examine this heterogenous group of studies from various perspectives and develop consensus on our interpretations and synthesis of findings by rich discussions.

Conclusion

We chose the iceberg metaphor for this study because it condensed our interpretations of the relevant research findings and the essence of ESs successful and unsuccessful support in evidence based vocational rehabilitation programs.

Our iceberg metaphor illustrates the importance of ESs being competent in addressing clients’ work performance difficulties related not only to their individual, illness-related factors, but also to more hidden psychosocial, behavioral, and environmental workplace factors. Given both the effects of a working alliance in general across mental health specialties along with the specific characteristics of the working alliance developed in SE, we suggest that employment specialist training and supervision be enriched by paying more attention to these important relational processes.