How Do “Mental Health Professionals” Who Are Also or Have Been “Mental Health Service Users” Construct Their Identities?

“Mental health professionals” are increasingly speaking out about their own experiences of using mental health services. However, research suggests that they face identity-related dilemmas because social conventions tend to assume two distinct identities: “professionals” as relatively socially powerful and “patients” as comparatively powerless. The aim of this study was, through discourse analysis, to explore how “mental health professionals” with “mental health service user” experience “construct” their identity. Discourse analysis views identity as fluid and continually renegotiated in social contexts. Ten participants were interviewed, and the interviews were transcribed and analyzed. Participants constructed their identity variously, including as separate “professional” and “patient” identities, switching between these in relation to different contexts, suggesting “unintegrated” identities. Participants also demonstrated personally valued “integrated” identities in relation to some professional contexts. Implications for clinical practice and future research are explored. Positive identity discourses that integrate experiences as a service user and a professional included “personhood” and insider “activist,” drawing in turn on discourses of “personal recovery,” “lived experience,” and “use of self.” These integrated identities can potentially be foregrounded to contribute to realizing the social value of service user and other lived experience in mental health workers, and highlighting positive and hopeful perspectives on mental distress.


Introduction
According to St. Claire and Clucas (2012) "individuals actively manage multiple identifications in ways that can have paradoxical implications for their health and wellbeing" (p. 86). This hypothesis is derived from social categorization theory, which holds that our identification with a particular group guides our behavior, increasing our conformity with that group (Turner, Hogg, Oakes, Reicher, & Wetherell, 1987). Jetten, Haslam, and Haslam (2012) have gathered evidence that identification with illness groups (which can follow diagnosis) affects people's behavior and attitudes, and can increase the salience of symptoms and reduce self-esteem through belonging to a stigmatized group (St. Claire & Clucas, 2012).
The findings of Leamy, Bird, Le Boutillier, and Slade (2011) about recovery in the context of severe mental health diagnoses are consistent with this concept of identity. From their extensive review of relevant qualitative literature, Leamy et al. (2011) reported that connection with other people and a valued identity were two of five key factors that people with diagnoses of severe mental health conditions talked about as constituting recovery. Identity referred not only to gaining or regaining a positive identity but also offloading a stigmatized one of "mental patient." Perhaps it is surprising, therefore, that growing numbers of "mental health professionals" with "mental health service user" experience are speaking out about their mental health service use. They are doing this verbally and through professional journals and autobiographies (e.g., Ahmed, 2007;Bassman, 2001;Beresford, 2005;Burling, 2005;Clarke, 2012;Deegan, 1987;Fox, 2002;Frese, 2000;Friedman, 2004;Harding, 2005;Kottsieper, 2009;Lemelin, 2006;Linehan, 2011;MacCulloch & Shattell, 2009;May, 2000; enfranchising the experience of mental distress. In the United Kingdom, this is perhaps further supported by the Health and Care Professions Council (2014) guideline that training of mental health professionals should involve service users as contributors for their relevant lived experience.
However, there is limited research exploring the impact of recovery-orientated care, with some researchers suggesting that the medical-psychiatric model still dominates (Hui & Stickley, 2007;Perkins & Slade, 2012;Warne & Stark, 2004). In relation to English mental health services, Roberts and Boardman (2014) have highlighted the continuing tendency for professionals to fail to work collaboratively with service users in managing risk, despite concern over several years about it from the Royal College of Psychiatrists (RCP;. Perkins and Slade (2012) suggested there is still too much emphasis on recovery "from" an illness, implying cure or containment, rather than recovery as "a way of life" (p. 13) as elucidated by Deegan (1987).
Therefore, service user professionals may face dilemmas when constructing their identities because within the medical-psychiatric model professional and patient identities are still often viewed as fundamentally different (ResearchNet, 2011). According to McCourt (1999), professionals are viewed as beyond distress and therefore to be a service user involves a radical role reversal (Rucinski & Cybulska, 1985). Shepherd et al. (2009Shepherd et al. ( , 2010 suggested that professionals can be unwilling to move away from their traditionally accepted "expert" role and that they hold negative attitudes toward service users and also toward professionals who are service users (Barrett, Pratt, Basto, & Gill, 2000;Basset, Campbell, & Anderson, 2006;Fisher, 1994;Hossack & Wall, 2005).
Such attitudes are apparent in much of the literature on professionals with service user experience: "Distressed professionals" are said to experience "impairment" (Sherman, 1996), and their service use therefore connotes a deficit more than an asset; "wounded healers" are those whose illness has left lasting effects on them, which is used with future clients (Jackson, 2001), and this seems to carry both positive and negative connotations. Potentially, the wounded healer is viewed as "not quite professional" There is a relatively new category of service users in recovery employed as "peer support workers" in mental health services (Perkins, Rinaldi, & Hardisty, 2010), which arguably carves out a special and positive role for these service users, valued for their ability to understand the experiences of other service users and to act as mentors in the recovery journey. However, research suggests some difficulties for peer workers in either retaining a separate and valued identity as service users, or taking on the identity of "professional" (Dyble, Tickle, & Collinson, 2014).
If the understandings of "service user" and "professional" tend to remain very different, then this presents a dilemma as to how professionals can talk about and openly value their service user experiences. Postmodernist research focuses on language and sees "talk" as an important aspect of identity construction (Davies & Harré, 1990;De Fina, Schiffrin, & Bamberg, 2006). Identity is seen as fluid and continually coconstructed and negotiated in social contexts (Davies & Harré, 1990;Potter & Wetherell, 1987).
However, socially understood identities can also become part of "taken-for-granted" discourses that are continually recycled and hard to modify because of their constant renewal in social interactions that enact the assumed social legitimacy or lack of it attached to a given identity. Benwell and Stokoe (2006) suggested that discourses-social episodes of speech occurring in everyday life (Harré & Langenhove, 1999)-regulate social power through socially constructed identities with associated social positions as more or less influential. The postmodernist methodology of discourse analysis (DA) has been suggested as useful when exploring mental distress because it allows us to question "taken-for-granted" understandings, takes a nontraditional research strategy, and values human and theoretical diversity (Harper, 1995). It therefore provides a useful basis for exploring the identity construction of professionals who have been service users, and the social power or lack of it that is enacted as part of such constructions.
Three empirical studies have explored the identity construction of service user professionals from a social constructionist perspective. Adame (2011) explored the identities of 11 self-described psychiatric "survivor" psychotherapists, defined as people who have survived abuses within the mental health system. Participants had strong "survivor identities" and felt their two identities ("survivor," "professional") complemented one another. However, regarding disclosure, they also spoke about experiencing stigma, isolation, questions regarding their competence, and not being seen as a "real" survivor when disclosing their "professional" identity to other psychiatric survivors. While drawing on the existing discourses of "wounded healer" and "impaired professional," this study highlights alternative contexts in which the "survivor" identity is positive. Yet again, it highlighted "them-and-us" dynamics, in which survivor professionals felt incompletely accepted in either the professional or the survivor movement contexts, as each is often positioned in opposition to the other's perspective. Adame (2014) presented a follow-up to her 2011 study, the focus in the 2014 report being the transition from service user to therapist identity. Five people were interviewed in depth and the key issues illustrated with the experience of one participant. The lack of offer of any human connection during hospitalization had left the participant wanting to create a better experience for others. After training as a counselor, the participant saw similarities between the survivor movement and humanistic psychology, with both valuing human connection. However, he set up an alternative practice. Integration of his survivor and professional identities seemed to be on the basis of shared values that may not sit easily within mainstream mental health services. This highlights splits within professional identities between medical-technical and humanistic-relational philosophies (Pilgrim, 2009). Joyce, Hazelton, and McMillan (2007) explored the workplace experiences of 29 nurses with mental health problems and found that participants' identity shifted from "nurse" to "patient" as they were identified as mentally distressed. They spoke about the mostly negative attitudes of colleagues; in hospital many felt the "patient role" was imposed upon them, and they were sometimes treated disrespectfully. Most participants spoke about their professional identity as not conforming to normality and therefore engendering discrimination. Again this seems to draw from the existing discourses of "impaired professional." Studies using more traditional epistemologies have reported that professionals with service user experience encounter dilemmas (Berry, Hayward, & Chandler, 2011;Cain, 2000;Charlemagne-Odle, Harmon, & Maltby, 2014;Gilroy, Carroll, & Murra, 2001;Stanley, Manthrope, & White, 2007;van Erp, Hendriksen-Favier, & Boer, 2010). These include issues regarding self-disclosure, experiencing stigma, prejudice and discrimination, being accused of overidentification and boundary violation, having reduced energy, confidence and emotional presence, and a discrepancy between their personal identity and their professional role. Participants also felt that their service user experiences enhanced their work, including useful self-disclosure, and increased emotional empathy and insight, the ability to hold hope for clients and experiencing some positive reactions from colleagues.
It could be suggested that difficulties in relation to these professionals' social identities arise because people are trying to construct an identity drawing on different discursive resources with different power implications (Parker, 1992;Sampson, 1993). Therefore, they are drawing on contradictory discourses (Billig, 1991;Billig et al., 1988) making their identity difficult to integrate. However, these studies have methodological limitations. Only three documented their analysis, leaving unanswered questions regarding the quality of the findings. Social constructionist analyses were rare.

Rationale
Despite growing numbers of mental health professionals disclosing their mental health service use, understanding of how they construct their identity is limited, with a lack of awareness of the influence of social structures and dynamics. Research suggests a range of dilemmas, but there is limited research that explores how people manage these and how this affects their identity construction, especially in the United Kingdom.
The aims of this study were to explore how a sample of U.K. service user professionals construct their identities, within a social constructionist epistemology, extending existing research taking into account some of the methodological limitations. By exploring how people talk into being their identities, we can explore and describe the range of identity constructions available to people, how they position the speaker and what they achieve for the speaker. It seemed timely to investigate this due to current policy calling for "lived experience" in mental health professionals and would allow for any new discourses to be systematically documented and heard more widely. This could have implications for practice in relation to the personal and professional development of service user professionals.

Research Questions
The study addressed the following research questions; Research Question 1: How does a sample of mental health U.K.-based service user professionals construct their identity in relation to different professional contexts? Research Question 2: What discursive resources are available to them when constructing their identities, exploring the wider discourses that these constructions are drawing on? Research Question 3: Is it possible to identify newly emerging discourses? Research Question 4: For each kind of construction identified, the researcher explored the social consequences of constructing these identities: How it may "position" them, facilitate, or constrain social action, and the relationship between the discursive resource and subjective experience?

Participants
The study included 10 participants (five male, five female; seven White British, two Asian, and one South African). Participants were from different professional disciplines (art therapy, nursing, management, social work, peer work, psychology, occupational therapy, and psychiatry), had all used outpatient and/or inpatient mental health services, for difficulties identified by them as including depression, suicidal ideation, paranoia, schizophrenia, bipolar disorder, psychosis, anxiety, and bulimia. Participants had different time scales since their diagnosis or last "episode," ranging from 15 years to 1 year. Eight participants considered themselves to have ongoing difficulties, with one participant currently taking antipsychotic medication. Eight participants were currently working; four full-time, one part-time, and three participants' working hours were unknown. Two participants were currently not working. Four of the participants were service users before becoming professionals and six were professionals before they were service users. Four were members of a service user researcher group embedded in a National Health Service (NHS) mental health trust. This group had an ethos of improving services by being curious about oneself and the context, and learning research skills together in an atmosphere of mutual support.

Design
DA allowed for a detailed exploration of the discourses available and used by participants, and how these constructed their identity in relation to different professional contexts. Interviews allowed for in-depth discussion and diverse forms of expression. The data were analyzed using a combination of the two main DA approaches: Foucauldian DA and discursive psychology. This enabled attention to be paid to both the negotiation and construction of meaning through language in everyday social interaction, achieving interpersonal objects, that is, identity (Wetherell, 1998;Willig, 2008) and the consequences of wider social and institutional discourse that are available for people to draw upon (Potter & Wetherell, 1995;Willig, 2008).

Procedure
Ethical approval was gained from Stanmore Research Ethics Committee. Participants were recruited using an advertisement in the United Kingdom's NHS premises, service user groups, recovery networks, and a trainee clinical psychology cohort. Participants were asked screening questions to check that they met the inclusion criteria. Participants identified themselves as being professionals (currently working or having worked in a professional caring role within the United Kingdom's NHS, voluntary sector, or privately) and service users (having felt they had no choice but to use mental health services at some point). Participants had to be employed within the last 18 months to be able to talk about their experiences in a professional context. Those with communication problems, non-English speakers, or in a crisis were excluded.
Interviews involved gaining written consent, conducting a semistructured interview, and debriefing. Interview topics included questions related to professional role, service user experience, being a professional with service user experience, professional contexts, and how participants and others talked about who they are. Interviews lasted on average 1 hr and 28 min, were recorded, transcribed, and analyzed.

Data Analysis
The transcripts were analyzed using the methodology recommended by Willig (2008) to 1. Find the discursive objects in the text-Transcripts were read and reread to find times when the participants referenced themselves or their identity. 2. Explore the discursive objects' construction-When participants referred to themselves, these constructions were explored to see how they were formed, their variability within the different contexts, and whether there were any conflicting discourses. 3. Situate the discursive constructions within wider discourses-Participants' identity constructions were explored to see which wider discourses their constructions seemed to draw on. 4. Explore the function, consequences, and implications of the discourse-By asking, What is/is not gained? What does it allow/not allow? What is its function? What is it in response to? What does it emphasize? How does it position the speaker/others and does it enable/disable social action? 5. Explore the relationship between the discourse and the subjective experience: For each identity construction, what could be felt, thought, and experienced within the construction?
To recognize extant discourses that participants may draw upon, the authors studied the writings of 25 people self-defined as mental health service user professionals about their experiences in their professional sphere. This would help to validate the analysis of the interviews and increase the potential to identify any novel discourses. However, care was taken not to impose extant discourses on participants' talk. Other measures taken to ensure the quality of the results (Mays & Pope, 2000) included keeping a research diary, discussing expectations and responses to the data and interrogating our initial readings to look for alternatives, and carrying out separate readings on some transcripts and then conferring. One of the authors (S.H.) self-identifies as a professional with service user experience and drew on this experience during the research.

Results
All participants seemed to draw on, presented, or had the separate identity constructions of "professional" or "patient" imposed upon them, switching between them, and demonstrating an "unintegrated" identity. All participants also demonstrated what appeared to be an "integrated identity," where their way of talking drew on experience as a service user and a professional simultaneously.
Participants used a range of discursive resources and presented their different identities to differing degrees depending on the context they were discussing. This section will present participants' "professional" and "patient" identity constructions and then their "unintegrated" and "integrated" identity constructions. All names have been changed to protect anonymity.

Separate Identities-"Professional" Identity
There seemed to be four discourses within the overarching one of professional identity: "shared language" (comprising "psychological" and "psychiatric-medical"), "professional power," "achievement-academic," and "accepted versus disallowed ways of being." Participants spoke about their disciplines, role, and jobs and had a sense of belonging and identifying with a particular professional group: Participants drew on a "shared language" discourse when talking about their profession without clarifying psychological or psychiatric-medical terms, as if aware that they were talking to another professional: Professionals were viewed as knowledgeable, valued, useful, competent, and responsible: "professional power." They were viewed as "experts" and listened to by others, leaving themselves feeling wanted and special: My identity as a [professional role] was I am superior. (John) A highly qualified professional . . . very competent. (Jo) Within this position professionals hold power over those they treat. Participants also spoke about hierarchy within the professional sphere that affords senior medical professionals more involvement in decision making, leaving those less senior feeling relatively devalued: Nursing staff in particular are very closed down by their hierarchy traditionally. Some get in places where they can operate but . . . they are quite threatened and ruled by fear. Participants drew on an "achieving-academic discourse": To be a professional you have to be academic and have significant achievements, adding to the idea that professionals have expertise, which affords them authority: I've done a lot of training myself, um professional development work. (John) I went to university and studied [subject]. (William) For some this was a demanding position because there was a sense of imperative to achieve. This led participants to fear being judged and feeling not good enough: Participants drew on a "discourse of expected versus disallowed ways of being" as a professional. Participants spoke about it not being acceptable to become angry as a professional because you were expected to be diplomatic, work within policy restrictions, and comply with the identity of a professional. Although having these restrictions left participants feeling relatively powerless, it could be seen as a way of protecting or maintaining their professional identity: If I was in a meeting I would put on the professional face. (Chris) If I am in a meeting with a very senior psychiatrist, I will keep my gob shut and almost pretend to go along with their stuff.
All participants drew on this construction as a professional, particularly when talking about working within professional contexts. As illustrated by the more subversive tone of William, a few participants spoke about rejecting their "professional" identity, possibly as a way to make sense of all their experiences. However, this meant that they were no longer identifying with the relatively powerful majority and instead risked being viewed as less competent: I'm unusual in that . . . I publicly say that I have recovered from all my academic and professional trainings. Some people find that funny. Some professionals get angry when I say things like that particularly at conferences . . . they will say, "What's your background young man?" (William)

Separate Identities-"Patient" Identity
Participants constructed a "patient" identity, and again several discourses were drawn upon: "professional control," "personal recovery," "antiprofessional," "survivor," and "needing labels." There was a discourse of "professional control" that someone in a position of authority needs to tell people they are unwell, label and treat them: My supervisor um, noticed that um, I was unwell. (Helen) The GP um, gave me, started getting me on pills, antidepressants and labeled me with depression. (Chris) Constructing the "patient" identity in this way suggests that those occupying this position are relatively powerless, have fewer rights and choices, and must comply with care directed by those in control. Individuals within this position may hold negative views about themselves and feel hopeless about the possibility of change: I was completely disempowered and not believed on the basis that I was ill. (Claire) Everything I did when I was sitting totally in the service user role and identity didn't have value. (Jo) This discourse of control by professionals echoes Deegan (1987) who experienced the imposition of "schizophrenia" (p. 4) as a self-characterization, and Bassman (2001), who wrote of learning to become "a good hospital patient" (p. 13).
Within the "patient identity" construction, participants of the present study also drew on a discourse of "personal recovery": Massively surreal, kind of journey, but it has made me determined. (Anna) I sometimes struggle with my own anxiety. The whole thing isn't going to die overnight. (William) Drawing on the "personal recovery" discourse appears to construct a more positive identity because it acknowledges ongoing difficulties, but reframes it as a journey rather than a "stuck" position. By constructing it in this way, service users have gained the chance to recover, and can be viewed as more capable and have more power to be involved in their care, increasing the individual's autonomy and self-esteem.
Within the "patient" identity construction, participants drew on an "antiprofessional" discourse. Participants spoke about disagreeing with the medical model, the labels and treatments they had received, and with professional practices and service structures: It's all this mental illness is a physical thing . . . chemical imbalance. Show me the brain scans of these people. Show me that their chemical imbalance has been tested and that they're being treated for their chemical imbalance. You can't do it. Participants drew on an "antiprofessional discourse" and used the word "you" as if using rhetorical devices specifically aimed at the interviewer: Not those crappy words that you use. Just human conversation, like I don't do all this, "You sit here, I sit there" business. (Chris) Taking this opposing position means that either something new can be created or, especially if a minority opposes established professional power they risk being ignored or dismissed, recalling the discourse of "expected versus disallowed ways of being a professional": Participants who drew strongly on an "antiprofessional" discourse appeared to feel as though they had survived the mental health system. They drew on a "survivor" discourse and some developed a "survivor" identity: Even though participants drew on an "antiprofessional" discourse and disagreed with diagnosis, they also drew on a discourse of "needing labels," in the sense of needing words to describe and understand experiences, which perhaps suggests an absence of accessible alternative discourses for understanding mental distress: We do need to have some kind of . . . words to describe things.
In the past I have felt reluctant to use those kinds of words [anxiety, depression] because I haven't felt like I've earned them. (Sarah) Rather than presenting an alternative construction of mental distress, some participants simply rejected the "illness" discourse and diagnostic labels. They did adopt an existing identity discourse (e.g., survivor), but without apparently reconstructing mental distress itself: I didn't accept that I had a mental illness and that was what was wrong with me. (Chris) Why do I have to accept a label that someone else is putting on me? They do feel like outside labels which don't fit me or the identity I was to forge. (Jo)

"Unintegrated" Identity
Participants spoke about finding it difficult to be both "professional" and "patient." Throughout the interviews, participants drew on, presented, and had the separate identity constructions imposed upon them in a way that they were "either/or," depending on the contexts discussed. This meant that participants moved from one to the other or switched between them, developing what seemed like an "unintegrated" identity construction: You've got your professional hat on. You can switch into your user hat. (Claire) The service user part, the psychotic part, the confused part and whatever part somebody else tells me. (John) One participant noted that he felt good about having two "hats," but for the majority of participants it seemed problematic. Within the "unintegrated" identity construction, participants drew on the discursive resources described above. However, because these constructions of "professional" and "patient" are usually so different, they do not come together easily, therefore dilemmas occurred leading to unintegration. Participants spoke about a "them-and-us" divide, with either the "professional" being the "good" identity and "patient" being the "bad" identity, or vice versa, as the reverse position can occur when service users collectively hold a "survivor" identity, drawing on an "activist" discourse with other service users (Adame, 2011;Weltz, 2003 Having had experienced some of the darkest and deepest forms of distress, a mad man versus someone, got a reputation for being highly professional, they're worlds apart unfortunately. (William) If the discourses of "professional" and "patient" are very different, one cannot easily occupy both positions. In the professional world, it seems to imply "impaired" professional rather than professional with added value. This was represented by participants talking about others assuming them to be unable to do their jobs while unwell. Some felt they lost their "professional" identity when they became unwell, as documented by Joyce et al. (2007). Some held the view that mental health services were not well equipped for professionals with mental health problems; some felt that professionals, including themselves, held negative views about such professionals; and some felt that awareness of their existence was lacking. This also recalls the discourse of "expected versus disallowed ways of being a professional": There is nowhere to be if you are feeling fragile. The message is don't come into work because if you can't cope with what's in front of you, you have no business being there. (Paul) They might judge me and say, "Oh you shouldn't be seeing clients." (John) In terms of anybody acknowledging my experience as anything valuable was nonexistent. (Jack) Participants seemed to draw on conflicting discourses that are dilemmatic, that is, drawing on an "antiprofessional" discourse while identifying as a professional. Participants spoke about this being a hard position to occupy and felt that they were viewed as different, and felt judged and powerless. They spoke about having to hide certain parts of themselves in certain contexts, and sometimes being pushed to behave in certain ways: I think being a health professional and being ill, a service user . . . is probably . . . the hardest thing. (Paul) Why am I being forced to either go professional or service user or carer? (Jo) Within this "unintegrated" identity construction, participants spoke about the service user movement, drawing on a "service user movement" discourse (Frese & Davis, 1997). Participants spoke about valuing, supporting, and admiring the work of service users, and about their own involvement from within their "patient" identity: Drawing on the "service user movement" discourse could be viewed as a radical position, opposing the discourse of "professional power," and runs the risk of being shut down by those who are more powerful. However, it could also lead to new valuing of "lived experience" and allows service users to speak where they may have been previously silenced.
Participants also drew on a "use of self" discourse, but within this "unintegrated" identity the "professional" identity construction did not allow service user professionals to disclose their mental distress to their patients as this would be "unprofessional": Participants did not disclose their mental distress within their professional contexts, only speaking openly to a few colleagues or within a particular context when they felt safe to do so: Some people know about my um, experiences of being unwell and being in hospital, some people don't. I'll do it in a way that I feel safe. (Anna) I was told . . . by a couple of colleagues . . . you never disclose this again, you don't talk about this to your colleagues. People take advantage. (Jack) Participants had to hide part of their identity, developing "unintegrated" identities. This is reminiscent of Bassman (2001) and Linehan (2011), both of whom waited until late in their professional careers before disclosing their mental distress, and Zerubavel and Wright (2012) who suggested that one's competence is less easily challenged by disclosure once it has become well established. Participants of the present study were aware of other service user professionals who did not disclose: I knew that there were other people . . . Later on I realized that she was an anomaly and not everyone else did share, even though they kind of professed to. (Sarah) For some there was also a sense that if they disclosed their "professional" identity with a group of service users they were viewed as not being "real" service users, as was highlighted by Adame (2011) All participants seemed to demonstrate an "unintegrated" identity, especially when they felt uncomfortable or uncertain of themselves. They seemed to be drawn into different positions by those around them, leading to the switching action between their separate identities. Participants felt that a more "integrated" identity would be better: I have got problems because I haven't managed to integrate fully both roles and feel comfortable in different environments. (William)

"Integrated" Identity
Participants also constructed what seemed to be an "integrated" identity, but for some this was less formed than the "unintegrated" identities. Within this "integrated" identity, participants were able to draw on all their experiences together, viewing them as complementary. For some, this seems to also encompass a new sense of self, which perhaps could be viewed as a "personhood" discourse: I don't feel that I need to demarcate, differentiate. I think both my roles . . . my experiences have contributed to me becoming who I am. (Helen) What makes me "me" is a combination of all of those things. (Anna) Within this "integrated" identity construction, participants drew positively on a "wounded healer" discourse. They spoke about going into the profession because of their own experiences of distress or caring and used their professional knowledge to make sense of their own experiences and when accessing services themselves: I think a lot of people go into mental health because they have a personal affinity um, for it anyway because of their um, insights of their own thoughts. (Claire) I think that [a continuum model] was a useful way of understanding some of the difficulties I had had in the past. (Sarah) Some participants felt that it was important for those working in services to have experienced some level of mental distress to fully understand or be able to work with service users: This combination of "lived experience" and "use of self" discourses made the integrated "wounded healer" identity more efficacious than that of other professionals, as one could provide better care: I've got a lot of compassion for people because I can recognize that these thoughts that they are believing . . . I can really support them in this. I'm not scared to work with really disturbing thoughts of a client, because I had pretty disturbing ones myself.
In fact they [service users] come to you and talk to you even more, they are more open to you than they are probably to the nurse or psychiatrist. (Jack) This echoes the "prosumer" discourse elucidated by Tsai (2002) and Yarek (2008), which highlighted friendship and reciprocity with service users while also fully in the professional role. Unlike the "unintegrated" identity construction, "use of self" in the "integrated" identity was open in that participants were able to disclose their mental distress experience, linked to a "coming out" discourse. However, there is still a tension, in relation to how others perceive service user professionals, and participants spoke about receiving a mixture of positive and negative responses when disclosing: So that [writing a paper] was in a way me coming out. (Helen) It's been over quite a long period of time. They were like, ok, yeah that's fine. (Paul) They didn't know what to say and they are psychologists and psychiatrists. It was really awkward. (Sarah) It's opened up doors for patients . . . to be able to return something to me like, "Oh how are you in yourself?" and "I understand you." They feel validated. (Claire) It might be suggested that participants could view their lived experience as placing them above other professionals. Yet, they spoke about their identity as being nothing special, drawing on a "normalizing" discourse. This meant that the "them-and-us" barriers set up by more traditional "professional" and "patient" identity constructions could be eroded because mental distress could now be viewed in a more ordinary light, and again a "personhood" discourse is drawn upon: It's just me, it's just who I am. Drawing on these discourses enabled participants to draw on both their personal and their professional knowledge while in their professional role in an integrated way, allowing for new skills and perspectives to develop and be valued. This potentially affords individuals occupying this position a space to speak out and draw on all these experiences. This may afford them more power, allowing them to contribute more fully to decision making. This "integrated" identity afforded participants the opportunity to challenge existing practices, drawing on an "activist" discourse and identity while remaining an "insider": Things don't progress unless you . . . challenge existing practice, and I'm not doing it in a ridiculous way. (Claire) I feel that I am on a bit of a path to change things um, and it's quite refreshing to do something that certainly, in my profession, hasn't been done before and to be part of that is really fantastic.
Another motivating factor for me in doing it and this is kind of my activist role. (Sarah) Participants from this position appeared to be able to engage in positive social action at an individual level, that is, making small changes to their practice, and at a wider level, where participants were afforded certain opportunities including writing personal accounts in professional journals (Anna and Helen), employing and supporting service user professionals (William and Paul), engaging in or conducting relevant research (Claire, Paul, Anna, Jack, and Jo), becoming involved in service user groups alongside those identifying solely as professionals (Sarah, Jack, Jo, and Anna), and representing service users and service user professionals within their professional role (Anna).
Some participants experienced many advantages of occupying their "integrated" identity position, including feeling proud, stronger, driven, determined, passionate, and valued: For others, it felt more like an "infiltrator" identity, doing things "under the radar," in the sense of a hidden, more "unintegrated" identity: I'm really proud to say that I um, unwittingly at times employed staff that have been service users in the past. All participants enacted an "integrated" identity at times in their talk; however, for some this identity seemed less well-formed than the "unintegrated" identities. Most participants felt that an "integrated" identity construction was the most helpful one but difficult to achieve.

Discussion
Participants were found to draw on different identity constructions, including "professional," "patient," "unintegrated," and "integrated." Participants drew on, presented, or experienced the two separate constructions of "patient" and "professional" as imposed on them, making them switch between the two, and constructing an "unintegrated" identity. At times, participants drew on all their experiences as valued and complementary, constructing an "integrated" identity. Participants' identity constructions positioned them differently, with some affording them more power and social action and others less. The "integrated" identities seemed less formed, suggesting that the discourses to construct this identity may be known but less used.
The dominant separate identity constructions seemed to position professionals as relatively powerful and expert and service users as relatively powerless and devalued. This corresponds to existing literature which suggests that within the dominant medical-psychiatric model (Walsh et al., 2008), service users tend to be positioned as being unable to recover and take control of their lives (Frese & Davis, 1997), developing a dominant "mentally ill patient" identity in contrast with the positioning of professionals as having an "expert" identity (Schiff, 2004). However within the "patient" identity, participants were also able to draw on a "personal recovery" discourse, enabling them to construct a more positive identity as suggested by the recovery literature (Deegan, 1987;Oades et al., 2009;Slade, 2013).
This study provides evidence for the dilemmas faced by professionals with service user experience, echoing existing literature from several countries, including the dilemma of disclosure (Schulze, 2007); the idea that professionals are or should be beyond distress (May, 2001;McCourt, 1999); negotiating the positive and negative impact of being a service user (Joyce et al., 2007); managing stigma, prejudice, and discrimination (Adame, 2011); and facing "them-andus" dichotomies (Adame, 2011(Adame, , 2014Schiff, 2004). This study provides evidence as to how these dilemmas may affect the identity construction of service user professionals, and suggests that in some contexts, they have "unintegrated" identities and switch between "professional" and "patient" depending on the discursive resources most salient in relation to those contexts.
This study also suggests the development of "integrated" identity constructions, on which participants were able to draw in relation to some professional contexts, incorporating positive identities of "wounded healer," "personhood," and "insider activist" in which mental distress is normalized as human experience, and recovery is "of life" rather than "from illness" (Perkins & Slade, 2012), echoing Deegan (1987). Professional and service user identities can complement one another and both be expressed as human experience within a "personhood" identity (Adame, 2014). They can enable empathy with, and access by other service users, as well as activism within mental health services. By systematically documenting how this integrated identity is talked into being, these less dominant discourses can be drawn into the foreground, allowing them to be heard more widely.

Limitations
Participants were self-selecting, making it difficult to transfer results to others who did not take part. However, the sample and their context have been described so as to situate them for the reader. The first author conducted respondent validation, but due to time constraints and the difficulty explaining DA to someone new to this methodology, feedback was limited. Harper (1995) suggested that DA runs the risk of "over-interpreting" data, and analysis not being sufficiently grounded; however, efforts were made to ensure the quality of the study. We have presented one possible reading of the data. Another limitation is that the interviews focused on only two of the multiple possible identities people manage during their lives (St. Claire & Clucas, 2012), and this will have guided participants' talk.

Future Research
Future research could explore how professionals with service user experience construct their identity further and explore when and where these particular identity constructions occur; for example, involving participants from professional disciplines more and less informed by the medical model, different professional contexts, client-colleague interactions, professionals with a greater range of service user experiences, those who have "come out" versus those who are hidden, peer support workers, those in user-led services, and community-based interventions based on people's interests in the arts or nature that may enable a wider range of identities to be expressed. As documented by Wong, Stanton, and Sands (2014) in a U.S. context, service users in recovery may join other service user or community-action groups from which they derive valued identity. Further research could help us understand such broader contexts in relation to how different identity constructions occur and how they might be changed. Participatory action research with professionals, service users, carers, and those with all these experiences could enable exploration of a range of discourses and attempt the co-construction of new and positive identities.

Clinical Implications
It seems important for mental health services and professionals to be aware that service user professionals exist and are becoming more vocal about their experiences. This may help challenge the idea that professionals are or should be "superhuman" (Deutsch, 1985;May, 2001), normalizing mental distress and developing a professional ethos that places more value and focus on both self-care and organizational support for all staff (Shepherd et al., 2010). With professionals speaking out about their service user experiences, this may help erode barriers between professionals and service users, increasing communication and developing new relationships between people who would like to see change (Adame, 2011(Adame, , 2014Roberts & Boardman, 2013, 2014. The normalizing of mental distress can also prevent the appearance of a new and separate group of people (professionals-with-addedvitamins, so to speak), different and superior to both service users and other professionals. The "personhood" identity and "personal recovery" discourses seem particularly helpful here, as they draw on broader "whole-person" sets of human experiences.
To reduce negative social positioning of service users, future work needs to consistently involve service users in the planning and delivery of services and mental health training (Andersean, Oades & Caputi, 2003;Health and Care Professions Council, 2014;Perkins & Slade, 2012) and hold the belief that service users can be professionals and professionals service users, and that this is unsurprising. Roberts and Boardman (2014) suggested that courses for professionals on empowerment and personal recovery that draw on participants' personal experiences of distress and difficulty as part of human existence may help reduce the distancing that can happen between professionals and service users.

Conclusion
This study explored how a U.K. sample of mental health service user professionals constructed their identities. Participants constructed separate "professional" and "patient" identities, switching between them in referring to different contexts, thereby developing an "unintegrated" identity. Participants also developed "integrated" identities in talking about some professional contexts. Although these were used less frequently, positive integrated identities included "personhood" and "insider activist," drawing on discourses of "personal recovery," and "use of self" in valuing and drawing on "lived experience."