Community Participation and Recovery for Mental Health Service Users: An Action Research Inquiry

Introduction: The social inclusion of individuals with mental health problems is an issue for mental health services, for the individuals who experience stigma, discrimination and exclusion, and for society at large. To develop community-orientated services that are capable of promoting inclusion it will, therefore, be advantageous to all parties to understand what service users find most helpful. Method: A 2-year action research project explored the recovery journeys of a group of assertive outreach service users who had progressed from being socially excluded and occupationally deprived to being participants in their local communities. The research aimed to understand how these outcomes were produced and to use this knowledge to inform local service development. Findings: This paper focuses on eight qualitative interviews, where service users recounted their stories of community participation and inclusion. The findings show how assertive outreach practitioners harnessed occupation as a basis for building relationships between practitioners and service users, and how this became a conduit towards participation in the mainstream community. Conclusion: Facilitating engagement in community-based occupations through creative collaboration helped participants reconnect with cherished roles, achieve long-standing goals and develop feelings of self-efficacy, belonging and wellbeing.


Introduction
Mental health service users are one of the most excluded and disenfranchised groups in society (Leff and Warner 2006). Social exclusion has been defined as non-participation in key activities of the society in which a person lives (Burchardt et al 2002). This suggests that supporting community participation can be a way of addressing this intractable problem (Office of the Deputy Prime Minister 2004).

Developing an inquiry
Practitioners in a Bristol assertive outreach (AO) team (which includes the author) observed that many service users had been supported to engage successfully with mainstream community occupations, propelling their own recovery journeys. A preparatory audit of AO case notes by the lead researcher confirmed this for about 45% of the AO caseload.
This phenomenon was explored to determine whether actionable learning about how these outcomes had been achieved could be applied more widely across local community mental health teams and day services, which aimed to develop socially inclusive practice by becoming more embedded in the community (National Social Inclusion Programme /Care Services Improvement Partnership 2006). In particular, the learning was relevant to the work of a local interagency community consortium, convened to coordinate the joint efforts of local day services and community partners to develop new communitybased occupational opportunities for mental health service users.
Community participation and recovery for mental health service users: an action research inquiry Method Action research (AR) methodology was chosen because it is rooted in real life practice and harnesses collaboration and mutual learning to produce practice-orientated results (Kemmis and McTaggart 2008). AR inquiries are often formalised versions of conundrums that practitioners have been struggling to solve. Here, practitioners were puzzled as to how hitherto 'hard to engage' individuals were now successfully engaging in community activities. 'Insider action research' such as this -where the researcher is embedded in the field of inquiry -acknowledges the inquirer's unique relationship to the inquiry, regards such indwelling as a strength and sees transparency about it as an issue of trustworthiness (Coghlan and Brannick 2010). The preunderstanding gained by counterposing insider knowledge with a critically reflective viewpoint is an asset in this AR.
The AR had two phases. First, semi-structured qualitative interviews explored service users' experiences of recovery and participation (aims 1 and 2). Second, the findings were fed into the consortium (aim 3) and two models of participatory AR -cooperative inquiry (Heron and Reason 2001) and appreciative inquiry (Cooperrider and Whitney 2005) explored consortium members' experiences of joint working over an 18-month period (aim 4). This paper focuses on the service user interviews. Fieldhouse and Onyett (2012) reflect on using AR methods with the consortium and Fieldhouse (2012) explores the practice implications of the consortium's joint working.

Social constructionism
Underpinning the inquiry was a social constructionist approach, which understands that reality is always reality as we know it and, therefore, subjective (Ekdawi et al 2000). This is true for all the key phenomena linked by the inquiry: recovery (Slade 2009), the sense of 'belonging' associated with social inclusion (Tew 2008) and meaningful occupation (Wilcock 2007a). Qualitative interviews were seen as the best way to capture rich description of these phenomena from a subjective viewpoint (Fontana and Fey 2008).

Ethical considerations
Recognising the vulnerability of AO service users, full consideration was given to how informed consent, the right to privacy and protection from harm could be maintained. A favourable ethical opinion was given by the local National Health Service research ethics committee.

Sampling strategy
Participants were selected by purposeful sampling (Patton 1990) to target the most information-rich service users in relation to the inquiry. Individuals must have been deemed occupationally 'hard to engage' when referred to AO and must have subsequently become successfully engaged while with the AO team. 'Hard to engage' was defined as being nonengaging with any of the local occupation-based adult services they had been referred to (such as vocational, rehabilitation or day services) or having occupation identified as an unmet need in their care programme approach (CPA) care plan.
Potential participants were excluded if they were assessed by their AO care coordinators as being too vulnerable to participate because of the distress an interview might cause them, or as not having the capacity to give informed consent to participate.

Data collection
Great care was taken to maximise the trustworthiness of the interviews. To address the inevitably imbalanced power dynamics (Fontana and Fey 2008), two service user researchers from Bristol MIND's User Focused Monitoring Project (UFM) joined the lead researcher as co-interviewers and data coanalysts. Semi-structured interviews were conducted in pairs using an interview schedule jointly constructed by the three interviewers. This had an overarching main question: I Looking back over the last few months/years, what do you think were the main events or milestones that have made a positive difference in your life? There were three possible follow-up questions based on the factors influencing engagement identified in the preparatory audit: one about 'doing things' (It seems you are far more active than you used to be. Can you describe some of the things you do now that mean most to you?); one about relationships (What was it about the relationship with … that was good or most important?); and one about identity (Has being more active changed the way you see yourself /your skills /your strengths? Can you tell us how/ what these are?).
It was acknowledged that the interview might evoke difficult memories, so a solution-focused approach positively reframed being 'hard to engage' (a problem) into being 'information-rich' (a potential solution) in relation to the learning that was sought. Interviewers emphasised their role as learners and the participant's role as the expert on his or her own situation.

Jon Fieldhouse
Interviews were audio-recorded based on written consent and were transcribed. Member checking occurred when participants were sent summaries of their interview transcripts for verification or amendment (Janesick 2000).

Project steering
To reinforce the commitment to accessing an authentic service user voice, the steering group included two service user representatives (who were not interviewees) and two trust-based service user development workers who supported them.

Inductive data analysis
To present the findings as authentic, it is necessary to vouch for how they were produced and to present the analytical process as transparently as possible.
The importance of the subjective viewpoint suggested thematic analysis because it is inductive or data driven. First, semantic codes organised the data under headings that used participants' verbatim language, then more interpretative themes captured the meanings within these data, going beyond the surface of the data while remaining grounded in it (Braun and Clarke 2006). Triangulation across three data co-analysts is regarded as a strength of this inquiry.

Immersion in the data and initial coding
The transcripts were read independently by each co-analyst and a qualitative data analysis software programme (MAX QDA) was used to code and cross-index 295 raw data extracts.

Exploring relationships between candidate themes
Propositional statements were written as descriptors for each code, including a subject, an object, and -most importantlya verb. These descriptors suggested how phenomena actively influenced one another, transforming the codes into candidate or potential themes, and showing them in relationship to each other.

Consolidating themes
Consolidation followed Patton's criteria (1990) that a robust theme shows internal homogeneity (cohering as a single entity) and external heterogeneity (being unique and clearly distinguishable from other themes). Thus certain themes were subdivided to capture important nuances of meaning. Seeing themes in relationship underlined the connections between themes for each individual. This illuminated personal narratives, making otherwise theoretical connections empirically and abundantly clear. It provided individual case examples, highlighting how positive outcomes had been achieved in the face of pressing mental health problems, as discussed later in relation to dual diagnosis.
Consolidating themes produced the configuration in Fig. 1, depicting four global themes written in full propositional style. There is an ongoing transaction across themes 1 to 3, evident in the fact that the majority of data extracts were located in the intersections (areas A to D). This highlighted the strong influence of the person-environment-occupation relationship. Theme 4 is the recovery journey, depicted as a movement (left to right) over time.
These themes, and their subthemes, are presented as Findings in Tables 1 and 2. Basic themes present simple premises drawn from verbatim data; organising themes arrange these into clusters of significance; and global themes present a compelling argument of what the data mean overall (Attride-Stirling 2001).

Findings Participants
Nineteen names were proposed by AO care coordinators. Five were excluded by the criteria described. Ten consented to be interviewed, although two later withdrew this consent without having to give a reason. Of the eight who were interviewed, all were male and six were from black and minority ethnic backgrounds or dual heritage. Their ages ranged from 24 to 57 years. Pseudonyms are used.
To generate actionable evidence, the findings had to avoid unnecessary complication (which would make them inapplicable), but not at the cost of oversimplifying the inherent complexity of the phenomena in question (which would render them meaningless). To this end, the three global themes (shown in Fig. 1) are presented as the main findings. The quotations presented in Tables 3 to 6 give more detail by illustrating the organising (and some basic) themes. The same numbering system for the themes is used across Tables 1 to 6.

Main findings
Participants enjoyed rewarding social relations through renewed engagement in mainstream occupations. The community offered a range of opportunities for this engagement, which held a unique appeal for participants. Participants' relationships with AO team members -based on occupations -was the conduit towards these opportunities and provided the context for participants' confident exploration of community activities and fuller participation in them. Organising themes (1) Rewarding social relationships were embedded (1.1) Participants had limited opportunities to pursue their goals and did not feel accepted by the in occupations community where they lived (1.2) New occupational opportunities tackled social isolation because new relationships were embedded in the 'doing' and felt good (1.3) Being engaged occupationally increased motivation, personal agency and will (1.4) Social connectedness was an indicator of participants' recovery -both for themselves personally and for their families

Community participation and recovery for mental health service users: an action research inquiry
(2) Accessing mainstream occupations provided key (2.1) Mainstream resources had particular qualities that supported participants' occupational engagement opportunities for social and occupational engagement (2.2) Occupationally focused work had an impact in tackling substance misuse problems (3) Certain qualities of the relationship which (3.1) The relationship was mutually respectful, robust and relaxed participants felt they had with AO supported (3.2) The relationship was based on occupations their confident exploration of new social and (3. 3) The relationship was goal-orientated and solution-focused occupational opportunities (3.4) The relationship was based on practical support (4) Participants described a recovery 'journey' (4.1) The journey was gradual (4.2) The journey was transformational AO = assertive outreach.

Discussion
This discussion examines four aspects of participants' experience: living with stigma; community participation; managing addiction; and enjoying improved physical health. A final fifth section reflects on AO as a service model focusing on 'hard to engage' individuals.

Living with stigma
Participants' stories were based on up to 5 years contact with the AO team, so the sense of living in an 'excluding' community was longstanding. For Matthew, this exclusion was acutely felt. He wanted to extend his social contacts beyond the drop-in centre he attended but he felt he should 'stay away from' ordinary people (Matthew's terms), as if he felt no eligibility to participate with them. This apparent internalised stigma -created by societal attitudes to mental illness and the trauma associated with exclusion -can have a long-term impact on self-image (Meehan 2007). Such psychosocial barriers are not as well understood as the barriers encountered by people with physical disabilities (Beresford et al 2010). They go beyond the restrictions imposed by a particular impairment, and are compounded by society's limited understanding of its role in the creation of them. The net result is individuals' pre-emptive 'self-exclusion' (Parr 2008).

Community participation
Participants identified several factors underpinning their gradually increasing community Jon Fieldhouse

3.3
Stanley: What's been helpful is that they've recognised your potential. Because they can see it in your work, but you don't see it yourself.

3.3.2
Interviewer: I'm wondering whether having people who you're doing the planning with, who you felt comfortable with and had a good relationship with … whether that affected the quality of the plan?
Gerry: Well, making sure that the psychiatrist understands basically what I've been doing … I'm coming on well, coming on well and that.
Interviewer: That sounds like an important job to have.

3.4.3
Matthew: I got to do it myself … but the assertive outreach makes it easier for me.
Interviewer: How do they make it easier for you?
Matthew: Oh! They got the car outside [laughs]. All I got to do is walk down to the car, get in, and we're away. AO = assertive outreach.

Scaffolding
'Scaffolding' is a term borrowed from Vygotsky (1978) to describe how skill acquisition happens through engagement with a challenge, facilitated by a temporarily constructed Community participation and recovery for mental health service users: an action research inquiry 'support' which is then removed when the individual can perform the skill himself or herself. Scaffolding underlines the importance of personal agency as the basis of change and is acknowledged to have particular relevance to the principles of recovery and the practice of occupational therapy (Davidson et al 2010). Here it describes a method of environmental adaptation: the creation of a flexible, temporary, affirming psychosocial space. It enabled individuals to derive peer support and to counter the more debilitating effects of stigma, and acted as a base from which to venture into mainstream occupations and the support networks that these hosted. The 'scaffold' was co-constructed through negotiation between practitioner and service user to serve a specific personal goal. It was constructed from any or all of the components in Fig 2. Each component is now explored.
generates it in greater abundance. 'Making a start' meant starting a process. Wilcock's (2007b) notion that the health-promoting power of occupation is most potent when it combines doing, being, becoming and belonging not only highlights that participation as an equal citizen is an issue of occupational justice (Stadnyck et al 2004) and human rights (Leff and Warner 2006) as well as a 'therapy issue', it also helps to explain the transformational recovery journey that participants described. The 'I' that drives the change and the 'I' that must accept the need for change can be a single continuous 'I'. In this way, the individual is supported in maintaining his or her personal integrity as well as developing positive feelings about the emerging new sense of self -a dialectic that is crucial to psychological therapy and to self-determination in particular (Deci and Ryan 2000). This self-efficacy is also a fundamental aspect of recovery (Slade 2009). Stanley, for example, described how resuming his artistic activities was not only a valued outcome, but fuel for further growth. It 'made him strong'. Significantly, when participants were asked what they attributed their progress to, most said themselves.
The important issue for participants was not that their psychiatric symptoms had to be alleviated first in order to become more active (symptomatology was not mentioned in any interviews), but that living a satisfying life as a person with mental health problems was an ongoing challenge that required social and practical support. Echoing Matthew's comments about going 'beyond drop-ins' and even beyond his wildest dreams, Stanley said that he was achieving goals he never thought possible. Using the delineations suggested by Lloyd et al (2008), participants' progress integrated functional, social and personal recovery processes.

Grading and pacing
Being attuned to latent goals allowed practitioners to approach entrenched problems in the lives of 'stuck' individuals and to work with the area of greatest potential for change. Grading and pacing -based on activity analysis and adaptation (Creek and Bullock 2008) -allowed the experience of success to filter in through habitual depressed thinking, internalised stigma and diminished self-belief, so that momentum could pick up from a new starting-point. Matthew's description (Table 5, Theme 3.4.3) of simply getting in the car and feeling 'we're away' indicates how powerful grading and 'quick wins' (Ryan and Morgan 2004a) can be. Grading and pacing addressed a wide range of personal goals (all of which subjectively felt 'undoable' at first), rendering them more 'doable'.

Therapeutic use of self
Each participant felt he was engaging with another person rather than with a mental health service. Practitioners' conscious use of personal qualities, attributes and responses to develop an intentional relationship with service users (Taylor 2008) -based on honesty, transparency, emotional authenticity and mutual respect -appears to have been a key feature of scaffolding.

Identifying personal investment in change: eliciting goals
Despite the descriptions of being 'stuck', participants wanted things to be different. They had latent goals, rediscovered through simple acts of 'doing', which allowed individuals to experience themselves differently and to bypass habitual beliefs about being 'un-includable', and this felt good. Wilcock (2007b) suggested that 'occupation' and 'health' are one and the same, particularly where the 'doing' promotes self-efficacy, fosters hope and accesses social capital. For Rahim, this appeared to be so. Occupational and social engagement were simultaneous and mutually reinforcing, all 'coming together in one package'.
Participants liked to do things, particularly with other people. The 'doing together' was how significant social relationships were enacted, expressed and developed. This highlights that whilst occupation requires motivation, it also Jon Fieldhouse AO's capped caseloads are likely to have been crucial here. Many therapeutic processes described by participants were underwritten by the team 'having time'. First, the close interpersonal relationship allowed practitioners to discern participants' latent goals and to act as holders of hope on the recovering individual's behalf, recognising potential when (as Stanley described) the individual had lost sight of it himself. Secondly, smaller caseloads meant that AO was not under pressure to discharge individuals when they fulfilled spurious markers of recovery, but could work towards real milestones of goal attainment. Finally, longer-term casework afforded a broader perspective of participants' repeating patterns of relapse and readmission, which could then be addressed. Stanley's 'wake-up call' (Table 5, Theme 3.1.2) was only possible because he had a robust relationship with practitioners, who could challenge him without destroying it.

Person-centred care planning
Effective care planning harnesses service users' own momentum and the solutions they consider in relation to their own perception of need. However, service-centred care planning often perceives need only in terms of what interventions are available within services (Ryan and Morgan 2004b). This pitfall was avoided by AO practitioners because they learned that participants' personal goals could be served (indeed, they had to be served) without the option of referral to specialised rehabilitation services, as these were invariably the same services that participants had disengaged from.

Community mapping
Participants valued engaging with mainstream occupations; going 'beyond drop-ins', as Matthew put it. It was important that community mapping prompted more than mere brokerage of community support as this not only risks fragmentation of care (Meehan and King 2007) but may be too narrowly focused on services as opposed to ordinary life opportunities (Bates and Seddon 2008).
Enhanced community mapping also improved cultural sensitivity: 75% of participants were from black or minority ethnic or dual heritage backgrounds and the mainstream community offered them a more diverse and personally meaningful range of occupational opportunities than existed within the day service programme. It is widely acknowledged that minorities are not well served by segregated services (Bates and Seddon 2008).

Being a 'travel companion'
Practitioners were able to 'go with' services users to create affirming environments wherever they were needed. It has been suggested that service users prefer a travel companion with whom they can share experiences and negotiate flexible ad hoc support, rather than a travel agent who merely makes 'bookings' or referrals (Deitchman 1980). Rather than seeing their role as being to discharge people to the community, the AO team introduced the community into the heart of people's CPA care plans, maximising the natural supports available to them. It was an essential dimension of care, not necessarily a discharge route away from services (although it could be that too). For many participants, knowing that the AO door would not slam shut behind them made it more likely that they would take a chance, walk through it and engage with their community. It was a matter of trust and time.

Creating affirming environments
Affirming environments improve individuals' engagement by creating a micro psychosocial environment characterised by empathic, non-judgemental relationships, and increased confidence to engage with others and experiment with new roles (Rebeiro 2001).
Affirming environments were co-created in various settings: a further education college (Stanley), a leisure centre (Ken), a horse-riding stable (Carl) and within different voluntary work settings, such as the Green Gym (Barry, Rahim, Matthew). Here, for example, AO team members initially gave service users lifts to and from sessions, worked alongside them at first, and gradually faded out this support as service users' confidence grew and the scaffolding could be dismantled. This offered a graded progression from 'the Green Gym as temporary affirming environment' to 'the Green Gym as ordinary community participation'.
In providing individualised support to service users who participated alongside the general public, this was, in 'inclusion traffic lights' terms, a green approach (Bates and Seddon 2008). It underlines that these individuals did not have to pass through a predetermined red /amber/green sequence in order to recover. Here, with support, they progressed directly from red to green.

Positive risk management
Participants' close relationship with AO fostered a positive risk-taking ethos based on ethics of care that emphasise the particular context of each decision about risk. This challenges the habitual risk averse position of services (Felton and Stacey 2008). Participants felt that they were trusted and given a chance.

Harnessing social capital
Rahim's pleasure in working alongside his 'friends' in the Green Gym underlines the importance of social connections built and maintained by the individual's own efforts (Tew 2008). This vital aspect of meaningful occupation affirms that one's life has a value for others as well as for oneself. Collective doing, such as in group-based voluntary work, allowed self-perceptions to be reappraised and recalibrated through the perceptions of fellow workers (Moghaddam 2009), fuelling a sense of competence and acceptance.
Ultimately, occupation can be seen as a community rehumanising process (Wilcock 2007a). It presents 'the community' as something tangible, with direct, navigable routes into its social capital and social networks. The community is revealed as a network of occupations, 'a myriad of roles, activities and possibilities' (Perkins and Repper 2003, p23).
Community participation and recovery for mental health service users: an action research inquiry 10. Advocacy, lobbying and partnership working Fieldhouse (2012) explored the work of the community consortium, concluding that effective community development work must be informed by an intimate micro-level knowledge of service users' daily occupational challenges, as much as a macro-level public health awareness of issues like social exclusion or occupational deprivation and alienation (Seebohm and Gilchrist 2008). Community development work unites these perspectives, bringing occupational therapy and occupational science together. This inquiry suggests that practitioners have much to contribute to, and much to gain from, this approach.

Living with addiction and contemplating change
Rahim, Carl and Stanley had long-standing problems with drug use. Stanley's story is used as a case example here because occupational engagement appeared to have the strongest impact on his addiction to crack-cocaine.
Supported by AO, Stanley's college attendance broadened his range of social contacts, prompting thoughts about change and enabling him to move from pre-contemplation (being unaware that a problem existed and oblivious to the concerns of people around him) to contemplation of change (Prochaska and DiClemente 1982), where he reflected on the consequences of addiction -including the impact on his relationship with his daughter: Stanley: Because when you are messing with drugs, you're not only hurting yourself, you hurt the others, the people around you … Big, big plus on that one. Giant plus … … And it's changed for me, and I definitely want a change for my daughter.
Interviewer: So you can see yourself fulfilling your role as a parent?
Working occupationally offered more than just an idea of how a drug-free life might be lived; it offered an actual experience, albeit a fleeting one, and a more assured basis from which to move forwards. In the third and fourth stages of the cycle of change -preparation and action -support networks offer vital reinforcement for consolidating change. Stanley's occupational engagement removed him from the networks that reinforced his drug-taking and brought him into contact with more supportive peers: They [Stanley's circle] see how I am, how I carry myself. I'm me, me -not saying 'do you want to buy a stereo', or 'do you want to do this' … I'm going up to them and saying 'look you wouldn't believe this, I'm in college' and 'I've got some work on the net and that' and I'm getting a lot of encouragement. And it's good. Fantastic.
Stanley's story is significant because it is estimated that a third of mental health service users have a substance misuse problem and perhaps as many as 50% of those with severe and enduring problems (Hawkings and Gilburt 2004). Yet the separate development of substance misuse and mental health services has created a 'split service', which is sometimes hard for referrers to negotiate. Individuals often fall between standard mental health services (because their needs are complex) and specialist substance misuse services which require an individual to be motivated to change before he or she is taken on. Occupationally focused work appears to foster that motivation and be capable of straddling splits in services.

Enjoying improved physical health
Much of this discussion has been about psychosocial change, but occupational engagement also means being physically active. Our selves are embodied. Mental and physical health problems frequently coexist and life expectancy is up to 10 years lower among people with a diagnosis of schizophrenia compared with the general public (Lester and Glasby 2006). Here, participants did not recover first and then become active; activity was the route to recovery. Simple things do need stating sometimes or they become invisible. For many participants, occupation and health felt one and the same. Carl felt 'better', Gerry 'livelier', Ken 'fitter', Jack was looking after himself better, Rahim enjoyed 'looking better' (attributing this to an improved appetite) and Stanley said there is 'nothing nicer' than having his health back.

Assertive outreach and 'hard to engage' service users
This inquiry endorses the view that AO can improve community living for people with severe and enduring mental health problems (Meyers 2010). It demonstrates that the term 'hard-to-engage' describes a feature of certain service users' relationship with services, not a characteristic of the service users themselves (Priebe et al 2005). It would be more accurate to see the underlying problem being 'un-engaging' services. The danger of services automatically referring 'hard to engage' individuals to AO is that it removes their obligation to understand their own role in service users' non-engagement. This can undermine the responsiveness of services, promote the stereotyping of individuals as 'hard-to-engage', reinforce the negative self-attributions that service users may already have, and thus perpetuate stigma and prejudice.

Critique
This inquiry focuses on service users' stories of success. It is acknowledged that different kinds of stories may have been elicited from the 55% of the AO caseload who did not engage in community activities. Although bias cannot be eliminated in qualitative approaches, it is hopefully clear that the inquiry did not set out simply to demonstrate good practice but aimed to understand this practice better by appreciatively exploring what had worked for service users. The positive outcomes were undisputable and the point of the inquiry was to unpack retrospectively how they had been achieved. In this respect, the role of purposeful sampling should be emphasised. Similarly, the appreciative approach, aiming to identify and nurture good practice (Zandee and Cooperrider 2008), should not be misconstrued.

Jon Fieldhouse
The inquiry would not have arisen without the researcher's insider view of the phenomenon of service users' engagement. The inquiry was rooted in practice and the preunderstanding described earlier was the key to unlocking particular meanings at interview and during data analysis. The presence of MIND UFM service user researchers as co-interviewers and data co-analysts also mitigated bias. In addition, the involvement of service users on the steering group indicates that every effort was made to elicit an authentic service user voice in the data and to involve service users as research partners.
In terms of service development, a further criticism might be that this study merely points to the good practices already promoted in The Ten Essential Shared Capabilities: A Framework for the Whole of the Mental Health Workforce (DH 2004) and The Capabilities for Inclusive Practice (DH 2007). However, an achievement of this inquiry is that it portrays these capabilities dynamically -that is, in use -and gives a voice to service users who report on their outcomes. Exploring these dynamics of practice also raises the issue of the economy of time that is often imposed on practitioners and service users, which can undermine such work.

Further research
Two ideas for further research emerge from this inquiry. One is an exploration of AO practitioners' views in response to the service user stories, to give a service provider perspective on the issues highlighted. The other is a single case study design to examine a phenomenon -such as working occupationally with dual diagnosis -in more detail. The longitudinal perspective afforded by service users' stories was illuminating, but the thematic analysis process effectively dismantled these narratives to generate themes across a group of participants. Although this brought some general issues into view, it might be useful to revisit the original data to explore each person's story as a coherent whole in a more deliberate way (Silverman 2010).

Conclusion
This inquiry shows how the intentional harnessing of naturally occurring resources in the mainstream community helped service users feel that they belonged in their community as competent, connected and contributing citizens. Crucially, this 'connectedness' was often felt to be absent in the community without an occupational route into it.
Practitioners facilitated this process by providing individualised support -through temporary scaffolding -in places where service users could participate alongside the general public. This facilitation, based on close interpersonal relationships with individuals, was focused on occupations, which then acted as conduits towards occupational engagement in the mainstream community.
Put simply, for participants in this inquiry, 'real life' occupations -accessed through care planning -not only achieved the outcomes that programme-based rehabilitation might have aimed to achieve but did so without segregating individuals and with a minimum of stigmatisation.