Innovating for a cause: the work and learning required to create a new approach to healthcare for homeless people

Abstract Innovation occupies a pivotal place in our understanding of knowledge-based economies, and this is raising questions about sources of innovation, how it originates, and the role played by employees, work practices and learning. This paper explores these issues through case study research into a new approach to providing healthcare for homeless people in England, and by bringing together conceptual insights from the employee-driven innovation literature, and more broadly from social and practice-based learning theory and organisational theory. Applying these perspectives to our case enables illumination of the innovation as a process – not an event – and as an ongoing set of organisational practices that transcend their origins. Through our analysis we argue that the notion of ‘a cause’ is helpful in elucidating the impetus and the commitment to making the innovation happen (and go on happening). Our findings are presented under three themes: ‘establishing a cause’, ‘organising for innovation’, and ‘innovative capability in practice’. Building on these, we have identified five key inter-related dimensions which help conceptualise the work and learning that it took to create and (re-)enact the innovation and that we suggest may have relevance for understanding and characterising other employee-led innovations in and perhaps beyond healthcare.


Introduction
Innovation has come to occupy a pivotal place in our understanding of knowledge-based economies over recent years. Indeed, the innovative knowledge-economy is now widely seen as key to building economically and socially prosperous societies. Whilst research and development is still regarded as the main route to innovation, there is increasing awareness of the importance of alternative sources of innovation such as employee driven innovation which emerges in and through work practices (Aho 2006;Høyrup et al. 2012;Melkas and Harmaakorpi 2012;Price, Boud, and Scheeres 2012).
Interest in how to foster and facilitate workplace innovation raises questions about the role of learning in this process (Ellström 2010;Evans 2012;Fenwick, Nerland, and Jensen 2012;Høyrup 2012;James, Guile, and Unwin 2013). Two broad and contrasting approaches to understanding workplace innovation can be distinguished by their distinctive theoretical orientations towards learning. The first conceives the original source of innovation as external to the workplace often, in the shape of major technological advances, including in the healthcare sector (May et al. 2005;Halford et al. 2010). Innovations are created and tested by external experts prior to being transferred via a uni-directional, top-down process for employees to implement. Here, learning is primarily conceived as an individualised matter of acquisition and transfer, with individuals learning to work in new ways, with new technologies, and with little or no consideration of the role that organisational and social processes play in mediating the relationship between innovation and learning (James, Guile, and Unwin 2011). A contrasting conception views everyday co-participation in work practices as the key resource for innovation, which emerges in reflexive relation to the problems and issues that arise in the workplace (Brown and Duguid 1991;Orr 1996;Høyrup 2012). Under this scenario employees may become innovators on the basis of their experience, and in-depth, and shared knowledge, and through their participation in the social relations and co-production of work. Hence, and importantly, the workforce as a whole is understood to have innovative capacity (Billett 2012;Price, Boud, and Scheeres 2012;Fuller and Unwin 2017). This perspective recognises that established organisations and teams can be generative of innovative activity, but it provides less insight in to the work and learning that it takes to create an innovation through new forms of organisation, social relations and practice.
Our paper explores the issues raised by this second conception by drawing on research into a new and innovative approach to providing healthcare for homeless people in England developed by 'Side by Side' (SBS), a charitable organisation. Our case, SBS, can be viewed as falling under the umbrella of 'social innovation' (Murray, Caulier-Grice, and Mulgan 2010;Mulgan 2012;Nicholls and Murdock 2012;Moulaert et al. 2013), primarily oriented to generating social rather than economic value, and defined by Murray and colleagues as: … innovations that are social both in their ends and in their means. Specifically, we define social innovations as new ideas (products, services and models) that simultaneously meet social needs and create new social relationships or collaborations. (2010,3) Whilst acknowledging the literature on social innovation as a general backdrop, the focus of our study is on understanding, the particularity of employee-led innovation, its origins and enactment in diverse kinds of organisational and workplace learning practices.
To do so, we bring together conceptual insights drawn from the employee-driven innovation literature, and more broadly from social and practice-based learning theory and organisational theory. Applying these perspectives to our analysis of a case of employee-driven innovation (SBS), enables us to illuminate the innovation as a process -not an event -and as an ongoing set of organisational practices that transcend their origins. Through our examination of the empirical evidence, the sense of 'a cause' emerged as an important aspect of our explanatory account. In what follows we will argue that this is helpful in elucidating both the value-based impetus to create (or rallying call for) a new approach to providing healthcare for homeless people and the collective commitment to making it happen (and go on happening) through what we term a 'community of practices' .
Following this introduction, the paper is organised in six sections. Firstly, we outline sources and understandings of innovation. Secondly, we discuss practice-based understandings of work, learning and innovation. In section three, we outline the rationale for innovation in the healthcare sector, and describe our case study methodology. The fourth section presents our findings under three themes: 'establishing a cause' , 'organising for innovation' , and 'innovative capability in practice' . Together, they illustrate the work and learning that it has taken to create and enact the innovative SBS approach. In section five, we present five inter-related dimensions that help characterise the innovation, and the final section provides our conclusions.

Sources and understandings of innovation
Much previous policy attention on innovation was grounded in understandings of the source of, and resources for, innovation as residing in specialist settings including university laboratories or dedicated innovation hubs (Aho 2006). In this research and technology-based model of innovation, creative ideas are developed and prototyped before being handed 'intact' to users. Implementation is seen as a matter of structural reform and take-up as requiring change amongst existing staff (Pope et al. 2013). This perspective is based on the idea that knowledge creation is the preserve of specialist institutions and highly qualified experts, assuming a uni-directional mode of learning transfer -from innovation producers to users. However, a policy shift away from this approach has been occurring. An OECD report based on evidence from countries and workplaces across the European Union noted: The underlying approach to innovation has been changing, shifting away from models largely focused on Research and Development (R&D) in knowledge-based globalised economies and giving more emphasis to other major sources of the innovation. Understanding how organizations build up resources for innovation has thus become a crucial challenge to find new ways of supporting innovation in all areas of activity. (OECD 2010, 9) This builds on longer trends in research. Management scientists have been at the forefront of identifying different sources of innovation, including by 'users' (von Hippel 1986;Lettl, Hienerth, and Gemuenden 2008), and 'amateurs' (Leadbeater and Miller 2004). More recently, there has been explicit interest in the role employees themselves can play in generating innovation (Høyrup 2012;Evans 2012). In this regard, Høyrup (2012, 7) has suggested the notion of 'employee-driven innovation' (EDI), arguing that: … as with any other kind of innovation, the result may be substantive products, services and/or process of an organization, and generation and/or implementation may be involved in the process.
In earlier work in management science, Victor and Boynton (1998) not only recognised that innovation could arise from diverse sources, including users, but also identified the scope and form of relational and distributed work organisation as key to the creation of innovative capability. They argued that ongoing collaborative and interactive relationships between producers and customers were needed successfully to (continuously) co-configure goods and services. Building and developing such transversal alliances was seen as a hallmark of innovative and adaptive work processes and outcomes (Engeström 2004), and draws attention to innovation as a social process.
If co-participation in work itself is viewed as a source of innovation, those lower down as well as higher up the organisational hierarchy are conceived as agentic and as having the capacity to innovate as part of everyday work practice (Brown and Duguid 1991;Billett 2012;Fenwick 2012), and as part of what Price, Boud, and Scheeres (2012) refer to as 'making and remaking their jobs' . In this regard, Ellström highlights employees' 'considerable creativity … when it comes to finding solutions to unexpected problems ' (2010, 6). In the next section, we outline how social practice-based understandings of work and workplace learning help explain the capability of, and potential for employee-led innovation.

Practice-based understandings of work, learning and innovation
Identifying and highlighting the work that employees actually do in contrast to formalised expressions of what they are expected to do is an important starting point for a practice-based perspective on the relationship between work, learning and innovation (Ellström 2010). For example, drawing on Orr's (1996) ethnographic study of photocopier technicians, Brown and Duguid (1991, 41) contrast 'canonical practice' (an organisation's 'espoused practice') with 'non-canonical practices' (the actual practices that staff engage in and learn through to get the work done). Creating job specifications based on canonical practice inevitably generates an over-simplified representation that fails to recognise and understand the learning, knowledge and skills employees deploy to navigate the demands of, and make adaptions, to their everyday work. Social Scientists are using practice theory (Schatski, Knorr Cetina, and von Savigny 2001;Schatzki 2006;Gherardi 2009Gherardi , 2000Fenwick, Nerland, and Jensen 2012;Fenwick and Nerland 2014) and workplace learning theory (Boud 2010;Evans 2012) to help overcome this shortcoming. For example, Price, Boud, and Scheeres (2012, 89) draw on Schatski's concept of enactment to refer to the relationship between purpose, activity and persons engaged in work, with practice being the glue that connects organisations and individuals and suggesting that 'learning is implicated in remaking practices' . Høyrup (2012) argues that it is impossible to separate innovation from workplace learning processes as they are interwoven and constitute preconditions for each other. Lave and Wenger (1991) laid the ground for social practice theory to include an explicit notion of learning. Their theorization of learning as co-participation in a changing community of practice foregrounds the collective or group as the unit of analysis, rather than the individual (and in contrast to an individualised theory of learning that underpins the top-down understanding of innovation outlined in the Introduction). For Lave and Wenger (1991) and Wenger (1998), it is the group's association with the social relations of production of goods or services that is the key to understanding learning through co-participation in practice. As part of the 'lived in world' , people are conceived as constituting and as constituted by the social world and not as separate beings who determine when they move in and out of the practice arena. As Lave explains: 'theories of situated everyday practice insist that persons acting and the social world of activity cannot be separated ' (1993, 5). Wenger (2009, 2) has suggested that 'Innovative capability and the spread of innovation are a property of social system that depends on its learning capability' . Whilst his argument focuses on the social and systemic features underpinning learning capability, he is less interested in the role of organisation and how the way work is organised can be more or less generative of innovative capability. Our research on employee-led innovation in healthcare is suggesting that a focus on organisational characteristics provides additional key insights into the relationship between work, learning and innovation that are not picked up by Wenger's approach.
Drawing on what Brown and Duguid (2001) call the 'prism of practice' , allows us to identify the diverse participants involved in the innovation and to explore their access to and involvement in the multi-layered organisational and pedagogical practices that characterise SBS. Through our aim to better understand why and how innovation can be employee-led, we have sought to build on previous practice-based insights in to the relationship between work, learning and innovation. Through our empirical analysis we take this a step further by identifying the role 'cause' plays in originating the innovation and through generating the spur that makes it keep on happening.

Healthcare context and case study
For some time, policymakers in the UK and some other countries have been promoting innovation in the healthcare sector (Department of Health 2011). The key drivers for this include: the ageing population, increasing numbers of people with multiple long term conditions, technological advances which enabling new treatments and ways of enabling services to be delivered differently, increasing patient expectations, and the need to control costs (a factor that has become paramount in the UK since the 2007/8 financial crash).
Healthcare work is normally delivered through bureaucratic forms of organisation and accountability designed to minimise risk. The aim of our research has been to better understand the conditions which support innovative capacity amongst the healthcare workforce. We have been interested in three key questions: (a) how is the workforce conceived in the healthcare sector?; (b) what role if any are ordinary employees expected to take in developing innovative ways of providing care?; and (c) are they conceived as part of the solution for addressing challenges or as a barrier to change? The following two quotations are illustrative. The first is from the former head of NHS England (the national agency for managing the UK's health service): Many of the problems which we suffer in the NHS are solvable if we use the intellectual capital of the 1.4 million people who work in the service. That's where the solution lies. (Sir Bruce Keogh, BBC radio interview, 29 May 2013) The second is from the Department of Health: … now and for the foreseeable future we must meet these demands from within our current real terms funding, while at the same time improving quality. This means that simply doing more of what we have always done is no longer an option. We need to do things differently. We need to radically transform the way we deliver services. Innovation is the way -the only way -we can meet these challenges. Innovation must become core business for the NHS. (Department of Health 2011, 4) Both these quotations signal the need for improvement and transformation in the way healthcare services are provided, and recognise the the need to activate the latent capacity of the workforce to achieve this aim. However, little is known about if or how this capacity might be effectively mobilised, not least in the highly regulated and resource-poor context of the National Health Service (NHS). Our research conducted between 2014 and 2016 was designed to help address this gap. Initial key informant interviews with 12 individuals from professional bodies (the British Medical Association and the Royal College of Nursing), from NHS England and the Department of Health, and from the Kings Fund policy think-tank provided context for the research and case study selection. We pursued a purposeful strategy to building our sample, opting to identify three cases of employee-led innovation that had brought about recognisable changes in care for particular patient groups, and that we could explore over time. In this paper, we focus on one of our case studies, SBS, which addresses the needs of homeless patients by focusing on how they are conceptualised and treated. Based in Greater London, it has developed a transformative approach to enhanced and integrated care co-ordination for homeless people admitted to hospital. This involves multidisciplinary, cross-sector teams working to ensure that homeless patients, often with highly complex needs, are cared for in a holistic way and discharged into an appropriate environment. Services are provided by doctors, nurses and care navigators (a new role which is discussed below) and are funded through contracts with the NHS. The SBS approach is now being replicated in other parts of England under its umbrella.

SBS
The case study followed a qualitative approach consisting of in-depth interviews with staff across SBS, ethnographic methods including observation of workplace practices, and analysis of documents published by the organisation. This involved 19 interviews and 62 h of observation (of day to day work supporting patients, multi-disciplinary team meetings, a planning session about service users' involvement in the annual conference, reflective practice sessions, and a faculty forum.
Interview transcriptions and field notes were initially coded for the broad topics underpinning our research, relating to work practices, organisation, learning, and employee-driven innovation. This preliminary analytic work was discussed in light of our experiences in the field in regular data analysis workshops, feeding into an iterative process of refining our coding and immersion in our evidence, which in turn facilitated the emergence of key themes.

Research findings
As a key part of our account, we draw on illustrative evidence from participants with different occupational roles and who bring diverse voices into the analysis. The findings are organised under three (inter-dependent) themes that focus on and illuminate the innovation as an ongoing process of enactment: (a) establishing a cause; (b) organising for innovation; and (c) innovative capability in practice.

Establishing a cause
SBS was formed in 2009 following the death of a recently discharged homeless person on the steps of a major London hospital. Professor X (Liam), a consultant at the hospital, was asked to lead an internal inquiry into the circumstances. Although he found no wrongdoing, he was struck by the way that the healthcare system appeared to be failing homeless people: I looked, and then I applied my professorial hat to it and asked the questions about mortality and morbidity. Found out the average age of death [for homeless people] in London was 42 … and the morbidity was appalling but no-one really knew because it's invisible. (Liam) The inquiry revealed that the homeless population attended hospital emergency departments five times more, were readmitted more than three times as often, and stayed in hospital three times longer than the housed population. Access to housing and social care relies on individuals being able to provide evidence that they have a long-term association with a local authority area. As this is often difficult for homeless people, many sleep rough between their hospital stays, resulting in frequent, unscheduled readmissions. Homelessness is associated with tri-morbidity, defined as the 'combination of physical ill health with mental ill health and drug and alcohol misuse' . The complexity associated with tri-morbidity requires careful care coordination that the existing approach was rarely able to offer.
Liam was profoundly affected by the vulnerability of homeless people and their shockingly poor healthcare outcomes. He started to look for people with whom he could collaborate to develop a new service. He identified Neil, a General Practitioner (GP), who had been providing healthcare services to the homeless community in his area. Liam said: … what persuaded me about him was that he had two rooms for in-patient homeless people on the ground floor with two kennels. (…) for the dogs. So, I said, 'This guy cares' . I wanted people who were compassionate. I wasn't going to work with people who weren't compassionate. (Liam) Liam and Neil were joined by a nurse (Frances) whom Liam had previously worked with and who was particularly experienced in encouraging patients to be involved in their own healthcare.
Care for the homeless has traditionally been based on the conventional medical model. The homeless person arriving in a hospital's Accident and Emergency Department will be treated for the medical condition they present. This usually ignores the complex and overlapping problems underpinning their condition. The conventional organisation of healthcare is designed with a 'standard patient' in view, and the shortcomings of this approach are exposed through the lens of the experiences of and outcomes for some of the most disadvantaged, vulnerable and unequal members of society.
Homeless people not only come in to contact with healthcare providers, but also with the wider social and welfare sector, including housing and drug and alcohol services. The risk, as Neil observes, is that no one person or organisation takes responsibility for the needs of the patient. He explained that the different parts of the health and social care system all find reasons not to undertake this role, with negative consequences for the ill person: Yes, you have a mental health problem but it's probably a personality disorder and anyway you're drinking too much. Yes, well, you could come to alcohol services but really while your behaviour is as bad as it is we don't know what to do you with. And what about your drug use? Yes well, drug services are interested but really they can't offer you any treatment until you start to think about your alcohol abuse. And what about your behaviour and is there a mental health problem? So everybody finds reasons not to help you and you get worse … (Neil, doctor).
The impetus then for developing an innovative model of healthcare for homeless people (and subsequently the creation of SBS) was the identification, naming and evidencing of the inadequate service and treatment available to the homeless community and its expression as a call to action or what we articulate as 'a cause' . Liam summed this up in a document prepared for NHS commissioners and service providers: The messy details of real people's lives and experience are often startlingly absent -no one mentions anyone with a name and no address. Martin Luther King famously said, 'Our lives begin to end the day we become silent about things that matter' . We who have the ability have the responsibility. Homeless health is a community problem, needing a community solution built on local ownership. This is a problem that cries out for compassion at the heart of the solution. The unacceptable has become the norm. We avert our gaze and have lost hope that anything can be done.
Dictionaries provide two dimensions to defining a cause: first, as a person or thing that gives rise to an action, phenomenon or condition, and second, as a principle, aim or movement to which people are committed, are prepared to advocate for and even to defend. Our evidence suggests that the cause of improving healthcare services for homeless people, one of the most marginalised and disadvantaged groups in society, gave rise to and provided the principle, or rallying call, for participants from diverse backgrounds to come together to create and recreate an innovation in homeless healthcare.
The notion of a cause taps into and draws on the originators' values, expertise and sense of professional responsibility, which have translated into the creation of a new multi-disciplinary, cross-sectoral approach focused on providing a compassionate, holistic healthcare service to a highly disadvantaged patient group. Learning from experience about the weaknesses of the existing service, provided a deep rationale for creating new structures and practices, and for mobilising agency and a sense of belonging amongst participants. We would argue then that the notion of a cause differs from and is more powerful than the more generic idea of an aim or direction driving innovation, in that it encompasses both the phenomenon (plight of homeless patients that gives rise to action) and a values-based principle (transforming and improving healthcare services for this group) to which participants have a shared emotional and ongoing commitment.
Finding, experiencing and establishing the cause is a process that, as we indicate in the next two sections, involves both translation in to new forms of work organisation and deliberative practices, and in to implicit work and learning processes that come to characterise SBS and through which this employee-led innovation is (re-)enacted.

Organising for innovation
Creating SBS and distinguishing it from the conventional medical approach included: (a) the founding of a new organisation; (b) the involvement of new kinds of participants and work roles; and (c) the creation of organisational practices enabling deliberative forms of co-participation. In seeking to understand the workplace as a learning environment, Fuller and Unwin (2004) explain how the way work is organised affects opportunities for learning. More expansive learning environments (and hence more likely to facilitate innovation) allow for 'substantial horizontal, cross-boundary activity, dialogue and problem-solving' practices (Fuller and Unwin 2004, 136), generating multi-dimensional, heterogeneous, distributed and reflexive forms of expertise. In contrast, more restrictive environments are characterised by forms of participation associated with narrowly defined jobs, fragmented tasks, constrained movements through time and space, limited diversity and opportunities for exercising autonomy and discretion. Our evidence highlights the relationship between the way the work in SBS is co-configured and its character as a learning environment.
A key decision was to create SBS as a charity which could raise funds, employ staff, and experiment with ways of working and job roles that would have been difficult within the constraints of the NHS. One of the doctors observed: The advantage of that [setting up a charity], it's kept us slightly at a tangent to the NHS so we're not entirely beholden to the NHS … How do you do a new thing? You can't prove the model until you've tried it and you can't try it without different funding arrangements … so it's [SBS] a body which can apply for funding, can pay staff to do these things. (Neil) The creation of a new organisation has given SBS what might be viewed as both insider and outsider status in relation to the NHS. It has insider credibility and legitimacy as a service commissioned by the NHS, and through its employment of registered doctors and nurses. As a charity, its autonomy flows from its charitable purpose and ability to operate alongside but not under the direct governance the NHS. Distinguishing itself from the conventional medical model and distancing itself (somewhat) from the NHS appeared to play a central role in the forming and reforming of SBS' organisational and community identity.
One of the most powerful ways in which SBS illustrates its employee-led innovative capability is through recognising the value of, and involving, diverse expertise, including non-clinical practitioners and those with experience of using healthcare services as homeless patients. A key premise is that there is significant value in learning from service users. This assumption is at the heart of the innovation and is operationalised in two ways. First, through the opportunity for ex-homeless people to become SBS employees through undertaking a newly created job role of 'care navigator' . These colleagues are considered to be 'experts by experience' derived from their own significant experience of homelessness. The care navigators are paid members of the SBS team working alongside doctors, nurses and other health and social care professionals. One of the doctors commented: That's a very novel thing and for you to get your head around what they [care navigators] actually are, what skills they possess and how they sit with everything because they're key to achieving the Service's goals really. (Sam) The experience and expertise of the wider homeless community is mobilised is through their recruitment as volunteers. A core member of the SBS team, Bill (who also came from a homeless background), works with this group to involve them in a range of activities. For example, they act as participant researchers collecting information from other homeless people about their experiences of using the healthcare system. They also feed into the development of standards of healthcare for homeless people. Bill's job is to generate the spaces and places of involvement and belonging that enable their voices to be heard through their contribution to the conversation about continuous service improvement between SBS and the wider health and social care service community. The volunteers have a clearly defined role in the making and remaking of SBS practice: Going through your 'lived' journey has given you lots of experiences. By working with [SBS] and sharing them with a wider audience you can be part of a process learning and teaching. You can help others to understand what it was like being homeless, what worked and what didn't when it came to your own health and recovery. (Bill writing in the SBS Experts by Experience Handbook) Whilst the core SBS team is relatively small, the approach extends across disciplines, sectors, and hierarchies to create a dynamic and extended collaborative network characterised by an adherence to, and involvement in, the 'cause' , and the sense of collective belonging to which this gives rise. Belief in the value of a cross-sector approach to improving homeless health and recognition that expertise is distributed across different disciplines, groups, and diverse agencies, are integral to the character of the innovation and provide the rationale for creating explicit, transversal practices. These include regular multi-disciplinary team meetings and the founding of a Faculty for Homeless and Inclusion Health (FHIH) With regard to the former, one of the doctors explained: So one of the things I found … was that one of the characteristics of successful interventions for homeless people was a weekly multi-agency meeting in which the CPNs [clinical practitioners] got together with the street outreach people, with the GP, with the nurse and everybody got together and talked through. (Neil) The weekly, multi-disciplinary, multi-agency team meetings provide a place to discuss individual cases from a variety of perspectives and in a collaborative way, and to consider how patients might be best supported on discharge from hospital.
The FHIH has been created as a space for those (clinical and non-clinical practitioners, and experts by experience) with a commitment to this area, to come together to share ideas, learn from each other's experiences and provide mutual support: … the network brings together clinicians and people with experience of homelessness in the Faculty. Members include bicycle paramedics, podiatrists, dentists, professors of epidemiology, psychiatry and infectious diseases, general practitioners, hospital consultants and practice, specialist and district nurses, physiotherapists, psychotherapists, psychologists, counsellors, drugs workers and alcohol specialists. (Faculty website accessed 19 July 2017) Open and free membership of the Faculty set it apart from the traditional medical Faculties of the Royal Colleges that oversee medicine in the UK, affirming and valorizing its distinctiveness and underpinning principles of cross-boundary, cross-sector, multi-disciplinary working, learning and involvement. Its practices include regular meetings, and an annual conference, providing fora for health and social care professionals and 'experts by experience' to come together, learn from each other, and where they can mobilise around the cause of service improvement. A field note from observation of a Faculty meeting indicated: There was a strong discourse about the importance of shared learning as a means of improving practice. There seemed to be a general commitment to ensuring the means to combine the efforts and energy of everybody working in this area, 'creating a voice' and moving from talking about issues to 'collective action' .
Another explicit practice is the monthly Reflective Practice meetings. These sessions are facilitated by a Psychologist and aim to provide a structured and 'safe' space for participants (belonging to the extended SBS community) to share their experiences and support each other in relation to the challenges they may be having in their work with homeless patients. The focus is on supporting emotional well-being: … we get to talk about patients and complex people and how we work and how we deal with some of the patients we get in, and what we actually feel like and what it is to go through when people make us stressful, how do we sort it. So we get that time for ourselves … so we hand over the bleep and the phone. It's quite good. (Care navigator, Bernie) Our observation of a Reflective Practice meeting indicated that it served the purpose of validating the SBS approach as different to the conventional model: Values and compassion seem to be a very important part of being part of [SBS]. In some ways, although it leads to frustration, the lack of compassion in other services is important to [SBS] as it marks them out as being positively different. One of the participants said something like: 'We become more compassionate than the norm, we're the abnorm!' … 'We have emotional attachment -incredible strength but it's abnormal and quite special' . (Observation field note) The explicit organisational arrangements and practices described here have generated the conditions for expansive trajectories of co-participation, co-figuration and belonging across disciplinary, sectoral and provider -user boundaries to emerge. In so doing they underpin and characterise the enactment and re-enactment of the innovation process. We now explore how this is complemented by and interwoven with day to day work and learning.

Innovative capability in practice
Our evidence indicates that the way work is organised and co-produced in everyday practices in and through SBS exemplifies its innovative capability and how the knowledge and agency of team members aligns with and is mobilised through commitment to the cause. Initially, we draw on the lived experience of one of the care navigators, Millie, to illustrate this. Here, she shares her early experience of becoming part of the team: Because I came from a background of homelessness and from drinking and things for twenty years, so when I came in I was a bit, like, 'Oh gosh, I wonder what people will think when they see …' But now I'm accepted and it's really nice … I don't feel there's them and us, I feel we are a team, and the doctor talks to me, he doesn't ever talk down.
Millie indicated that as 'experts by experience' , the care navigators can empathise with patients as well as knowing that it is possible for their situation to improve. She said: 'it can be done, I'm living proof' . The care navigators undertake a wide range of activities working with clinical and non-clinical colleagues across the hospital and beyond. They advocate for patients during their stay in and discharge from hospital, also moving across sectoral boundaries, for example, by accompanying individuals to meetings with officials about their housing status; helping them find accommodation, liaising with family members, and supporting them in their rehabilitation.
In this example, Millie talks about a homeless person who has presented at the hospital but has no record of living in the area: And I think it was just -I was blessed the day I went up [with him to the local authority housing office], I think the woman felt sorrier for me actually. I explained the case and said what had happened, and she actually got him into a hostel and he is there now, and my other care navigator … has gone down with him today to do the housing benefit.
Participants accounts indicate that much of their day-to-day practice is non-routine, necessitating a creative 'can do' approach. As Millie observed: 'there are just so many different things someone could come in with. Each day is completely different, you don't know what to expect' .
Our ethnographic approach to data collection enabled us to observe some of the ways in which stories are used to reinforce the value of the cause, and as a situated pedagogical practice (Wenger 1998). The SBS team regularly face situations that would be unusual for staff working in conventional medical practice settings. For example, one of the nurses told the story about a doctor who refused to see a patient: I remember this chap came in and he was, yuck, and I was heaving because he smelt so bad. And the consultant said, 'I'm not seeing you. You can go' . I said, 'but he is ill' . 'No, he is fine. You can go' . (Frances) Finding the patient's access to healthcare blocked, Frances thought creatively about how to solve the problem recounting how she took him for a shower and found him some clean clothes: I put him back in to his cubicle and said to the consultant, 'will you come and see the patient?' He said, 'I'm not coming to see him' . I said, 'you'll come and see him now' . So he comes in says 'well this isn't the patient' . I said 'Because I've given him a bath … Is it only because I've given him a bath that you will come and see him?' He said, 'yes' . It's awful, absolutely awful but he was very ill this chap. (Frances) This example shows that taking up and enacting the cause requires ingenuity and determination, and in some cases resistance to hierarchical authority and status.
The co-location of SBS members in a shared office created the conditions and resources for co-participation in work and learning. An observation field note states: The team seem to spend a fair amount of time together in the office and were able to discuss the needs of different patients and the situations they were in. These conversations involved a lot of information sharing and learning from each other's experiences and expertise, and tasks were shared out according to who had the expertise to deal with them. (Field note April 2014).
Everyday SBS practice also involved ways of relating through co-participation across time and space, for example, when working on the move. Staff spent time with their clients taking them to appointments, including by car. This provided a fruitful opportunity for relationship building, information gathering and negotiation of meaning, as this observation field note indicates: In the car [care navigator] asked about the benefits he [the patient] is claiming and was concerned that he should be claiming for everything that he is entitled to … [case worker] found the time in the car to be very useful, as it gives him an opportunity to talk to them about their situation in an informal setting … in the car people are less guarded and he is able to bring up sensitive issues more subtly. (Field note April 2014) The innovative capability of SBS is continually refreshed through the kinds of everyday opportunities for workplace learning associated with colleagues' participation in the dynamic and transversal social relations and organisational practices through which the innovation is continually re-enacted and the cause is served.

Dimensions and processes of employee-led innovation
Arising and abstracting from our empirical account, we have identified five key inter-related dimensions which underpin and help conceptualise the work and learning that it took to create and (re-)enact SBS and that have potential to be generalizable to other employee-led innovations. We argue that these act as processes which continue to mobilise participants' involvement and commitment, and the innovative capability of SBS. They align with and contribute to our overall theoretical objective to understand employee-led innovation not as a one-time event but as a dynamic, relational and socialised process in SBS, healthcare and perhaps beyond in to other sectors.
The first dimension is 'disrupting' . The death of a homeless man in close proximity to a major London hospital provided the catalyst for a senior consultant to ask fundamental questions about the treatment of homeless people by the health service and expose, through the collection of extant data, their over-representation in hospital admissions and extremely poor health outcomes compared with the housed population. One of the originators (Neil) referred to how their 'shroud waving statistics' provided a warrant for disrupting the taken for granted conventional approach to healthcare provision, which failed to distinguish and respond to the special and complex needs of this highly vulnerable and disadvantaged patient group. Evidence and critical thinking created a space for identifying and articulating the way that homeless people were experiencing health care services, helping to develop the case for innovation, and its support. 'Disrupting' the existing situation, opened up a debate about the shortcomings of how homeless patients are treated by the healthcare system. It was a key initiating feature of the innovation but this aspect of the cause continues to drive participants' day to day capacity to challenge, disturb and resist aspects of the health and wider welfare system that are perceived not to be acting in the best interests of their patients.
The second dimension is 'othering' . In drawing conclusions about the failings of healthcare provision for homeless people, it was critical to conceive a clearly distinctive and contrasting alternative. Having a moral purpose (a cause) infused the development of the innovation with an appeal to compassion as an underpinning organising principle. The imperative to provide a compassionate service to homeless people enabled it to attract diverse participants into the SBS community and feeds into its ongoing discursive differentiation from the previous approach. This dimension was also entwined with the idea of modelling, whereby Liam identified the importance of involving key individuals, a doctor and nurse, Neil and Frances who epitomised, embodied and modelled the values, experience, and attributes that would need to be mobilised in the co-conception and development of a compassionate approach. The process of othering illustrates that differentiation is fundamental to the dynamic of the innovation, and sense of belonging and lived experiences of the participants (and that are not available to non-participants).
The third dimension relates to the innovation's insider-outsider identity. Being separate from the NHS, but involving registered clinicians, facilitated experimentation and ways of practising which legitimated challenge to the existing approach. The inclusion of 'experts by experience' as staff as well as through the volunteer programme turned on its head the stigmatised societal view of homeless people as outsiders on the margins of society, by recasting, including and valuing them as experts in the lived experience of homelessness and healthcare services, with this as their warrant for inhabiting an insider status in the innovation. The creation of the FHIH appropriated the nomenclature of medical faculties and their discourses of and authority for specialism in the guise of insider, but constructed a very different outsider perspective with membership open and free to anyone, clinical or non-clinical (including homeless people) with a commitment to 'the cause' . Playing with the insider-outsider dynamic provided the scope and capability for SBS to enact a hybrid identity.
The fourth dimension is 'storying' . This captures how the discursive resources underpinning the development and articulation of the SBS concept and negotiating its meaning, enabled it to presuppose the form, processes, practices and expectations of the new approach. As we have tried to show, SBS was created in reflexive relation with the ongoing processes of establishing and enacting the cause, and in the context of the stories told (and retold) about the contradictions and weaknesses perceived in the existing model of healthcare and how a different approach was possible. Telling stories as a pedagogical practice continues to be an important element in participants' everyday workplace learning.
Through examination of our findings it has become clear that without the perspective of organisation, understandings of work, learning and innovation are limited both analytically and prescriptively. Our analysis has shown how choices about the form of organisation and its practices are generative of diverse kinds of co-participation. This brings us to the fifth dimension -'transversality' . SBS has challenged and crossed traditional boundaries, evidenced in a variety of ways including multi-disciplinary working; the inclusive and distributed concept of expertise; and, working across health and social care, local authority and welfare sectors. Wenger (2009, 17) makes the point that 'maximising learning capability requires all sorts of transversal processes that cut across dimensions' (emphasis in original). In our case study we found that SBS' transversal design and character was associated with expansive organisational and pedagogical features underpinning its innovative capability. These included its extended relationships and explicit and implicit opportunities for co-participation and co-production (e.g. the multi-disciplinary team meetings, reflective practice sessions, FHIC forum) as well as the day to day social relations of the innovation characterised by working and learning with others across sectoral and disciplinary boundaries, including from service users in ways that facilitated the circulation of ideas, dialogue and multiple ways of knowing, whilst reducing the risk of stasis.
In this section, we have identified a range of inter-related dimensions that help explain the origins, and ongoing work and learning processes that characterise the innovative capacity of SBS, and that we suggest could help conceptualise other employee-led innovations in the healthcare, and possibly in the wider social sector.

Conclusion
The purpose of this paper has been to identify and understand the work and learning that it has taken to create a new employee-led approach to and way of providing healthcare services for homeless people. To achieve this we have brought together and utilised conceptual insights from employee-driven innovation, and social and practice-based learning and organisational theories to provide a framework for understanding and explaining our evidence. The notion of a cause, understood both as something that gives rise to a call for action with which people from diverse backgrounds and forms of expertise wish to align, and also as the principle that makes something else happen, has been helpful. The cause of improving healthcare for homeless people has provided the impetus for participants in SBS to transform health services for a highly disadvantaged and marginalised group. In this way our analysis augments complementary social practice-based understandings of work and learning by focusing the role of cause in illuminating how SBS, as an employee-led innovation, is collectively made and remade in and through new and changing forms of practice.
Invoking an employee-led narrative of innovation and learning steered us to adopting a practice-based orientation as an important conceptual lens, and highlighted the complementarity between a social practice-based theorization of workplace learning, and SBS (and employee-led innovation more generally) as a social and collective process. Whilst Liam's experience of the death of the homeless person outside his hospital acted as a catalyst, enacting and re-enacting the innovation involves the mobilisation of people who align to the cause and co-commit to improving (continuously) the experiences of, and outcomes for, a highly vulnerable and stigmatised patient group.
We drew on insights from workplace learning and organisational studies to ensure that work, the way it is organised and co-configured as a social process, was foregrounded in our empirical investigation and analysis These perspectives have highlighted the importance of distributed, multi-disciplinary, cross-sector working practices, and the involvement of diverse kinds of expertise, including experts by experience, as characterising SBS' innovative capability and its nature as a learning environment. It has also informed our understanding of the relationship between work, learning and innovation, supporting Brown and Duguid's contention that: To see working, learning and innovating are interrelated and compatible, and thus potentially complementary not conflicting forces, requires a distinct conceptual shift. (1991,40) The shift they refer to requires the adoption of a social, situated theory of learning. This rejects the notion that learning and knowledge can be straightforwardly transferred between contexts and settings, and from individuals who have the knowledge and learning to those without. Rather, as articulated by Lave and Wenger (1991) and Lave (1993), learning is understood as a social process inextricably linked to meaning making through co-participation in changing social practices, of the kind we have outlined. These provide the resources, opportunities and conditions that constitute what participants learn. Taking a practice-based perspective on the relationship between work, learning and innovation, has provided a lens through which to identify diverse deliberative and everyday practices that characterise SBS. We argue that these are mutually co-constitutive as part of a fundamentally interwoven and dynamic relationship between context (the experience of and outcomes for homeless patients in the UK), community (SBS participants) and practice (organisational and pedagogical actions/practices). In this regard, we have found Gherardi's focus on the 'practices of a community' helpful (2009, 528): The attention thus shifts to how practical knowledge is enacted in situated contexts of action, and how it changes in relation to the context of action that the action itself creates.
We suggest that the notion of a cause provides a bridge between the generic notion of an aim as driving (any kind of ) innovation and the more specific values-based driver for employee-led innovation in healthcare. Our evidence has illustrated the way in which the underpinning cause has provided an essential appeal to participants with different positions and roles within SBS, and from diverse social, occupational and disciplinary backgrounds and perspectives to learn from their experiences in collective pursuit of the improvement of healthcare for the homeless population. In this way, SBS has forged a new community (drawing from a range of other communities) founded on a shared valorization of the expertise associated with different members and on their collective involvement and commitment. Hence, rather than choosing between Lave and Wenger's emphasis on 'community of practice' or Gherardi's on 'practices of a community' , we suggest that a twin, inter-dependent focus on both community and practices bridges the gap between their two perspectives, and is needed to understand the innovation.
From this perspective, we conclude that SBS is making and remaking its own history as a dynamic and expansive community of practices.