The challenges of joint working: lessons from the Supporting People Health Pilot evaluation.

PURPOSE
This paper reports the findings of the evaluation of the Supporting People Health Pilots programme, which was established to demonstrate the policy links between housing support services and health and social care services by encouraging the development of integrated services. The paper highlights the challenges of working across housing, health and social care boundaries.


METHOD
THE EVALUATION OF THE SIX HEALTH PILOTS RESTED ON TWO MAIN SOURCES OF DATA COLLECTION: Quarterly Project Evaluation Reports collected process data as well as reporting progress against aims and objectives. Semi-structured interviews-conducted across all key professional stakeholder groups and agencies and with people who used services-explored their experiences of these new services.


RESULTS
THE ABILITY OF PILOTS TO WORK ACROSS ORGANISATIONAL BOUNDARIES TO ACHIEVE THEIR AIMS AND OBJECTIVES WAS ASSOCIATED NOT ONLY WITH AGENCIES SHARING AN UNDERSTANDING OF THE PURPOSE OF THE JOINT VENTURE, A HISTORY OF JOINT WORKING AND CLEAR AND EFFICIENT GOVERNANCE ARRANGEMENTS BUT ON TWO OTHER CHARACTERISTICS: the extent and nature of statutory sector participation and, whether or not the service is defined by a history of voluntary sector involvement. In particular the pilots demonstrated how voluntary sector agencies appeared to be less constrained by organisational priorities and professional agenda and more able to respond flexibly to meet the complex needs of individuals.


CONCLUSION AND DISCUSSION
The pilots demonstrate that integrating services to support people with complex needs works best when the service is determined by the characteristics of those who use the service rather than pre-existing organisational structures.


Introduction
The importance of housing support is widely accepted within the English policy context. Its significance was reinforced with the launch, in 2003, of the Supporting People Programme, which put housing related support at the centre of the government's strategy to enable vulnerable people to live independently. The programme also reflected another key theme of government policy-joint working. The Supporting People programme is a working partnership between local government and other statutory and non-statutory agencies. This article presents some of the findings from our evaluation of six pilots, which was designed to explore how organisations work across boundaries. We illustrate the challenges of providing integrated care across housing and health care services. We highlight what helps and what hinders joint working across agencies and sectors, at both a strategic and operational level, and emphasise the important role that voluntary sector agencies can play in supporting people with complex needs.

The policy context
Housing was recognised as a cornerstone of social care in England in the 1990, NHS and Community Care Act. Attention initially focused on housing adapt-This article is published in a peer reviewed section of the International Journal of Integrated Care ation and the development of independent living as a means of helping specific groups of people-older people and people with learning disabilities-to live independently in the community w1x. Over time a more holistic notion of 'housing support' has developed. This encompasses the provision of social housing, in some cases with 'extra care', together with 'floating' (peripatetic) services designed to enable people to live on their own. Housing support is provided to a wide range of people including young parents and exoffenders, and its importance to adult social care was identified in the 2005 Green Paper w2x.
Although the relationship between housing and health is well-established, the structural links between housing services and the NHS have historically been poor w3x. Recent moves towards a more explicit public health agenda have, however, been mirrored in public policy, with a clear role set out for housing and housing support. For example, Easterlow and Smith suggest that contemporary public health policy focuses on promoting 'healthy public policies, inter-sectoral alliances and community care, all of which have housing at their core' w4x. The White Paper 'Our health, our care, our say' notes that, wherever possible, people should 'have the option to stay in their own homes,' and that this may necessitate the provision of intensive support at home allowing more people 'to continue to live in their own homes for longer' w5x.
The Supporting People programme encompasses these trends. The programme was introduced in England in 2003 as a means of facilitating independent living in the community for groups that require low-intensity support and also for those that are socially excluded, at risk or hard to reach through existing service provision. Its broad aim is to provide housingrelated support to help people to stay in their own homes or to move towards having their own homes, to increase independence and the capacity for selfcare. The programme reflects wider policy aims associated with preventive action, tackling social exclusion, co-ordinating services around the needs of individuals, promoting choice and increasing the role of the voluntary sector w6x.
The Supporting People programme was launched by the then Office of the Deputy Prime Minister in April 2003 and is managed by local authorities. It is designed to be delivered through a working partnership of local government, health services, voluntary sector organisations, probation, housing associations and support agencies. Together these organisations commission services that will enable vulnerable people to develop and sustain their capacity to live independently. Supporting People services, by their very nature, are intended to provide support to individuals across organisational boundaries.
Although the partnership ethos is central to the programme, the relationship between health care and Supporting People services has never been considered strong w7x. Indeed there has been a perception amongst policy makers that health planners and practitioners were not engaged in local Supporting People partnerships, perhaps because they did not understand its relevance to the health agenda. To remedy this the Office of the Deputy Prime Minister announced its intentions to establish the Supporting People Health Pilot programme in the summer of 2003, with the intention of demonstrating: the policy links between Supporting People and health and social care, the potential benefits from collaboration and how practitioners and agencies could work together to support the health needs of particularly vulnerable groups.

The joint working context
Joint working is a longstanding policy theme. Within health and social care in England it has recently been associated with attempts to modernise service provision w8x. Policy reform has focused on: enabling the statutory sector to pool budgets and jointly commission services w9x; establishing new models of provision as a way of ameliorating some of the difficulties of working across organisational boundaries (for example the creation of Care Trusts w10x), and encouraging the development of integrated services w11x. Recent reforms have encouraged greater contribution from the voluntary sector ensuring that partnerships reach out beyond the statutory sector w12x.
There is a rich vein of research focusing on joint working between health and social care services w13-17x, that highlights a number of difficulties associated with organisational, cultural and professional and, contextual issues. Although the literature on joint working between housing and health is less well-developed, similar difficulties have been noted w3, 18-21x. Kodner and Spreeuwenberg (2002) defined integration as a set of methods and models of organisational service delivery designed to create connectivity and collaboration within and between different sectors, with the aim of enhancing the quality of life for people with chronic or complex problems w22x. This conceptualisation mirrors the basic aims of the Supporting People health pilot programme, which sought to co-ordinate services across housing, health and social care boundaries as a means to support vulnerable people to live independently in the community. It is the experiences of those involved in these endeavours that the article aims to explore.

The pilots
The Office of the Deputy Prime Minister received bids from 122 partnerships in England wishing to be Health Pilots. The six pilots selected by the Office of the Deputy Prime Minister represented a wide range of people who use Supporting People services and a range of agencies from the statutory, independent and voluntary sectors. As a precondition of funding each pilot was required to identify aims and objectives, which would demonstrate how the new service could support national health targets. Details of the pilots are provided in Table 1.

Methodology
Evaluating a diverse range of projects presents a series of challenges. The approach adopted sought to understand both the process and outcome of the intervention as a means to determine what works, for whom and in what circumstances w23x. The evaluation sought to identify common themes and issues as the basis of an overall evaluation of the initiative per se, but took a tailored approach to the evaluation of each individual project.
The methodology rested on two assumptions. First, that the views of all key stakeholders, including those using services, were important. Secondly, that as far as possible relevant data should be collected on as contemporaneous a basis as possible. Two main sources of data collection were used: Quarterly Project Evaluation Reports from each pilot and semi-structured interviews with key stakeholders, including people who use services. The Project Reports provided data about process and implementation as well as reporting progress against aims and objectives related This article is published in a peer reviewed section of the International Journal of Integrated Care to health target(s) which pilots had been required to identify.
Each pilot was visited on three occasions: at their inception; mid way and towards the end of the initiative. During these visits we conducted interviews with between six and eight interviewees across all key professional stakeholder groups, including commissioners, managers of services, representatives of partner agencies as well as project workers. Interviewees were selected on the basis that they were involved in the development of the pilot andyor that they were centrally involved in the work of the pilot. Interviewees were asked their views on whether or not the pilot was achieving its aims and objectives and to describe the factors that supported or hindered efforts to work across housing, health and social care boundaries. People who used the services were also interviewed during each visit. The interviews allowed us to explore the development of the Health Pilots from a range of perspectives w24x. Interviews were transcribed and analysed thematically, emerging themes were discussed within the research team and, as a means to increase the saliency of the analysis, we checked the authenticity of our findings with representatives of the pilots at regular workshops w25x. In this way we were able to identify trends that are generalisable across the pilot sites and beyond w26x.

Results
In the following sections we discuss the themes emerging from the data, highlighting the challenges entailed in providing integrated care across housing and health and social care services.

Understanding the aims and objectives
Previous research has consistently emphasised the importance of agencies understanding the basic aims of their joint effort w17x. The experiences of the pilots support this. They illustrate that joint working was most effective when professionals not only understood the aims and objectives of what they were doing, but appreciated the fundamental role of housing as a means to support the individual people they worked with. In essence, dry exhortation to work in partnership was an insufficient catalyst; the potential benefits of joint working had to be apparent to individual professionals through the experience of the people they worked with.
For example, the idea behind the 'On-Track' pilot came from professionals working in local health and voluntary services who identified a gap in services for young people with dual diagnosis (people with mental health and substance misuse needs). These young people were 'falling through the net', being passed between youth to adult services and between drug and mental health services. Professionals working in different sectors recognised the need for stable housing and intensive support in order to engage young people with dual diagnosis in relevant health care services. They shared an understanding that only rarely could they achieve their own organisational aims and objectives when working in isolation. The pilot was therefore designed to establish a housing related support service that would help young people to engage with appropriate health services. The aim, as one partner described it, was to provide practical help to This level of understanding about the central role of housing was not initially evident across all pilots. For example, the 'Place to Live' pilot was inspired by the Valuing People White Paper w27x, which calls for people with learning difficulties to be given more control over where they chose to live. Although originating from discussions within a multi-disciplinary group, the original proposal was written by social care professionals with little or no involvement from community nursing staff whose participation was crucial to ensuring that individuals moving into supported housing had access to appropriate health care service. The initiative was seen as a social care project. As one social worker reflected: ''things should have been done differently during the planning stage; we needed to take more people on board, it was important for everyone to own the project.'' As well as lacking a shared understanding of what the pilot was trying to achieve, few community nurses appeared to recognise the role of housing to wellbeing. For example, one health professional described how social workers and community nurses had very different ideas about what constituted a 'health need' and what constituted a 'social need'. As a result few referrals were initially made by community nurses for housing assessments for the people they supported.
The 'Place to Live' pilot also illustrated that staff working in allied services, particularly the homelessness unit, neither appreciated the housing rights of people with learning disabilities nor understood the links between housing and well-being. This was a theme echoed elsewhere in the evaluation. Four of the six pilots identified the need for staff working in homelessness units or hostels to have training about the housing and support needs of vulnerable people. For example, housing department staff associated with the 'On-Track' pilot did not appreciate the need to house young people with dual diagnosis away from housing estates known to have a drug problem. Training resulted in improved working between these agencies and also improved the support they provided to specific individuals. However, in keeping with previous research w28x this raises serious questions about the role of generic housing officers who, within the current policy framework, have to work with a range of individuals with complex lives and health problems, which have a major impact on their ability to live independently. Inevitably this requires housing officers to work more closely with other services but without specific knowledge of the needs of different groups, or a specialist service to provide this support, this type of co-ordination may prove difficult to achieve.

Having a history of joint working
The importance of building on existing relationships between agencies was evident within the evaluation.
For example, at a strategic level the 'Housing Support Outreach and Referral' pilot evolved from a long history of partnership working in the field of HIV services between two London Boroughs. Not only had the Boroughs previously commissioned HIV services together but they had also worked in partnership with the Primary Care Trust, which commissioned voluntary sector services. This approach was generally regarded as an effective way of addressing a complex problem. As a result the pilot did not have to spend time developing an ethos of joint working 'from scratch.' The two Boroughs commissioned the Terrence Higgins TrustyLighthouse to develop the housing support service. Although the organisation had no experience of providing housing support services they had an established record of providing a range of services to people living with HIV and a wealth of experience of working with acute health care providers. This led one partner to comment ''I know (the) Terrence Higgins Trust, I know their roles, that helps, we expect realistic and appropriate things from each other.'' The pilot was, therefore, developed within a context that valued joint working as a means to support people living with HIV and benefited from a history of collaboration amongst the central partners.
The 'SWAN NEST' pilot also profited from a history of joint working, building on the success of the SWAN programme, a multi agency initiative originally developed to address community safety issues. As well as providing accommodation and 'floating' tenancy support, pilot staff also assisted women to engage with health services as part of a strategy to exit sex work. All of the agencies involved recognised that they could only set up the scheme by working in partnership and that joint working had been made easier because as one partner reflected, they had had a ''positive experience of joint working in the past, we trust each other, you will deliver because you have in the past.'' This shared history, however, was based on individuals who had worked together over a long period of time. This made the partnership fragile. Reorganisation of statutory and non-statutory agencies was a constant threat and partners recognised that changes in personnel could unsteady the partnership. Consequently no one took the partnership for granted and partners were willing to invest considerable time and energy into the development and subsequent management of the new service.

The management of inter-professional working
One of the key themes to emerge from the evaluation was the need for joint working to be based on clear arrangements in respect of governance and managerial responsibility, both at strategic and operational levels. Not only does this help establish the democratic accountability of partnerships w29x but transparent arrangements, agreed by all partners, ensure that staff understand to whom they are accountable and enable the work to be managed effectively. Someone needs to be ultimately accountable for the project. However, as previous research indicates w30x, insufficient thought is often given to the complexity of managing complex initiatives. For example, at several pilots the difficulties of sharing information across organisational boundaries were not resolved because it was not clear where responsibility for this lay.
The governance arrangements for the 'SWAN NEST' pilot provide a model for how these complex processes can be made straightforward. Progress was reported to the existing SWAN Partnership Steering group, which met bi-monthly and comprised of senior representatives from the main partner agencies, including the borough council, the Primary Care Trust, the general practice and the specialist housing support agency. Additionally, because the Primary Care Trust held the contract for the pilot with the Office of the Deputy Prime Minister, the Assistant Director for Public Health reported progress to the Primary Care Trust board. These arrangements ensured not only that one individual took overall responsibility for the pilot but also that there was an effective forum in which to tackle problems across the partnership. The pilot also had an operational manager who provided the link between strategic and front line working. This was crucial to the pilot's success at meeting their aims and objectives.
The pilots also demonstrated that operational staff working across organisational boundaries need specialist supervision, as well as managerial supervision. This became evident early on, specifically in pilots that were working with people with particularly complex needs and chaotic life styles such as sex workers and homeless people with HIV. Staff at these pilots had to work intensively with individuals in order to link them into a variety of general and specialist health services and other agencies such as housing and probation. Not only did this require them to have a detailed knowledge of a range of services it also required them to have an understanding of how best to support individual clients. Through the provision of specialist supervision, pilots were able to ensure that the practice of individual workers was effective and safe. It also provided them with time to 'off load' and reflect on the difficult nature of the work they were doing.
For example the 'Housing Support Outreach and Referral' pilot held regular team meetings, sometimes on a daily basis depending on the cases they were covering. As one of the workers described it, Not all of the pilots recognised the need for professional supervision systems. The 'On-Track' pilot, for example, did not initially establish formal supervision systems with the result that project workers were anxious about the appropriateness of the support they were providing. These systems play a crucial role in maintaining professional standards w31x and as more integrated services develop agencies will need to provide matrix supervision systems that allow workers to receive professional support for the different elements of their work. After several months the 'On-Track' pilot ensured that workers had access to supervision from housing managers as well as from specialist drug agencies. Whilst these systems are already common place in many voluntary sector agencies it will be important to ensure that as the sector begins to play a bigger role in the provision of support to vulnerable groups these systems are in place in all agencies. One way to do so would be for commissioners of services to require these arrangements as a contractual condition w31x.
The role of the voluntary sector 9.1 In this evaluation, effective joint working (defined as the pilots ability to achieve its aims and objectives) was associated not only with an understanding of the purpose of the joint venture, a history of joint working, and clear and effective governance arrangements, but on two other characteristics: the extent and nature of statutory sector participation and whether or not the service is defined, by a history of voluntary sector involvement.
Those pilots working in service areas with little or no tradition of statutory sector provision-for example with sex workers, or where services have developed more recently (HIV services)-appeared to have less difficulty working across organisational or professional boundaries. Indeed these pilots seemed to be based on a profound sense of 'needing' to do something to fill a gap in provision. The HIV sector, for example, has a strong ethos of partnership working across the statutory and voluntary sectors, which appears to lend itself towards a more flexible approach to supporting vulnerable people. As one health partner from the In contrast, although the core partners in pilots working in the fields of learning disabilities and older people services displayed a high level of commitment to joint working, this did not always appear to be as widespread within the agencies concerned. For example, the 'Place to Live' pilot was based in an integrated team but the different professional groups were not co-located, nor did they meet together on a regular basis or share an understanding of the relationship between housing and well-being. All of these factors are associated with successful joint working w17x and without them, the pilot struggled initially to develop an ethos of joint working. Ironically, although there is a long history of community nurses and social workers working in the field of learning disability and, more recently in integrated teams, this seems to have led to the emergence of strong professional boundaries that may inhibit multi-professional working. As one social worker reflected ''in integrated teams people can get precious about their roles.'' The pilots demonstrate the important contribution that the voluntary sector can make in supporting vulnerable people to live independently in the community. The involvement of the voluntary sector brought additional credibility to the work of several pilots. As well as harnessing the expertise that exists within the voluntary sector, pilots were able to draw on their networks. For example, the involvement of the Terrence Higgins Trust in the 'Housing Support, Outreach and Referral' pilot provided extra credibility because the Trust was well known amongst people with HIV and brought with it an extended network of voluntary services that the Trust was involved with. Consequently the service was able to refer people to a wide range of voluntary organisations, such as community transport services and charitable food projects. As previous research has found w21x it is support with these practical tasks that helps enable service users to live independently.
The development of new services in the voluntary sector also provided powerful models of how services could be provided. For example, a health professional involved in the 'On-Track' pilot remarked that the pilot had demonstrated how, ''The traditional ways of providing services from the statutory sector aren't always the best and that other providers can do it more successfully.'' Importantly those pilots that were based outside of the statutory sector demonstrated a high degree of flexibility in the way in which they supported vulnerable people to live independently. Pilots worked intensively with individuals to identify what they wanted to address in order to live independently and supported them to achieve the goals they set for themselves. This person-centred approach helped people engage and maintain engagement with services, which they had often failed to do in the past and echoes findings from previous research which suggests flexible and personcentred housing support services are more likely to offer an effective solution to social exclusion w21x. The appropriateness of this approach was alluded to in interviews conducted with people who used services.

Difficulties associated with working across statutory, non-statutory boundaries
The involvement of voluntary sector agencies in the Health Pilots was not without difficulty particularly with regard to how professionals from different sectors work together. Several pilot workers reported that statutory sector colleagues were reluctant to work in partnership with them. For example, despite the tradition of cross sector working within HIV services some tensions were reported from the 'Housing Support Outreach and Referral' pilot. These were most notable amongst staff working in a homeless persons unit who were initially reluctant to identify a link worker between the pilot and the unit. This reluctance appeared to be based on a perception that voluntary sector organisations were not as 'professional' as statutory services. One of the outreach workers commented that some ''statutory services see us as dogooders, they don't see us as a professional service or as an equal.'' These professional rivalries were most notable when pilots tried to establish effective ways of sharing information with colleagues in the statutory sector. These systems are particularly important when people have complex needs and chaotic lifestyles. In these circumstances services need to be co-ordinated in a timely manner and based on up-to-date information.
Initially some statutory sector agencies were reported to be unwilling to share information with colleagues in the non-statutory sector. To resolve this problem several pilots, including 'SWAN NEST' and 'On-Track' built on local practice, for example adapting existing 'release of information forms' which service users were asked to sign as proof that they had agreed to the pilot contacting other agencies as a means to seek or share relevant information. Occasionally the authenticity of these forms became the focus of disagreement between agencies. Whilst these disagreements were always resolved they reflect what Secker and Hill have referred to as 'a reluctance or structural inability to share information and a lack of clarity about the constraints of confidentiality' w32, p 348x. These disputes support the need for national guidance and protocols, which would remove the potential for disagreement and would smooth the path of those working at the front line.

Discussion
The Supporting People health pilot evaluation provided an opportunity to explore joint working between health, social care and housing across a number of different contexts. Consequently the evaluation was able to highlight themes that help explain why joint working is successful in relation to some service areas and not others.
The experiences of the pilots echo themes identified elsewhere in the literature. They suggest for example that whilst some agencies and professionals understood the central role of housing to health and wellbeing and were willing to work across organisational boundaries to support people to live independently other pilots found it difficult to develop a shared purpose. The 'Place to Live' pilot established within an integrated learning disability team struggled initially to move beyond what Secker and Hill have termed 'rigid demarcations and role boundary conflicts' w32x.
The provision of training eventually helped to build a sense of common purpose between the different professional groups and improved the co-ordination of work within the team. However, the difficulties suggest that the processes of 'integration' will not in itself remove the historical boundaries between professions and improve joint working.
The evaluation drew attention to the impact of voluntary sector involvement in the creation of integrated services to work with people with complex needs. Importantly the findings suggest that establishing the pilots in the voluntary sector meant the new services were not constrained by statutory sector models of provision and consequently were more able to respond flexibly to the needs of individuals rather than being controlled by a professional agenda and organisational priorities. Whilst current rhetoric emphasises the need for services to focus on the individual circumstances of service-users rather than conforming to a 'one size fits all' approach this ideal is notoriously difficult to achieve w33x. However, in this study those pilots based in the voluntary sector appeared to be less burdened by complex bureaucratic structures and more able than those in the statutory sector to build services around individual need. Basing new services in the voluntary sector also brought additional credibility to new services, particularly if it was a national or locally recognised agency and allowed access to networks and expertise that exists outside of the statutory sector.
The experience of the pilots suggests that some people, particularly those with complex needs may find it easier to engage with a service primarily because it is based in the voluntary sector. Indeed some people find voluntary sector services more accessible because they were not based around statutory professions whilst others need the enhanced responsiveness that they perceive in the voluntary sector.
These findings reflect Kodner and Spreeuwenberg's assertion that the logic of integration should be determined by the characteristics of specific groups rather than existing organisational structures w22x. The 'On-Track', 'Swan Nest' and 'Housing Support and Outreach and Referral' pilots were not imposed as prescriptive models of joint working. Rather they were established because professionals, as well as their managers, identified a gap in provision for these specific groups or recognised existing services were failing to address the complex housing and support needs of these individuals.
However, achieving the greater flexibility required in integrated services necessitates changes in strategic and operational management and lines of accountability and the development of a more 'whole systems' way of thinking about service delivery w34x. Whilst some pilots were able to develop clear and effective governance arrangements that crossed agency and sector boundaries these developments inevitably pose challenges. For example unequal relationships between statutory commissioners and voluntary provider organisations may undermine the spirit of openness that is required for agencies to develop truly person-centred services and therefore, may have an adverse effect on the ability to form partnerships. Indeed Hudson suggests that recent reforms to the commissioning process may impede the effectiveness of 'whole systems working' w35x. Additionally, there is a danger that the regulation and inspection regime required by commissioners, and New Labour more broadly, may restrict the ability of voluntary sector agencies to work flexibly w36x.

Conclusions
The Supporting People Health Pilots were established to encourage greater involvement of health and social care professionals in Supporting People partnerships, as well as demonstrating the potential benefits to health and social care from joint working. The pilots demonstrate how housing support services can be developed to enable vulnerable people to live independently in the community. They illustrate how agencies and professionals can work across organisational boundaries, ensuring greater access to a wider range of health care services for particularly marginalised groups. The pilots demonstrate that integrating services to support people with complex needs works best when the service is determined by the characteristics of those who use the service rather than pre-existing organisational structures.
Two new themes emerged which help explain why agencies work together more effectively in some services than in others: the degree to which statutory services are involved in the service and the involvement of voluntary sector agencies. However, the involvement of the voluntary sector was not without difficulty. These findings have resonance for broader policy agenda beyond Supporting People, particularly recent calls to increase the role of the voluntary sector in health and social care services.