The development of integrated health care models in Scotland.

Integrated health care is a key policy aim of Scotland's newly devolved government. 'Partnership working' is the mechanism that has been selected to achieve this goal. Three illustrative examples of health care integration models developed in Scotland are considered; system organisation and structure; Local Health Care Co-operatives (LHCCs); and Managed Clinical Networks. Using these examples the paper explores the nature of 'partnership' and asks if it can deliver integrated care.

Introduction

Achieving integration of care services is a key policy objective of Scotland's newly devolved government [1] and is intended to reduce the frustration, the delay, the inefficiency, and the gaps that frequently exist in care systems. These are long-standing problems. They have their origins in the way policy has been made, in the way different services are funded, planned, and managed; weaknesses in budgetary and information systems; communication failures and organisational and individual behaviours.

Successive governments in Britain have sought to address these issues in a variety of ways. As long ago as the 1970s UK policy makers invented joint finance, a dedicated sum of money to be invested jointly by health and social services [2]. More recently emphasis has been placed on a greater devolution of budgets and decision-making, and on the pooling of resources [3]. The aim is a more flexible response to the needs of individuals who are more concerned with the provision of service than with the provider of service.

Joint working, or partnership, as it is now fashionable to call it has become a dominant concept running through much of public services policy in Scotland as well as other UK administrations. Partnership is the essence of New Labour's ‘Third Way’ [4]. Making a success of partnership is crucial to the implementation of these policies and so this is a subject of great importance to individuals who need care, to those whose job it is to fund, design and deliver it and to their political masters. The question demanding an answer is whether partnership can deliver integrated care? Can this concept succeed in Scotland where managed competition is judged to have failed, for the promotion of partnership as an organisational model derives from the Labour Party's rejection of the Conservative Party's NHS internal market [5]. This paper discusses the nature of ‘partnership’ and explores the extent to which it can deliver integrated health care through three illustrative examples of integrative policy developed in Scotland since 1997. Further details on the health care system in Scotland (including a map of the administrative authorities) can be found at http://www.show.scot.nhs.uk/, and detailed health and health care statistics at http://www.show.scot.nhs.uk/isd

Integration

The dictionary defines integration as ‘the act of making a whole out of parts; the consolidation and harmonising of parts’. Within the Scottish health and social care context this ‘harmonisation of parts’ has attracted various labels: ‘joined up services’; ‘the patient's journey’; ‘clinical or care pathways’; ‘seamlessness’; and ‘care networks’ are a few examples. Their common denominator is the purposeful working together of independent elements in the belief that the resulting whole is greater than the sum of the individual parts [6].

Internationally, the goal of greater integration of services is being actively pursued in a wide variety of settings. For instance, in New Zealand the current government, like its Scottish counterpart, has rejected managed competition and launched a substantial range of integration projects [7]. On the other hand, in the United States Health Maintenance Organisations (HMOs) actively pursue ‘managed care’ central to which is the process of care integration as a route to commercial success [8]. In both of these very different examples the twin objectives are reduced cost and improved quality of care. Integration does not require merger or take-over, and can be achieved through contractual relationships [9]. In a wider sense they are attempts to reconcile increasing demands with limited resources.

Given this global interest in integrated care it is unsurprising that it is also at centre of current reforms of the NHS. But within the UK there is a diversity of models to achieve this objective with different mixes of the three approaches described above. In England Primary Care Groups, and Primary Care Trusts with budget holding responsibilities are seen as key drivers [10] (http://www.doh.gov.uk/). Similarly, in Wales the Welsh Assembly's recently published plan envisages 22 Local Health Groups centred on primary care as the engine of integration that will enable the abolition of Health Authorities [11] (http://www.wales.gov.uk/). These models are quite different from the approach in Scotland that relies heavily on the idea of ‘partnership’ amongst public sector organisations (specifically health boards and NHS Trusts) working together as ‘Boards of Governance’ (see below). The recently published Scottish Health Plan [12] (http://www.scotland.gov.uk/) that announced these administrative reforms also attempts to promote integration by building on Scotland's well-established programme of clinical audit and clinical guideline development [13]. Compared with elsewhere in the UK Scotland has not adopted the financial incentives associated with budgetary devolution to primary care [14] reflecting the experience of GP fund holding which was seen as a divisive and bureaucratic model [5]. It is important therefore to understand the concept of partnership that is now seen as the key to integrated care in Scotland's health and related care services.

Partnership

Partnership is an attractive concept, and is a common feature of so many human relations that it is hard to argue against the idea. Some partnerships bring great benefits to the partners, but not all of them are successful [15]. So, what makes an effective partnership? As in all human relations partnerships depend on the development of trust; partnerships cannot be created overnight and herein may be a difficulty with the current application of partnership in Scotland. It has almost become a panacea, a universal remedy for all ills and because of political realities it has to deliver results quickly. Partnership working is now central to the Scottish Executive's plans to ‘modernise’ the NHS, to its health improvement objectives, and to its social inclusion strategy [16]. There is a wide range of partners and partnerships that require participation by NHS leaders with the aim of achieving integration of services and organisational endeavour. The remainder of this paper examines some of them to illustrate the breadth of the application of this concept bearing in mind that the purpose of partnership is to add value to the work of the partners. Like a bridge, it should enable partners to reach a destination that would otherwise be beyond them [15]. It should make difficult decisions about health care easier, and not easy decisions more difficult.

The integration of health and related care in Scotland

The publication of Designed to Care [17] in December 1997 represented a decisive shift against competition as the instrument for service quality improvement and cost control. The goal was the promotion of both horizontal and vertical integration in health care delivery. Subsequent policy developed by the Scottish Executive has built on this framework, and indeed the most recent Health Plan Our National Health argues that its proposals finish off the transformation begun by the 1997 reforms [12]. In this paper three health service examples of integrating policies are presented and have been selected to illustrate the range of policy developed and the complexity of achieving integrated care. They are: system structure and organisation; management decentralisation; and clinical networks. The paper does not address integrative policies designed to achieve improved population health status [18] or integration between health and social care [19]. Equally no attempt is made to locate health policy in the broader context of government social policy that has placed great emphasis on the co-ordination of public agencies or as it is often described ‘joined up government’ [20].

System structure and organisation

Figure 1 describes the organisational structure introduced to the NHS in Scotland in 1997 and its revision proposed by the Scottish Health Plan published at the end of 2000. That it is possible to present the NHSScotland (the new name for the health service in Scotland) in these terms is an indication of the shift from an internal market that could not be easily represented in an organisational chart. Although the NHS internal market might be more accurately described as a publicly funded health care system with some competition, the trend from 1997 has been to a planned public service nationally organised, locally delivered. An important characteristic of the post internal market health service has been the renaissance of planning, a function that fell in to disrepute for most of the 1990s being replaced by ‘purchasing’ and ‘commissioning’. One of the first acts of the incoming (Labour) Scottish Health Minister in 1997 was to require Health Boards and Trusts to prepare 5 year Health Improvement Plans (HIPs) as a means of re-integrating the components parts of the NHS [21] with an emphasis on ‘local health systems’ made up of 15 health boards responsible for public health and planning, working with NHS Trusts responsible for service delivery. Subsequently, Designed to Care announced the merger of the existing 47 Trusts into 28 new trusts to foster integration in acute hospital services (Acute Trusts) and between primary care and specialist services for the elderly, the mentally ill and the learning disabled (Primary Care Trusts). Vertical integration between these two types of Trusts was to be achieved through the mechanism of a ‘Joint Investment Fund’ known as JIF to be the responsibility of the Primary Care Trust. Its aim was to enable the transfer of services where appropriate from secondary care to primary care through a process of clinical discussion, service redesign and resource reallocation. The JIF was not a separate additional sum of money as the idea was to move resources from one sector to the other once agreement had been reached on the new service pattern. It was intended to retain the influence of primary care over secondary care, recognising that GP fund holding had led to an expanded range of primary care services and greater responsiveness on the part of acute hospitals [22].

Figure 1

The Scottish Executive has now judged that the arrangements in Designed to Care did not go far enough to abolish the culture of the internal market and has announced in its health plan ‘Our National Health’ its intention to establish new NHS Boards based on the existing 15 ‘local health systems’ that bring the leaders of Health Boards and Trusts together with additional members drawn from local government and staff representatives as a ‘Board of Governance’ (Figure 1).

NHS Trusts are to retain their role as employers and their existing operational freedoms but the emphasis in this model is on collective responsibility, the Board of Governance being accountable to Ministers for the delivery of local health plans informed by national guidance on NHS priorities and national service frameworks (e.g. for cancer).

The Health Plan also announced the end of the JIF although only limited formal evaluation of JIFs has taken place. The principal available evidence of their impact is contained in a prepared report by a ‘Support Group’ set up to identify good practice [23]. The learning points from this report are summarised in Table 1. The messages appear to have much in common with the lessons from previous efforts to encourage joint working—time, trust, openness, leadership and commitment [4]. The decision to abolish the JIF is also important for what it tells us about the difficulties of re-engineering the boundary between care levels by a reliance on partnership working without economic incentives. In the wake of the abolition of JIFs there have been calls from within the government's own supporters to create financial levers as a means to encourage integration at the interface of primary and secondary care [24].

Table 1

Learning Points from the Joint Investment Fund (JIF)

The process worked well:

1. Where there is a history of good relationships.

2. Where the leadership of the Primary Care Trust is committed to making JIF work and has afforded it high priority.

3. Where attention has not been diverted by other major changes/crises.

The process was ‘slow’:

4. Where the Health Board and Trusts faced financial pressures.

5. Where there is no shared vision of the JIF.

6. Where the JIF is afforded low priority.

7. Where there is scepticism about whether resources can be moved from the acute sector.

8. Where relationships between local ‘partners’ were historically poor and the cultural change required by JIF seemed impossible.

Source: Scottish Executive Department of Health, 1999 [24].

It is not clear if the Scottish Executive will respond to this suggestion or if it intends to replace the JIF with some other specific mechanism (e.g. ‘collaborative contracting’ as advocated by some [25]) though this seems unlikely. The Health Plan emphasises service ‘redesign’ as a key activity for the new Boards of Governance and it appears that revised planning, funding and accountability arrangements are envisaged as the driving force for integration. Details of how these arrangements will work in practice are currently emerging [26] with the aim of implementing them from the autumn of 2001, and so it will be some time before their impact on the integration of services can be judged.

Decentralised management: local health care co-operatives

The creation of Local Health Care Co-operatives (LHCCs) as part of the internal structure of Primary Care Trusts (PCTs), was a key proposal of Designed to Care. They were intended to be local integrating organisations, bringing together primary and community health services with a range of specialist services (for the mentally ill, the elderly and the learning disabled) whose focus is increasingly on care delivered in or close to people's homes. Participation in LHCCs by general practitioners (GPs) is voluntary and although legislation only came into force to establish PCTs in April 1999, 952 practices participate in them; only 20 have decided not to become involved. In total there are currently 79 LHCCs serving populations varying in size from under 10,000 to as many as 172,000. Does this variation in size matter? Evidence from England on the effects of size on the functioning of Primary Care Groups suggests that bigger may be better for some functions but by no means all [27].

As LHCCs were created as integral parts of PCTs to whom they are accountable, they were given considerable discretion to adopt governance arrangements suited to local circumstances. Most have a multi-disciplinary management board typically drawn from medicine, nursing, pharmacists, the professions allied to medicine, and the public. Activities undertaken by LHCCs across Scotland and the extent to which integration of care has been achieved are distinguished by their variability [28]. Most progress has been made in respect to the development and co-ordination of extended primary care teams, including arrangements for ‘out of hours’ primary care. There is limited evidence so far of integration with specialist services within PCTs and partnership working between LHCCs and the secondary, acute sector does not appear to be well developed (see above), though there are exceptions [29].

In some regards it is too soon to judge the effectiveness of LHCCs as they have been evolving for only two years and there has been limited evaluative research published evidence on them [30]. A Scottish Executive web-site provides links to a network of LHCCs and the work of a ‘Best Practice Group’ (http://www.scotland.gov.uk/.) and a forthcoming report by Audit Scotland [31] should give added insight to their functioning. Despite the absence of formal evidence, recent debate in the Scottish Parliament is indicative of a substantial measure of political support for LHCCs from all political parties in Scotland. Partially this is recognition of the relative newness of LHCCs but it also reflects a growing view that LHCCs can play an important part in the evolution of a care hierarchy that supports local community health and well being through the integration of care.

Table 2 is an illustration of a model that is beginning to emerge. It takes as its starting point the view that the traditional hierarchy of primary, secondary and tertiary care fails to acknowledge the potential to do more than ever before close to or in the patient's own home. The absence of locally integrated primary and social care inhibits the realisation of this potential and can contribute to the spectacle of hospitals under pressure, unable to admit or discharge patients fast enough to keep pace with demand. Ironically, in Scotland patients who have waited to get in to hospital also have to wait to get out [32].

Table 2

A new hierarchy of care

Community Health and Well Being

[A non-medical emphasis on the control of local health hazards, and the promotion of positive health through public health programmes linked to community plans]

Self Care

[Enabling people to look after themselves with the assistance of carefully designed information and educational materials, including advice offered through services delivered on line or through digital TV]

NHS 24

[A nurse-led triage system to direct patients unable to care for themselves to the most appropriate member of the extended primary care team or in emergency to the ambulance service or hospital]

Extended Primary Care

[Stronger teams of primary care professionals including doctors, nurses, midwives, pharmacists, social workers etc able to meet the vast majority of patients care needs]

Intermediate Care

[Focussed on community hospitals, nursing, residential care and the patient's own home; utilising the skills of ‘intermediate care physicians’, nurses, therapists and social workers IC offers locally provided ‘step-up, step-down’ services including investigation, rehabilitation, and respite, principally but not exclusively for the elderly]

Secondary Care

[Linked through managed clinical networks, and supporting the work of the levels below]

Tertiary Care

[Linked through managed clinical networks, as centres of highly specialised advice and care]

A feature of this model is that it attempts to promote horizontal integration of primary care and related services and vertical integration with secondary services through ‘intermediate care’ and ‘managed clinical networks’. Developing intermediate care is a prominent objective in the plans of the NHS in England [11]. It is advocated as a way of easing pressure on both primary care and acute hospital services. The National Services Framework for Older People [33] has defined intermediate care as a new range of services, ‘at home or in designated care settings, to promote their independence by providing enhanced services…to prevent unnecessary hospital admission and effective rehabilitation services to enable early discharge from hospital and to prevent premature or unnecessary admission to long-term residential care’. The Framework emphasises that such services should be, ‘integrated within a whole system of care including primary and secondary health care, health and social care, the statutory and independent sectors’ [33]. Whilst these proposals are not without their critics, some of whom fear that they may lead to a downgrading of specialist rehabilitation services [34], they determinedly set out to achieve continuity of care by integrating the work of many agencies and professional groups. The Scottish Health Plan has not explicitly embraced these ideas but there is a similar direction of travel. In some respects Scotland's network of community hospitals [35] offers a firmer base on which to build than is the case in England where such provision has been greatly reduced over a number years to achieve financial savings. Community Hospitals and specialist services for the elderly are part of the responsibilities of Primary Care Trusts of which LHCCs are an integral part so there is obvious potential to accelerate the development of integrated care within a single organisation. Similarly, it is possible to envisage PCTs and LHCCs developing new models of intermediate care services for the mentally ill whose care is also part of their remit.

How might this model be developed in practice? Experience in New Zealand and England where increasing management and financial responsibility is transferred to primary care practitioners offers one way forward. LHCCs would be encouraged to progress through a series of development stages that bring greater budget-holding responsibilities over the care hierarchy. This is an approach that encourages the devolution of decision-making and complements professional incentives with economic levers in a way that the JIF failed to do. As LHCCs demonstrate that they can discharge the responsibilities transferred to them they earn increasing freedom to redesign the local health care system in collaboration with their colleagues in secondary care.

Clinical networks

A second feature of the model at Table 2 is that secondary and tertiary care is secured for LHCC populations by means of ‘managed clinical networks’. This is an idea developed in the Scottish Office Review of Acute Hospital Services [36]. The Review was established in the wake of a period of intense pressure on hospitals in the winter of 1996 that had led to professional and political concerns that a rising tide of emergency medical patients were not receiving the care they needed [37, 38]. Once the immediate problems subsided Government Ministers decided that it was necessary to conduct a wider examination of the future pattern of hospital services and the review was asked to consider their development over a 5–10 year period.

Although the number of District General Hospitals in Scotland had increased over the previous decade in order to serve larger towns, the pursuit of greater efficiency in their operation (e.g. day case surgery) had led paradoxically to an overall decline in the number of available hospital beds in the country. Changes in the medical workforce—reduced hours of working for junior doctors and new training arrangements—and the growing specialisation of clinical practice meant that it was becoming increasingly difficult to maintain services of acceptable quality in some hospitals. These problems were most serious, but not confined, to smaller hospitals in rural and remote areas. The NHS internal market had also seriously undermined a co-ordinated approach to service, workforce, and capital planning. Traditional responses to these problems in the UK have led to the centralisation of services in the belief that economies of scale are achieved, that the critical clinical mass required for the maintenance of quality is created, and that it is the best way to make use of scarce specialist skills. The Review found the evidence for these alleged benefits to be ambiguous and thus sought alternative strategies not least because centralisation approaches are unpopular with communities who resent the loss of access (and employment) and are sceptical that the motive of public officials is simply to ‘cut’ services for financial reasons [39]. Given the nature of Scotland's geography, and the widely distributed population beyond the ‘central belt’ that extends from Edinburgh in the east to Glasgow in the west, these are understandable concerns. Managed Clinical Networks were proposed by the Review as an evolutionary concept (rather than a panacea) building on existing professional relationships and referral patterns with the objective of securing access to locally delivered, quality assured care through the managed integration of hitherto separate clinical services.

MCNs have been defined as,

‘Linked groups of health professionals and organisations from primary, secondary and tertiary care working in a co-ordinated manner unconstrained by existing professional and Health Board boundaries to ensure the equitable provision of high quality clinically effective services throughout Scotland’ [40].

MCNs can be of various types, concerned for instance with an individual speciality (e.g. neurology) or disease (e.g. peripheral vascular disease) both of which were proposed as pilots for model in the Review and are now in the process of implementation and evaluation. The concept of clinical networking is not new, but the distinguishing feature of this model is the emphasis on the active management of the connections in the network. Guidance issued by the government subsequent to the Review emphasised this point by setting out a series of ‘core principles’ to be adopted by all aspiring networks (Table 3).

Table 3

Managed clinical networks—core principles

1. There should be clarity about Network management arrangements

2. Networks should have a defined structure, setting out the points at which the service is to be delivered, and the connections between them

3. Clear statements should be made of the specific clinical and service improvements that patients can expect

4. Networks should use an evidence base (e.g. clinical guidelines developed by Scotland's medical royal colleges known as ‘SIGN’) and be committed to the expansion of the evidence base through appropriate research and development

5. Membership of networks should be multi-disciplinary and multi-professional and include patient representation

6. A clear policy on the dissemination of information to patients and the nature of that information should be in place

7. All health professionals in the network should practice in accord with the evidence base and the general principles covering the network

8. An integral quality assurance programme acceptable to the Clinical Standards Board for Scotland (an accrediting body) should be in operation

9. The network should exploit educational and training potential within it

10. Audit data should be produced to defined standards and network members should participate in the review of the result

11. Clinical staff in the network should circulate to improve patient access and enable the maintenance of professional skills.

Source: Scottish Office MEL (1999) 10 [40].

In addition to the two demonstration models referred to above, MCNs are currently being developed in cancer, coronary heart disease [41], diabetes, renal transplantation, palliative care, and for the treatment of cleft lip and palate. Table 4 summarises the structure and clinical activities of the South East of Scotland Cancer Network (SCAN) as this is one of the most advanced examples. Full details of this network can be found at http://www.scan.scot.nhs.uk/ the SCAN website, which is about to be launched. Similarly, The Tayside Regional Diabetes Network is now well established and provides integrated care linking primary and secondary care for 10,000 diabetic patients in and around the city of Dundee. Full details can be found at http://www.diabetes-healthnet.ac.uk/.

Table 4

The South East of Scotland Cancer Network—SCAN

1. SCAN is an organisation of 9 NHS Trusts located in 4 health board areas serving a population of about 1.4 million people.

2. It is focussed on networks for the 4 common cancers—lung, colorectal, breast, and gynaecological.

3. A network for palliative care is being established.

4. Each cancer network has a multidisciplinary management group chaired by a cancer clinician.

5. Each network is implementing relevant SIGN clinical guidelines and QA standards required by the Clinical Standards Board for Scotland.

6. Each network has a clinical audit facilitator co-ordinated through the Scottish Cancer Therapy Network.

7. Referral protocols for each network are being implemented.

Source: SCAN Annual Report 2000 [42].

Experience in Scotland with MCNs illustrates the importance of reforming administrative arrangements as an integrating strategy. As the whole idea of these networks is that they should operate across institutional and other boundaries they challenge existing budgetary flows and capital planning processes. They demand greater mobility by key clinical staff, requiring them and managers to have a loyalty to a network as well as an institution, and raises difficult questions about who is responsible for their ‘clinical governance’ [43] since the network is a ‘virtual’ organisation. The answer in Scotland is to trace the accountability of professionals participating in networks to their employing Trust that, together with the local Health Board, should approve the creation of networks. The problems described are not insuperable but they do require a significant organisational development effort to overcome them as has been acknowledged in the Scottish Health Plan [12].

Discussion

This paper has sought to illustrate some of the current trends in health care integration in Scotland. The rejection of managed competition, as the driving force for improvement in health care services, owes much to the view that in the NHS it led to inequity and bureaucracy. ‘Partnership’ is now prescribed as the medicine to rid the NHS of these ills and to restore a national, integrated health service. There is no doubt about the commitment of the present administration to this remedy and their policies are backed by unprecedented increases in spending on health services [44]. There is a wide degree of professional support within the NHS for the general direction of these policies and the Executive's political opponents are advocating similar approaches. Although it is only four years since the beginning of the end of the internal market it is appropriate to ask a number of questions about the likelihood of model that has replaced it delivering integrated care.

The first question concerns the nature of incentives. Partnership is a complex concept and whilst it has found favour as an idea the Scottish model has, compared with other parts of the UK, fewer economic incentives to encourage its development. Is this important? If economic incentives are relatively limited (e.g. to remaining within strict financial targets) how is integration through partnership to flourish? The lesson from Managed Clinical Networks is that professional collaboration can compensate as evidenced by the Tayside Regional Diabetes Network and SCAN. But building these networks challenges many established practices that take time to alter. In the context of primary care in New Zealand, it has been reported that professional incentives have proved to be more effective than commercial incentives in modifying professional behaviour [7]. If the scale of engagement of GPs in LHCCs compared to GP fund holding is a measure then Scotland's experience lends support to this view. On the other hand New Zealand GPs who participate in associations of independent practitioners have had budgets for an expanding range of services devolved to them. The trick appears to be to develop an approach that builds on professional relationships by progressively extending the influence of primary care practitioners over other parts of the health care system as reward for demonstrable competence in their discharge of increased management and financial autonomy.

A second question concerns time scales. All the evidence is that partnership requires time to develop but public health care systems are in the centre of political debate and policies must deliver benefits quickly. Key indicators of success are often measures of waiting time or waiting list size. If politicians are to retain the support of their electorates they must be able to demonstrate that partnership and integration deliver tangible improvements in access. Such concerns may be relatively less important as motivators for clinicians who value integration for the professional relationships it may bring and who also have longer term interests in clinical outcomes. Can politicians keep faith with a concept that may not deliver quickly enough for the electoral cycle?

A third question concerns the relative role of local and central bodies to achieve health care integration. The NHS is the largest public service for which the Executive is directly responsible consuming about a third of its total spending. The success or otherwise of the Executive's policies in health are therefore of considerable significance for the government as a whole. They are also significant as an indicator of the impact of political devolution [45]. Political pressures to achieve improvements in the patient's experience of the NHS could lead in turn to impatience with local organisations and growth in control of their activity by the Scottish Executive. Is this the direction of travel set out in ‘Our National Health’? Are the new structural arrangements the zenith of partnership or a statement that it has not developed fast enough or consistently enough from the publication of Designed to Care? If there were a trend to greater central control would this help or hinder integration? A recent critique of NHS policy in England includes the observation that ‘a centralised approach is likely to be punitive and coercive. It disempowers people in the field. And it is likely to do little to motivate the best performers to improve’ [46]. On the other hand the evidence of JIF, LHCCs and MCNs is that a reliance on local actors can lead to widespread variation in the pace of implementation of national policy. This may not matter in a market model where it might be a stimulus to improvement but it has political consequences in public services that purport to be nationally organised and universally available.

A fourth question concerns the availability of resources to support health care integration. The nature and scale of the changes illustrated in this paper should not be under estimated. They affect almost all patients in some way, and directly impact on the working practices of all professional groups. They challenge existing institutional structures and cultures. Change of this scale requires resources for individual and organisational development. Similarly, integrated health care demands integrated information systems, especially systems relating to patient care records. There are important and sensitive issues about the confidentiality of this information in an era of electronic communications [47] but without its availability the benefits of integration may, as in other areas of medicine, be realised more slowly [48]. Recent developments in NHSScotland's communications infrastructure [49] have linked all GP practices with each other and with NHS hospitals but in the past the NHS has found it difficult to sustain investment in these ‘backroom’ functions because of financial pressures in clinical areas. As the experience of MCNs has shown there is a strong argument to be made for investment in them as an ingredient in any recipe for integrated health care.

Finally there are some important questions about the relationship between policy and health services research and development. As this paper has demonstrated the dynamics of health service policy in Scotland are driven to a considerable extent by political objectives and timetables. Anyone searching for some kind of pure evidence-based health service policy will be disappointed. That is not to say that evidence plays no part, rather that evidence may be partial or incomplete and that other political considerations are important in framing policy questions and determining the way they are answered. For this reason evaluation should be an integral part of policy implementation as in the case of Managed Clinical Networks, but as the examples of JIF and LHCCs show it can be difficult to draw conclusions on the impact of policy when there is limited published evaluation available. This has been described as the ‘development gap’ [50], a weakness that undermines successful policy implementation. To make good this deficit requires ‘development forums’ that bring together all those in a particular area or region with an interest in health service policy—from within the health service, but also including the voluntary sector, local government, and industry. Such a forum has been established recently in Scotland with similar aims in mind [51]. The centrality of integration in current health care policy in Scotland suggests it should be a subject for inclusion in its programme.

Conclusion

Scotland's health services continue to undergo substantial organisational change at the heart of which is the objective of integrating health and related care services. This whole systems approach is ambitious in its scale and timetable. Partnership working is the means by which the Scottish Executive has chosen to pursue integration. It is an approach that carries some risks. It relies on the construction of integrated policies at national and local level, which is no easy undertaking. It demands that numerous organisations see their role as but one part of a larger whole when there are strong local and organisational identities. Above all it requires a culture of trust and co-operation to replace behaviour influenced by economic incentives, which will take time to mature. As the examples in this paper have tried to show there is no shortage of policy ideas and models, but as in all public policy the greatest challenge lies in turning the ideas in to working reality. Only time will tell if the approach now being pursued in Scotland has sufficient incentives to do so. In the meantime it is important to seize the learning opportunities that these policies present both through the evaluation of individual models and through comparative analysis. Health care models and policy ideas are transferred from one country to another at an increasingly rapid rate [52]. In the past it has been relatively unimportant to explore variations in the content of health care policy in the UK since the similarities outweighed the differences, but political devolution has created the potential for greater diversity. Increasingly, it is evident that policy divergence is now occurring [51] and it is time to embrace the study of health care policy in the individual countries of the UK as an added dimension to international comparative analysis. Integrated health care is an excellent candidate as a starting point.