Clinical and serviceintegrationThe route to improved outcomesNatasha Curry and Chris Ham

The King’s Fund seeks tounderstand how the healthsystem in England can beimproved. Using that insight, wehelp to shape policy, transformservices and bring aboutbehaviour change. Our workincludes research, analysis,leadership development andservice improvement. We alsooffer a wide range of resourcesto help everyone working inhealth to share knowledge,learning and ideas. King’s Fund 2010First published 2010 by The King’s FundCharity registration number: 1126980All rights reserved, including the right of reproduction in whole or in part in any formISBN 978 1 85717 605 6A catalogue record for this publication is available from the British LibraryAvailable from:The King’s Fund11–13 Cavendish SquareLondon W1G 0ANTel: 020 7307 2569Fax: 020 7307 2801Email: publicationsEdited by Jane SugarmanTypeset by Peter Powell Origination & Print LimitedPrinted in the UK by The King’s Fund

ContentsAbout the authors and acknowledgementsExecutive summary The King’s Fund 2010vviiIntroduction1Definitions and forms of integration3Macro-level integration9Meso-level integration21Micro-level integration33Implications for the NHS43References48

About the authorsNatasha Curry joined The King’s Fund in 2005 and has undertaken research in a numberof areas, including long-term conditions, commissioning, and choice. She led The King’sFund’s evaluation of practice-based commissioning in four PCTs which was publishedin 2008. She is currently developing a piece of work that is examining the implication ofthe NHS reforms for the voluntary sector in health. She also manages the Fund’s workon predicting the risk of unplanned admission to hospital. Natasha previously worked asa consultant at Matrix, a research and consultancy company, prior to which she was theevaluation officer at the Chinese National Healthy Living Centre.Chris Ham took up his post as Chief Executive of The King’s Fund in April 2010.He has been professor of health policy and management at the University of Birmingham,England since 1992. From 2000 to 2004 he was seconded to the Department of Health,where he was director of the strategy unit, working with ministers on NHS reform.Chris is the author of 20 books and numerous articles about health policy andmanagement. His work focuses on the use of research evidence to inform policy andmanagement decisions in areas such as health care reform, chronic care, primary care,integrated care, performance improvement and leadership.Chris has advised the World Health Organization (WHO) and the World Bank andhas served as a consultant to governments in a number of countries. He is an honoraryfellow of the Royal College of Physicians of London and of the Royal College of GeneralPractitioners, a companion of the Institute of Healthcare Management and a visitingprofessor at the University of Surrey. In 2004 he was awarded a CBE for his services to theNational Health Service.AcknowledgementsWe would like to thank our colleagues, Anna Dixon and Nick Goodwin, for providingcomments on earlier drafts, and Peter Colclough, Jon Glasby, Dennis Kodner and SteveShortell for acting as external reviewers. We alone are responsible for the final text. The King’s Fund 2010v

Executive summaryIntegration can take a variety of forms, involving either providers, or providers andcommissioners, who work together to deliver better outcomes at the macro, meso andmicro levels. There are many examples of integrated systems operating at the macro level in the UnitedStates and wide variations in how these systems are organised. The integrated systemsreviewed in this paper, such as Kaiser Permanente and Geisinger Health System, demonstratehigh levels of performance on many indicators for the populations that they serve.Common characteristics of these integrated systems contributing to their performanceinclude multispecialty group practice, aligned incentives, the use of informationtechnology (IT) and guidelines, accountability for performance and defined populations,a physician–management partnership, effective leadership and a collaborative culture.Integration at the meso level focuses on the needs of particular groups of patientsand populations, such as older people and patients with one or more long-termconditions. Evidence from North America and Europe shows that integrated healthand social care systems for older people demonstrate positive results on manyindicators. There is also evidence that disease management for patients with longterm conditions can deliver benefits on some indicators. The evidence on approachessuch as chains of care and managed clinical networks is inconclusive.Integration at the micro level encompasses a diverse range of approaches, many ofwhich seek to improve care co-ordination for individual patients and carers. Theseapproaches include care planning, case management, patient-centred medicalhomes, virtual wards, personal budgets, IT, telehealth and telecare. There is evidenceto support the use of all these approaches, although the findings of evaluationsare inconsistent, for example, in relation to case management. Interventions usingmultiple strategies to strengthen care co-ordination appear to be more successfulthan those using single strategies.The evidence brought together in this paper shows that moves to achieve closer integration ofcare in the English NHS should continue. Organisational integration alone is unlikely to deliverbetter outcomes and effort must focus on clinical and service integration. Action is needed at themacro, meso and micro levels, and multiple strategies should be pursued at all three levels.General practice commissioning offers a platform on which to develop integrationprovided that practices involved in commissioning consortia are encouraged tocommission and provide services in collaboration with clinicians in community healthservices and secondary care.Policy-makers should encourage the emergence of clinically integrated groups andintegrated provider networks based on patient choice wherever possible and linkedthrough contractual integration. The King’s Fund 2010vii

IntroductionPolicy-makers have used a variety of mechanisms for reforming the NHS in Englandin the past decade, including targets and performance management, regulation andinspection, and choice and competition. The coalition government elected in May 2010has put forward proposals to extend choice and competition and to reduce reliance ontargets and performance management as part of a far-reaching programme of reform.These proposals are designed to put patients at the centre of the NHS and improve outcomes.Alongside the emphasis on choice and competition, there has been increasing interest inintegrated care. The policy document that signalled the importance of integrated care wasthe final report of the NHS Next Stage Review led by Lord Darzi which noted:We will empower clinicians further to provide more integrated services for patients bypiloting new integrated care organisations (ICOs) bringing together health and socialcare professionals from a range of organisations – community services, hospitals,local authorities and others, depending on local needs. The aim of these ICOs willbe to achieve more personal, responsive care and better health outcomes for a localpopulation (based on the registered patient lists for groups of GP practices).(Department of Health 2008, p 65)Subsequently, 16 areas were selected for inclusion in a pilot programme and these arebeing evaluated to explore how services have been integrated and the impact they havehad on patients and service use. The 16 areas include some of the NHS organisationsinvolved in adapting lessons from Kaiser Permanente’s integrated way of working overthe past decade (Ham 2010b). Alongside the pilot programme, other areas have also takenthe initiative to integrate care and have sought to do so in the face of policies that have notalways supported integration (Ham and Smith 2010).One of the questions that arises from the change of government is whether the interestin integrated care will continue in view of the even greater emphasis being placed onchoice and competition. On one reading, integrated care could act as a barrier to choiceand competition if it were to entail establishment of organisations that take on theappearance of monopoly providers of care in their areas. An alternative argument is thatintegrated care organisations could be in the vanguard of the disruptive innovationsneeded to improve performance, especially if there is competition among integrated careorganisations (Christensen et al 2008). As this argument implies, there is no inherentcontradiction between integration and competition provided that patients are able toexercise choice either within or between integrated care organisations.To make these points is to emphasise the need for a more nuanced debate about thedirection of reform that recognises the possibility of integration and competitionboth having a part to play in improving performance. This debate should be informedby evidence on the performance of integrated systems and the many ways in whichintegrated care can be taken forward. It should also be informed by greater clarity on themeanings of integration, integrated care and integrated care organisations, because theseterms are often used synonymously but may have different meanings. The King’s Fund 20101

Clinical and service integrationThis paper contributes to that debate by describing and summarising relevant evidenceabout high-profile integrated systems and outlining examples of integrated care forparticular care groups or people with the same diseases or conditions. It also reviewsways of achieving closer integration for individual service users and carers through careco-ordination and other approaches. It is not intended to be an exhaustive review ofthese issues; rather it offers a selective summary of experience and evidence, focusing onexamples of integrated care that appear to have most relevance to the NHS in Englandin the context of the coalition government’s programme. The paper is aimed primarily atpolicy-makers and NHS leaders working on these issues, in the hope that the evidencebrought together here will help to inform the future direction of reform.The paper starts by clarifying the meanings of different terms and the many forms ofintegration in health care.2 The King’s Fund 2010

Definitions and formsof integrationThere are many competing definitions of integration and integrated care. A review byKodner and Spreeuwenberg (2002, p 3) led these authors to suggest that:Integration is a coherent set of methods and models on the funding, administrative,organisational, service delivery and clinical levels designed to create connectivity,alignment and collaboration within and between the cure and care sectors. The goal ofthese methods and models is to enhance quality of care and quality of life, consumersatisfaction and system efficiency for patients with complex, long term problemscutting across multiple services, providers and settings. The result of such multipronged efforts to promote integration for the benefit of these special patient groups iscalled ‘integrated care’.It follows from this definition that integration is concerned with the processes of bringingorganisations and professionals together, with the aim of improving outcomes for patientsand service users through the delivery of integrated care.Many advocates of integration see it as a potential solution to fragmentation, definedas the breakdown in communication and collaboration in providing services to anindividual which results in ‘deficiencies in timeliness, quality, safety, efficiency andpatient-centredness’ (Wagner 2009). MacAdam (2008), for example, writes aboutintegration as ‘frameworks of care’ that reduce fragmentation and duplication of healthcare, which can lead to poor patient outcomes, inefficient services and wasted resources.Fragmentation is often the result of organisations, professionals and services operatingindependently of each other, with adverse consequences for service users.Leutz (1999) has suggested that there are different degrees of integration, ranging fromlinkage through co-ordination to bringing together services into one organisation.Linkage involves organisations agreeing to collaborate to improve outcomes; coordination entails organisations putting in place defined structures and processes toovercome fragmentation; the most radical form of integration involves establishing newprogrammes and units in which resources are pooled and information shared. Whateverthe degree of integration, Lewis et al (2010, p 11) emphasise that ‘the primary purposeof integrated care should be to improve the quality of patient care and patient experienceand increase the cost-effectiveness of care. As such, integrated care is provided with botha rationale and a common basis for judging its impact’.Typologies of integrationOne of the most comprehensive typologies is that developed by Lewis et al (2010), whichbuilds on work by Fulop and colleagues (see Figure 1 overleaf). The King’s Fund 20103

Clinical and service integrationFigure 1 Fulop’s typologies of integrated care (from Lewis et al 2010) Organisational integration, whereorganisations are brought togetherformally by mergers or through‘collectives’ and/or virtually throughco-ordinated provider networks or viacontracts between separateorganisations brokered by a purchaser.Systemic integration Functional integration, where non-clinicalsupport and back-office functions areintegrated, such as electronicpatient tionIntegratedcare to mative integration Service integration, where differentclinical services provided are integratedat an organisational level, such asthrough teams of multidisciplinaryprofessionals. Clinical integration, where care byprofessionals and providers to patientsis integrated into a single or coherentprocess within and/or across professions,such as through use of shared guidelinesand protocols. Normative integration, where an ethosof shared values and commitment toco-ordinating work enables trust andcollaboration in delivering health care. Systemic integration, where there iscoherence of rules and policies at allorganisational levels. This issometimes termed an ‘integrateddelivery system’.Source: Adapted from Fulop et al (2005)A distinction can be made between horizontal and vertical integration. Horizontalintegration occurs when two or more organisations or services delivering care at a similarlevel come together. Examples include mergers of acute hospitals as well as the formationof organisations such as care trusts that bring together health and social care. Verticalintegration occurs when two or more organisations or services delivering care at differentlevels come together. Examples include mergers of acute hospitals and community healthservices, and tertiary care providers working with secondary care providers.Both horizontal and vertical integration may be real or virtual: real integration entailsmergers between organisations, whereas virtual integration takes the form of alliances,partnerships and networks created by a number of organisations. Virtual integrationmay occur along a continuum, ranging from formalised networks based on explicitgovernance arrangements at one extreme to loose alliances or federations at the other.Virtual integration is often underpinned by contracts or service agreements betweenorganisations, as in the supply chains found in many manufacturing industries.It can therefore be seen as a form of contractual integration rather thanorganisational integration.4 The King’s Fund 2010

Definitions and forms of integrationExamples of integration in the NHSTorbay Care Trust was formed in 2005 and brings together responsibilities forcommissioning of and provision for adult social care and community health services.The formation of the care trust was facilitated by a history of partnership working, along-standing commitment to integrated care and, at a practical level, co-terminousboundaries between the council and the primary care trust (PCT). In order tomeet the needs of older people, Torbay has established five integrated health andsocial care teams organised in localities aligned with general practices. The teamsseek to proactively manage vulnerable service users with the intention of reducinghospitalisation and, where hospitalisation has occurred, to facilitate re-ablement.Brent Integrated Diabetes Care aims to improve outcomes for people with diabetesthrough closer working among staff in hospitals, the community and general practice.Much diabetes care is provided in general practice, supported by a diabetes specialistnurse. An intermediate specialist care service acts as an interface between primaryand secondary care, and cares for patients with poorly controlled diabetes or thoserecently discharged from hospital. A rapid access clinic has been established, whichaims to treat those individuals who have minor issues but would otherwise have hadto wait for a specialist appointment. As a result, only patients with very complexneeds are seen by specialists in secondary care.South East London Cancer Network was set up in 2001 with the aim of ensuringequitable access to high-quality care for all patients in the area wherever they aretreated. The network seeks to provide seamless, multidisciplinary and multiagencycare as close to a patient’s home as is safe and cost-effective. Cancer networks wereidentified in the National Cancer Plan as the framework through which cancerservices should be delivered. The South East London network covers six PCTs, sixacute trusts and a range of palliative care providers. All member organisations signup to a common set of values around collaborative working, and agree to share goodpractice, information and experience.North East London Foundation Trust took over Barking and Dagenham PCT’sprovider arm in July 2009. Under the previous government’s TransformingCommunity Services policy, PCTs have been required to divest themselves of theirprovider services to further formalise the purchaser/provider separation. Althoughpatients continue to receive care at the same sites, the management of services hasshifted to the trust which will continue to manage them until March 2011. This is anexample of vertical integration and the first such case to be reviewed (and approved)by the NHS Competition and Cooperation Panel.Evidence from the United States indicates that organisational integration may occur inthe absence of clinical and service integration. As Burns and Pauly (2002, p 134) foundin their review, ‘the structures that were put in place to integrate different providers oftenfailed to fundamentally alter the manner in which physicians practiced medicine andcollaborated with other health care professionals’. The consequence was that ‘integratedstructures rarely integrated the actual delivery of patient care’ (Burns and Pauly 2002,p 134). This observation is supported by a recent review of organisations claiming todeliver integrated care to older people in North America which found that only halfactually provided more co-ordinated care for older people and their carers (MacAdam2008). Alongside organisational integration, therefore, it is important to consider theextent to which care is effectively co-ordinated. The King’s Fund 20105

Clinical and service integrationFigure 2 Conceptualisation of integrated care in terms of organisational form (fromDonaldson in Ham and de Silva 2009)HIGHSingle provider,weak internalco-ordinationSingle provider,strong internalco-ordinationSiloed providersMultiple wellconnectedproviders/clinical networksExtent oforganisationalmergerLOWCo-ordination of careWEAKSTRONGThe relationship between organisational integration and care co-ordination is illustratedin Figure 2, which indicates that organisational integration in itself may be insufficientto overcome fragmentation of care. It also suggests that high levels of care co-ordinationcan be achieved both within integrated organisations and between different organisationsworking together in networks.This brings out a further important distinction, relating to the level of care that is the focusof integration. As discussed in more detail below, integration may be pursued at macro,meso and micro levels. Care co-ordination is one way of achieving integration at themicro level by ensuring that service users experience seamless care. Care co-ordinationdepends less on organisational integration than on clinical and service integration,because the experience of service users is influenced more by the nature of team workingand the adoption of shared guidelines and policies than by the nature of organisationalarrangements. This has important implications for the NHS and we return to them in thefinal section.Implicit in the discussion so far is that integration is concerned with the provision of care.Although this is often the focus for both policy-makers and practitioners, the extent towhich the provision of care should be integrated with or separated from responsibilityfor health care funding or commissioning is also an important issue of debate. Until theintroduction of the internal market reforms in the 1990s, the NHS was an example of ahealth care system in which funding and provision were largely integrated in the sameorganisations, such as area health authorities and district health authorities.In line with developments in other countries, the reforms in the 1990s resulted inseparation between provision and commissioning (as funding has become known)in England. The merits of maintaining this separation continue to be discussed,with some commentators arguing that organisations in the United States combiningprovision and commissioning perform better than those in which these functions areseparate (Christensen et al 2008). The argument for bringing together provision andcommissioning in the same organisations has proved more persuasive in Scotland andWales, which have reverted to an integrated structure since political devolution.6 The King’s Fund 2010

Definitions and forms of integrationPolicy-makers in England continue to promote separation, most recently by requiringPCTs to divest themselves of their responsibilities for service provision.The final point to make in clarifying definitions and forms is that integration may alsoinvolve funders or commissioners working together to deliver integrated care andimprove outcomes. Commissioner integration has been pursued in various forms sincethe late 1970s, when joint financing was introduced between health and social care. Morerecently, the flexibilities made available under the Health Act 1999 (now the Health andSocial Care Act 2001) have enabled NHS organisations and local authorities to set uparrangements under which one authority transfers resources to the other to undertakelead commissioning of health and social care, and to transfer resources into a single budgetmanaged by one of the authorities on behalf of both (known as pooled budgets). It is alsopossible to establish care trusts to achieve organisational integration of health and socialcare budgets and services. One example is Torbay, which is described in more detail later(p 25).Where they have been taken up, these flexibilities have been used mainly in relation toservices for people with mental illness, learning disabilities and physical disabilities, aswell as for older people. Although many of the examples discussed in this paper relateto the integration of health care services, some of the evidence on health and socialcare integration has also been reviewed, especially concerning services for frail olderpeople in whom the risks of fragmentation are most apparent. In relation to the typologyoutlined above, commissioner integration focuses mainly on the normative and systemicdimensions of integration.Levels of integration and evidence of impactExamples of integration are reviewed at three levels: The macro level is one at which providers, either together or with commissioners,seek to deliver integrated care to the populations that they serve. Examples includehealth maintenance organisations such as Kaiser Permanente and Geisinger HealthSystem, and integrated medical groups.The meso level is one at which providers, either together or with commissioners, seekto deliver integrated care for a particular care group or populations with the samedisease or conditions, through the redesign of care pathways and other approaches.Examples include initiatives to integrate care for older people in North Americaand Europe, disease management programmes, chains of care and managedclinical networks.The micro level is one at which providers, either together or with commissioners,seek to deliver integrated care for individual service users and their carers throughcare co-ordination, care planning, use of technology and other approaches.Although we have distinguished between these three levels for the sake of analysis, inpractice they are often used in combination; this is in recognition of the fact that changesat the macro level, on their own, are limited in their ability to make a difference forservice users and also to address the weaknesses of care fragmentation. For example,organisations such as Kaiser Permanente and the US Veterans Health Administrationseek to leverage the benefits of organisational integration by focusing on populationmanagement and care co-ordination. As discussed in the final section, integration isunlikely to deliver on its promise of improving outcomes unless there is action at all levels.There are a multitude of difficulties associated with measuring the impact of efforts toachieve closer integration. First, the aims of integration may be manifold, so the criteria The King’s Fund 20107

Clinical and service integrationagainst which success is measured vary widely. Second, even where intentions are clearand consistent, the target populations, size of intervention group and context may bedifferent and difficult to compare. Third, some of the intended outcomes of integrationare not easily measurable. For these reasons, assessing the impact of integration remainsa significant challenge (Goodwin et al 2008) even if there are promising signs thatintegration can have positive effects and the evidence base is ‘good enough to be bothintriguing and frustrating’ (Tollen 2008).With these caveats in mind the following sections summarise the evidence and highlightthe emerging findings.8 The King’s Fund 2010

Macro-level integrationSome of the best examples of integration at the macro level are to be found in theUnited States where there is evidence that integrated systems, which have high levels oforganisation, often perform better than the fragmented forms of care that predominatethere (Shih et al 2008). These systems seek to integrate one or more of the three coreelements that underpin health care in the United States, to overcome the issues ofduplication, poor-quality care and inefficient use of resources which can arise due tofragmentation of care. These three elements are the health plan (or commissioner to useUK terminology), the physicians who provide outpatient care and have admitting rights,and the hospitals that deliver inpatient care (Gleave 2009).Some systems have integrated all three elements and represent examples of completeprovider and commissioner integration; others have focused mainly on providerintegration, for example, by bringing together physicians and hospitals; there are alsoexamples of physicians coming together in integrated medical groups. In some cases,medical groups come together to work at a sufficient scale to be able to take onresponsibility for commissioning all or part of the care for the populations whom they serve.The diverse forms of integration in the United States add to the complexity of assessingtheir impact and distinguishing the different elements that have a bearing on theirperformance, let alone assessing whether one form of integration is superior to the others.Kaiser PermanenteKaiser Permanente is the largest non-profit-making health maintenance organisationin the United States, serving 8.7 million people in eight regions. Kaiser Permanente is avirtually integrated system in which the health plans, hospitals and medical groups ineach region remain distinct organisations and co-operate closely using exclusive andinterdependent contracts. The exclusivity of the contract means that the medical groupsdo not see patients from other health plans and members of the health plan generallyobtain all their care from Permanente physicians. The Permanente Medical Groupsreceive a capitation payment to provide care to members in Kaiser facilities and, as such,take responsibility for clinical care, quality improvement, resource management, and thedesign and operation of care delivery in each region. The mutual interdependency of thethree parts of the system means that no single part can afford to let the others fail; thisacts as an incentive for partnership working.In all regions, there is an emphasis on keeping patients healthy, consistent with Kaiser’smission as a health maintenance organisation. Key principles of their approach aresummarised in the box overleaf. Although these principles hold for all regions, eachregion has the autonomy to deliver care according to local need (McCarthy et al 2008).For example, Kaiser Permanente in California owns and runs hospitals, whereas inColorado it is made up of a health plan and medical group that contract with non-Kaiserhospitals with which they have a long-term relati

Natasha Curry joined The King's Fund in 2005 and has undertaken research in a number of areas, including long-term conditions, commissioning, and choice. She led The King's . Chris has advised the World Health Organization (WHO) and the World Bank and has served as a consultant to governments in a number of countries. He is an honorary