A literature review to explore integrated care for older people

Abstract Purpose This paper reports on some of the findings of a literature review commissioned to explore integrated care for older people. Methods The process of revising included finding and selecting literature from multidisciplinary sources, and encompassed both published papers and ‘grey’ literature, i.e. material which had not been reviewed for publication. Results The study found that thinking has moved on from a focus on the problems of accessing services to exploring ways in which they may function in an integrated way. Conclusions The study shows how thinking on integrated care for older people has developed, and knowledge of micro, mezzo and macro strategies is now more available.


Introduction
This paper reports on some of the findings of a literature review, which formed part of an empirical study into integrated care for older people in Northern Ireland [1].The study was commissioned by a group of service providers from health and social care in the same locality, who were beginning to work together in an integrated way to provide care for older people, a field beset by the historical divisions in the UK between health and social care, and between care and services, such as housing and transport. The empirical study collected data from staff and service-users to explore their experiences of care, and the ways in which integration had been developed. This literature review was designed to support and expand this activity, by providing a foundation for enquiry and analysis, identifying key concepts and definitions, and informing the development of the questions that the empirical study would seek to answer.
Discussions with the commissioning team identified a number of key issues that they wished to address.The project had been stimulated by the recognition that existing services fell short of providing appropriate, timely and co-ordinated care for older people. There was a consensus that older people have complex and interacting needs, and they often require treatment and care from a range of professionals and carers, services and agencies at the same time. Furthermore, it was recognised that older people accessed the support that they needed from a wide range of statutory, independent and voluntary sector services. When so many staff, services, sectors and agencies are involved it was felt that it was all too easy for gaps in care, fragmentation of care, lack of co-ordination between services, or duplication of services to occur.
In addition to concerns about the impact that these factors have on the quality of care that older people receive, there had been increasing awareness that they have an impact on the efficiency and effectiveness of the available services. For example, an older person may have an extended period of hospitalisation resulting from difficulties in arranging the necessary services to support them in the community rather than their need for specialist in-patient care. In this situation, ineffective use is made of the available hospital service and the overall cost of care is increased because in-patient care is more costly than care in the community.
From this accumulated knowledge about the difficulties that beset the care system the idea had emerged that collaborative working and integration of the various parts of the care system will minimise the problems that older people and service providers encounter. There appeared to be a general agreement in the commissioning team that system-integration is beneficial both to the service user and to service providers. For service users, strategies that reduce the complexity of accessing health and social care and enhance the provision of the services that they require were viewed as beneficial. For service providers the espoused benefits include cost-effectiveness, reduction in length of hospital stay, reduction in inappropriate hospitalisation and decrease in admission to long-term care. Hence, the focus of attention has shifted from attempting to understand the problem to seeking to find ways to promote integration across the care system. In the discussion of the review findings, we outline the strategies that have been explored and then we attempt to identify the issues that remain unanswered.

Methodology -searching for and reviewing the literature
Finding and analysing literature is particularly difficult where integrated care is concerned. To address the breadth of research and debate across different service sectors, professional and agency boundaries and academic disciplines makes the searching process more complex, given that the review needs to locate material across a wide range of different professional, academic and organisational bodies of knowledge. In the case of this literature review, material had to be searched for which would reflect different forms of integration, including: This review, therefore, had to be broad in its scope and range, choosing keywords to search with to reflect the different words and terms used in different disciplines. The process of analysing and synthesising the material was also more complex, with many different research goals and research models to include.
The usual approach to a literature review, therefore, had to be modified. Identifying and locating material was dependent on both expertise in retrieving and locating material through 'insider knowledge' [2] of the research team and the experts that they were able to consult, and expertise in using data searching tools and processes. Analysing the material was similarly dependant on understanding the different professional and policy developments, which the material reflected and referred to. Not only did the content of the material have to be analysed and evaluated in terms of research methods and findings, but also the context of the material and all of its interdisciplinary, inter-sectoral dimensions had to be understood.
The literature search that was carried out focused on accessing both published material and grey literature that related to the topics and issues that were identified by the commissioners, the expert panel and key informants. Key informants were identified by the research team through discussion with the commissioners and from their own networking.

Retrieval and acquisition of material
Material was retrieved from both electronic databases and from the grey literature. A wide range of electronic databases were searched to reflect the wide scope of integrated care, e.g. databases covering health management, nursing, medicine and social care. The search was limited to English language articles as the health and social care context tends to differ quite markedly between different countries. A different search process was needed for 'grey literature', i.e. material, which has not gone through a peer-reviewed publication process, for example, inhouse reports and reviews, informal evaluations and progress reports, items in local newspapers and practice journals. Traditional (non-electronic) methods were used, including looking in the reference lists of articles and books to see what information sources the authors used and asking experts in the topic for their recommendations of grey literature. Members of the expert panel also suggested items of grey literature. The Web sites of key organisations in the field of health service management were also checked for relevant literature.
The selection process went through two stages. The first stage, carried out by the information specialist on the project team, entailed discarding duplicates and the obvious 'false hits' from the database searches, i.e. items that were on different topics. The second stage was carried out by the two subject specialists on the research team and comprised an iterative approach with the specialists making independent judgements, then comparing their choices and reaching consensus. The selection criteria were based on the knowledge that the research team had of the literature review process and of the topic being explored. However, they encompassed the following aspects: • the UK context, with selection of occasional substantive items covering a foreign context which could be applicable to the UK • intervention studies or evaluations of services and initiatives • substantive, detailed descriptions of services and initiatives • academic literature, with selection of occasional substantive items from the professional literature or with selection of items from the professional literature where coverage of the issue was sparse The grey literature also underwent this second stage selection process.
In total 4,222 items were retrieved and 148 of these were selected for review.

Analysis of the literature
We used a process of thematic content analysis, i.e. recording the aspect of integration of care for older people that was addressed in the paper. Each item was therefore, read and annotated, using a standard data extraction form, according to characteristics such as the form of integration that was addressed (e.g. between service sectors), the dimension of integration that was addressed (e.g. the interface between agencies/services/care sectors, the type of publication (e.g. research journal or newsletter), for research material details about method, design, subjects/participants were recorded, the country of origin, key issues or outcomes that were addressed and key phrases that were identified in the item. From this thematic content analysis, the team identified different strategies, which had been reported on in the literature. These were divided into three levels by the team-macro strategies, which had taken, place at a societal level, mezzo strategies, which had occurred at a service system, level, and micro strategies which had occurred at an individual service user level.

Central themes in the literature
The following discussion outlines some of the key findings from the literature analysis, giving examples of strategies at all three levels, with a main focus on the situation in similar service systems to those found in Northern Ireland.

Macro strategies (societal level)
An example of a macro strategy was offered by Stuart and Weinrich [3], who highlighted the outcome of integrated national policy and service development in Denmark. The Danish welfare system is distinguished by the principle of comprehensive, universal, tax-financed health and welfare services; and a high degree of decentralization in both decision-making and financing of these services at the local level. During the last 20 years the longterm care policy has led to the development of integrated, institutional and community-based services for older people. In this situation, the systemwide integration of services was supported by national policy that was both responsive to local need and fostered the development of home and community based care.
The situation in Denmark indicates that consistent long-term national policy is important to the development of integrated services. In many situations, however, short-termism dominates national policy, which hinders the progress toward integration. The imperatives for national healthcare policy vary significantly across states, but those that are driving toward the development of effective integrated systems are seeking common goals-integrated delivery systems, 'one-stop' information gathering, informational linking across the care continuum, effective communication and cost tracking.
Within the United Kingdom, health and social care policy has promoted the integration of services for older people and the various countries have responded to this agenda in different ways. Northern Ireland, for example, has been held up as a model for the integration of health and social care. Since 1974, health and social services have been commissioned and provided through Health and Social Services Boards. This model was found to be advantageous in facilitating integrated provision of services during the inquiry that was undertaken by the Boards. In contrast, recent reports have been critical of the level of integration and co-ordination of service delivery. This is illustrated by the following two reports. A review of community care in Northern Ireland in 2002 [4], highlighted that there were practices and services that demonstrated effective well-coordinated delivery of services, however, there were limited mechanisms to publicise these or to roll them out regionally. This contributed to a situation where insufficient team working occurred and problems arose at the interface between hospital and community such as delays in getting people admitted to hospital services and being discharged from hospital. The second study reported on the findings of a postal survey of 73% of old age psychiatrists in Northern Ireland and England [5] and concluded that integrated structures were more concerned with integrated management systems and less with integrated practicebased activities.
In England, historical, professional, administrative and financial barriers have been identified as contributing to divisions between the centrally funded NHS and locally run social services. These divisions have resulted in an interminable struggle between the two authorities to delineate their respective responsibilities.As a response to this, recent health and social care policy has promoted integration across service sectors and health and social care organisations in an attempt to reduce these divisions. This has been drawn together in the NHS Plan for England [6], which sets the governments proposals for restructuring the NHS over the next 10 years. Partnership rather than structural reorganisation has become a key tenet of modern health and social care policy, and it is presented as the means to improve relationships between services and sectors. This was encapsulated in new statutory duties that have been imposed on health and social care organisations. For example, the new primary care organisations (PCG/T) now have a statutory duty to work with other organisations and they are responsible for commissioning hospital services and for developing integrated primary and community health services. Substantial financial resources have become available to support new local health and social service collaborations.This removes some of the structural barriers to integration and provides the means for local strategic partnerships to develop.
In a review of two policy initiatives that were concerned with integrating health and social care services-primary care groups and trusts and the Health Act flexibilities-Glendinning [7] found that removing or at least relaxing structural, organisational and financial boundaries assisted progress towards integration. She also argued that removal of the structural barriers to integration was not a sufficient condition for integration to take place. There were internal barriers, such as turf wars over professional domains, which remained if they were not addressed. These findings supported the conclusions that Wistow [8] reached, namely shared vision, trust and inter-professional co-operation are required in addition to mere structural integration.
This brief review of the situation in Northern Ireland and England suggests that national policy is influencing the progress that is being made toward integration. Much of the policy thrust, however, has been at the level of inter-organisational working and it appears that it is assumed that inter/intra-professional partnership will follow interagency working. This is not necessarily the case. The findings highlighted above suggest that macro strategies can remove the barriers to interagency working, but inter/intra-professional partnerships need to be supported if they are to be effective. Macro strategies, therefore, need to be in place alongside mezzo and micro strategies for integration to occur.

Mezzo strategies (organizational level)
Health and social care organisations are designed to provide specialised services for the unique and common problems that occur across the life span.The types of difficulties that older people experience when using multiple services within care organisations are understood and recent efforts have focused on developing strategies to promote within-organisation integration. These strategies can be broadly described as those that aim to reduce organizational compart-mentalisation through vertical and horizontal integration; and those strategies that change working arrangements to promote collaboration and inter-disciplinarity between the staff that provide the service.

Vertical and horizontal integration strategies
Throughout the international literature the benefits of service and organiational integration has been espoused (see for example [9][10][11][12][13]). This approach focuses on changing organisational structures and processes in an attempt to make linkages across the boundaries and the hierarchies that beset health and social care organisations. In Britain, according to Hudson [14], the major thrust of public policy has been at the level of promoting interorganisational working rather than at the level of developing interprofessional partnerships. Policy measures to facilitate this have included statutory obligations for health and social care organisations to 'work co-operatively' with each other, pilot schemes, and financial incentives to promote partnership working. This has resulted in the innovative reorganisation of services such as the restructuring of different types of care, such as acute and long term care, under one administrative umbrella, and/or pooled funding arrangements resulting in client centred care management across all levels of service. To date, the emphasis in the literature has been on describing such changes and there is insufficient evidence to estimate the benefits, harms and costs of these developments. Attention has also been given to the re-structuring and development of services within care organisations to reduce bottle-necks and gaps in service provision for older people. For example, intensive home care, medical assessment units, transitional care services, rehabilitation services, intermediate care and early discharge and supported discharge programs are a few of the types of services that have developed in recent years (see for example [15][16][17][18][19][20]). Although individual studies do highlight the effectiveness and cost-effectiveness of these services, a lack of synthesis of this literature exists as a result of the range of methodologies that have been used to study the impact of these changes.
Alongside attempts to enhance connectivity within organisations by modifying or changing structural arrangements there has been a concerted effort to develop organisational processes to enhance efficiency and effectiveness. This is illustrated through the attention that has been given to communication and information systems in recent years. These systems have been attributed as making a significant contribution to the problems experienced by older people who use health and social care services. A vast literature has developed around this subject and the central points that emerge from this are the importance of 'one-stop' information gathering, the need for integrated information (financial and clinical) systems, accessible and usable information and management systems, and informational linking and sharing across agencies, the care continuum and across care sectors.
A pan European initiative, EPIC, drew on the characteristics, described above, to create information systems that aimed to improve the quality of care of vulnerable people living in the community. These systems were intended to be prototype information systems for integrated care, which were capable of sharing information across health and social care professionals. One aspect of this project was the development of an assessment instrument, which incorporated all of the features of multi-disciplinary assessment of older people into one tool that enabled a large number of professionals and different agencies to share information. The difficulties of doing this have been recognised for some time as there are a plethora of assessment tools and instruments that address 4 domains: functional capacity; cognitive/mood/psychosocial domains; social/environment circumstances; and clinico/medical condition [21]. The instrument developed in the EPIC project, known as EASY, provided a way for a single comprehensive assessment to underpin the delivery of individualised and coordinated care [22][23][24][25].

Changing working arrangements
Multidisciplinary/interdisciplinary and multiprofessional/interprofessional working are terms in current use. The benefits of this way of working are evident in the studies that seek to evaluate their effectiveness. In this review the majority of the literature focused on the impact of the changed working arrangements on the way that professionals worked with each other. For example, Ross and Tissier [26] used a multi-method case study design to investigate the development of a care management initiative, whereby a social worker care manager worked with 2 designated GP practices and took joint responsibility with a district nurse for assessment and care management referrals from the practices. They found that the team arrangements enabled the different types of professionals to discuss referrals and to negotiate the division of labour, which enhanced information flow and avoided duplication of effort. The GPs valued the improved communication within the team and only having a single point of contact for feedback on referrals and summaries of care plans. The study also highlighted the difficulties that were inherent in working across the boundaries between health and social care in Britain-there was reluctance in some sections of social services to accept the authority of the district nurse to arrange the services that the client required and there were incompatible information systems between health and social services that limited the sharing of information.
With respect to organisational outcomes there are indications that collaborative working arrangements have a positive impact on the achievement of organisational objectives. One study, by Sommers et al. [27], investigated the effect that a team of professionals (physician, nurse and social worker), which focused on coaching patients and setting treatment plans for chronic disease self-management, had on utilisation of services and patient-reported health status. The findings indicated that interdisciplinary collaboration had the potential to lower hospitalisation rates and reduce office visits to physicians, whilst maintaining functioning (there was some evidence for the improvement in patient-perceived health status) in older people requiring chronic disease management.
Whilst there is a wealth of literature that suggests that multidisciplinary/interdisciplinary and multiprofessional/interprofessional working has beneficial outcomes for patients, professionals and health and social care organisations, there is also the suggestion that these arrangements are not always harmonious. The individuals that work in these situations wrestle with the difficulties that are inherent in working across more than one service or organisation, and in working across more than one professional practice. Perhaps these difficulties would only be overcome when the structural and processorial strategies, and changes to working practices develop in a way that they are able to respond compatibly to the same agendas. In their review of the literature Bebbington et al. [28] identified the following features as facilitative in promoting effective interdisciplinary working: the co-location of staff, informal ties between the professionals, shared training, single line management arrangements that provide for both operational issues (clear goals, accountability) and clinical supervision, good liaison arrangements, stable conditions of employment and increased flexibility in working arrangements.

Micro strategies (individual service user level)
In response to the difficulties that older people experience in their journey through care services, strategies that operate at the service user level have been developed. In the main this type of strategy assists those who are planning the care and services for older people to identify appropriate services and facilitate access to them.The majority of initiatives that fall into this category are an addition to the services that already exist and they fall into 2 broad classifications.
Firstly, there are strategies that map out the older person's journey through service(s) and in doing this they provide an approach for the co-ordination of all the services, which can be accessed by the service user within individual care organisations or within the care system that care organisations belong to. Care pathways and integrated care pathway are examples of this type of strategy. There are a number of studies that investigate the effect of these strategies on care and service provision (see for example [29][30][31]). In the action research study reported by Walker and Haslett [31], the impact of the implementation of a continuity of care model in an extended care centre was investigated. In one aspect of this study, the patient's journey though in-patient services was examined.This led to the identification of the processes that were involved in different episodes of care. From this information, it was possible to identify practices that could be changed that would reduce the patients overall length of stay. Following changes to the service it was found that there was improved teamwork and improvements, in key performance areas-length of stay in in-patient services reduced initially; however, the improved through put was slowed down by an increased shift of acute patients from the hospital to the extended care centre. The authors argued that the performance of the whole hospital network benefited from the work that was undertaken on the patient pathways.
The second category of the service user strategies concerns those approaches that support the older person in their negotiation with, and access to relevant services. To do this roles have been created whereby the post holder works across organisational boundaries and supports the older person as they make the transition from one care setting to another.These roles include discharge managers, case managers where the emphasis is on the co-ordination of the care package, care management co-ordinators and liaison nurses. There is a vast literature describing the way that these posts have developed as a way to respond to the problems of poorly coordinated or tardy discharge planning, and delays in starting or completing needs assessment. Such posts aim to promote open, honest, continuous and timely communication between health care professionals, older people and their family, doing this to effect a coordinated package of care.
The impact of these roles, however, is equivocal. A few studies demonstrate the positive impact that these roles have on outcomes of care-for example Van Achterberg et al. [32] report that a co-ordinator of care results in more continuity of care; Hofmeyer and Pallan [33] suggest that a hospital liaison nurse has a central role in enhancing communication between hospital and community nurses; and Group [34], through a systematic review of whether a policy of early discharge with support could be as effective and efficient as conventional care, concluded that early discharge services for a selected group of stroke patients can reduce the length of hospital stay. This positive conclusion is balanced by the finding that there is a lack of conclusive information on the impact of such services on patient and carer outcomes, and patient and carer preferences.The latter point could be generalized to the body of literature that examines the impact of these roles. Although there has been great commitment to developing these types of innovative roles, their effectiveness remains unclear.

Conclusions
The literature that was selected for this review highlights the extent of the work that has been undertaken in this field. Much of the work has focused on understanding the problems that older people experience in accessing and using care services, which has culminated in diverse efforts to overcome the problems. There has been significant development in health and social care policy, however, that is supportive and facilitative of the integration of services that older people use. The majority of attempts to integrate health and social care services take the form of modifications to what already exists and the development of ways to enable older people to negotiate what already exists, rather than radical change to the whole system. This may be in the form of making linkages across the boundaries that are inherent in the system, making linkages between the hierarchies that best health and social care organizations, and facilitating improved and effective communication between organizations and the staff that work in them. These approaches enhance the connectivity between the different parts of health and social care and are in the main small and incremental changes.
There has, however, been little work concerned with integration across the whole continuum of care. In addition, less attention has been given to all the services that older people require to live fulfilled and independent lives. For example integration across health, social care, housing and transport equally affect the way that older people live, yet these services tend to operate in parallel rather than in partnership with each other. Presumably the lack of attention to whole system change is due to need for long term commitment that this requires. The experience from Denmark indicates that given commitment, investment and national leadership, system-wide integration is possible. In countries, such as Britain, where these policies are largely short term and consequently subject to change, it is difficult to evaluate the impact that they are having. This highlights the need for longitudinal studies in this area that both inform and are informed by developing policy.
It is clear that care organisations are changing in response to national and local policy, and, at the same time to the needs of service users. Consequently there are multiple changes taking place in any care organisation at the same time. At one level of analysis this may appear to be in response to clear objectives set by government departments and to be well co-ordinated. At another level of analysis, however, the changes that are taking place are chaotic and unplanned. One outcome of this situation has been the development of a raft of mezzo and micro strategies that aim to integrate the systems, services and the operational processes within care organizations. There has been much effort put into describing the impact that these developments have had on changing practice yet there has been little effort given to systematically investigating the impact that these changes are having on the whole system of care on patient and carer outcomes. This points to the need for large, systemwide studies that link all three levels of strategy together in a way, which can usefully and constructively inform people working towards change across the whole system of care.