Factors influencing effective data sharing between health care and social care regarding the care of older people: a qualitative evidence synthesis

Article history

The research reported in this issue of the journal was funded by the HSDR programme or one of its preceding programmes as award number NIHR135660 as part of a series of evidence syntheses under award NIHR130538. The contractual start date was in October 2022. The draft manuscript began editorial review in August 2023 and was accepted for publication in January 2024. The authors have been wholly responsible for all data collection, analysis and interpretation, and for writing up their work. The HSDR editors and production house have tried to ensure the accuracy of the authors’ manuscript and would like to thank the reviewers for their constructive comments on the draft document. However, they do not accept liability for damages or losses arising from material published in this article.

Copyright statement

Copyright © 2024 de Bell et al. This work was produced by de Bell et al. under the terms of a commissioning contract issued by the Secretary of State for Health and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.

2024 de Bell et al.

Integrating health and social care

The need for co-operation and co-ordination between health and social care has been recognised for a long time. In the UK, integrated working has been required by various pieces of legislation; these have included the Health and Social Care Act 2012 and the Care Act 2014. 17 More recently, there has been a move to Integrated Care Systems,17 which are supported by the NHS Long Term Plan. 4 Different ways of joint working have been tested through pilot programmes, including the Integrated Care Pilots, launched in 2008; the Integrated Care and Support Pioneers, which began in 2013; and the 2015 New Care Model Vanguards. 8 Within these programmes, information technology and data-sharing have consistently been identified as issues. 8,11 Informational continuity – the availability of relevant data or information on a patient to any professional involved in their care – is essential for care co-ordination. 9 Failure to ensure informational continuity and care co-ordination ‘… is likely to result in inefficient use of resources and relies on people or their family or carers, to coordinate care themselves’. 18

Data-sharing between health and social care

Although data-sharing is an aspect of communication, it has its own specificities. Initiatives and interventions aiming to improve communication between health and social care professionals may not necessarily lead to improvement in data-sharing. Data-sharing always has interorganisational and interprofessional aspects: even when information is shared between individual professionals, specific conditions (e.g. policy, legal and ethical frameworks, inter-institutional agreements and professional beliefs) need to be in place to make the exchange of information acceptable and desirable. Data-sharing is also a complex process that involves collecting, coding, storing and updating information which is then shared with or retrieved by other stakeholders, who interpret and make use of it. Stakeholders in the system are informationally dependent in the sense that they require data generated by other stakeholders to complete specific tasks and co-ordinate activities. They may differ in terms of information needs (e.g. content, format, access) as well as usage and contribution to the system’s maintenance (e.g. collecting, uploading and updating information).

Data-sharing can take different forms19 depending on whether it:

• is between two or multiple stakeholders (e.g. telephone call vs. multidisciplinary team meetings or shared electronic record systems)

• involves direct synchronous communication (e.g. face to face, telephone calls, multidisciplinary team meetings) or happens indirectly/asynchronously (e.g. making data available to other users on a shared electronic record system)

• is formal (e.g. multidisciplinary team meetings) or informal (e.g. ‘corridor’ conversations)

• involves patients and their families (e.g. patient-held paper-based records, patient access to an electronic record system)

• involves technology (e.g. telephone, fax, e-mail, electronic record systems)

Different methods of data-sharing are often used in complementary ways (e.g. multidisciplinary team meetings and phone calls). The effectiveness of the process is further mediated by policy context, institutional boundaries and cultures, professional relationships and technological developments.

The provision of information systems that support data-sharing across organisational and professional boundaries has been identified as a key element in providing integrated care10,16 and is a long-standing policy objective in the UK. 20 Advances in technology increasingly enable electronic data-sharing; the NHS is tasked with using data and technology to improve health outcomes and is aiming to introduce a digital patient record accessible to health and social care professionals. 19

However, there are a range of challenges in sharing data between health and social care, occurring at both interorganisational and interprofessional levels. 6 Legislation such as the General Data Protection Regulation means that governance agreements need to be in place to allow data-sharing and also leads to concerns about what type of data can be shared and who it can be shared with. 6,8 Patient information is often inaccessible as different professions may not have access to the same records, or because of inconsistent documentation across organisations and professional boundaries. 6,21 Additionally, professional hierarchies lead to power asymmetries which can form a barrier to effective data-sharing. 16

Why it is important to do this review

The overall topic and area of uncertainty that the review focuses on was identified as a priority within a James Lind Alliance research prioritisation project. The overarching topic of the research prioritisation exercise was: How can we best provide sustainable care and support to help older people live happier and more fulfilling lives? The third of the ‘Top 10’ research priorities, prioritised by care workers, carers and older people, was:

How can social care and health services, including the voluntary sector, work together more effectively to meet the needs of older people?

This was viewed as a priority in order to ensure:

1. care workers and health professionals know about all the care and support available in their area and can signpost older people and their families to services

2. assessments in health services lead to the provision of appropriate social care and someone takes responsibility to check that all needs are met

3. funding and resources are distributed across all sectors to avoid voluntary services being forced to provide social care ‘on the cheap’

4. social care workers are members of multidisciplinary teams caring for older people in hospital

5. voluntary sector services are valued and respected for the essential care they provide

6. health professionals and care workers co-ordinate their care successfully to provide the best possible care for the older person

7. health and social care services communicate with each other, refer older people to each other’s services and provide seamless care. 32

The researchers involved in the priority-setting project stated:

there is a large existing evidence base (and several evidence syntheses) on integrated working but it is hard for practitioners to make sense of it. The key question is how to mobilise existing knowledge about integrated working… not just amongst health and Local Authorities, but also social care providers. 32

Overall, the project concluded that:

A new evidence synthesis is also needed on the mechanisms/interventions that local areas implement to improve communication between health services, social care services and social care providers. 32

This review addresses this need. Initial scoping searches found a wide-ranging body of evidence regarding communication between health and social care, including strategies aimed at organisations and individual professionals. Consultation with key stakeholders was used to focus the review on a specific aspect of communication: data-sharing.

Inclusion criteria

The criteria for the inclusion of studies in the review are described below and summarised in Table 1. Further detail can be found in Appendix 1, Table 8.

Search methods and sources

Screening and study selection

Stage 2: full text

We obtained the full text of papers where either reviewer judged the title and abstract to meet the criteria. Two reviewers (SDB, ZZ) assessed the full-text publication of each record independently for inclusion, with disagreements settled through discussion and, where necessary, with the involvement of a third reviewer. The study selection process was detailed using a Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA)-style flow chart (Figure 1), with a reason reported for exclusion of each record assessed at full text. 45

FIGURE 1.

PRISMA flow diagram of the screening process.

Data extraction and management

We considered multiple reports or publications which used the same data to be a single study. This meant that 49 studies met the inclusion criteria (reported in 54 papers, with 1 study reported in 6 papers and 1 study in 2 papers). In qualitative evidence syntheses, this poses a problem; unlike reviews of effectiveness, they do not aim for an exhaustive sample but to include variation in concepts. Also, a volume of qualitative data too large to allow familiarity with the content can reduce the quality of the synthesis. 46 We therefore used purposive sampling to identify the papers which would contain the most relevant data for the analysis. Consultation with stakeholders (as detailed in Stakeholder engagement) indicated their interest in data-sharing in a wide range of settings and populations, so we decided to use maximum variation sampling to ensure the broadest range of possible studies were included. 46 We began by mapping the included studies, identifying their aims and alignment with our review objectives, population of focus and the richness of the data in the study. 46,47 The richness of data was defined according to the scale developed by Ames et al. 47 where ‘thin’ studies had very little, and often only descriptive, qualitative data relating to our review objectives, while ‘rich’ studies had a large amount and depth of qualitative data relating to our objectives. Whether a study was ‘rich’ or ‘thin’ was decided independently by two reviewers (SDB, ZZ), with disagreements settled by discussion.

This exercise indicated that the majority of studies did not have aims which were closely aligned with our objectives and that over half the included studies had ‘thin’ data. We included all studies with ‘rich’ data (n = 24) in the analysis (and all studies with aims that corresponded more directly with the review objectives fell in this category). Studies with ‘thin’ data (n = 25) are reported in Appendix 3.

From the final sample, we extracted data on:

1. the characteristics of the study (e.g. study reference, aim, methods, service user population, study participants and findings)

To fully capture the characteristics of the included studies, we developed and piloted a data extraction form in Microsoft Excel^® (Microsoft Corporation, Redmond, WA, USA) based on templates developed by the National Institute for Health and Clinical Excellence. 48,49 This summarised contextual and methodological information as identified above and can be found in Report Supplementary Material 1.

1. themes identified by the study authors during the analysis (passages from the papers associated with the identified themes, including participants’ accounts and the authors’ interpretations, i.e. the Results, Discussion and Conclusions sections of the included studies)

Since the extraction of data related to the themes was part of the data analysis, we detail this in Data analysis. The full texts of all included studies were uploaded into NVivo v.12 (QSR International, Warrington, UK), and all data extraction tasks related to specific themes were managed using this software.

One reviewer (SDB or ZZ) performed data extraction for each study, with their data checked by a second reviewer (ZZ or SDB). Disagreements were settled through discussion.

Quality assessment

Quality appraisal of included studies was performed by one reviewer (SDB or ZZ) and checked by a second (ZZ or SDB), with disagreements settled by discussion and, if required, a third reviewer (RA).

Studies were not excluded based on quality. However, the methodological quality of the included studies and the quality of reporting were considered in the interpretation of results. 50,51

Data analysis

Framework analysis is a systematic method of analysing primary qualitative data. 55 This method has been developed for application in systematic reviews of qualitative evidence, where it is known as framework synthesis. 56 It is increasingly valued in the study of complex interventions and health systems,57 as it offers a highly structured but flexible approach to data analysis and can be used to map and compare the concepts under study, including identifying associations between themes. 56,58

Conducting framework synthesis involves five distinct stages:51

• familiarisation with the topic

• development of a ‘framework’ (i.e. initial set of themes)

• indexing, where studies are screened and relevant data extracted using the initial framework

• charting, where themes are revised according to data in the studies

• mapping and interpretation

Indexing and charting

After identifying and screening studies (as detailed in Search methods and sources and Screening and study selection), we moved to the indexing and charting stages of the synthesis. This involved coding the data from included studies (as defined in Data extraction and management) line by line using an iterative approach. 48 The initial framework served as a starting point for the coding process, with new codes generated to capture further details in the data not covered by the original framework. There was constant comparison of codes across studies,51 with the initial framework developed and changed to accommodate new data and evolving understanding of the phenomenon under study (data-sharing between care organisations and professionals).

The included studies were divided randomly between the reviewers, with half being coded first by SDB and the coding checked by ZZ, and the other half coded first by ZZ and checked by SDB. Data and themes were further explored and clarified through discussion with the wider review team. 62

External engagement

The focus of this review – data-sharing between health and social care organisations and professionals in relation to the services they provide to older people – is naturally fraught with tensions and controversies. This is evidenced by the long history of local and national initiatives targeting collaboration between health and social care organisations. The results from the current review are intended to be of value to and impact on the lives or professional practices of various stakeholder groups, including older people and their carers, health and social care professionals, voluntary organisations, healthcare commissioners, social care commissioners, and policy-makers.

To ensure we adequately understood the complexity, and sometimes the technical nature of the topic, and considered different perspectives and interests, the review team engaged with professional stakeholders and public and patient involvement and engagement (PPIE) representatives at different stages of the work.

Departures from the protocol

After screening results from all other databases that had been searched, we decided to refine the EMBASE search. This was due to the high volume of references which would have needed to be screened and because EMBASE has rarely been found to retrieve unique results in previous health evidence syntheses. 41,42 Further detail is given in Screening and study selection.

Due to the large number of relevant full-text articles initially identified for inclusion, we decided to use a purposive sampling method to focus the analysis on a smaller number of the most relevant studies. This is described more fully in Data extraction and management.

Results of the search and studies included in the review

A summary of the search and screening process is provided in Figure 1. The bibliographic database searches retrieved 13,092 records, with another 312 records then identified through citation-chasing and searches of websites. Following deduplication, we double-screened 7853 records from database searches and 312 records from other sources at title and abstract. This identified 192 reports which were eligible to be assessed at full text. Of these, 49 studies met our inclusion criteria, reported in 54 papers, with 1 study reported in 6 papers and 1 study in 2 papers. After purposive sampling, 24 studies were included in the analysis. Studies that met the inclusion criteria but were not included in the analysis (‘thin studies’) are reported in Appendix 3.

Studies excluded after screening at full text are listed in Report Supplementary Material 2, along with reasons for exclusion. The primary reasons for exclusion at this stage were that the population of focus was not older people (n = 28) or that the topic was not data-sharing (n = 50).

Summary of included studies

Of the 24 studies included in the review, reported in 29 papers and one report, the main settings or contexts in which data-sharing took place were primary care (n = 4 studies),9,12,20,63 the community (n = 8),18,64–70 transitional care (n = 5),14,23,71–73 palliative care (n = 5)19,21,74–76 or care home settings (n = 2). 77,78 Few studies (n = 5)14,20,66,71,77 reported demographic details regarding the service user population that data-sharing was intended to benefit. Most defined their population of focus as older people with care needs (Table 4), although some were conducted in specific populations, for example, adults with dementia (n = 2). 66,68 More details on service user populations, where provided, are given in the descriptions of the study clusters below.

In 11 studies, study participants were health and social care professionals,12,14,19,20,67–69,73–75,78 11 studies included patients and carers as well as professionals,9,18,21,23,63–66,71,72,77,79–84 in 1 study participants were not clearly reported,76 while 1 study was conducted solely with older people. 70 Sample sizes ranged from 2 to 220 participants. A total of 13 studies9,14,18,20,53,64,66,68,70,74,76–78 used a mixed-methods approach and 11 were solely qualitative. 12,19,21,23,65,67,69,71,72,73,75,79,80–83 Interviews were used to collect data in all studies, with most obtaining additional data using other methods such as focus groups9,21,23,63,68,70,77,79–84 and participant observation. 23,63,65,69,71–73,75,77,80–85 Eight studies conducted document analysis or audits, for example of care plans. 9,14,20,64,68,69,72,76

Data were shared through face-to-face communication (between two or more care professionals, including formal communication, such as multidisciplinary team meetings, and informal communication e.g. ‘corridor’ conversations), technology-assisted interpersonal communication (e.g. telephone, e-mail, fax) and shared electronic record systems. The former two methods often involved paper-based records. In five studies, all of these methods were used. 12,14,23,64,69,80–84 Two studies focused only on shared electronic systems,21,68,79 one on multidisciplinary team meetings,72 and in two, the data were stored on paper-based forms. 65,74 In all other studies, multiple (but not all) methods of data-sharing were used (Table 5). Both health and social care data were shared in most included studies. 9,12,14,18,20,21,23,65,68–71,74,76,79–84

There were 13 studies investigating current practice,9,12,18,19,21,63,67,69,71,72,75,76,78,79 while 11 studies evaluated the implementation of initiatives focused either solely on improving data-sharing or on improving care delivery, of which data-sharing was a part (Table 4). 14,20,23,64–66,68,70,73,74,77,80–84 A range of health and social care staff shared data in the included studies (Table 5). Social workers were the social care professionals most often involved in data-sharing, with other professionals including social care assessors, representatives from the voluntary sector (e.g. dementia advisers) and care home staff. From the healthcare sector, nurses (district, community and/or specialist) shared data in the majority of studies, while other professionals included doctors (GPs, geriatricians and other specialists), occupational therapists, physiotherapists, mental health practitioners, paramedics and pharmacists.

Quality of included studies

Findings

Initially, we coded different aspects of data-sharing and related mediating factors under broad headings derived from sociotechnical theory, as described in Chapter 3, Methods. As the analysis progressed and specific patterns in the data were identified, we observed that factors affecting data-sharing could be grouped into two broad categories: general factors common to all settings and uses of data-sharing, and factors specific to a particular care delivery process or setting. The latter was coded as ‘purpose of data-sharing’ within the overarching theme of Goals; five purposes were identified: joint assessment,14,20,64,65,67–69,72 integrated case management,9,12,18,63,64,66,68,70 transitions from hospital to home,14,18,23,69,71–73,81–84 data-sharing regarding residents of care homes18,21,73,74,77–79 and data-sharing for palliative care. 19,21,74–77,79 Studies were grouped into five clusters based on these purposes. There was some overlap between clusters; for example, joint assessment formed part of the process of integrated case management in some studies but, on the whole, each cluster focused on the process of data-sharing in relation to a specific care delivery process or context. 64,68 Table 7 shows the allocation of studies and the total number of papers and studies in each cluster.

Our decision to think about the papers in clusters was based on the diversity of contexts in which data-sharing was described within the broad evidence base. Once the studies had been grouped into five clusters based on the purpose of data-sharing, the framework then provided a way of identifying common factors which were perceived to influence data-sharing across the clusters, as well as areas for which there were key differences between clusters. The final framework (Figure 2), which was developed iteratively from our initial framework (see Chapter 3, Wallace criteria), contained four themes – Goals, as defined above and Relationships, encompassing relationships between professionals as well as factors such as professional hierarchies; Processes and procedures, which included policy context, interorganisational relationships, and ways of working (e.g. guidance and training); and Technology and infrastructure, which details technical factors including the accessibility of different methods of data-sharing, as well as issues such as their accuracy and security.

FIGURE 2.

Themes in the final framework in relation to the clusters of studies.

We used the Wallace criteria scores for each study (Table 6) to determine whether the quality of studies varied by cluster. We added up the number of ‘yes’ scores (criteria met) in each cluster and divided this by the number of studies included in that cluster. This gave us the mean number of criteria met per study for each cluster, which ranged from 7.63 to 8.83 (mean 8.3, standard deviation 0.45). This means that no clusters included only low-quality studies and the mean quality of studies (using the number of criteria met as a crude measure) was relatively high (8/12) across clusters.

We also considered Goals to encompass the perceived implications, or intended benefits, of data-sharing. We briefly summarise these benefits because an understanding of the aims of data-sharing in the included studies provides context to the factors perceived to influence data-sharing. Data-sharing provided a comprehensive ‘picture of the patient’ to all professionals involved in their care,67,76–78 while specific benefits were identified for some methods of data-sharing; for example, electronic systems improved transparency as professionals could access prior decisions in the system. 20 Data-sharing initiatives could change ways of working within formal working structures, improving collaboration,20,77 and lead to more timely and efficient delivery of services. 9,19,20,63 Additionally, data-sharing had the potential to reduce staff workload and support more co-ordinated and person-centred care;9,19,21,70,78,81 for example, patients did not have to repeat their stories or conversations about their care. 21,69,76

Below, we use the framework to explore themes which were common to each cluster of studies, before providing more detail on findings from each theme which are specific to each of the five clusters. Quotations are used to illustrate points; where these are from participants in the primary studies, this is indicated.

Summary of findings

Twenty-four studies were included in the review. Most studies focused on populations of older people with complex needs, for example people with multimorbidity; some included specific populations, such as people with dementia, end-of-life patients or people living in care homes. Social workers were the professionals most often involved in data-sharing from the social care sector, while healthcare professionals included nurses, doctors and occupational therapists, among others. Data-sharing tended to occur using multiple methods; in most cases, data were drawn from paper-based or local electronic systems, while shared electronic systems were less common and generally limited to pilot projects and initiatives.

The framework used to synthesise the studies comprised four overarching themes: Goals, Relationships, Processes and procedures, and Technology and infrastructure. Within Goals, we identified five purposes of data-sharing: joint (health and social care) assessment; integrated case management; transitions from hospital to home; care homes; and palliative care. Studies were grouped into five clusters based on these purposes of data-sharing. Within the other three themes, some of the factors perceived to influence effective data-sharing were more general and applied across these clusters of studies (e.g. the impact of professional hierarchies), while others were context-specific (e.g. patients’ and families’ perceptions of EPaCCS as a ‘death register’).

In relation to our original objectives, which were to identify factors that could potentially influence effective data-sharing between health and social care (1) organisations and (2) individuals and (3) in relation to new data-sharing initiatives, we found that evidence in included studies was interlinked. Effective and seamless data-sharing between individual health and social care professionals depended on intraorganisational factors – for example policies and protocols, resources (e.g. means and allocated time), organisational culture – and interorganisational arrangements (agreements, channels of communication and technology) which, in their turn, were mediated by macro-level factors (e.g. policy initiatives, legislation, funding). On the other hand, the success of interorganisational arrangements for data-sharing, such as shared electronic records, was highly dependent on the behaviour of individual health and social care professionals as determined by their needs, perceptions, knowledge and skills. For instance, failure to allocate time for uploading data on a shared database led to higher-priority tasks taking precedence and poor and inconsistent quality of records. When the needs, knowledge and perceptions of (different groups of) database users were not considered during the development process, they were reluctant to engage with the system. The reasons included finding it unhelpful to their day-to-day tasks; not having the necessary knowledge and skills to use it effectively; not being encouraged and enabled to do so or because they couldn’t ‘see the bigger picture’ (i.e. the importance of the information they shared to other decision-makers and the overall impact on patients). Data-sharing initiatives that failed to appreciate the complexity and systemic nature of the process or were too vulnerable to contextual factors, such as short-lived policy programmes and funding, struggled to achieve their goals and lead to sustainable changes in practice.

Conceptual model of data-sharing between health and social care

From our analysis, we developed a conceptual model of data-sharing between health and social care providers which also considers the involvement of patients/users and their informal carers. By ‘data-sharing’ we mean the collection, coding, storing and updating of information about individual patients/service users which is then conveyed, transmitted or made available to other care providers (or individual care professionals belonging to other institutions) across the health/social care boundary. Such data-sharing systems may or may not involve patients/users and their informal carers (either in terms of holding, passing on or having access to their own data). Below we provide a brief description of the model, which is also illustrated in Figure 3.

FIGURE 3.

Diagram summarising the main factors that affect data-sharing at macro, meso and micro levels. Light blue arrows represent the level of alignment of vision, policy and practice between stakeholders while navy arrows represent interpersonal and interprofessional relationships as a mediating factor; the orange arrows of different thickness represent the fact that stakeholders differ in terms of information needs, usage and contribution to the data-sharing system. Patients/clients and informal carers are represented as separate stakeholders as their relationship is likely to affect their participation in data-sharing.

The overarching aim of data-sharing between health and social care providers is to enable care co-ordination by ensuring informational continuity. The specific Goals of data-sharing and related metrics of effectiveness will depend on the specific context in which it takes place. For instance, the aim of EPaCCS is ‘… to capture patient wishes and preferred place of death and improve coordination of care in real time, through enabling the sharing of information across health and social care services’,21 with the most prominent measure of effect being the proportion of users who die in their preferred place of death. 76

Various macro-, meso- and micro-level factors could have a bearing on the effectiveness and acceptability of initiatives aiming to improve data-sharing (either alone or within a broader remit). At a macro level, dominant policy discourses and related initiatives (e.g. integrated care) play an important role in creating a policy and service delivery context in which health and social care providers feel enabled and incentivised to create conditions for data-sharing at a ground level. Related to this is the provision of resources, with the need for long-term sustainable funding mechanisms and the availability of a legal framework that enables data-sharing. Frequent policy changes, ‘pilot fatigue’ (i.e. negativity towards ‘yet another’ new initiative), a lack of sustainable funding and the need to create workarounds to deal with the inadequacies of the current legislation were often identified as barriers to successful implementation of data-sharing initiatives.

At a meso level, factors relating to the different institutional cultures and practices that health and social care organisations have developed over the years need to be changed and aligned across organisational boundaries. Organisations need to develop a shared vision of care provision, often in relation to a specific setting (e.g. transition from hospital to community) or patient group (e.g. patients with palliative care needs) and update their data-related policies and protocols to reflect this new vision. In particular, they need policies and protocols relating to the collection, recording and sharing of information about individual patients/users. Such changes require a systems view of the data-sharing process and co-production involving all system users. This helps to depart from the traditional model where different care providers are concerned only with their own informational needs while data-sharing is seen as a subsidiary activity of little importance. Co-production could help users better understand each other’s decision-making processes and informational needs as well as dealing with issues such as differences in culture and language.

Another crucial factor at this level is the availability of resources to support the transition to new ways of working and their long-term sustainability. These resources include technology and infrastructure (e.g. interoperable or shared electronic record systems; office space and technology for team meetings and colocation), redesigning protocols to allow sufficient time for data-sharing and providing training and managerial support. The latter two relate not only to new data-sharing processes but to support for a broader cultural change, in line with the shared vision of care agreed between participating organisations at the macro level.

Co-production concerns not only the initial design of the system but also its further calibration and development. Evidence suggests that data-sharing systems, such as shared electronic patient records, rarely behave or are used as expected. They may have a wide range of undesired ‘side effects’ (e.g. encourage a ‘task-focused’ care)78 and without robust audit mechanisms in place may have disastrous consequences. 89 Co-production could also help with another resource-related issue: the different levels of contribution to and usage of the system by different stakeholders. ‘Information-rich’ users might be reluctant to engage with a data-sharing system if they have to collect additional information, enter data in multiple systems and frequently update records, without the provision of additional resources.

New technology, such as shared electronic patient records, is often seen as a silver bullet for data-sharing. Yet, there is very little evidence that a single technology could solve all data-sharing problems, many of which relate to human behaviour. In line with the sociotechnical theory underpinning our analysis, the development of a data-sharing system should be guided by the joint optimisation principle, where technology and human elements are considered together, and their interaction and ‘goodness of fit’ determines the performance of the system as a whole. 28 Different combinations of data-sharing methods (e.g. shared records, multidisciplinary team meetings, colocation) and technological solutions (shared electronic systems, video-conferencing etc.) could provide ‘best fit’ in different settings.

At the micro level, factors relate to the behaviour of health and social care professionals, and patients and informal carers, reflecting their beliefs and values as well as knowledge and skills. For instance, to be able to share data effectively, health and social care professionals need to develop a systems awareness of the data-sharing system and the system of care provision more generally. Understanding the decisions other professionals make and the information they need could guide them in collecting and sharing data. Although changes in organisational culture and ways of working (meso level) are expected to change individuals’ behaviour, this will depend on their consistency over time, the provision of resources and leadership, and the time required for such changes to take place.

Fostering interpersonal relationships, through forums such as multidisciplinary team meetings and colocation, is an important mediating factor that seems to have a considerable impact on the success of data-sharing interventions. These relationships help address interprofessional barriers, such as professional hierarchies and prejudices, and seem to play an important role in developing solutions when the ‘official system’ is transitioning or not working as expected. They are also important for the involvement of patients and informal carers who may have concerns about engaging with the system and participating in the data-sharing process.

There is limited evidence of the involvement of patients/users and their informal carers in the data-sharing process, and where evidence does exist, it is limited to specific contexts, such as palliative care. There is no question that the involvement of service users in the development of such systems is highly desirable. However, while having access to their health and social care records and directly contributing information may lead to empowerment, engagement and better co-ordinated and more person-centred care, it will also require considerable changes in current health and social care practices, to avoid misuse, misunderstanding and undesired effects. Future research is required to identify effective system designs and understand the implications of such involvement.

Findings in context

Data-sharing, and evidence about data-sharing, should be considered within the broader context of health and social care integration and collaborative working. From an analysis of four international case studies of integration in health care, Rosen et al. 90 suggest that there are six processes – clinical, organisational, informational, financial, administrative and normative – that support integration within teams and between organisations. Among these, informational integration was identified as a particularly challenging process. 90 Furthermore, all of these processes interact, with wider literature suggesting that a lack of integration at one level, such as the absence of an information system shared between organisations, can restrict integration at other levels, such as the delivery of co-ordinated care at a clinical level. 10 Within studies included in the review, and across the clusters, different forms of data-sharing had their own advantages and disadvantages. As can be seen in the discussion of different methods of data-sharing below, these processes were present, and their interaction contributed to the success of different methods in different contexts.

Multidisciplinary team meetings provided a systematic method of data-sharing, with all participants receiving the same information. They were also a forum for relationship-building (e.g. developing trust) and made participants aware of their informational interdependence and specific information needs, thus facilitating data-sharing. Individual relationships are particularly important in the context of data-sharing, as power (im)balances between professions can result in individuals deciding what they think others need to know. 91 Reducing and resolving these divisions facilitate data-sharing, with an essential component being the development of a shared language and common understanding of terms and processes. 15,16 Multidisciplinary team meetings enabled health and social care professionals to present information in a form accessible to, and usable by, other team members (e.g. by ‘translating’ professional jargon); to provide context to the information they shared; to focus on important details and explain the implications of specific information; and allowed team members to ask ‘data keepers’ for clarification and further information. On the other hand, the effectiveness of multidisciplinary team meetings was limited by a number of practical and relational factors, such as lack of time and geographical distance (an obstacle that could now be overcome with the new information and communication technologies), interprofessional relationships (with the presence of occupational hierarchies at the professional level being a problem identified in other studies6,16), team dynamics and poor leadership. Also, multidisciplinary team meetings are, by definition, costly in terms of the numbers of staff committing time to them.

Direct communication between individual health and social care professionals (verbally or by telephone, fax or e-mail) provided a more flexible and effective way of data-sharing, and often complemented multidisciplinary team meetings. Similarly to the latter, it allowed care professionals to ‘negotiate’ the information they shared and received. The effectiveness of direct data-sharing depended on relational factors, such as trust, understanding and acceptance of each other’s role, skills and expertise. Of the processes identified above, leadership and effective governance arrangements were found to be particularly important,90 and we found that individuals acted as ‘change agents’ in facilitating data-sharing in studies included in the review. While institutional arrangements that sanctioned and facilitated data-sharing (e.g. arrangements around confidentiality and data governance) were significant, relationship-building was a greater facilitator. There were practical limitations to direct communication (e.g. paramedics were often unable to get hold of the patient’s GP to check end-of-life arrangements).

Colocation facilitated direct interprofessional communication and offered similar advantages, for example: creating an opportunity to build relationships and better understand other stakeholders’ roles and institutional environment; offering more opportunities and time to discuss patients with each other. At the same time, there was a risk of the colocated professionals losing their professional autonomy and having their role misinterpreted and transformed to fit in the culture of the ‘host institution’ (e.g. social workers colocated to GP practices felt sometimes as if they were there to serve GPs’ requests rather than as autonomous care professionals).

There is a typical assumption that information technology will improve data-sharing, and while shared electronic record systems were welcomed by most participants, they were not seen as a solution to all data-sharing problems. They were perceived as helpful in co-ordinating the care of specific patient groups (e.g. end-of-life patients) and offered various advantages, such as direct access to data generated by multiple stakeholders; remote and mobile access (highly important for some providers, such as ambulance and out-of-hours services); and continuous update of patient records by multiple stakeholders including, in some cases, patients and families (e.g. end-of-life care).

However, research included in our synthesis also provided evidence that they were challenging to implement and even more challenging to maintain over time. Sociotechnical systems theory emphasises the importance of the interaction between technical and social,29 and the scepticism in the older literature on electronic systems is often linked to poor design and technical problems. 16 While the importance of technical aspects such as interoperability, cybersecurity and access to mobile internet was acknowledged in the included studies, they also reported on the impact of the social factors discussed earlier, such as:

• cultural issues; for example, different levels of access and editing rights could reflect professional hierarchies and prejudice rather than stemming from pragmatic or professional reasons

• complex data management involving multiple stakeholders updating the system and thus increasing the risk of human error, incompleteness and failure to keep the records updated

• users’ perceptions and attitudes, such as trust in the validity and accuracy of the data

• possible misinterpretation due to limited or no contextual information and lack of an ‘interpreter’ (an advantage of multidisciplinary team meetings and direct interprofessional communication)

• an inadequate legal framework

Our findings correspond with those of the wider literature, which suggest that when new technology is underused or poorly embedded or implemented in a workplace, it does not result in the transformation which is intended or assumed by policy. 11,23 Influencing factors revealed by the included studies could be mapped onto different stages and elements of the data-sharing process, again indicating the importance of interaction between different processes, for example informational and administrative. 90 Collecting, coding and storing data to be shared with other stakeholders required understanding of their information needs and requirements (e.g. what decisions the data will be used for, by whom, level of detail, language, format). Data shared directly by health and social care professionals (via interpersonal communication and multidisciplinary team meetings) are usually retrieved from each profession’s own local paper-based or electronic record systems and have been collected, coded and stored to meet their own information needs. External users (i.e. users from outside of the data-recording organisation) have no direct access to such patient records and rely on those who retrieve the relevant data to present it in a way that makes it usable, for example the presentation of information related to a service user’s health condition by a district nurse to a social worker in a multidisciplinary team meeting. This may involve a process of translation, clarification and contextualisation with the need, as noted above, for a shared language. Along with accessing the information, participants also learn about each other and build relationships, supporting the development of a shared language and understanding.

In comparison, data collected and uploaded to a shared electronic record system sometimes include limited contextual information, and electronic systems (with standard and limited data fields) may offer fewer opportunities for clarification, so data are therefore at higher risk of misinterpretation. For instance, data collection may not be standardised and may involve the use of heterogenous methods and practices, resulting in inconsistent data. For example, one study reported that assessment tools and definitions used by care homes to collect data on their residents were highly heterogenous. 78 Another study reported different assessment protocols followed by social workers and district nurses within the joint assessment process. 69 Users of the system might be ignorant of such inconsistencies, which will lead to further variation in practice. Information may also be recorded in such a way that it makes it ‘unusable’ by other stakeholders (e.g. paramedics often found ACP documentation too vague and ambiguous to help them make decisions about resuscitation or hospital admission of end-of-life patients). Therefore, shared electronic record systems should be designed to meet the specific information needs of all potential users, involving them in the conceptualisation, definition and development of the system from the very start. As the system is continuously updated, with different users contributing information, having a shared vision, systems awareness of the care provision, understanding of each other’s information needs and learning from each other’s experience become crucial for maintaining its function.

Not only do different care providers have different information needs, but they also differ in terms of usage and expected contribution to a joint records system. For instance, ambulance and out-of-hours services considered access to EPaCCS highly beneficial when attending a patient receiving palliative care as this could help them make an informed decision about resuscitation and hospital admission. Yet they access such information infrequently while contributing limited or no information at all that could benefit other users. On the other hand, ‘information-rich’ professionals, such as GPs or care home staff, might make regular updates of a patient record, including information that will benefit mostly other users. Again, this is different from usual practice where professionals only collect and store information they need in a format that serves their information needs and share with others only what is available. This ‘contribution/usage imbalance’ was one of the reasons that GPs were less enthusiastic about the implementation of EPaCCS, especially when the system was not interoperable with the other electronic record systems they were using (so they had to enter the same information multiple times).

Furthermore, some participants (notably GPs and care home staff) expressed concerns that ‘information hungry’ systems requiring extensive data collection are likely to encourage a bureaucratic care culture, and therefore distract them from focusing on patients/clients. Care home staff were concerned that focusing on data related to specific tasks could lead to ‘task-oriented’ care, while limiting the type of information that is collected (e.g. restricting electronic data collection to prepopulated fields rather than contextual information on a patient’s story). Focusing on medical information could also encourage further medicalisation and formalisation of care with less attention being paid to the patient’s quality of life, experience and overall well-being. As the provision of person-centred care is a key aim within the NHS,4 data-sharing initiatives need to ensure they are enabling this.

Access-related issues also differed between direct interprofessional data-sharing and shared electronic record systems. When sharing data directly (between individuals and in multidisciplinary team meetings), health and social care professionals felt in control of what was shared and could use their professional judgement to decide on the relevance and appropriateness of disclosing specific information. This provided some reassurance that even when they were uncertain about the legality of such decisions, they could justify them from an ethical point of view. Effective sharing, especially of sensitive data, through such forums required time to gain trust and understanding of each other’s role and appreciate the informational interdependence of care providers. Building personal relations facilitated the process as care professionals shared data because they had a shared vision of care and were convinced that it would help patients/clients, despite legal uncertainties or gaps in institutional policies.

In contrast, shared electronic record systems (if used without other channels of communication) typically lack such a ‘human interface’ that could facilitate more nuanced or tailored data-sharing across professional and institutional boundaries. This needs to be considered in their design but also in the broader context of interprofessional and interorganisational communication by providing ways to counteract such deficiencies. They also face challenges related to access and editing rights. Decisions over whether to make all patient data accessible to all users or limit access and editing rights not only create serious legal challenges76 but present ethical dilemmas to health and social care professionals who might be uncertain about what data should and should not be uploaded to the system – especially when the system is also accessible to patients and families. On the other hand, limiting the access and editing rights of certain professional groups could perpetuate existing professional prejudices and hierarchies, stymie engagement with the system and limit its effectiveness. Accessing data that have not been ‘translated’ or summarised for specific users (e.g. social workers accessing medical data) could also increase the risk of misinterpretation and error.

While the design of electronic systems centres on their use by professionals, the people whom these systems are intended to benefit – the patients and their informal carers – are often forgotten. Very few studies included patients’ and informal carers’ perceptions of data-sharing by professionals, or of access to their own records. However, where this was investigated, patients and carers usually had a positive attitude towards data being shared between health and social care providers, assumed this was already taking place and, compared to care professionals, were less concerned about confidentiality and data protection.

Paper-based patient-held records were considered potentially helpful by patients and families but less so by care providers. There were issues with updating the record and using it to make clinical decisions. An example of this was paper-based ACP records which contain information about patient preferences in relation to end-of-life care. Being placed on such a record was often seen by patients and families as being placed on the ‘death register’; keeping the record ‘visible’ in the patient’s home was also undesirable and perceived as a reminder of imminent death. Even when ACP documentation existed, paramedics struggled to find it in the limited time they had, and often found the information too vague and ambiguous to help them make a decision; confirming its validity and meaning (e.g. by calling the patient’s GP) was rarely an option.

While paper-based patient-held records are designed mainly for the benefit of the patient/family and have a relatively limited role in care professionals’ decision-making (e.g. whether or not to carry out cardiopulmonary resuscitation), patient and/or carer access to a shared electronic record system raises a whole new set of questions, for instance: which part of the record the patient should be allowed to access and edit; in what circumstances informal carers should have access to the patient’s record; and whether such information needs to be presented in a language that is accessible to patients and families. Additionally, patient/carer access to relevant information in these records may have an impact on the confidence and quality of care professionals’ decision-making. 19,74,77 Some EPaCCS are already providing access for patients who are able to enter their preferences about end-of-life care, but no research investigating this issue met the criteria for inclusion in this review.

Implications/recommendations for future research

Technology is advancing rapidly, facilitating innovations which may support more widespread and more effective data-sharing. With the NHS Long Term Plan setting out goals for the use of technology in the NHS in England, including the use of electronic patient records systems in all care settings,4 further research is needed to ensure the successful use of different types of technology. Our findings suggest that while shared electronic patient record systems have been in use for some time now (e.g. in palliative care), the evidence on their implementation and use is limited and of poor quality. 76 Generating better-quality evidence that captures both objective outcomes and stakeholders’ experience could inform the effective implementation of electronic systems more widely.

There are other uses of technology, such as holding meetings virtually, which have been driven by the COVID-19 pandemic. 92 Technology could be key to resolving some of the barriers identified in this review, for example by allowing community professionals to attend multidisciplinary team meetings held on hospital sites. However, more research is needed to assess whether technologies implemented as pragmatic solutions to problems raised by the pandemic, with a focus on factors such as speed, are also the most effective solutions to the barriers to data-sharing. For example, studies on the use of technology in primary care have tended to focus on GPs; the experiences of other care professionals need to be investigated as they may be different. 93 Similarly, research from primary care indicates that certain types of technology are preferred by professionals (e.g. telephone consultations), despite the range of technologies available. 93 Understanding how health and social care professionals are sharing data post COVID-19, and why these methods are being used, including the impact of policies such as the Digital and Data Strategy for Health and Social Care in Wales,94 could ensure that current methods are achieving their intended outcomes and inform the development of initiatives to improve the effectiveness of data-sharing.

Another more recent technological advancement, artificial intelligence (AI), is also likely to change the landscape of care co-ordination and data-sharing. For instance, the option to interact with AI-powered databases using natural language will change the way health and social care professionals collect, retrieve and share data, and overcome long-standing problems, such as the need for technical skills to interact with electronic databases. Rather than being yet another digital tool, AI will soon become a central player in the process of collecting, storing, retrieving and sharing data as well as using data to make decisions about the person’s life. An example of this is the use of smart homes for people with various chronic conditions, where AI can constantly collect data (monitoring) and make decisions related to safety, well-being, medical and social care needs. Therefore, future studies looking at the role of AI in health and social care need to consider its impact on care co-ordination and data-sharing, not only in terms of the technical aspects of the process but also with respect to its impact on human relations.

With few studies reporting on patient experiences of different types and levels of data-sharing, this area clearly needs further investigation. In particular, there is very little evidence on shared patient record systems where patients and informal carers have access to and are able to update their records. Limited data from EPaCCS projects show the complexity and challenges involved in creating and maintaining such systems. However, further research is needed to understand the impact they might have on health and social care practices (e.g. the way in which care professionals record and share data).

Similarly, there was a gap in the evidence regarding data-sharing between healthcare and private and voluntary sector organisations, and a related lack of evidence from care homes. The ageing population means that more people will need social care, or become care home residents, in the future, and are likely to receive care from professionals employed by private or voluntary sector organisations. Consultation with our stakeholder group also identified this as an area for which they wanted more information. As discussed in this review, public sector health and social care organisations face difficulties sharing data; given the additional barriers private and voluntary sector organisations face, research is needed on successful data-sharing initiatives including these types of organisations. For example, NHS England’s Enhanced Health in Care Homes model has been shown to result in 23% fewer emergency admissions among care home residents in test sites. 4 The model mandates better communication (including data-sharing) within the care system, so evaluation could indicate factors involved in successful data-sharing between public and private and voluntary sector organisations. There is also a need to explore what works and for whom – whether organisational, whole teams, or for individual professionals or patients – to inform the design and implementation of systems that consider equity and digital inclusion. 92

The reviewed literature suggests that data-sharing is often perceived as something that ‘comes naturally’ to care professionals and the only thing needed is motivation. However, the complexity of data-sharing, and the various challenges and uncertainties associated with the process, indicates that preparing care workers and supporting them in the process is of paramount importance. The scope of training should be extended to include elements such as shared vision of care, systems awareness and legal aspects of data-sharing. We found very little evidence on successful training and management programmes to support the process of care co-ordination and data-sharing, but plenty of evidence that failure to provide this is likely to lead to poor outcomes.

Last but not least is the importance of further research into the legal aspects of data-sharing. Lack of clarity in this area was identified as a major problem across settings and levels of communication, from the experience of care professionals to difficulties encountered by multiagency initiatives such as EPaCCS, where ‘… IG rules need to be “bent” or controversially interpreted so that EPaCCS projects are not obstructed or abandoned, while the lawfulness of decisions is ensured within broader legal and governance frameworks’. 76

Implications for policy and practice

Our findings indicate that no single factor or change is enough to facilitate effective data-sharing; a combination of approaches and a systems view of the whole care management and delivery process are needed to enable the effective sharing of information across service and professional boundaries. 23 Initiatives to improve data-sharing should consider this in their design, ensuring that professionals are able to share information in multiple ways. Additionally, while some agreement regarding governance was necessary between health and social care organisations, the evidence in our review showed that organisational support for relationship-building was a greater facilitator of data-sharing than formal organisational or information systems integration. The use of methods of data-sharing that allow relationship-building is therefore important, as is the development of knowledge and understanding, whether of different professional roles or of processes and ways of working.

When designing new initiatives or systems, studies where staff were consulted were more likely to lead to effective data-sharing, a finding echoed in the healthcare system more widely; for example, a report by the Health Foundation found that although the experiences of NHS staff with the increased use of technology during the COVID-19 pandemic were generally positive, they were particularly positive for those who had been personally involved in implementation. 92 Following from this finding, co-production approaches may be a component in the success of new initiatives. 95 Co-production means the design, implementation and long-term management of initiatives or systems collaboratively by all those who will use or be affected by them. 95 We found that processes at interorganisational and interprofessional levels influenced data-sharing, suggesting that co-production would need to occur at both levels. At the organisational level, co-production might involve agreeing on the processes of collecting, storing and sharing data. At the professional level, co-production approaches can determine what works in the specific contexts in which teams or individuals are working,96 which for data-sharing initiatives might regard the provision of adequate resources such as time (e.g. to learn new systems), functional technological solutions and training. Additionally, involvement in intervention development can result in a sense of ownership, with studies of co-production of other workplace interventions, such as to reduce sedentary behaviour, indicating this improved acceptability and uptake. 97,98 We found allowing time for solutions to be developed to problems that may arise during the implementation of data-sharing initiatives was important; a sense of ownership of the intervention might aid professionals in this iterative process. The wider co-production literature suggests that the success of co-production can be challenged by institutional rigidity and differing values between organisations. 96 These were also barriers to data-sharing, particularly at the organisational level, so would need to be considered if co-production approaches were used to develop data-sharing initiatives.

Limitations

We found a relatively large number of qualitative studies that met our inclusion criteria, which meant that we used a purposive sampling approach. While studies were included on the basis of their data richness, it is possible this approach means that we missed insights contained in some ‘thin’ studies; for example, where they focused on settings for which we found fewer data-rich studies, such as care homes. The ‘thin’ studies identified from our search are accessible in a list in Appendix 3. We found several reports from the grey literature which met our inclusion criteria (both rich and thin studies); these came from citation-chasing and, while they were also identified during our searches of the websites of relevant organisations, we did not find any additional grey literature meeting our inclusion criteria. Although the fact we found grey literature already included in the review while searching websites suggests we were using suitable search terms, it is possible that some grey literature was missed due to the limited search functions offered by organisations’ websites.

Studies tended to focus on data-sharing within the context of wider working practices or interventions, for example integrated care programmes. While this context is part of the complex story surrounding data-sharing, it meant that only a small number of studies specifically investigated data-sharing. As the included studies contained few findings explicitly about data-sharing, we had relatively little information to include in the analysis for some factors perceived to influence effective data-sharing, which sometimes made it difficult to draw out detailed findings. A related limitation is that we found few studies focusing on patient and carer experiences of data-sharing that met our inclusion criteria. While studies of the views of professionals, who are the ones sharing data, are likely to contain the most relevant information on this topic, this evidence gap is significant, as the opinions of patients may influence the effectiveness or implementation of data-sharing between health and social care, for example the perception of EPaCCS as a ‘death register’.

No publication year limit was placed on studies included in the review and, surprisingly, we found no relevant studies on data-sharing during or post the COVID-19 pandemic. As noted above, technology and the routine use of technology has advanced rapidly and the COVID-19 pandemic has had a significant impact on working practices, so this lack of evidence is a limitation of the review. Yet despite the increasing priority being placed on the use of electronic systems in health and social care,19 a key finding of the review is that technology is neither a panacea nor a simple substitute for interpersonal data-sharing, and that multiple methods of data-sharing are needed. While the results of earlier studies in the review may be less relevant today, they still contain information that could inform data-sharing practices.

The methodological quality of the included studies was relatively good and similar across clusters. This provides some reassurance that both the findings related to more general factors and those related to cluster-specific factors are based on good-quality evidence. However, the reviewed evidence has important limitations that need to be considered when interpreting the results. Most of the studies failed to provide evidence of reflexivity and many had limitations or uncertainty related to adequacy of samples and methods of data collection. For some of these studies, this might be a reporting issue rather than a failure to adhere to the accepted standards for conducting qualitative research. Others, however, seem to have genuine limitations which cast doubt on the validity of their findings. While we believe that the overall conclusions of the review are valid and reflect observations that are consistent across studies and settings, specific findings might be biased, incomplete and based on interpretations that failed to consider the complexity of data-sharing and the full range of perspectives relevant to this topic. Therefore, the validity of specific findings (e.g. patients’ and carers’ perceptions of palliative care co-ordination systems as ‘death registers’) should be judged in the context of the primary studies that report them and not taken for granted because they are included in the review and ‘make sense’.

We used framework synthesis as our method of analysis. Framework synthesis was suitable given that our topic – data-sharing – is complex, as it allows flexibility, using both a deductive and inductive approach to identifying themes across included studies. 56 However, the construction of the initial framework relies on existing literature and theories which can lead to limitations, as pre-existing frameworks may contain gaps, with categories not filled by included studies, or information in included studies not fitting into categories in the framework. 57 While iterative development ensured all information in included studies fitted into the framework, there may be gaps where we did not find studies, for example data-sharing regarding mental health. Our consultations with stakeholders were intended to understand whether important areas may have been missed, and if so, to document and identify these as areas for further research.

Equality, diversity and inclusion

We used the NIHR-INCLUDE guidelines99 to reflect on equality, diversity and inclusion while designing the protocol for the review and writing the final report. Our population focus was older people, with age being a PROGRESS-Plus characteristic100,101 that has been identified as being associated with discrimination. 91 Few studies explored older people’s experiences of data-sharing in relation to their health and/or social care records (although this might be a side effect of our inclusion criteria, as acknowledged in Limitations), or the impacts of poor or improved data-sharing, so we have identified this as an area for further study. While not all data-sharing uses technology, both members of the public and NHS staff aged over 55 years have been found to be more likely to report negative experiences of using technology in the NHS. 92 Technological competence of staff was discussed in one included study,23 and the need for training was also identified within the theme of Processes and procedures. Further research into digital inclusion of different care professionals may be needed, as it is important that staff and patients are supported to prevent digital exclusion in the workplace.

Another equity issue highlighted by our research is the common imbalance between those care professionals who produce and enter data into electronic record systems, and those who use the systems and their data. There seemed to be few mechanisms for incentivising or compensating those care professionals who might be expected to invest time in providing data for systems, yet who may not be the primary beneficiaries of having a shared or integrated patient/user record system.

Public and patient involvement and engagement

We consulted PPIE representatives while searching for studies and developing the framework for analysis. They shared their views of data-sharing between health and social care and their experiences, for example, of arrangements being made for social care provision after a stay in hospital. While these discussions made some contribution to our understanding of the topic and key issues regarding data-sharing which informed the framework, they highlighted the challenges of discussing a topic which focuses primarily on professional working practices with service users. PPIE representatives generally thought that data should be shared between health and social care and provided examples of how it had impacted them when this had been done well or had not happened. However, they were unaware of the methods that professionals had used to share data. This raised questions in the research team around the value of eliciting service user views on the topic, as PPIE representatives were more concerned with whether the relevant data had been shared than the method used to do this. We acknowledge that we were unable to involve PPIE representatives in all stages of the review, due to a lack of time and resource; additional meetings might have enabled richer discussions.

Later consultation focused on the plain language summary of the review and ensuring that this conveyed the findings of the review in an accessible way.

Acknowledgements

The Exeter HS&DR Evidence Synthesis Centre team would like to acknowledge and thank Health and Care Research Wales and Social Care Wales for their support in developing this review and the members of the PSAG and PPIE group for their valuable feedback. We would also like to thank Sue Whiffin and Jenny Lowe for administrative support throughout this review.

This report presents independent research funded by the National Institute for Health Research (NIHR). This review (award NIHR135660) was commissioned by the Health Services and Delivery Research (HSDR) programme as part of a series of evidence syntheses under the award NIHR130538 and supported by the NIHR Applied Research Collaboration South West Peninsula. For more information, visit https://fundingawards.nihr.ac.uk/award/NIHR135660 and https://fundingawards.nihr.ac.uk/award/NIHR130538.

CRediT contribution statement

Siân de Bell (https://orcid.org/0000-0001-7356-3849): Conceptualisation (equal), Formal analysis (equal), Writing – original draft (lead), Writing – editing and reviewing (equal). Zhivko Zhelev (https://orcid.org/0000-0002-0106-2401): Conceptualisation (equal), Formal analysis (equal), Visualisation (lead), Writing – original draft (supporting), Writing – editing and reviewing (equal). Alison Bethel (https://orcid.org/0000-0002-0963-9201): Conceptualisation (equal), Writing – editing and reviewing (equal). Jo Thompson Coon (https://orcid.org/0000-0002-5161-0234): Conceptualisation (equal), Funding acquisition (equal), Project administration (equal), Writing – editing and reviewing (equal). Rob Anderson (https://orcid.org/0000-0002-3523-8559): Conceptualisation (equal), Funding acquisition (equal), Project administration (equal), Writing – editing and reviewing (equal).

Disclosure of interests

Full disclosure of interests: Completed ICMJE forms for all authors, including all related interests, are available in the toolkit on the NIHR Journals Library report publication page at https://doi.org/10.3310/TTWG4738.

Primary conflicts of interest: Siân de Bell declares none. Zhivko Zhelev declares none. Jo Thompson Coon is a member of the NIHR HTA General Board (2019–23) and was a member of NIHR PRU Commissioning Panel 2022. Jo Thompson Coon and Alison Bethel are also supported by the National Institute for Health Applied Research Collaboration South West Peninsula. Rob Anderson declares none.

Data-sharing statement

This is an evidence synthesis study based on published primary research; it did not generate new data. All data extracted from the primary studies, along with links to each publication, can be found in the Appendices to the report. Further information can be obtained from the corresponding author.

Ethics statement

This was an evidence review, based on published primary studies, so ethical approval was not required.

Information governance statement

This study did not involve primary research or, therefore, the handling of any personal information.

Department of Health and Social Care disclaimer

This publication presents independent research commissioned by the National Institute for Health and Care Research (NIHR). The views and opinions expressed by the interviewees in this publication are those of the interviewees and do not necessarily reflect those of the authors, those of the NHS, the NIHR, MRC, NIHR Coordinating Centre, the HSDR programme or the Department of Health and Social Care.

This monograph was published based on current knowledge at the time and date of publication. NIHR is committed to being inclusive and will continually monitor best practice and guidance in relation to terminology and language to ensure that we remain relevant to our stakeholders.

Disclaimers

This article presents independent research funded by the National Institute for Health and Care Research (NIHR). The views and opinions expressed by authors in this publication are those of the authors and do not necessarily reflect those of the NHS, the NIHR, the HSDR programme or the Department of Health and Social Care. If there are verbatim quotations included in this publication the views and opinions expressed by the interviewees are those of the interviewees and do not necessarily reflect those of the authors, those of the NHS, the NIHR, the HSDR programme or the Department of Health and Social Care.

CINAHL Ultimate

(((TI semi-structured OR AB semi-structured OR SU semi-structured) OR (TI semistructured OR AB semistructured OR SU semistructured) OR (TI unstructured OR AB unstructured OR SU unstructured) OR (TI informal OR AB informal OR SU informal) OR (TI in-depth OR AB in-depth OR SU in-depth) OR (TI indepth OR AB indepth OR SU indepth) OR (TI face-to-face OR AB face-to-face OR SU face-to-face) OR (TI structured OR AB structured OR SU structured) OR (TI guide OR AB guide OR SU guide)) N3 ((TI discussion* OR AB discussion* OR SU discussion*) OR (TI questionnaire* OR AB questionnaire* OR SU questionnaire*)))

OR

((TI “focus group*” OR AB “focus group*” OR SU “focus group*”) OR (TI qualitative OR AB qualitative OR SU qualitative) OR (TI ethnograph* OR AB ethnograph* OR SU ethnograph*) OR (TI fieldwork OR AB fieldwork OR SU fieldwork) OR (TI “field work” OR AB “field work” OR SU “field work”) OR (TI “key informant*” OR AB “key informant*” OR SU “key informant*”) OR (TI interview* OR AB interview* OR SU interview*))

OR

(MH “Qualitative Studies+”)

AND

(((TI share* OR AB share*) OR (TI sharing OR AB sharing) OR (TI integrat* OR AB integrat*)) N2 ((TI data OR AB data) OR (TI organisat* OR AB organisat*) OR (TI organizat* OR AB organizat*)))

OR

((TI integrat* OR AB integrat*) N1 ((TI care* OR AB care*) OR (TI health* OR AB health*)))

OR

(((TI interprofession* OR AB interprofession*) OR (TI ‘inter profession*’ OR AB ‘inter profession*’) OR (TI organisat* OR AB organisat*) OR (TI organizat* OR AB organizat*) OR (TI interagency* OR AB interagency*) OR (TI agenc* OR AB agenc*)) N1 ((TI communicat* OR AB communicat*) OR (TI share* OR AB share*) OR (TI sharing OR AB sharing) OR (TI integrat* OR AB integrat*)))

OR

(((TI data OR AB data) OR (TI digital* OR AB digital*)) N1 ((TI link* OR AB link*) OR (TI system* OR AB system*) OR (TI access OR AB access) OR (TI information* OR AB information*) OR (TI exchange* OR AB exchange*)))

AND

AF “united kingdom” OR SU “united kingdom” OR TI “united kingdom” OR AB “united kingdom” OR TI britain OR AB britain OR SU britain OR AF britain OR TI england OR AB england OR SU england OR AF England

OR

AF wales OR SU wales OR TI wales OR AB wales OR TI scotland OR AB scotland OR SU scotland OR AF scotland OR TI “northern ireland” OR AB “northern ireland” OR SU “northern ireland” OR AF “northern ireland”

OR

(MH “United Kingdom+”) OR (MH “Great Britain+”)

EMBASE search strategy

Database: EMBASE <1974 to 2023 March 21>

Search strategy:

------------ ------------- ------------- ------------ ---------- ---------------

1. exp qualitative research/ (113,853)

2. ((“semi-structured” or semistructured or unstructured or informal or “in-depth” or indepth or “face-to-face” or structured or guide) adj3 (discussion* or questionnaire*)).tw. (49,954)

3. (focus group* or qualitative or ethnograph* or fieldwork or field work or key informant* or interview* or experience*).tw. (2,512,755)

4. 1 or 2 or 3 (2,546,302)

5. ((share* or sharing or integrat*) adj2 (data or organisat* or organizat*)).tw. (34,626)

6. (integrat* adj2 (care* or health*)).tw. (38,525)

7. ((interprofession* or inter profession* or organisat* or organizat* or interagency* or agenc*) adj2 (communicat* or share* or sharing or integrat*)).tw. (5901)

8. ((data or digital*) adj2 (link* or system* or access or information* or exchange*)).tw. (123,953)

9. (electronic* adj2 record*).tw. (112,902)

10. (patient* adj2 record*).tw. (118,943)

11. (share* or sharing or integrat* or access*).tw. (2,118,154)

12. 9 or 10 (217,083)

13. ((share* or sharing or integrat* or access*) adj2 ((electronic* adj2 record*) or (patient* adj2 record*))).tw. (3086)

14. 5 or 6 or 7 or 8 or 13 (197,865)

15. interdisciplinary communication/ (13,450)

16. 14 or 15 (210,857)

17. 4 and 16 (35,185)

18. exp United Kingdom/ (459,762)

19. (national health service* or nhs*).ti,ab,in,ad. (465,861)

20. (english not ((published or publication* or translat* or written or language* or speak* or literature or citation*) adj5 english)).ti,ab. (58,823)

21. (gb or “g.b.” or britain* or (british* not “british columbia”) or uk or “u.k.” or united kingdom* or (england* not “new england”) or northern ireland* or northern irish* or scotland* or scottish* or ((wales or “south wales”) not “new south wales”) or welsh*).ti,ab,jx,in,ad. (3,712,789)

22. (bath or “bath’s” or ((birmingham not alabama*) or (“birmingham’s” not alabama*) or bradford or “bradford’s” or brighton or “brighton’s” or bristol or “bristol’s” or carlisle* or “carlisle’s” or (cambridge not (massachusetts* or boston* or harvard*)) or (“cambridge’s” not (massachusetts* or boston* or harvard*)) or (canterbury not zealand*) or (“canterbury’s” not zealand*) or chelmsford or “chelmsford’s” or chester or “chester’s” or chichester or “chichester’s” or coventry or “coventry’s” or derby or “derby’s” or (durham not (carolina* or nc)) or (“durham’s” not (carolina* or nc)) or ely or “ely’s” or exeter or “exeter’s” or gloucester or “gloucester’s” or hereford or “hereford’s” or hull or “hull’s” or lancaster or “lancaster’s” or leeds* or leicester or “leicester’s” or (lincoln not nebraska*) or (“lincoln’s” not nebraska*) or (liverpool not (new south wales* or nsw)) or (“liverpool’s” not (new south wales* or nsw)) or ((london not (ontario* or ont or toronto*)) or (“london’s” not (ontario* or ont or toronto*)) or manchester or “manchester’s” or (newcastle not (new south wales* or nsw)) or (“newcastle’s” not (new south wales* or nsw)) or norwich or “norwich’s” or nottingham or “nottingham’s” or oxford or “oxford’s” or peterborough or “peterborough’s” or plymouth or “plymouth’s” or portsmouth or “portsmouth’s” or preston or “preston’s” or ripon or “ripon’s” or salford or “salford’s” or salisbury or “salisbury’s” or sheffield or “sheffield’s” or southampton or “southampton’s” or st albans or stoke or “stoke’s” or sunderland or “sunderland’s” or truro or “truro’s” or wakefield or “wakefield’s” or wells or westminster or “westminster’s” or winchester or “winchester’s” or wolverhampton or “wolverhampton’s” or (worcester not (massachusetts* or boston* or harvard*)) or (“worcester’s” not (massachusetts* or boston* or harvard*)) or (york not (“new york*” or ny or ontario* or ont or toronto*)) or (“york’s” not (“new york*”or ny or ontario* or ont or toronto*))))).ti,ab,in,ad. (2,893,846)

23. (bangor or “bangor’s” or cardiff or “cardiff’s” or newport or “newport’s” or st asaph or “st asaph’s” or st davids or swansea or “swansea’s”).ti,ab,in,ad. (119,041)

24. (aberdeen or “aberdeen’s” or dundee or “dundee’s” or edinburgh or “edinburgh’s” or glasgow or “glasgow’s” or inverness or (perth not australia*) or (“perth’s” not australia*) or stirling or “stirling’s”).ti,ab,in,ad. (398,020)

25. (armagh or “armagh’s” or belfast or “belfast’s” or lisburn or “lisburn’s” or londonderry or “londonderry’s” or derry or “derry’s” or newry or “newry’s”).ti,ab,in,ad. (55,330)

26. or/18-25 (4,533,426)

27. (exp “arctic and antarctic”/or exp oceanic regions/ or exp western hemisphere/or exp africa/ or exp asia/) not (exp united kingdom/ or europe/) (3,484,679)

28. 26 not 27 (4,278,686)

29. 17 and 28 (5868)

30. (share* or sharing or integrat* or data or organisat* or organizat* or care* or health* or interprofession* or inter profession* or interagency* or agenc* or communicat* or data or digital* or link* or system* or access or information* or exchange*).ti. (3,815,833)

31. 29 and 30 (2725)

MEDLINE search strategy

Database: Ovid MEDLINE(R) ALL <1946 to March 22, 2023>

Search strategy:

HMIC search strategy

Database: HMIC Health Management Information Consortium <1979 to January 2023>

Search strategy:

--------- ---------- ----------- ----------- ------------ ----------- -----------

1. (focus group* or qualitative or ethnograph* or fieldwork or field work or key informant* or interview*).tw. (25,366)

2. ((“semi-structured” or semistructured or unstructured or informal or “in-depth” or indepth or “face-to-face” or structured or guide) adj3 (discussion* or questionnaire*)).tw. (875)

3. exp qualitative research/or mixed-methods research/ or qualitative analysis/ or qualitative techniques/(1710)

4. 1 or 2 or 3 (25,870)

5. ((share* or sharing or integrat*) adj3 (data or organisat* or organizat*)).tw. (745)

6. (integrat* adj2 (care* or health*)).tw. (3081)

7. ((interprofession* or inter profession* or organisat* or organizat* or interagency* or agenc*) adj2 (communicat* or share* or sharing or integrat*)).tw. (532)

8. ((data or digital*) adj2 (link* or system* or access or information* or exchange*)).tw. (1709)

9. exp Access to information/(2159)

10. exp Information transfer/(2053)

11. (electronic* adj2 record*).tw. (1372)

12. (patient* adj2 record*).tw. (1564)

13. (share* or sharing or integrat* or access*).tw. (35,836)

14. 11 or 12 (2411)

15. ((share* or sharing or integrat* or access*) adj2 ((electronic* adj2 record*) or (patient* adj2 record*))).tw. (159)

16. 5 or 6 or 7 or 8 or 9 or 10 or 15 (9552)

17. 4 and 16 (1001)

18. (share* or sharing or integrat* or data or organisat* or organizat* or care* or health* or interprofession* or inter profession* or interagency* or agenc* or communicat* or data or digital* or link* or system* or access or information* or exchange*).ti. (146,652)

19. 17 and 18 (556)

SPP search strategy

Database: SPP <202210>

Search strategy:

--------- ----------- ----------- ---------- ---------- ----------- --------

1. (focus group* or qualitative or ethnograph* or fieldwork or field work or key informant* or interview*).tw. (43,060)

2. ((“semi-structured” or semistructured or unstructured or informal or “in-depth” or indepth or “face-to-face” or structured or guide) adj3 (discussion* or questionnaire*)).tw. (607)

3. ((share* or sharing or integrat*) adj3 (data or organisat* or organizat*)).tw. (936)

4. (integrat* adj2 (care* or health*)).tw. (3534)

5. ((interprofession* or inter profession* or organisat* or organizat* or interagency* or agenc*) adj2 (communicat* or share* or sharing or integrat*)).tw. (811)

6. ((data or digital*) adj2 (link* or system* or access or information* or exchange*)).tw. (1410)

7. (electronic* adj2 record*).tw. (206)

8. (patient* adj2 record*).tw. (178)

9. (share* or sharing or integrat* or access*).tw. (60,232)

10. 7 or 8 (340)

11. ((share* or sharing or integrat* or access*) adj2 ((electronic* adj2 record*) or (patient* adj2 record*))).tw. (27)

12. 1 or 2 (43,306)

13. 4 or 5 or 6 or 11 (5629)

14. 12 and 13 (564)

PQDT search strategy

(((((TI,AB(semi-structured) OR TI,AB(semistructured) OR TI,AB(unstructured) OR TI,AB(informal) OR TI,AB(in-depth) OR TI,AB(indepth) OR TI,AB(face-to-face) OR TI,AB(structured) OR TI,AB(guide)) NEAR/3 (TI,AB(discussion*) OR TI,AB(questionnaire*))) OR (TI,AB((“focus group” OR “focus groups”)) OR TI,AB(qualitative) OR TI,AB(ethnograph*) OR TI,AB(fieldwork) OR TI,AB(“field work”) OR TI,AB((“key informant” OR “key informants”)) OR TI,AB(interview*))) AND (((TI,AB(share*) OR TI,AB(sharing) OR TI,AB(integrat*)) NEAR/2 (TI,AB(data) OR TI,AB(organisat*) OR TI,AB(organizat*))) OR ((TI,AB(interprofession*) OR TI,AB((“inter professional”)) OR TI,AB(organisat*) OR TI,AB(organizat*) OR TI,AB(interagency*) OR TI,AB(agenc*)) NEAR/2 (TI,AB(communicat*) OR TI,AB(share*) OR TI,AB(sharing) OR TI,AB(integrat*))) OR ((TI,AB(data) OR TI,AB(digital*)) NEAR/2 (TI,AB(link*) OR TI,AB(system*) OR TI,AB(access) OR TI,AB(information*) OR TI,AB(exchange*))) OR (((TI,AB(electronic*) NEAR/2 TI,AB(record*)) OR (TI,AB(patient*) NEAR/2 TI,AB(record*))) NEAR/2 (TI,AB(share*) OR TI,AB(sharing) OR TI,AB(integrat*) OR TI,AB(access*))) OR (TI,AB(information) NEAR/2 (TI,AB(share) OR TI,AB(sharing))))) AND ulo((gb OR g.b. OR britain OR uk OR u.k. OR “united kingdom*” OR england OR “northern ireland*” OR scotland* OR wales))) AND (title(health) OR title(social))

Web of Science (all databases listed) search strategy

# Web of Science search strategy (v0.1)

# Database: Web of Science Core Collection

# Entitlements:

1. - WOS.SCI: 1900 to 2023

2. - WOS.AHCI: 1975 to 2023

3. - WOS.ESCI: 2015 to 2023

4. - WOS.ISTP: 1990 to 2023

5. - WOS.SSCI: 1900 to 2023

6. - WOS.ISSHP: 1990 to 2023

# Searches:

1. TS=((semi-structured OR semistructured OR unstructured OR informal OR in-depth OR indepth OR face-to-face OR structured OR guide) NEAR/3 (discussion* OR questionnaire*))

2. TS=(“focus group*” or qualitative* or enthograph* or fieldwork or “field work” or “key informant*” or interview*)

3. #2 OR #1

4. (TI=((share* OR sharing OR integrat*) NEAR/2 (data OR organisat* OR organizat*)) OR AB=((share* OR sharing OR integrat*) NEAR/2 (data OR organisat* OR organizat*))) Date Run: Wed Mar 22 2023 11:45:16 GMT+0000 (Greenwich Mean Time) Results: 101793

5. (TI=(integrat* NEAR/2 (care* OR health*)) OR AB=(integrat* NEAR/2 (care* OR health*)))

6. (TI=((interprofession* OR “inter profession*” OR organisat* OR organizat* OR interagency*OR agenc*) NEAR/2 (communicat* OR share* OR sharing OR integrat*)) OR AB=((interprofession* OR “inter profession*” OR organisat* OR organizat* OR interagency* OR agenc*) NEAR/2 (communicat* OR share* OR sharing OR integrat*)))

7. (TI=((data OR digital*) NEAR/2 (link* OR system* OR access OR information* OR exchange*)) OR AB=((data OR digital*) NEAR/2 (link* OR system* OR access OR information* OR exchange*)))

8. (TI=(electronic* NEAR/2 record*) OR AB=(electronic* NEAR/2 record*))

9. (TI=(patient* NEAR/2 record*) OR AB=(patient* NEAR/2 record*))

10. (TI=(share* OR sharing OR integrat* OR access*) OR AB=(share* OR sharing OR integrat*OR access*))

11. #8 OR #9

12. #10 AND #11

13. #12 OR #7 OR #6 OR #5 OR #4

14. #13 AND #3

15. TS=(“national health service” or nhs)

16. OO=(“national health service” or nhs)

17. AB=((english not ((published or publication* or translat* or written or language* or speak* or literature or citation*) NEAR/5 (english))))

18. TS=(gb or “g.b.” or britain or (british not “british columbia”) or uk or “u.k.” or united kingdom*or (england not “new england”) or northern ireland* or nothern irish* or scotland* or scottish* or ((wales or “south wales”) not “new south wales”) or welsh*)

19. CU=(gb or “g.b.” or britain or (british not “british columbia”) or uk or “u.k.” or united kingdom*or (england not “new england”) or northern ireland* or nothern irish* or scotland* or scottish* or ((wales or “south wales”) not “new south wales”) or welsh*)

20. SO=(gb or “g.b.” or britain or (british not “british columbia”) or uk or “u.k.” or united kingdom*or (england not “new england”) or northern ireland* or nothern irish* or scotland* or scottish* or ((wales or “south wales”) not “new south wales”) or welsh*)

21. CI=(bath or “bath’s” or ((birmingham not alabama*) or (“birmingham’s” not alabama*) or bradford or “bradford’s” or brighton or “brighton’s” or bristol or “bristol’s” or carlisle* or “carlisle’s” or (cambridge not (massachusetts* or boston* or harvard*)) or (“cambridge’s” not (massachusetts* or boston* or harvard*)) or (canterbury not zealand*) or (“canterbury’s” not zealand*) or chelmsford or “chelmsford’s” or chester or “chester’s” or chichester or “chichester’s” or coventry or “coventry’s” or derby or “derby’s” or (durham not (carolina* or nc)) or (“durham’s” not (carolina* or nc)) or ely or “ely’s” or exeter or “exeter’s” or gloucester or “gloucester’s” or hereford or “hereford’s” or hull or “hull’s” or lancaster or “lancaster’s” or leeds* or leicester or “leicester’s” or (lincoln not nebraska*) or (“lincoln’s” not nebraska*) or (liverpool not (new south wales* or nsw)) or (“liverpool’s” not (new south wales* or nsw)) or ((london not (ontario* or ont or toronto*)) or (“london’s” not (ontario* or ont or toronto*)) or manchester or “manchester’s” or (newcastle not (new south wales* or nsw)) or (“newcastle’s” not (new south wales* or nsw)) or norwich or “norwich’s” or nottingham or “nottingham’s” or oxford or “oxford’s” or peterborough or “peterborough’s” or plymouth or “plymouth’s” or portsmouth or “portsmouth’s” or preston or “preston’s” or ripon or “ripon’s” or salford or “salford’s” or salisbury or “salisbury’s” or sheffield or “sheffield’s” or southampton or “southampton’s” or st albans or stoke or “stoke’s” or sunderland or “sunderland’s” or truro or “truro’s” or wakefield or “wakefield’s” or wells or westminster or “westminster’s” or winchester or “winchester’s” or wolverhampton or “wolverhampton’s” or (worcester not (massachusetts* or boston* or harvard*)) or (“worcester’s” not (massachusetts*or boston* or harvard*)) or (york not (“new york*” or ny or ontario* or ont or toronto*)) or (“york’s” not (“new york*” or ny or ontario* or ont or toronto*)))))

22. OO=(bath or “bath’s” or ((birmingham not alabama*) or (“birmingham’s” not alabama*) or bradford or “bradford’s” or brighton or “brighton’s” or bristol or “bristol’s” or carlisle* or “carlisle’s” or (cambridge not (massachusetts* or boston* or harvard*)) or (“cambridge’s” not (massachusetts* or boston* or harvard*)) or (canterbury not zealand*) or (“canterbury’s” not zealand*) or chelmsford or “chelmsford’s” or chester or “chester’s” or chichester or “chichester’s” or coventry or “coventry’s” or derby or “derby’s” or (durham not (carolina* or nc)) or (“durham’s” not (carolina* or nc)) or ely or “ely’s” or exeter or “exeter’s” or gloucester or “gloucester’s” or hereford or “hereford’s” or hull or “hull’s” or lancaster or “lancaster’s” or leeds* or leicester or “leicester’s” or (lincoln not nebraska*) or (“lincoln’s” not nebraska*) or (liverpool not (new south wales* or nsw)) or (“liverpool’s” not (new south wales* or nsw)) or ((london not (ontario* or ont or toronto*)) or (“london’s” not (ontario* or ont or toronto*)) or manchester or “manchester’s” or (newcastle not (new south wales* or nsw)) or (“newcastle’s” not (new south wales* or nsw)) or norwich or “norwich’s” or nottingham or “nottingham’s” or oxford or “oxford’s” or peterborough or “peterborough’s” or plymouth or “plymouth’s” or portsmouth or “portsmouth’s” or preston or “preston’s” or ripon or “ripon’s” or salford or “salford’s” or salisbury or “salisbury’s” or sheffield or “sheffield’s” or southampton or “southampton’s” or st albans or stoke or “stoke’s” or sunderland or “sunderland’s” or truro or “truro’s” or wakefield or “wakefield’s” or wells or westminster or “westminster’s” or winchester or “winchester’s” or wolverhampton or “wolverhampton’s” or (worcester not (massachusetts* or boston* or harvard*)) or (“worcester’s” not (massachusetts* or boston* or harvard*)) or (york not (“new york*” or ny or ontario* or ont or toronto*)) or (“york’s” not (“new york*” or ny or ontario* or ont or toronto*)))))

23. CI=(bangor or “bangor’s” or cardiff or “cardiff’s” or newport or “newport’s” or st asaph or “st asaph’s” or st davids or swansea or “swansea’s”)

24. OO=(bangor or “bangor’s” or cardiff or “cardiff’s” or newport or “newport’s” or st asaph or “st asaph’s” or st davids or swansea or “swansea’s”)

25. CI=(aberdeen or “aberdeen’s” or dundee or “dundee’s” or edinburgh or “edinburgh’s” or glasgow or “glasgow’s” or inverness or (perth not australia*) or (“perth’s” not australia*) or stirling or “stirling’s”)

26. OO=(aberdeen or “aberdeen’s” or dundee or “dundee’s” or edinburgh or “edinburgh’s” or glasgow or “glasgow’s” or inverness or (perth not australia*) or (“perth’s” not australia*) or stirling or “stirling’s”)

27. CI=(armagh or “armagh’s” or belfast or “belfast’s” or lisburn or “lisburn’s” or londonderry or “londonderry’s” or derry or “derry’s” or newry or “newry’s”)

28. OO=(armagh or “armagh’s” or belfast or “belfast’s” or lisburn or “lisburn’s” or londonderry or “londonderry’s” or derry or “derry’s” or newry or “newry’s”)

29. #15 OR #16 OR #17 OR #18 OR #19 OR #20 OR #21 OR #22 OR #23 OR #24 OR #25 OR #26 OR #27 OR #28

30. #29 AND #14

31. TS=(health* NEAR/5 (care* or social))

32. #31 AND #30

33. TS=(older NEAR/2 (people* or person* or adult* or patient*))

34. TS=(elderly or senior* or aging or ageing or geriatric*)

35. #33 OR #34 Editions:WOS.SCI,WOS.SSCI,WOS.AHCI,WOS.ESCI Date Run:Wed Mar 22 2023 12:13:39 GMT+0000 (Greenwich Mean Time) Results:4280476

36. #35 AND #30

37. #32 OR #36

38. TI=(share* or sharing or integrat* or data or communicat* or information*)

39. #38 AND #30

40. #39 OR #37