Introduction and background
Since the late 1980s, policy in England has rested on the assumption that an individual requiring support has rights to closely-tailored, responsive, flexible, personal packages of care (Carr, Reference Carr2010). The concept of case management emerged within these parameters, with its United Kingdom roots in local authority social services (Challis and Davies, Reference Challis and Davies1986) and community mental health provision (Challis et al., Reference Challis, Jacobs, Hughes, Stewart, Venables and Weiner2005). Initially, case management, rapidly renamed care management, was conceived of as a mainly social work task. Despite the evidence identifying the value of specific targeted approaches to implementing care management (Challis et al., Reference Challis, Darton, Hughes, Huxley and Stewart2001), more generalised and less intensive approaches developed. Ambitions to take account of service users’ wishes, resources and needs were diluted by care managers’ lack of direct commissioning or purchasing power and by high thresholds of eligibility for publicly funded social care. Only 4% of the care managers’ time was spent liaising with National Health Service (NHS) staff (Weinberg et al., Reference Weinberg, Williamson, Challis and Hughes2003).
However, some efforts to integrate health and social care services have been tried, with nurses undertaking care management roles, particularly in adult mental health services. Challis et al. (Reference Challis, Jacobs, Hughes, Stewart, Venables and Weiner2005) found that closer communication across the professions resulted generally in less time needing to be spent on care planning and case review in the minority of social services departments that have NHS staff working alongside social workers. However, the Department of Health (DH, 2004a) has introduced in England a new case manager role for nurses in NHS primary care, the ‘community matron’. This new Nurse Case (rather than care) Manager (NCM) role was charged with managing people with multiple long-term conditions who were at risk of unplanned hospital admission (DH, 2004b). A national target was set for NHS Primary Care Trusts (PCTs) to appoint 3000 community matrons by 2007, later extended to 2008 (DH, 2004c). The employment of community matrons did not reach the target numbers for 2008 (Keen, Reference Keen2008), but there was widespread adoption of the role, in different ways, at local level.
Currently, the role of care management varies in local social services agencies, some care managers being located at community-based and hospital-based older people's teams, and at community-based teams for adults with mental health problems (Jacobs et al., Reference Jacobs, Challis, Stewart and Weiner2006). International descriptions of case management show similar variations (Verkade et al., Reference Verkade, Van Meijel, Brink, Van Os-Medendorp, Bauke Koekkoek and Francke2010). An older person therefore might be care or case ‘managed’ by a variety of professionals (Pickard, Reference Pickard2009) or feel that the social work service is a ‘lottery’ (Manthorpe et al., Reference Manthorpe, Moriarty, Rapaport, Clough, Cornes, Bright and Iliffe2008). The extent of actual overlap in England was recently investigated by Challis et al. (Reference Challis, Hughes, Berzins, Reilly, Abell and Stewart2010), who noted distinct similarities in the goals and objectives of NHS case management and local authority (social services) care management but also some differences:
The principal differences reflected policy guidance with the latter (social services) emphasising inappropriate care home admission and a care management approach to the majority of users and the former (NHS) a greater focus on improved health outcomes for patients and a more differentiated response to need apparent in the levels and qualifications of staff (Challis et al., Reference Challis, Hughes, Berzins, Reilly, Abell and Stewart2010: 7)
The overall aim of the study from which this paper is drawn was to describe which nurses ‘did’ case management in NHS primary care, what they did, what the recipients of case management thought of it and what impact it had on other stakeholders, including local authority social services. Our study used the term case manager for a professional (in this instance a nurse) holding responsibility for overseeing and/or delivering the processes of case finding, assessment, care delivery, monitoring and review for, and with, a patient and their family carer(s) (Goodman et al., Reference Goodman, Drennan, Davies, Masey, Gage, Scott, Manthorpe, Brearley and Iliffe2010: 13). We undertook a literature review, a national survey of nurse case management in PCTs in England and case studies of the services in three purposively selected PCTs. This paper draws on findings from the case studies.
A detailed report on the methodology of the overall study can be found elsewhere (Goodman et al., Reference Goodman, Drennan, Davies, Masey, Gage, Scott, Manthorpe, Brearley and Iliffe2010), along with a description of the policy context (Drennan et al., Reference Drennan, Goodman, Manthorpe, Davies, Scott, Gage and Iliffe2011), the relationship between NCMs and general practice (Iliffe et al., Reference Iliffe, Drennan, Manthorpe, Gage, Davies, Masey, Scott, Brearley and Goodman2011) and an economic evaluation of nurse case management (Gage et al., under review). This paper explores the potential impact of NCMs on older people and on interfaces with social services.
Methods
Building on a review of the policy and research literature, and a national survey of the implementation of the NCM role (as described in the full report, see Goodman et al., Reference Goodman, Drennan, Davies, Masey, Gage, Scott, Manthorpe, Brearley and Iliffe2010; Drennan et al., Reference Drennan, Goodman, Manthorpe, Davies, Scott, Gage and Iliffe2011), this study adopted a case study approach (Yin, Reference Yin2003) because this method is highly suited to exploring innovations in their context. Case study methods are appropriate when investigators desire or are forced by circumstances to define research topics broadly, to cover contextual or complex multivariate conditions and to rely on multiple sources of evidence (Yin, Reference Yin1999; Reference Yin2003). They enable researchers to understand emerging problems and their practical solutions in the system under study, and to gain insights that are potentially profitable in optimising future development and policy advice (Swanborn, Reference Swanborn1996; Reference Swanborn2010). Case studies can be exploratory, descriptive or explanatory, and findings from explanatory case studies can be amassed for cross-case analyses.
Three sites participated with differing socioeconomic characteristics. The study took place between 2008 and 2009. Figure 1 shows the derivation of the case study samples.
Figure 1 Derivation of the case study samples. PCTs = primary care trusts.
Recruiting the case study sites
During the national survey, five PCTs in different regions of England expressed an interest in participating in the next phase of the study. After consideration of population diversity, sociodemographic characteristics and health economies, three PCTs were recruited to form the case studies reported here to ensure a spread of sites.
In Site 1, an inner urban area of a major city, the PCT was coterminous with an inner London local authority (population just over 200 000). It is an area with high levels of deprivation, but also with areas of affluence. Its Index of Multiple Deprivation (IMD) score for 2004 placed it among the worst 20 out of the 354 local authorities in England. The PCT was responsible for the commissioning of all local health services. The local authority was responsible for other local publicly funded services, such as social care.
Site 2 PCT covered a rural county (population 250 000) containing agricultural and service industries. It is relatively affluent compared with national averages, with no localities classed as significantly deprived (IMD score). A two-tier division of responsibility for publicly funded services existed between the two District Councils and the County Council. The County Council was responsible for Adult Social Services and the Districts for services such as housing. The PCT was responsible for commissioning all local health services.
Site 3 PCT, a coastal conurbation, was coterminous with a local authority (population 250 000). The site was fast growing and economically strong in 2007. Although there was relative affluence in many parts, there were also areas of deprivation. Four localities were among the 25% of the most deprived localities in the country (IMD score). The PCT was responsible for the commissioning of all local health services. The local authority was responsible for publicly funded services, such as social care and housing.
General recruitment
All potential participants were provided with information sheets, given time to consider whether to participate and had the opportunity to raise any questions or concerns. Written or verbal consent was obtained.
Recruiting NCMs
Senior managers in the three PCTs were requested to ask NCMs whom they considered to be providing a good standard of care to indicate their willingness to participate and be contacted by the study team. The managers were not involved once the NCMs had expressed interest in possible participation. The aim was to recruit four NCMs in each study site: a total of 12. NCMs were followed up by interview over a period of at least nine months. In this paper, we refer to all as NCMs to avoid confusion of different terminology.
Recruiting older people
The last five patients referred to each of the four participating NCMs in the three study sites were identified by the NCMs, and were asked if they would be interested in participating in the study. The aim was to recruit up to 60 patient participants. Patients were excluded from recruitment if they had severe mental health problems, were under the age of 60 or were at the end of life. The NCMs made the initial contact with each patient and asked for permission for researchers to contact the patient and explain the study.
Carer recruitment
At the first interview, each patient participant was asked to identify whether there was a family member (or friend) who provided support with activities of daily living. The researcher provided these nominated carers with the study information sheet and consent form, allowing two days before re-contacting to see whether the carer would be interested in participating. We emphasised that the purpose of carer participation was to reflect on the services that the patient and carer received and not to discuss personal information about the patient. Patients and carers were followed up by interviews over a period of nine months.
Stakeholder recruitment
We aimed to recruit up to 10 stakeholders in each site to investigate how NCMs were perceived from organisational, service and user perspectives. The aim was to recruit informants with a wide range of perspectives, including NHS and social care practitioners (eg, medical consultants, general practitioners (GPs), social workers), managers, commissioners and patient representatives. These stakeholders were identified by the NCMs in the study, by reviewing PCT documents on long-term care strategies, and by invitation to local patient representative groups. Stakeholders were interviewed only once in the study period.
Interviews
Semi-structured, face-to-face interviews with each participating NCM were carried out to gather data on their professional history and education, patient caseload, working practices, the team and organisation where they were located, collaboration and communication methods with other professionals, organisations and services, views on nurse case management and on any other issues of importance to them.
Semi-structured face-to-face interviews with older people and carers were conducted, and were followed up by telephone interviews. These interviews gathered data on how the older people came to need case management, their health and personal circumstances, the range of activities the NCM carried out with them, including the amount of contact time, the services received and resources received, including any that were not arranged or coordinated by the NCM.
Semi-structured interviews were undertaken with stakeholders either face-to-face or by telephone as preferred. The topics covered in this interview included perceptions of different types of nurse case management, local influences on the development of nurse case management, experience of the contribution and impact of nurse case management and factors supporting or inhibiting it.
Interviews were recorded, transcribed and stored in NViVO (version 7) qualitative data handling software. Tapes were then erased. The process of analysis of the interview data used a template methodology (Crabtree and Miller, Reference Crabtree and Miller1992) and was undertaken by two members of the research team. This incorporated themes arising from the earlier national survey, whereas a method of constant comparison was used to identify any new themes (Strauss, Reference Strauss1987).
Ethics approval
The study received full ethical approvals from the NHS Southampton & South West Hampshire Research Ethics Committee (July 2007), the University of Hertfordshire and the research governance bodies of the study sites.
Findings
Study participants
A range of different ‘models’ of nurse case management was identified among the 12 NCMs in the three study sites. All were registered nurses with a spread of different roles. Four were community matrons (full-time dedicated NCMs with small caseloads of less than 50 older people); three were District Nurses with large pre-existing caseloads, who used case management methods selectively in their work; three were Clinical Nurse Specialists, who focused on patients with a specific major health problem (like heart failure); one was a Nurse Practitioner, based in a general practice; and one was a Care Home Case Manager whose caseload included older people living in care homes. The District Nurses and the Nurse Practitioner had very large pre-existing caseloads (110–610) and, not surprisingly, reported spending a relatively small proportion of their time on case management.
In total, 51 older people receiving NCMs services were recruited across the three study sites (21 male and 30 female), ranging in age between 64 and 98 years. Only a minority, 10, received social care funded by their local authority, typically from a home care worker. The majority of participants were homeowners (n = 33; 65%), whereas just under a third rented their accommodation (n = 15; 29%). The characteristics of the sample differed slightly but not significantly by site, but were broadly typical in terms of ethnicity and socioeconomic status of the demographic profile of the area where they lived. Of 19 family carers identified by the older people, 11 agreed to be interviewed.
Thirty stakeholders were interviewed across the three study sites, including four from voluntary organisations, 12 GPs, two hospital specialists, five managers from community nursing services, two managers from social services and five health and social care commissioners. The views of the GPs have been reported elsewhere (Iliffe et al., Reference Iliffe, Drennan, Manthorpe, Gage, Davies, Masey, Scott, Brearley and Goodman2011); this paper supplements, where relevant, accounts of the experiences of older people with the views of participants from the voluntary sector, the community nursing service and social service managers and of commissioners.
The tracking of older people receiving attention from NCMs allowed detailed case histories to be acquired; we present four of these here as illustrations of what NCMs actually do and focus on the interface with social services (when this was present). We then describe the perceptions that NCMs had of their work, particularly its ‘social’ dimensions. Finally, we set these findings in the context of the stakeholder interviews about the place of nurse case management in the wider provision of health and social care.
Older people's experiences of nurse case management
We illustrate the interfaces of case management and social care services with four vignettes taken from the case study sites (all vignettes are anonymised). In three of the vignettes, the essence of the interviews is crystallised for brevity. However, in the case of Mrs Rose, where there were changes over the period of four months, we provide some chronology of events.
Vignette 1
Vignette 2
Vignette 3
Vignette 4
These four examples illustrate that some older people and their carers were long-term users of health services and accessed multiple forms of local authority services, not social care services alone. Some had been disabled for long periods (Mr Green for 20 years), and were accustomed to various forms of professional attention and resources. This experience informed their judgments about the role of NCMs. In essence, they valued nurse case management for:
-
a) The NCM's clinical expertise (eg, Mrs Green and Mr Grey's daughter),
-
b) The NCM's assistance in enabling continuity of care and acting as intermediary with multiple services, such as social care (eg, Mrs Rose),
-
c) The therapeutic effect of the NCM's psychosocial support (eg, Mrs Green and Mr Grey's daughter).
For older people and carers, some of the commonly reported outcomes of nurse case management were increased confidence in managing their conditions, arising from the NCMs’ advice on how to cope with them better, feeling that their priorities were generally addressed, giving them and their carer time and relating to a professional who knew their ‘story’. The desirability for this was confirmed by members of the older people's group who formed part of the stakeholder discussions; they commented that a NCM could be very helpful because they could ‘take on’ social services matters:
The concept is marvellous. I'm not clear how extensive this case management is though? How much can they do for people? I know one of our members has a nurse who seems to look after social services and all kinds of things like that. That surprised me, the social services thing, I didn't know nurses could do that, but it makes sense. (Participant in older people's group)
Case managers’ relationships with social services
At the centre of the nurses’ narratives was their dual role of providing direct (face-to-face) care and ensuring that older people received services to enhance their health and wellbeing. They distinguished between ‘hands on’ work and case management work. This had implications on how nurses interpreted the priorities and focus of their activities and who else they involved in their patients’ care. Liaison and advocacy were sometimes described as ‘not nursing’ but, nevertheless, an essential part of the case manager role. One case manager (names are anonymised), working under the title community matron, said:
I see my role as mainly ensuring that patients get the care they need whether it be from the district nursing team, specialists, (local authority) adult services or other services such as charities, palliative care services or Macmillan (cancer specialist) nurses… A lot of what I do is overseeing the care of patients, I liaise a lot with GPs, sometimes I feel like it's more of an advocacy role than pure nursing. (Karen – community matron)
There were several examples of where a case manager would actively seek greater support, such as additional social care, for a carer and the older person:
He's (older person) had a bad time and his wife was struggling to cope. Case managing him means I can make sure she gets the help she needs too. (Linda – NCM)
One NCM had taken up the role because her previous work as a District Nurse team leader had removed her from patient contact. In contrast, another nurse was rejecting the more extensive community matron form of case management because it did not conform to her understanding of nursing:
I didn't like the work as a community matron – it was too much like being a social worker. I prefer to have a clinical role. (Myra – Clinical nurse specialist)
Although there was a potential tension between providing ‘hands on’ care and acting on behalf of the older person, most of the nurses recognised that acting as a case manager meant coordinating and managing services, as well as providing them:
For me it's about making sure that each patient, treated as an individual, gets what they need in the right way for them. (Noreen – community matron)
This was not always easy; one NCM described how it had taken three days to organise a change in a person's medication. This had involved making sure that the older person understood the new regime, liaising with social care workers and the District Nurse visiting the person regularly to provide home care and treatment, respectively, and finding a pharmacist who would set up the right medication pack.
Nonetheless, despite the case managers’ best efforts, some older people declined their recommendations. Home care was the service that was most frequently declined; nine out of 51 patients rejected this potential help, mainly on the grounds that they did not want to lose their independence. Such a means-tested service would also have incurred financial charges for many if taken up from the local authority, or older people or their families would have had to make arrangements and pay for home care privately.
Extension to nursing roles
Although there were examples of deliberate crossing of boundaries, some nurses also talked of potential role enlargement by undertaking work covering social care needs:
We have a standard assessment tool to use for new patients. This looks at all the issues the patient may have, including the health and social needs. (Oonagh – community matron)
In the early part of this study, there were few examples of joint assessments or visits to older people with other professionals, mostly from the NHS. None of the NCMs, over the nine months of the case study research, used an integrated system of sharing documentation with social care services. However, by the end of the nine months, because their role and work were becoming better known among health, social care and third sector organisations, case managers spoke of growing confidence in how they worked with others and how this fostered their willingness to involve other services.
Stakeholder perspectives
Although older people receiving attention from NCMs, and the nurses themselves, were able to describe the social dimensions of the care given to individuals, the stakeholders had little to say about the role of nurse case management in the wider provision of services. Their contributions suggested that the local NHS and social care systems in all three case study sites were fragmented, with each part only seeing the potential of case management from its own perspective, and not from that of the patient, or from the whole care system.
For example, social services interviewees were uncertain as to what the term ‘case management by nurses’ really meant, particularly in relation to ‘care management by social workers’. They wanted to know when a service user's ‘case worker’ should be a social worker and when it should be a nurse. They could envisage the potential benefits of nurses as case managers in situations where a person had complex health needs or when social care staff did not possess (or were thought by others not to possess) necessary clinical knowledge. Not surprisingly, they argued for good communication between health care and social care professionals supporting the same older person, irrespective of job titles or roles. They cited multi-disciplinary teams, such as community rehabilitation teams and intermediate care teams, as good models for supporting people to remain in their own homes. They were clear that the role of NCMs had not been devised on such a joint basis. The commissioners were concerned with effectiveness in meeting the needs of people with long-term conditions and were not particularly wedded to the notion that nurses should perform case management. One said of community matrons (the local form of NCMs):
It is not likely that the community matron service will be increased and we are worried that as they leave, for whatever reason, they many not be replaced – case management is seen as low priority because it caters for so few people at such high cost.
Discussion
What this study shows
NCMs may identify unmet social needs among their patients and seek help for them from local authority social services departments, potentially supplementing the work of social services and increasing social workers’ workload and making demands on budgets. However, they may also be able to carry out tasks (particularly providing psychosocial support), which avert demands for social care, thus substituting for social workers. Furthermore, they may have a preventive role in reducing the risks of an older person's emergency admission to hospital, thereby reducing the number of referrals to social services from hospitals. Because some models of nurse case management were partly chosen as a solution to the problem of unplanned hospital admission, and did not evolve from a joint understanding of need across the health-social care boundary, social workers are not able to estimate the contribution of substitution, supplementation and prevention from nurse case management. As collaborative working grows, this estimation may become easier, but uncertainties about the value of nurse case management among managers and commissioners may jeopardise its future.
Strengths and limitations of the study
Although the sample sizes were small, the use of different data sources allowed us to compare the needs and experiences of the older people and their carers. The all-encompassing feature of a case study is its focus on a single phenomenon within its real-life context. Explanatory case studies can suggest important clues to causal relationships, but not with the certainty of true experiments (Yin, Reference Yin2003). We accept that case-to-case transfer (Firestone, Reference Firestone1993), transferability (Lincoln and Guba, Reference Lincoln and Guba1985) and usability (Swanborn, Reference Swanborn1996) of findings from case studies are dependent on the readers’ judgments, not ours.
Further resources would have enabled greater numbers of interviews with stakeholders and with practitioners working in areas such as supported housing and in home care services. Nonetheless, the study acquired a rich set of data, especially from older people who were often in poor health.
How these findings fit in
The policy aspiration that NCMs would be able to overcome the fragmentation of care for people with complex disabilities and illnesses appears to have been achieved in part but for very few people, with the financial cost borne by the NHS (Gage et al., under review). However, the promotion of NCM roles had none of the hallmarks of integration so often espoused by policy makers, which stress the potential for coordination across health and social care (Wagner, Reference Wagner1998). Similarly, it did not appear to have built on the many examples of integrated and co-located social work and nursing services (see Kharicha et al., Reference Kharicha, Levin, Iliffe and Davey2004).
NCMs may see themselves as working with people with high levels of need as defined by NHS criteria rather than wider definitions of health and wellbeing (McLeod et al., Reference McLeod, Bywaters, Tanner and Hirsch2008). Many of the 51 older people and 11 carers in this study were in the position of being at high levels of need (or potential demand); however, they were not using social services because they had strong support from others, notably their families, and also wider social and service networks, such as neighbours or housing staff. This is illustrated in the examples of Mr Green and his daughter and in the case of Mr Grey and his wife. Mrs Rose, receiving social care services of many types, still perceived herself wanting a more regular and supportive relationship with the NCM.
It was interesting that older people saw NCMs bridging clinical and social services worlds, rather than simply a means of avoiding unnecessary admissions to hospitals, which is often a key criterion of success for case management (Reilly et al., Reference Reilly, Abell, Brand, Hughes, Berzins and Challis2011). Social care sometimes appears to be only conceived of as help with personal care and risky situations. As Bornat and Bytheway (Reference Bornat and Bytheway2010) observed, engagement with care managers is: ‘a comparatively rare, and possibly brief, experience for older people, largely limited to those previously living in what have been deemed particularly risky circumstances and who are now receiving care in what are intended to be comparatively secure environments’ (p. 1132). This is not all that is needed by older people. Moreover, social care is not always regarded as socially acceptable, may not be ‘worth the money’ and may jeopardise informal care.
At organisational level there are suggestions that individual NCMs had to establish their position and embed their work where it was most needed or it was feasible to do so, because the role rapidly developed following central NHS prompting rather than local negotiation and needs assessments. Joint planning and joint working may be tested when one party creates new roles on its own. For local authority commissioners and providers of social care, NCMs may offer preventive support, may be well-informed and expert advocates and may share their clinical skills with social care staff. There were examples of this in the case studies reported above. However, other models of support to the same groups of older people (recently discharged from hospital) are also compelling. McLeod et al. (Reference McLeod, Bywaters, Tanner and Hirsch2008: 87), for example, found that volunteers undertaking post-discharge support undertook ‘advocacy, education and interpersonal contact attuned to addressing psychological barriers to greater social integration’.
These views therefore partially chime with those found in Challis et al.'s (Reference Challis, Hughes, Berzins, Reilly, Abell and Stewart2010) study that explored general case and case management over the same time period as this study and which noted ‘little evidence of integration between the two services’ (p. 10):
being a nurse-led service (case management), links within the primary care sector were strong, reflecting its origins. On the other hand, there were poorer links with both other health services and social care although examples of pilot initiatives to promote joint working were found (Challis et al., Reference Challis, Hughes, Berzins, Reilly, Abell and Stewart2010: 201).
They also found that a ‘hallmark of joint working’, the use of a single case file shared between all professionals, was rare and that methods of assessment varied so much that attempts at targeting were inconsistent (Abell et al., Reference Abell, Hughes, Reilly, Berzins and Challis2010). Our study did not locate any examples of shared assessment or records. These may be barometers by which joint working and services designed to enhance existing relationships can be judged.
Implications for practice
The implications of nurse case management roles are fivefold. First, the role of NCMs is highly variable locally, and professionals working in local agencies will need to be familiar with these new colleagues’ remits, possibly shaping them where possible. Second, NCMs who make referrals to social services of people needing support may possess good-quality information that can inform social workers’ assessments. A modus vivendi of information exchange will doubtlessly emerge, but this may need to be refined into protocols. Third, nurses and social workers arranging hospital discharge should ensure bridges are constructed with case managers, because the latter may be able to draw on other NHS support at this vulnerable time. Fourth, developing shared recording systems around this group may promote communication, especially around review and monitoring by NCMs. Finally, NCMs provided great support to some carers who may fear that professional intervention might lead to separation or long-term care for their relative (see Sullivan, Reference Sullivan2009). NCMs may further be able to explain new systems of personalised social care (see Carr, Reference Carr2010) to carers who may perceive that calling on social services constitutes an admission of failure.
Conclusion
This study found that older people value what NCMs do, especially the bridging of health and social care. Although NCMs struggle with their new role, particularly in social domains, they nonetheless are able to supplement or substitute for social services in different contexts, and even have the potential to prevent older people from needing social services support by preventing emergency hospital admissions. Their potential may not be fully realised if, as the stakeholders from local health and social care economies in this study reported, the NCM role came from outside and was not grown organically within their services (Iliffe et al., Reference Iliffe, Drennan, Manthorpe, Gage, Davies, Masey, Scott, Brearley and Goodman2011).
NCMs attempted to integrate health and social care from within the NHS and they provided a form of support and care for older people who otherwise were often not receiving attention from local authority services. In practice, the role's implementation and its activities became highly variable, leading to the risk of confusion, duplication and gaps between agencies and professionals. Nonetheless, there are important lessons for other professionals in considering the value of case management and the ways in which this role may be revisited in more personalised social care and health care systems.
Acknowledgements and disclaimer
We are grateful to all study participants and to the members of the study's advisory group. This study was funded by the National Institute for Health Research Service Delivery and Organisation programme (NIHR SDO; reference 08/1605/122), but the views expressed in this article are those of the authors alone and should not be interpreted as those of the NIHR, the NHS or the participating sites.
