Abstract
Objectives
Whole system integration of health and social care has been positioned as key to improving care, increasing efficiency and controlling costs. However, evidence for the benefits of whole system integration is scarce. Drawing on organizational theory, this study uses the implementation of remote care services, viewed as an enabler for whole system working, to explore the reality of achieving this policy objective.
Methods
Qualitative, longitudinal data were collected across nine UK sites adopting remote care over three years. Three sites formed the Department of Health's Whole Systems Demonstrator (WSD) programme for remote care. In addition, the implementation of remote care was explored in six other sites unconstrained by the randomized control trial procedures of the WSD programme. The methods were ethnographic (including 235 hours of observations and 184 interviews). Participants were health and social care staff and Government policy makers.
Results
Remote care did not lead to system redesign; however, local ‘ownership’ of new services did lead to more collaborative practices across the care system. Lack of integration was an enduring and endemic challenge across all sites, relating to differences in statutory responsibilities, absence of shared budgets and hybrid organizational roles, differences in work practices and organizational philosophies, and ambiguity around what ‘whole system working’ actually entailed.
Conclusions
Policy initiatives like the WSD programme provide opportunities to phase in collaborative practices and create an awareness of the need for joint working. However, the progress observed suggests that the concept of whole system redesign around remote care is currently unrealistic.
Background
Health systems worldwide are entering a period of unprecedented challenge because of the ever-increasing numbers of people with chronic, complex health needs. 1 This challenge is exacerbated by fragmentation, duplication and poor coordination across care services, which may compromise patient safety, system efficiency and health outcomes. 2 In coping with escalating demand, UK policy has advocated that the design of care services should be underpinned by a ‘whole system’ approach. 3 There are a number of proclaimed benefits to ‘joining up’ individual parts of the system, such as sharing the costs of investment through pooled budgets, reduction of service duplication and patient level benefits via a holistic approach.4,5
Despite recognition of system benefits and risks that can be leveraged through inter-agency collaboration, in practice the uptake of a whole system approach has proved elusive and challenging,6,7 not least because of the number of distinct and often competing policy agendas. Health care is a contested territory in a state of flux, with care providers under pressure to achieve and adhere to distinct performance targets whilst self-organizing to ensure compatibility with the ever-changing policy environment. 8 Politicians and policy makers are trying to introduce market style reforms (i.e. audit and performance management regimes), drive down costs and increase competition at the same time as moving towards a holistic approach, by strengthening collaboration. 9 In the UK, a tension is emerging for instance between the desire to achieve cross-system collaboration and current systems of payments for health care providers and other policy drivers such as personalization of care, integrated care and adherence to performance targets.8,10–12
Our focus in this article is on the introduction of a specific policy directive, the UK Whole System Demonstrator (WSD) programme. This was set up to explore the concept of ‘whole system working’ using remote care (telehealth and telecare) technologies as a driver for change.
Barriers to a whole system approach
One of the barriers to ‘whole system’ working is identifying what it entails in practice. This remains undefined, with currently no single or shared operational definition. 5 Terms such as ‘partnership working’, ‘joint working’, ‘seamless services’ and ‘integrated care’ are all associated with this concept and have populated policy documents and managerial discourse. However, these also remain undefined.13,14 The only tenuous common thread is the perennial quest for care coordination and service integration around patient care. 7 This ambiguity means the evidence base for the benefits of whole system working is equally sparse and ambiguous. Hunter and Perkins 15 note the same gap between evidence and practice in relation to partnership working. They call for greater emphasis on research that investigates the magnitude of change that these different patterns of governance can achieve, in reducing inequalities and improving public health outcomes.
This evidence deficit is somewhat paradoxical if one considers that the UK government has advocated the benefits of a whole system approach in the delivery of public health, and committed itself to producing evidence-based policy and a strong ‘audit culture’– placing emphasis on ‘what works’. 15 Within this orderly framework, evidence precedes policy and drives best practice. With the focus primarily on control and prediction, methods of data collection and evaluation frequently adopt a highly positivistic and often reductionist approach (i.e. randomized control trials (RCTs)) unappreciative of the unexpected, possibility of experimentation (trial and error), and emergent order in parts of the care system. Evidence-based policy combined with a reductionist approach may not be the best approach for evaluating complex innovations and ideas within health and social care. 16 This is especially so for those innovations spanning boundaries between different components of the system. Implementation of new ways of managing care and organizing services under such conditions needs to at least be guided by intuition, creativity and experimentation. 8
In the UK and elsewhere, the way health and social care services are formally structured, funded and managed is one of the main challenges to the successful delivery and sustainability of integration. ‘Silo thinking’ around budget allocations, performance targets and professional self-interests have the power to slow down integration in favour of concentrating on individual parts of the health and care system.17,18 Achieving integration requires a receptive environment to collaborative practice such as integrated IT systems, data sharing and the use of statutory standards within and across organizations (i.e. pooled budgets, whole system commissioning, joint managerial roles and single employment frameworks). There is also limited knowledge regarding how diverse groups in complex multiprofessional settings make sense of the concept of ‘whole system working’, 11 whether they are willing and ready to embrace this idea, and how this impacts on any implementation initiatives. Previous research indicates that the idea of making decisions in a ‘joined up’ manner may be viewed as threatening. 10
Managing long-term conditions using a whole system approach
In 2006, the White Paper invited: ‘all health partners to work together in a system-wide approach to developing urgent care services including better care for patients with long-term conditions, shifting care from acute hospitals to the community, promoting better public health, integration with social care and improving access to GPs in hours’. 3
This policy set out the broad objectives to deliver new models of integrated care that could move health care away from hospitals and high-cost settings, to community-based patient-centred services that could work equally well across the different agencies involved. Within this directive, remote care technology was widely seen as a solution to the needs of the ageing chronically ill, and viewed as a vehicle stimulating whole system redesign.3,4,19
Remote care can be described as a complex innovation as its implementation cuts across government and local policy making and crosses a number of boundaries: organizational, professional and technological. Telehealth and telecare shift the location of care from hospital and residential home settings to a patient's own home to promote self-management with the support of ICT-enabled systems. In addition, this technology shifts work practices across different professions, with for example community matrons taking workload from GPs in relation to daily patient monitoring. This shift requires bringing together stakeholders from different parts of the care system. However, the split between telecare and telehealth technologies reflects current fragmentation in the UK care system around ‘intended beneficiaries’ in terms of service receivers and service providers. 20
Telecare (focusing on risk management of elderly people in their homes) is largely provided by UK local authorities’ social care departments whereas telehealth (focusing on monitoring people's vital signs) is largely provided by the National Health Service (NHS). The implementation of remote care is influenced by the complexities of health and social care provision, including financial characteristics and their potential inconsistencies (e.g. remote care may reduce hospital admissions but this is detrimental to their income). 21 Incentives for collaborative working are limited in the absence of robust evidence on the system-wide impact of remote care on costs, improvement of quality and effectiveness of care.
In 2007, the UK Department of Health introduced the call proposal for the WSD programme in order to demonstrate the extent to which remote technology could ‘glue’ together parts of the care system to support people with longer term, complex health and social care needs. 22 It has been argued that attempts to put a whole system approach into practice and (re)engineer integration across health and social care services can be achieved through multicomponent policy change programmes. 11
The WSD programme
The WSD programme is believed to be the largest multisite evaluation of the implementation, impact and acceptability of telehealth and telecare technologies. The study involved 6191 patients and 238 general practitioner practices. The programme was officially launched in 2008 and completed in 2011. 23 On the conclusion of programme, the government recently called for the expansion of remote care technology to benefit another 3 million people. 24 We were part of the evaluation team following the progress of the three WSD programme trial sites over the duration of the programme, focusing on the organizational aspects of delivery.
The programme's aim was to use remote care as a driver for ‘wholescale redesign of services for those with long-term conditions’, 22 supporting the development of data sharing systems, interlinked care processes and integrated care teams for those health and social care service users. In the original protocol, three types of participants were to be recruited: those eligible for telecare, those eligible for telehealth (i.e. technology to assist in the management of COPD, diabetes, heart failure), and a mixed care need group, eligible for both types of services. The inclusion of this latter group offered a particularly interesting case for redesign, by combining telecare and telehealth services across the care system.
Another aim was to produce ‘credible evidence that comprehensive integrated care approaches combined with the use of advanced assistive technologies, do both benefit individuals and deliver gains in cost effectiveness of care’. 22 In order to gather robust evidence, the three WSD programme sites were subjected to a large pragmatic cluster-randomized controlled trial. 17
The aims of our part of the overall evaluation were to examine whether remote care could stimulate whole system change and more broadly if remote care implementation changed the current forms of service delivery. In doing this, we employed an inductive qualitative approach to explore the meanings of whole system working in policy and practice settings. We employed a sensemaking perspective 25 to gain insights into how whole system working concept was perceived and enacted by a range of stakeholders (policy makers, health and social care staff). Theories of sensemaking have long been used in management literature to understand responses to planned or unplanned organizational change outcomes. 26 Sensemaking is a narrative process that introduces a sense of order and stability following change, by creating shared understandings around often only vaguely defined strategic or policy agendas. 27 Our aim was to draw on sensemaking theory to understand how interconnections between policy, organizational environments, context (e.g. history of remote care at each site, RCT) and the innovation (remote care) were understood across multiple levels of the system.
Methods
Our research design was based on an inductive, qualitative analysis of longitudinal case study data in Primary Care Trusts (PCTs) and Local Authorities implementing remote care between June 2008 and July 2011. Three of our total of nine cases formed the UK Department of Health's WSD programme.
After nine months of the WSD programme, the Department of Health acknowledged that the requirements of the RCT may hinder the wider generalizability of implementation lessons across the three WSD programme sites and we were commissioned to expand the study across six additional UK sites, outside the trial. We have called these six sites ‘non-WSD’ programme sites. Both the WSD and non-WSD programme sites were all seen as ‘front-runners’ in the implementation of remote care, in the sense that their previous experience meant they stood a reasonable chance of scaling-up the innovation.
To capture how the implementation unfolded, we drew on a wide range of data sources and materials and carried interviews at different time points. Each site had a dedicated management team charged with rolling out the remote care initiative, recruiting staff and reviewing progress. Teams also facilitated and co-ordinated installation of the technology, recruitment, assessment and monitoring of patients/clients within or outside the trial context. Across the three WSD programme sites we carried out 115 semi-structured interviews (174 hours of observations and reviewed 92 strategic documents) with health and social care staff within the WSD programme management teams and policy advisors. In the six non-WSD programme sites we carried out 69 interviews with key local health and social stakeholders, 60 hours of observations and reviewed 37 strategic documents. We explored the impact of remote care agenda on local implementation practices and its success in stimulating whole system redesign. We purposively interviewed everyone involved in the strategic management and/or operational side of implementation. It should be noted that the teams in the ‘non-WSD’ programme sites were smaller in size and the data collection was only twice a year. There was much less activity in terms of remote care implementation in these sites, hence the need for continuous data collection was not deemed necessary.
For analysis, we drew on theories of innovation adoption 28 and sensemaking. 27 Our ethnographic data collection and analysis involved an iterative development following the grounded theory approach. 29 Our data were analysed in three stages, developing comprehensive coding and theme abstraction that explored implementation processes at multiple organizational and individual levels. To ensure reliability, two members of our team independently read the interview transcripts to agree on the emerging themes.
Before data collection began, the study was approved by Liverpool Research Ethics Committee (ref: 08/H1005/4). The data presented is part of a larger body of work currently being written up.
Results
Misalignment between vision and enactment
In the WSD programme, the UK Department of Health envisaged whole system change and described this approach as ‘truly integrated services’ and a ‘radical and sustained shift in the way in which services are delivered’.
22
Implicit was the virtue of this new way of working and the Government's determination to direct care organizations to move away from the old forms of organizing. The new policy required assembling multiple parts of the care system to create a ‘Gestalt’ whole. Remote care technology and the services around it were viewed as a key enabler: The idea was that WSD was meant to be a demonstration of how the health and social care could work together as a system, as a whole system, facilitated by all the technology. The vision was that the technology would bring … people’s experience of separate services, together through this (Policy advisor).
System change takes a long time; it’s a resource that needs funding. The vast majority of our funding went on providing services to these people, and you don’t have the drivers that fundamentally changed the way whole services originally were played. It’s hard to do that without information. You need to do that with a top-down strategy … We thought it might happen at the microcosm within the project, but in reality we need it to happen across the board. It has to happen at a strategic level (Policy advisor).
In WSD we delivered what we said we would deliver; the evidence of whether the principle of remote care is one that we should pursue further. The thing that I don’t think we did do was, from the whole systems point of view, fundamentally change the way people work. I don’t think what that has led to is system change (Policy advisor).
It may be an optimistic misnomer to say that the WSD was going … to invoke a whole systems change. It had to be a bit of a walled garden, to preserve the integrity of the research, it had to maintain itself as a walled garden, so how is it going to influence the whole system? It can’t. I think that was optimistic. (WSD manager)
There was certainly a period where we were rather locked into WSD being a discrete as opposed to being an integrated whole system approach – an add-on service that was influenced quite strongly by the need to not distort the evaluation process. So, we carried on doing some things that we knew were either expensive or just under par … (WSD manager)
I think again the principle of whole system working was largely geared around as having people in the design of the trial that would benefit from both telehealth and telecare, and coordinated intervention. As it turned out we found very few of those, because of our eligibility criteria probably. So because we dropped that on the trial, the emphasis on whole system working was reduced (WSD manager).
It always was the case that clinicians needed to be convinced, but now the clinicians will have the budgets, or be leading on commissioning, I think it’s an added benefit of having a randomised control trial (WSD manager).
Don’t say you’re going to do a whole system demonstrator and then do a randomised control trial instead or if you’re going to, recognise what the difference is going to be. It cannot do both of those two things. It has to do one or the other. In the WSD we specifically could not use the lessons learned in order to change the system because to change the system would have invalidated the RCT (WSD manager).
Wider barriers
Our analysis highlighted other inherent barriers to realizing the whole system redesign such as a lack of alignment in funding models, and a lack of professional receptivity and readiness. However, the majority of the issues mentioned by the WSD programme sites as reasons for explaining the problems in implementing remote care were also apparent in the non-WSD programme sites. Differences in funding models and budget constraints were considered across all sites as a chief inhibitor. Lack of alignment in the ways in which care is paid for at the point of delivery has always been problematic, with health care free at the point of delivery whilst social care is means-tested. As one health care professional mentioned ‘as nurses we can refer for a personal falls alarm but we can't refer directly for any of the other telecare peripheries. That needs to go through social care … then I have to ask for a care manager to go out and make that assessment’.
In addition, the way in which the costs and benefits of remote care are spread across health and social care providers raised concerns about a potentially increased burden in the primary and social care sectors due to a reduced rate of hospital admissions. The question of ‘which organization pays’ for the service and ‘which benefits’ was seen as a challenge to the mainstream implementation of remote care beyond the trial phase. Integrated budgets were seen as a possible solution, but integrating different parts of the system in the absence of a formalized overarching whole system structure was seen as highly challenging. Systemic barriers needed to be addressed for any attempt to succeed.
Telecare is often a paid-for service and telehealth can’t be because it’s on the NHS. So, tell me how that’s going to work if there was integrated telecare and telehealth services. It would be quite funny trying to even design it. There are a lot of problems that are government level problems and system level problems. There is not a whole system that overarches, from the financial management to the performance point of view (WSD manager).
And there have been lots of efforts to enable health and social care to work together, pooled budgets, and various legal mechanisms to do that. But I suspect that they’re different professional tribes, different managerial, different practice sort of tribes out there. They are not trained together, they don’t use the same language and have a whole picture necessarily when they have to meet (non-WSD).
If you are working in highly complex organisational structures, we do not default to we must get the whole system working together. There are some circumstances where it is not possible to do that, and therefore you have to refine it down, make it much simpler and work with those players who actually want to work together (non-WSD).
As a whole, it’s quite difficult at the moment, because we’re going through a reorganisation, and my role will change in April, so it is difficult at the moment. We are going to be a separate organisation to the PCT, so we will have our own name, our own branding from April. We will have a new structure. The titles will all be changing (WSD manager).
Whole system working: Ambiguity and diversity
Participants across all nine sites were asked about their understanding and experiences of whole system working. The results indicated a diverse range of perspectives regarding the meanings associated with this term and its enactment, reflecting local strategic decisions, managerial interpretations and the interests of specific groups. For policy makers and some practitioners, whole system working was synonymous with notions of integrated care and care coordination:
It would mean that there would be an integrated service that would provide a holistic support for an individual and not start slicing them in different ways for different purposes.
And what whole system hopefully will do is get everybody to talk to one another in the same language, because it gets confusing for the patients, because they’ve got so many people saying so many different things. The classic is client and patient … we need to have a vocabulary for all … one dictionary if you like.
I think there’s a lot about sort of bringing organisations together at the high level. It is an organisational transformation unless health and social care become one organisation with the same systems and financial budgets it is going to be very difficult.
Despite our analysis revealing that the whole system vision was largely not enacted, there were pockets of increased collaboration around remote care services. Remote care was successful in breaking down boundaries by (a) increasing awareness of the interdependencies between and within the different agents of the system and (b) identifying duplications and gaps in services: I think it certainly has achieved quite a lot of boundary breaking-down, and now it’s less … it’s not so much of a taboo working with social care (health care professional-WSD).
I will talk to somebody of a similar level in another organisation and they cascade the information upwards and downwards. ... the whole system for me in that context has meant I talk to operational staff, management staff, their management, my management, policy staff, I talk to people in other organisations (health care professional-WSD).
When I first started, in 2006, there were very few senior officials or ministers who knew what telehealth or telecare was. Now all of them speak about it regularly. Every policy document in some way references it so I think the programme had a significant influence just in terms of raising awareness (Policy advisor).
Discussion
A ‘whole system’ approach has become a popular policy aspiration, a management imperative and a new service delivery ethos, in health care. We set out to explore the practicality of this approach by examining one of the few implementation initiatives which specifically set out to empirically test this intuitively attractive idea, the Whole System Demonstrators programme. 22 We explored how stakeholders across the health and social care landscape, involved in the deployment of remote care technology, made sense of the whole system concept. Our findings across the nine sites suggest that ‘whole system’ approaches lack operational clarity, with no shared definition of what this idea means in practice, and ambiguity around how this vision might be achieved. Investigating policy makers’ and practitioners’ perceptions of the success of whole system working delivery in their localities, our results show that a whole system approach has not been realized. Nor was whole system working a neutral concept with a shared desirable outcome. Instead it is a concept open to variation, negotiation and multiple interpretations.
The way different stakeholders made sense of a whole system approach revealed endemic barriers to change their respective needs, anticipated gains and losses. 10 For many, whole system working was an ‘integrated care’ model, with an end-state placing emphasis on the holistic needs of patients and a ‘seamless’ service experience; not a statutory transformational change process focused on wider system benefits. Remote care did make inroads in achieving increased collaborative working. Organizational collaboration was enhanced and issues of ‘professional tribalism’ were temporarily suspended by the need to plan and implement the WSD programme and the other local remote care implementation initiatives. However, operational changes put in motion during these initiatives tended to fade over time. 31 Perhaps what is important is what the system and individuals learn about the opportunities and challenges of working together. Remote care technology alone was insufficient to be a catalyst for such large-scale transformational change. It needed to be accompanied by senior management support, new financial structures, integrated IT systems, joint work practices, and new hybrid roles.4,13 In the WSD programme sites these supports were abandoned in the early stages of the programme, with the focus changing to evaluating benefits evidence of the technology through the RCT. We also observed how structures (e.g. joint leadership) that were put in place to support integration activity in the non-WSD programme sites also failed to endure beyond the pilot phase of telehealth projects, in part due to a lack of convincing evidence for the benefits.
In the WSD programme sites, evidence of remote care effectiveness has been successfully gathered. 30 However, this evidence is for telehealth and telecare in isolation from each other. Hence, the opportunity to see how a combined integrated health service model might create much larger benefits and break down the ‘frontier mentality’ 13 has been lost. This separation, and a differentiated disease-focused approach, 19 both in WSD programme and in the non-WSD programme sites, has reinforced old silos and served to create new ones. Efforts to integrate the services became unattainable as priorities shifted, with most boundary spanning roles being replaced by separate management and operational roles within telecare and telehealth.
For the majority of our respondents the WSD programme was ‘an interesting experiment’ into doing something ‘very challenging’, combining remote care service development with an RCT service evaluation. The appropriateness of the RCT as a method for evaluating the implementation of remote care services and ‘resolving’ issues around whole system redesign was questioned by all sites. To ensure the trial established causal relationships – hence robust evidence – the approach to implementation had to be kept unchanged and arguably detached from the organizational reality of changes in context. The WSD programme evaluation and trial protocol, with its aim to control for the impact of context, thus inhibited systemic learning and responsiveness, highlighting the challenges in developing evidence-based policy for highly complex innovations. This echoes the views of other commentators who argue that health technology service evaluations are not suitable for controlled study designs especially when the innovation is evaluated whilst implemented. 32
If a whole system approach to health care is deemed necessary, a blueprint for how this might be achieved in the future (care pathways, management structures, financial agreements, information exchange systems) is needed, with a policy framework that provides clear definitions and supportive processes. Policy levers need to be put in place to incentivize sometimes competing highly specialized care providers to collaborate. However, this may be constrained not only by lack of evidence but also by conflicting perspectives on what ‘evidence’ in this domain might comprise. Currently, there are no metrics to evaluate the success of a whole system approach, or performance outcomes against which organizations can jointly measure progress. 11
With many participants expressing deep-seated differences in the extent to which they are ready to work together, it is evident that whole system transformation needs to be owned and led by local organizations of care to succeed. We also need local and national policy congruence and intent if progress is going to be made. Sitting back and hoping that through the implementation of new technology this type of disruptive and transformational change will organically and magically appear is not a viable option. Rather than positioning remote care as a vehicle to deliver whole scale redesign, perhaps a more pragmatic message is that plans to implement remote care could represent a useful stimulus for ‘diagnosing’ the challenges around implementing whole system change.
Footnotes
Acknowledgements
We thank all those who participated in the study. We also thank all the members of the Whole System Demonstrator Research Team for their helpful comments and contribution in this study. The team includes Martin Knapp, Catherine Henderson, Martin Bardsley, Adam Steventon, John Billings, Jennifer Dixon, Hellen Doll, Martin Cartwright, Shashivadan P Hirani, Anne Rogers, Caroline Sanders, Peter Bower, Raymond Fitzpatrick, Virginia MacNeill, Jennifer Beecham, Jose-Luis Fernandez, Andrew Bowen, Michelle Beynon, Lorna Rixon, Luis A. Silva and Stanton Newman (Principal Investigator for the Whole System Evaluation Team).
Funding
The study was funded by the Department of Health as an independent evaluation of the Whole System Demonstrator programme. The views expressed are those of the authors and do not necessarily reflect those of the Department of Health. The study was approved by Liverpool Research Ethics Committee (ref: 08/H1005/4).
Conflict of interest
All authors declare that they have no competing interests.