Changing health care with,for,or against the public: an empirical investigation into the place of the public in health service reconfiguration

Abstract

Objectives

This study sought to understand the different approaches taken to involving the public in service reconfiguration in the four United Kingdom health systems.

Methods

This was a multi-method study involving policy document analysis and qualitative semi-structured interviews in England, Northern Ireland, Scotland and Wales.

Results

Despite the diversity of local situations, interview participants tended to use three frames within which they understood the politics of service reconfigurations: an adversarial approach which assumed conflict over scarce resources (change against the public); a communications approach which defined the problem as educating the public on the desirability of change (change for the public); and a collaborative approach which attempted to integrate the public early into discussions about the shape and nature of desirable services (change with the public). These three framings involved different levels of managerial time, energy, and resources and called on different skill sets, most notably marketing and communications for the communications approach and community engagement for the collaborative approach.

Conclusions

We argue that these framings of public involvement co-exist within organisations. Health system leaders, in framing service reconfiguration as adversarial, communicative or collaborative, are deciding between conceptions of the relationship between health care organisations and their publics in ways that shape the nature of the debates that follow. Understanding the reasons why organisations adopt these frames would be a fruitful way to advance both theory and practice.

Introduction

Transforming health services is widely seen as a difficult political exercise. ¹ One source of difficulty is in persuading patients and the wider public that service changes are improvements, not ‘cuts’ or reductions. This challenge has been documented within the context of the four countries of the United Kingdom (UK) in particular, ² but also reported elsewhere, such as Ireland ³ and Canada. ¹ ‘The public’ has been widely identified as a problem for making significant changes to health care. ⁴ In response, studies have suggested involvement practices which might allow change to proceed, including: involving a wide range of stakeholders, early and prolonged engagement, and emphasizing clinical rationales for change processes.^5,6 These recommendations are in line with a wider view of public involvement as a coherent intervention with shared meaning which can bridge the troubled waters of service reconfiguration. The terminology of ‘involvement’ nevertheless masks a wide and even conflicting range of practices and purposes.

In this paper, we argue that techniques of public involvement reflect how organisations understand themselves and their relationship to the population they serve. We conducted a multi-method study comparing public involvement in service change across the four countries of the UK: England, Northern Ireland, Scotland and Wales. We demonstrate the presence and salience of distinct underlying approaches to public involvement in decision making, which are reflected in distinct and only rarely overlapping academic literatures on change management, public involvement in health care, and democratic decision making.

Methods

The study involved document analysis and semi-structured qualitative interviews at the national and local level in each health system. We began by identifying and analysing written guidance and regulations around involving the public in major service change hosted on publicly available websites for the relevant health departments or government agency in each system. We used the current guidance for our study period, 2016–17. We defined major service change as a range of changes from whole hospital closures to ward closures or reductions in the opening hours of emergency departments. This definition allowed us to acknowledge the subjectivity of what constitutes ‘major’ in different contexts. ⁷

On the basis of this analysis, we then identified key individuals and organisations at the national level to approach for interview. This included people who had authored significant pieces of written guidance and who were responsible for these areas of policy within health departments or key agencies in each of the four UK countries. We also snowballed from initial contacts. Between May 2016 and February 2017, we conducted semi-structured interviews with a total of 26 key individuals working on or around policy for public involvement in major service change. National level interview participants represented a range of government organisations in each country, as well as experts (academics or voluntary sector stakeholders) who had held formal roles advising on policy, and management consultants (Table 1). We were unable to recruit participants from the national Board of Community Health Councils and the National Clinical Forum Wales (both of which we made repeated efforts to contact).

Table 1.

Number of national and local interviews conducted in the four health systems of the UK.

Professional local of interview participants	Number of interview participants	Subtotal	Total
Scotland			9
National interviews		6
Experts with policy roles	2
Officials	4
Local interviews		3
NHS managers	2
Public campaigner	1
England			11
National interviews		7
Experts with policy roles	1
Officials	4
Management consultants	2
Local interviews		4
NHS managers	3
Public campaigner	1
Northern Ireland			12
National interviews		7
Experts with policy roles	1
Officials	6
Local interviews		5
NHS managers	4
Public campaigners	1
Wales			12
National interviews		6
Experts with policy roles	2
Officials	4
Local interviews		6
NHS managers	2
Public campaigner	4
Total			44

Of 26 national interview participants, nine were interviewed by author 2 and 17 by author 1, working with the same topic guide. Eighteen participants were interviewed in person and eight by telephone. At their request, four participants were interviewed in pairs with a colleague. Interviews were audio-recorded except for two cases where the device failed and one where the interview participant did not agree to being recorded; in these cases, we took detailed notes. The topic guide covered the participant’s role in public involvement and major service change, key developments in policy for involving the public in service change, perceived strengths and weaknesses of the current system, and examples of proposals where involving the public in major service change had gone ‘well’ or ‘badly’.

In the case study phase, we used the examples provided by interview participants to select two change proposals in each of the four systems. In each health system we sought to include one process perceived by our national interview participants as successful, and one perceived as unsuccessful (Table 2). We conducted interviews with 18 people across the case studies, including staff working in local National Health Service (NHS) organisations (such as purchaser or provider organisations) and members of the public who had been involved in or campaigned against the proposal. Seven were conducted by author 2 and eleven by author 3. Three were conducted by telephone and 15 in person. The topic guide covered experiences of the change process, perceptions of what had gone well or badly, and participants’ awareness (if any) of the role of national guidance within the change process. Local recruitment was more challenging than that conducted at the national level. Staff involved in service changes that were perceived as having gone badly were often reluctant to take part and it was more difficult to identify members of the public active in the change processes. We used media searches (particularly local newspaper coverage) and official documents (for example, consultation reports, board papers for local NHS organisations, and official reports from agencies who assure the quality of involvement) to contextualise our understanding of the local cases.

Table 2.

Key characteristic of cases of service change in the four countries of the UK.

Name of case	Population served by affected services	Area type	Change proposed	Change process outcome
Wales A	50–100,000	Rural	Removal of services and Accident & Emergency	Change abandoned.
Wales B	Under 50,000	Rural	Removal of minor injuries and services including inpatient beds.	Change partially implemented.
Northern Ireland A	50-100,000	Urban	Closure of Minor Injuries unit.	Change implemented.
Northern Ireland B	Under 50,000	Semi-urban	Closure of specialist unit and Accident & Emergency	Change abandoned.
England A	250,000–500,000	Urban	Closure of Accident & Emergency.	Change abandoned.
England B	500,000–1 million	Predominantly rural	Downgrade multiple Accident & Emergency to Minor Injuries units.Build new hospital.	Change implemented.
Scotland A	1 million +	Urban	Closure of multiple acute hospitals.Build new hospital.	Change implemented.
Scotland B	Under 50,000	Rural	Removal of inpatient beds and reduce Accident & Emergency to minor injuries service.	Change ongoing with continuing contention.

Interviews usually lasted about 45 minutes. Transcripts or written notes of interviews were analysed using the Framework method, ⁸ chosen for its particular suitability for multi-researcher teams and policy-relevant research. Following data familiarisation, the whole research team identified a working list of coding categories. Authors 2 and 3 independently coded a selection of transcripts then revisited the list, amending and supplementing the themes. The resulting working analytical framework was then applied to the full dataset, coding in NVivo. Following analysis, we produced draft findings and recommendations which we shared with interview participants in late 2016. This process was focused on checking for comprehensiveness and error reduction, ⁹ as well as allowing us to test our interpretations, and tease out points of disagreement or uncertainty.

Ethical approval for the study was formally granted by the Usher Research Ethics Group, University of Edinburgh and all interview participants received an information sheet and consent form. To maintain confidentiality, we have grouped interview participants into generic categories of professional location in order to avoid individuals being identifiable. Within these groups, we have specified the location (England, Northern Ireland, Scotland or Wales) and allocated numbers to individuals.

Results

Drawing on existing literature, we expected to encounter some ‘good practice’ as described above, and some projects where involving the public had been simply ignored or neglected.^5,6 In practice, we encountered a wider range of choices that had been made in how to do this work, and competing rationales for those choices. A key starting point for the analysis was the observation of a national-level official of their sense of two competing approaches to the work of public involvement in contentious change.

The communications approach would be very risk averse, a kind of actually if you tell people now what you’re planning then they’re going to react badly to that and that’s not a good thing to do. So, we shouldn’t tell anybody our plans and we should keep our communication very much around kind of surface level positive stories… So, it’s more about trying to put a positive spin on it, whereas I think from our approach would be let’s be open with people that actually there’s not enough money, you’d get better outcomes and stuff. Tell people much more of the facts. (Official 3, England)

This statement articulates what became the key analytic contribution of this article: the assertion that there are coherent and different logics at work. We identified a tripartite typology: the adversarial, communications, and collaborative approaches. We describe each in turn.

Adversarial: Change against the public

In this interpretation, there is a fundamental conflict between escalating demand and strained budgets. This lent itself to doing the bare minimum to meet statutory obligations in place in each of the four countries to consult the public, in order to avoid costly judicial review or unpredictable political intervention. Some of the most forceful expressions of this view came after the formal interview and recording had ended, or with a specific request that the participants not be quoted. None of the health systems’ official documents used this rhetoric.

While involving clinicians closely engaged in service change projects was often seen as an effective strategy, clinicians and clinician-managers were some of the most vocal advocates of an adversarial approach. In some cases, the adversarial approach also stemmed from a fundamental distrust of processes of public involvement, seen to exacerbate inequalities by giving undue weight to the views of those who ‘shout loudest’. Here, organisations can be understood as defending a notion of ‘the public interest’ against actual, organised publics. ¹⁰

These debates point to deeper differences in the intent and purpose behind ‘doing’ public involvement work. We heard conflicting interpretations of the most basic answers to what this work is for. For some participants, change in the NHS was essential:

We can’t continue to provide the current service model that we have across so many sites and expect the best outcome, the best experience of care for individuals. If that’s the model you want then you’re going to have to settle for a second-rate health service, that’s the trade off. (Official 4, Northern Ireland)

This also linked to a fairly common argument not just that change was essential, but that public expectations for the availability of local services were an obstacle to change:

Everybody wants their health services on their own doorstep, but we don’t like paying for it, so there must be an element of realism. (Official 2, England)

Communications: Change for the public

A second group of interview participants outlined an approach which sought to ‘educate’ the public in the case for change, often through what the researchers observed to be sophisticated communications campaigns and astute use of clinical leadership. Fundamentally, for this approach, the goals of the change programme are non-negotiable and the key task of public involvement is persuasion, not shared decision-making.

This work was perceived as requiring a professional skillset which was not widely available within NHS organisations, and its proponents often referred to organisational change management literature, which emphasises consistent messaging above all. This approach was, in its own way, committed to hearing from the public, and its proponents often prided themselves on reaching the population through its innovative communications activities, including social media work and ‘roadshow’ type travelling events. This was also the approach associated with marketing and communications professionals, who brought to projects their own particular understandings.

In the earliest phase, often described as ‘engagement’, it was notable that communications-driven service change projects tended to convene small publics for their purposes, sometimes offering financial incentives for members of the public to take part in focus groups. In such projects the language of marketing was prevalent:

You’ve got that clinical evidence, you’ve got that market research from a public perception point of view, you then go out and do some pre-engagements where you’re testing your messaging, you’re testing the appetite for change but you’re also identifying where the pitfalls are going to come through consultation phase. (NHS Manager 1, England)

Outside England, this language was less prevalent, but the requirement to develop “a good story to tell as to why they were favouring that particular proposal” (Official 4, Scotland) was raised in all systems. However, there were differences in whether ‘early engagement’ referred to involving the public in actively producing and refining this ‘story’, or simply to extract public views on a given ‘story’ through market research processes. Some engagement events would be highly-planned, and in very stage-managed projects practitioners often exhibited a distrust and discomfort with public attendees perceived as ‘activists’, attributing their opposition to a personal preference for conflict not a grievance or concern:

  …  they’re often campaigners/activists by nature which means they’re not really resolution focused … You know, they shout for resolution, but it’s not really their focus because their natural state becomes one of activism and … without it, they lose a role. (Management consultant 1, England)

Engagement events tended to require expert facilitation, often paying external management consultants. In some cases, events used creative methods, such as theatre, with actors ‘acting out’ difficult messages, to persuade the audience, to stimulate but also to direct debate.

Once proposals for change were in the public domain, a key debate concerned whether consultation events should be drop in (with a high staff-citizen ratio), or town hall style, as summarised by one participant:

[A] lot of people, staff in particular, don’t like public meetings. They’ve got a bad history of them and what they think is that basically they’re going to get a load of folk in a room who are going to shout at them cause they’re not going to be happy and that’s going to be difficult. And most people will say, actually it’s not helpful cause you only hear the loud voices and they all wind each other up and therefore it’s not necessarily terribly productive. That’s one argument. A counter argument from communities is that they quite like to put people on, you know, give them a good old warming up and they feel it’s reasonable to put people through their paces to see how good the answers are. And an additional argument would be is that they think it’s reasonable for everybody to hear the same thing under one roof, so we can all hear that. (NHS manager 2, Scotland)

Traditional town hall style meetings enabled members of the public to recognise shared feelings and organise around proposals, with less opportunity for staff to have individual or small group conversations to explain and persuade. It seemed that town hall type meetings were often replaced by facilitated discussions. One campaigner who had attended multiple facilitated consultation events argued:

The consultation is stage-managed … It’s impossible within that process to try to speak out or challenge or be a dissenting voice, you’re just muffled around like they do, I don’t know, like a spider wraps that stuff round … you know, that’s what you feel like, you’re like a little fly caught and wrapped up round and you’ve got no voice. (Public campaigner 1, England)

This tension between creating the conditions for persuasion and dialogue, and making space for dissent to be heard, seemed a key challenge in these practices. The vision of dialogue rejected persuasion in favour of the ‘force of the better argument’, whereas for many study participants the NHS had a fundamental duty to ‘bring people with us’ and persuade them of the need for change. Participants reverted to the change management notion of ‘consistent messaging’: doubts and problems should be resolved within the organisation, while external work relied on “putting a positive spin on it” (Official 3, England).

There were examples of service change projects in which involvement was a rushed add-on to internal processes of change management, sometimes because organisations were operating in complex policy environments with conflicting imperatives:

[T]hey know that they need to [reconfigure services], they know all the reasons why, they’ve got all of the evidence of why they need to do it and they want decent public participation, and the message that they’re being told is well actually because we’re doing these [Sustainability and Transformation Plans] and everything’s tied up in them locally, actually don’t do your engagement just on this because that’ll just confuse people locally, so just hold off and don’t do anything for three months, and then when they do get to it they’ll have to do it really quickly. (Official 3, England)

In this quote, national demands for transformation of services both delayed, and then rushed, processes of engagement which local staff were committed to doing. ¹¹ In this sense, the logic of a communications approach was imposed by higher-level decisions on staff who preferred a more collaborative approach.

Collaborative: Change with the public

A third approach built on the idea of public expertise about services and communities as an asset for service change and emphasised a dialogic approach. Success here was understood as a change proposal having been not merely ‘sold to’ but shaped by local communities. This approach has its roots in community development literatures and strong resonance with collaborative approaches to governance. ¹² It is normatively and intellectually closely identified with the public participation in health literature, and its proponents make a clear normative and practical case for collaboration despite what they acknowledge to be substantial time and resource costs. A typical collaborative engagement process begins with a set of open questions about the shape of local services, rather than with a defined proposal on which to consult.

The majority of our study participants argued that attention should be paid to public involvement early in a change process, which coheres with a collaborative approach. However there were different views on what activities this early engagement should include. Relationship building could proceed through regular attendance at existing community groups and fora, listening to substantive community concerns on a day-to-day level, and not just when making change:

We would make the offer which is the fact I’m on a permanent contract and going nowhere and I’d be delighted to come back out and see you in a couple of months’ time or a couple of weeks’ time if you think that would be more appropriate … If it takes us ten years we can meet every year … People want to believe and want to have trust in the public sector, and that requires them to be able to make manifest that in a relationship as opposed to a one-off type of thing. So I think if you are able to set out that this is the beginning of a relationship and no I’m not going to go away. (NHS manager 2, Scotland)

Practices of (both literally and figuratively) ‘going to where people are’ contrasted with the option of building a trusted ‘brand’: practitioners emphasised activities such as using charitable funds to sponsor sports teams because it built “a different relationship … So we’ll be more visible and hopefully people will think we’re more responsible” (NHS manager 1, England).

Some participants argued that dialogue required a degree of openness to discussion on the part of NHS organisations and members of the public. This was fundamentally at odds with the marketing focus of the communications approach:

The health boards shouldn’t really be pushing or selling the proposal, they should be justifying how they’ve arrived at it and explain why they think it’s the best option, but they shouldn’t be selling it. Because it’s about them hearing what the issues are and being able to respond, and if there are gaps then identifying how that might be considered or addressed. Whereas the community, the public thinks if we raise our voices against then that’s sufficient. But it’s not, they’ve got to bring a wee bit more to the table than that. (Official 5, Scotland)

On the other hand, participants noted the difficulties of ensuring constant engagement. Many pointed out that public engagement is time and resource intensive (requiring both dedicated staff and senior staff time), depends on personal trust relationships that can be hard to sustain, is not always representative, might have no immediate payoff, and can seem irrelevant to actual financial or operational demands. All of these are reasons why many organizations’ leaders might not emphasize constant public engagement.

Discussion

This study describes three ‘ideal type’ approaches to public engagement in major health service change. Acknowledging that any given change process is complex, bringing together different actors’ beliefs and choices in a given organisational and policy context, we believe that these types are internally coherent. For example, if we believe publics have valuable and relevant knowledge about health services, we are unlikely to devise a project which focuses exclusively on ‘selling’ our vision of future services. Furthermore, these beliefs both grow from and create particular understandings of the relationship between health care organisations and the populations they serve.

Processes of public involvement can contain multiple, even conflicting sets of individual skills and professional commitments. ¹³ We recognise that there are no simple causal lines between approach and outcome but introduce these three approaches into a field of practice where competing aims and methods are often masked by the ‘warmly persuasive words’ of community. ¹⁴ Approaches to public participation are often placed on a spectrum from informing to empowering. ¹⁵ In our study, all reviewed cases clustered towards the informing side of this spectrum. However several of what we categorise as adversarial or communications projects did not appear focused even on informing the public. Some of the most passionate advocates of a communications-led approach among our study participants appeared to have minimal interest in empowering the public. At the same time, communications-led approaches did have internally coherent rationales and some innovative methods, which are both worthy of study in their own right and widely supported within some health policy circles.

Simply declaring that involving the public in major service change is difficult and fraught denies the diversity and internal consistency of the service changes we studied.^16,17 Several communications-driven projects were nationally reported at conferences and think tank-hosted events as ‘best practice’ in making transformational change. This suggests that the basic statutory demand to involve the public in decisions about designing services is interpreted differently through different professional competencies.^13,18 It reflects not simply failure to involve, but adherence to different models of the relationship between a health care organisation and its population. ¹²

We noted earlier that the three approaches we describe are reflected in distinct and only rarely overlapping academic literatures on change management, public involvement in health care, and democratic decision making. Change management scholars define success instrumentally as goal-achievement, ¹⁹ and this assumes the appropriateness of the change.^1,3,20 Conversely, participation scholars do not seek managerial goal achievement, but demonstrated public impact upon change projects; success is a proposal which is challenged and shaped through dialogue with the population.^21,22. Each literature is, in its own way, deeply normative,^23,24 although the standards by which they evaluate change projects are in direct opposition. While only rarely explicitly recognized, ⁵ the conflict between these two objectives reverberated through our fieldwork. A third body of literature within political science rejects the notion of the public as a single actor to be involved. Rather, populations are diverse and people have shifting views that are shaped by elites’ agenda setting and framing techniques. ²⁵ In this view, change against or for the public is a more realistic prospect than change that is dependent on deliberative processes or identifiable, stable, collective preferences. This literature has been critiqued for giving insufficient attention to procedures that do indeed allow publics to deliberate and reach coherent views. ²⁴

Conclusion

We found three conflicting frames for public engagement which are used by organisations in the UK when contemplating change: adversarial, communicative, and collaborative. While our data is drawn from the UK’s four health systems, we suspect that these frames would have wider resonance internationally. Since health care organisations are by definition the first movers in reconfiguration proposals, these approaches shape subsequent events, including conflict, implementation, and participants’ understanding in the future. While reconfigurations themselves are as diverse as the different parts of the health system, there are also clear standardising forces that mean that the three basic approaches are widely used. Examples include legal requirements for consultation, which set the minimum required for a likely adversarial approach, and the extensive guides to good practice in marketing (for a communications approach) and public participation (for a collaborative approach).

Evaluating the three approaches in terms of ‘what works’ for the adoption and implementation of change ²⁶ is difficult due to the local specificity of health services and conflicts as well as the normative and often conflicting assumptions that drive much of the literature. Our argument does have immediate implications for the large literatures on change management and public participation. Public participation literature tends to assume, and evaluate, initiatives based on the extent of democratic empowerment that they promote. ²⁷ Awareness that a communications approach can explicitly include participation without empowerment expands understandings of what organisations are really trying to do. Instead of evaluating reconfigurations against frameworks in public participation, or assuming that change is good, we argue that local framings of the problem, that is, how practitioners and members of the public are interpreting the tasks and goals of involving the public in change, need to be studied empirically and carefully.^28--30 Understanding the empirics better is not the end point, but merely the beginning of debates about these practices.