Context and history: autonomy and decision space

The Cambridge online dictionary (Cambridge online) defines autonomy as follows:

• • the right of an organization, country, or region to be independent and govern itself,

• • the ability to make your own decisions without being controlled by anyone else.

Thus, the concept of autonomy carries within it two aspects: freedom to make decisions or act; and freedom from external control (Verhoest et al., Reference Verhoest, Peters, Bouckaert and Verschuere2004). Exworthy et al. (Reference Exworthy, Frosini and Jones2011), explore autonomy in the context of NHS Foundation Trust hospitals in England and suggest that ‘freedom from’ implies a degree of vertical decentralisation, with authority devolved from the centre to the local level, whilst ‘freedom to’ also encompasses the notion of local organisations free to act in ways responsive to local needs. Exworthy et al. (Reference Exworthy, Frosini and Jones2011) argue that the latter includes the idea of freedom within ‘horizontal’ local organisational inter-dependencies.

Bossert (Reference Bossert1998) provides an analytical framework that can be used to evaluate the decentralisation of health systems. Bossert explains the range of choices available to local decision-makers along a series of functional dimensions shaping local decision-making which, in turn, shape local performance. He terms this ‘decision space’, and uses the concept to analyse the three key elements of decentralisation:

• • ‘the amount of choice that is transferred from central institutions to institutions at the periphery of health systems;

• • what choices local officials make with their increased discretion; and

• • what effect these choices have on the performance of the health system’ (Bossert, Reference Bossert1998: 1513).

Thus, according to Bossert, ‘decision space’ represents the freedom granted by central government to local organisations (in this case, local NHS actors) to make choices (Figure 1).

Figure 1.

Decision space approach

(adapted from Bossert, 1998: 1514)

Whether or not change occurs depends upon the use made of this ‘space’ by local actors, which in turn is affected by the characteristics of the local organisations. We focus upon the first of Bossert's elements, exploring the perception that local actors have about the extent to which they can make choices.

Exworthy and Frosini (Reference Exworthy and Frosini2008), however, argue that, in addition to this ‘vertical’ granting of decision space (i.e. decentralisation), the exercise of choice by managers in local organisations is shaped by, and depends upon, other actors in the local area. They argue that the realisation of decentralisation is dependent on a local organisation's ability to collaborate with other agencies, over which they have no direct or immediate authority. Yet, at the same time, they must also compete with these agencies for resources (e.g. financial resources from government and human resources from the labour market). Thus, the ‘decision space’ manifested locally arises from the interactions between ‘vertically’ granted autonomy and the ‘horizontal’ realities of acting within a specific local organisational context.

Autonomy and decision space in the NHS

As discussed in the introduction, the history of the NHS is characterised by repeated rhetorical shifts between the need for decentralisation (lauded as increasing local autonomy) and the need for central control (presented as reinforcing public accountability and fiscal prudence). Thus, for example, the 1983 so-called ‘Griffiths reforms’ under the Conservative Thatcher government introduced the idea of locally managed ‘Units’ with general managers and were couched in terms of delegation to these local organisations, but subsequent years saw the introduction of stringent performance review systems, pulling power back towards the centre (Klein, Reference Klein2013: 114). Allen (Reference Allen2006) explores the more recent history of NHS centralisation/decentralisation, and characterises it as a story of ambivalence, with rhetorical commitment to decentralisation manifest in the introduction of market mechanisms and autonomous providers, limited in practice by stronger performance management from the centre. The NHS Plan (Secretary of State for Health, 2000) introduced the idea of ‘earned’ autonomy, by which ‘high performing’ NHS Trusts were granted a number of freedoms and flexibilities (Mannion et al., Reference Mannion, Goddard and Bate2007), but in practice these ‘freedoms’ proved something of a mirage, as the need to meet top-down targets tended to dominate subsequently (Bevan and Hood, Reference Bevan and Hood2006; Macfarlane et al., Reference Macfarlane, Exworthy, Wilmott and Greenhalgh2011). Thus, for example, Hoque et al. (Reference Hoque, Davis and Humphreys2004), studied NHS acute hospitals and found that many were ambivalent about exercising decision-making autonomy due to the fear of being made ‘scapegoats’ if something went wrong.

NHS Foundation Trusts were established in the mid-2000s. These secondary care service providing organisations were intended to be quasi-independent of central government, with additional powers to re-invest surplus income and with local representation in their governance structures (Department of Health, 2005; Dixon et al., Reference Dixon, Storey and Alvarez Rosete2010). Allen et al. (Reference Allen, Keen, Wright, Dempster, Townsend, Hutchings, Street and Verzulli2012) studied the development and governance of NHS Foundation Trusts, finding that they had used their increased vertical autonomy to introduce more business-like practices, with regulators only intervening when problems were reported. However, national-level targets continued to have a significant impact on their work, and they were also constrained horizontally by the need to maintain good relationships with local organisations. It was intended that Foundation Trusts would manifest additional local accountability, with governance by local ‘members’, but this has proved problematic to realise in practice (Allen et al., Reference Allen, Keen, Wright, Dempster, Townsend, Hutchings, Street and Verzulli2012). Exworthy et al. (Reference Exworthy, Frosini and Jones2011) studied Foundation Trusts’ willingness and ability to exercise autonomy at three levels – macro, meso, and micro – and found that, whilst they had theoretically gained autonomy from central control, in practice, new forms of economic regulation had partially replaced the previous NHS hierarchy, with Foundation Trusts held to account by Monitor (a health care provider financial regulator, since April 2016 known as NHS Improvement) through a variety of performance and assessment mechanisms. Furthermore, local financial inter-dependencies and the need for local collective action constrained their horizontal autonomy within their local health economies.

Exworthy and Frosini (Reference Exworthy and Frosini2008) used a similar framework to examine autonomy and decision space in Primary Care Trusts (then responsible for commissioning the majority of NHS services). They explored both the ability and willingness of managers to use their decision space, and found that, amongst other factors, the need to act in a collegiate manner within their local health economies and avoid destabilising significant local providers constrained their autonomy. Furthermore, Primary Care Trust managers were found to be somewhat risk averse, reluctant to use their ‘decision space’ in case the Department of Health did not approve of their actions. Checkland et al. (Reference Checkland, Harrison, Snow, McDermott and Coleman2012) confirmed this, highlighting the difficulties associated with a market-driven model of commissioning in a system where local actors felt loyalty to one another and to their local area.

Thus, increasing the autonomy of local organisations in the NHS is not straightforward. The granting of administrative freedom to make decisions by no means ensures that this freedom can and will be used, and there is some evidence that an increasing focus on regulation and accountability regimes (Peckham et al., Reference Peckham, Exworthy, Greener and Powell2005), associated with the granting of greater local autonomy, has tended to push managers to act in ways which are risk averse and consonant with the demands of the centre (Macfarlane et al., Reference Macfarlane, Exworthy, Wilmott and Greenhalgh2011). In the rest of this paper, we explore the perceptions of CCG clinician leaders and managers about their autonomy, and analyse the extent to which they believe that they have experienced the greater autonomy promised by the HSCA12. In doing this, we explore their perceptions of vertical autonomy – freedom from central control – and horizontal autonomy – freedom to act within their local health economy.

Vertical autonomy – decentralisation

As CCGs took on their new responsibilities, interviewees in both Areas described themselves as having enhanced autonomy compared with their experiences of previous accountability regimes:

CCG respondents felt they had autonomy primarily because of their position as statutory bodies, which gave them a sense of legitimacy:

Notably, these accounts do not suggest reluctance to exercise autonomy. Our CCG respondents emphasised their appetite for their new role, highlighting an early sense of local power and authority:

However, CCG interviewees also explained that their initial sense of increased autonomy had waned over time. For example, CCGs are subject to ‘assurance’ from NHS England, via an assurance framework (NHS England, 2013), which focuses upon their quality of internal management and financial outcomes, alongside measures of performance such as waiting times, etc. They are also expected to address national priorities, and meet targets over which they potentially have little control:

Moreover, over and above the formal and timetabled mechanisms associated with the assurance framework, CCGs were increasingly being asked for ad hoc information to provide evidence for others in the system that all was well:

Some respondents expressed disappointment in the nature of the assurance regime, finding it to be hierarchical rather than collaborative or developmental:

External political pressure was identified as a cyclical factor acting to increase constraint. When asked in mid-2015 whether commissioners felt they had autonomy within their day-to-day commissioning activities, this was a CCG commissioner's response:

Some people stated that they saw the picture positively, i.e. they understand the contexts and the framework, and could make it work by doing what they had to do while also pursuing the things they wanted to do:

Notably, this manager associated his sense of autonomy with his own tendency and ability to plan his work carefully so as to remain within the boundaries of the system, which required care and skill. In contrast, a number of CCGs stated that the ‘must do’ things, such as fulfilling short notice requests for performance data reports from NHS England, compromised their capacity to successfully get on with their commissioning work, thereby reducing their autonomy. This was seen as being driven, in part, by the increasing financial challenges in the system:

Horizontal autonomy – the local environment

CCGs are operating within an increasingly complex local environment. Following the structural changes enacted in the HSCA12, CCGs are accountable to a plethora of organisations (Checkland et al., Reference Checkland, Allen, Coleman, Segar, McDermott, Harrison, Petsoulas and Peckham2013). Since 2014, this local environment has further increased in complexity. CCGs have to provide formal accounts to bodies such as local authority Health and Wellbeing Boards, Monitor and local government Health Scrutiny Committees, but are also required or expected to engage with a large number of new bodies and arrangements. The HSCA12 significantly reduced the funding available for the management of CCGs, with the intention that managerial support for commissioning would be ‘bought in’ from quasi-independent ‘Commissioning Support Units’ (CSUs). Instead of doing the commissioning work themselves, CCGs are required to procure some services from CSUs (Petsoulas et al., Reference Petsoulas, Allen, Checkland, Coleman, Segar, Peckham and McDermott2014). Respondents told us that these arrangements added additional layers of complexity to their work:

In addition, over the course of the study, CCGs were expected to engage with a range of new bodies which had not been established in the HSCA12, but which were developed on a seemingly ad hoc basis to solve particular issues. These include: System Resilience Groups, set up to provide a forum in which local commissioners and providers can work together to meet system-wide challenges such as winter hospital demand pressures; Urgent and Emergency Care Networks, intended to support urgent care providers in meeting rising demand; Specialised Commissioning Collaboratives, bringing together local commissioners to work with NHS England to commission highly specialised services; and Sustainability and Transformation Plans (latterly renamed Partnerships, STPs), ‘place based’ inter-organisational arrangements intended to reintroduce some of the regional system coordination lost with the abolition of Strategic Health Authorities. In the STP areas, all local health and care organisations (including Local Authorities, CCGs, NHS England local staff, and secondary and community care providers) are expected to work together to ensure the overall financial sustainability of their area, producing an STP Plan in order to qualify to access funding to reduce their financial deficits (NHS England, 2015). Alongside these initiatives, in 2014 NHS England published the Five Year Forward View (NHS England, 2014). This set out NHS England's aspirations for funding and service development, and, inter alia, established a number of ‘New Models of Care’ known as ‘Vanguards’ (NHS England, 2016). These are collaborative ventures between commissioners and providers which seek to design and implement new ways of delivering care crossing existing boundaries between primary, secondary, community and social care. Vanguards were commissioned via a national process, and the relationship between local Vanguards and their responsible CCGs are complex and variable, with some closely involved whilst others are not. Finally, many CCGs have decided to work together and with relevant providers informally on a variety of scales and in different combinations. These are sometimes referred to as ‘networks’ or ‘sectors’, and may, for example, include combining forces to jointly commission some services, or working across the boundaries between commissioners and providers to tackle complex service issues. Taken together, and with Exworthy and Frosini's (Reference Exworthy and Frosini2008) ‘horizontal’ dimension of autonomy in mind, this sprawling range of inter-organisational arrangements clearly suggests that CCGs’ freedom to act, even if they have formal autonomy from the centre, is likely to be limited by the need to co-ordinate, collaborate and interact with other local actors.

The development of these inter-organisational arrangements was emergent and on-going during data collection. In this constantly changing situation, interviewees often struggled to express what particular groups or initiatives were intended to achieve or what their role in them was meant to be.

The initiatives mentioned exist at different scales and cover different populations. One CCG clinician was critical of the Urgent and Emergency Care Network model on the basis that the Network with which his CCG was associated did not correspond to any other relevant geographical boundaries. It represented yet another responsibility that the CCG was required to discharge at a broad scale, which contrasted particularly with the local focus that CCGs were created to attend to:

Not only were these developing arrangements complicated to negotiate, they also had an impact on other organisations trying to work in partnership with the NHS:

As well as being more structurally complex the new system was also functionally more complex. Prior to the HSCA12, Primary Care Trusts were responsible for the majority of commissioning activity. Following the HSCA12, responsibilities were split between CCGs, Local Authorities, NHS England and Public Health England. This placed an onus on CCGs to work closely with all these organisations in order to ensure co-ordination across care ‘pathways’:

This manager looked back nostalgically to the days in which an over-arching organisation provided some degree of co-ordination, in contrast to a system with many autonomous organisations all with statutory authority: