%20-%20https://sketch.com%20--%3e%3ctitle%3e3ECBBF50-FD08-412A-943D-2CFC5E025A16%3c/title%3e%3cdesc%3eCreated%20with%20sketchtool.%3c/desc%3e%3cg%20id='Stylesheet'%20stroke='none'%20stroke-width='1'%20fill='none'%20fill-rule='evenodd'%3e%3cg%20transform='translate(-926.000000,%20-668.000000)'%20id='BUTTONS'%3e%3cg%20transform='translate(100.000000,%20597.000000)'%3e%3cg%20id='BTN-/-RIGHTS'%20transform='translate(816.000000,%2065.000000)'%3e%3cg%20id='ICON-/-Rights'%20transform='translate(10.000000,%206.000000)'%3e%3cpath%20d='M13.5752841,0%20L1.34943182,0%20C0.59375,0%200,0.586666667%200,1.33333333%20L0,22.6666667%20C0,23.4133333%200.620738636,24%201.34943182,24%20L17.6505682,24%20C18.40625,24%2019,23.3866667%2019,22.6666667%20L19,5.36%20L13.5752841,0%20Z%20M13.5752841,1.89333333%20L17.0838068,5.36%20L13.5752841,5.36%20L13.5752841,1.89333333%20Z%20M1.34943182,22.6666667%20L1.34943182,1.33333333%20L12.1988636,1.33333333%20L12.1988636,5.36%20C12.1988636,6.10666667%2012.7926136,6.69333333%2013.5482955,6.69333333%20L17.6235795,6.69333333%20L17.6505682,22.6666667%20L1.34943182,22.6666667%20Z'%20id='Shape'%20fill='%23595959'%20fill-rule='nonzero'%3e%3c/path%3e%3ccircle%20id='Oval'%20stroke='%23595959'%20stroke-width='1.5'%20fill='%23595959'%20cx='9.5'%20cy='13.5355339'%20r='5.5'%3e%3c/circle%3e%3cpolyline%20id='Path-2'%20stroke='%23FFFFFF'%20stroke-width='1.5'%20transform='translate(9.500000,%2013.035534)%20rotate(45.000000)%20translate(-9.500000,%20-13.035534)%20'%20points='8%2016.0355339%2011%2016.0355339%2011%2010.0355339'%3e%3c/polyline%3e%3c/g%3e%3c/g%3e%3c/g%3e%3c/g%3e%3c/g%3e%3c/svg%3e)
Introduction
The structure and financing of health systems are key elements that influence their shape, performance and coverage. Whilst there are those that argue for integration between the different elements of services as a key goal (Ham, Reference Ham2009; Pollock, Reference Pollock and Price2011), others have claimed that the separation of the purchasing of health services from their provision is a prerequisite for improvements in value for money (WHO, 2000; Figueras et al., Reference Figueras, Robinson and Jakubowski2005). Indeed, the 2010 World Health Organisation annual report explicitly argues that this purchaser–provider separation is a powerful weapon in the fight to achieve universal coverage of care, albeit one that must be wielded with care (WHO, 2010: 90). The core logic of this prescription is that it detaches decisions about service provision from the self-interest of providers, especially clinical professionals (Harrison, Reference Harrison1991).
This approach has been adopted in the English National Health Service (NHS) for some twenty years, taking the form of purchasing of secondary, tertiary and community care by primary care organisations responsible for defined local populations. (Under intra-UK political devolution, different arrangements exist in Wales, Scotland and Northern Ireland.) By 2006, some 75 per cent of the total health budget had been devolved to primary care organisations (Figueras et al., Reference Figueras, Robinson and Jakubowski2005: 19), with England divided into 152 Primary Care Trust (PCT) territories, each with a population of persons registered with general medical practitioners located inside the territory for whom to purchase (or commission) care. PCTs had some ability to determine local care priorities, and to specify the quality standards to be met by providers. PCTs could also agree to delegate ‘indicative’ (non-binding) budgets to local consortia of general medical practices to commission services for their registered patients, termed ‘Practice-based Commissioning’ (PBC). Secondary and tertiary care could be commissioned from publicly and/or privately owned hospitals, implying competition for patients, though not based on price. Most patient cases are categorised to one of some 1,300 Healthcare Resource Groups (HRGs, analogous to the Diagnosis-related Groups employed in the US), and each HRG attracts a nationally specified standardised payment, or ‘tariff’ (Department of Health, 2002) since patients in each category are assumed to represent an identical workload. However, this ‘Payment by Results’ (PbR) system permits competition in terms of perceived service quality, since most non-emergency patients may choose to be referred to any hospital that is recognised by a national regulator (the Care Quality Commission). Following the election of a coalition government in 2010, the split between commissioning/purchasing and providing services has been further entrenched in England by an on-going, major reorganisation that replaces PCTs with ‘Clinical Commissioning Groups’, led by GPs (Department of Health, 2010).
There is relatively little published empirical evidence about the impact of commissioning in the last decade or so; that is, the period following the introduction of the standard PbR tariff. This regulatory change moved the focus of commissioning away from agreeing contracts for volumes of care or for prices towards a discussion of pathways of care and quality. The few studies of PCT commissioning suggest that such organisations are perceived as weak in relation to providers (Abbott et al., Reference Abbott, Procter and Iacovou2009) and find difficulty in incorporating quality or clinical performance considerations into their work (Allen, Reference Allen2002, Exworthy, Reference Exworthy1998), in moving away from historical patterns of service (Bate et al., Reference Bate, Donaldson and Murtagh2007) or in taking explicit prioritisation decisions (Bravo Vergel and Ferguson, Reference Bravo Vergel and Ferguson2006). PBC has been investigated in two major studies (Curry et al., Reference Curry, Goodwin, Naylor and Robertson2008; Coleman et al., Reference Coleman, Checkland, Harrison and Dowswell2009b), and although these reached somewhat different conclusions about the value of such commissioning, neither demonstrated that commissioning had led to major changes in, or major disinvestments from, services. It is important to note that commissioning and competition are different albeit potentially related mechanisms: whereas the outcome of commissioning derives from commissioners’ decisions about health needs, the design of services and the capacities of providers, the outcome of competition may derive from transaction rules, provider behaviour, patient choice, and the characteristics of local markets. Thus, although there is some evidence that the quality of care is higher in areas where an effective choice of secondary care provider exists (Propper et al., Reference Propper, Burgess and Green2004), this does not constitute evidence about the success or otherwise of commissioning activity.
These limitations in research evidence have not prevented the development of a received wisdom of commissioning ‘failure’ amongst politicians and policy commentators. The last year has seen the publication of a number of reports and commentaries examining NHS commissioning (Health Select Committee, 2010; Paton, Reference Paton2010; Smith et al., Reference Smith, Curry, Mays and Dixon2010). Despite varying analytical approaches and data sources, there is a great deal of consistency in the indictment of commissioning offered by these three publications, which assert two broad lines of explanation for this perceived failure.
The first of these is that the rules governing transactions between commissioners and providers in the PbR system are weighted in favour of hospitals and therefore do not permit PCTs fully to exercise their functions. Thus, the Health Committee identified a lack of adequate ‘levers’ with which commissioners might influence providers of care (Health Select Committee, 2010: 35), whilst the Nuffield Trust saw PCTs as less autonomous from central direction than many providers (and thus more risk-averse) and also less legitimate in the public eye (Smith et al., Reference Smith, Curry, Mays and Dixon2010: 22). More specifically, Nuffield noted that PCTs have little control over the volumes of referrals made by GPs or the choices of hospital made by patients, and that the PbR regime makes it difficult for commissioners to relocate hospital care into community settings (Smith et al., Reference Smith, Curry, Mays and Dixon2010: 28–9). There is little doubt that this line of explanation is valid in its own terms. Although PCTs have cash-limited annual budgets, and must remain within budget, they are required to pay for the care provided for their patients wherever it occurs, and are not permitted unilaterally to control patient volumes through contractual caps on providers. Hospitals are therefore incentivised to maximise their activity within physical capacity. PCTs’ main option for influencing volumes and expenditure lies in incentivising PBC consortia through the sharing of savings on indicative budgets resulting from reduced referrals and/or the development of new ‘pathways’ to treat patients in community settings (Checkland et al., Reference Checkland, Harrison and Coleman2009b).
The second broad line of explanation deployed in recent indictments of commissioning focuses on the capacity of staff engaged in commissioning. Both the Health Committee (Health Select Committee, 2010: 3) and Nuffield (Smith et al., Reference Smith, Curry, Mays and Dixon2010: 22–3) identified a lack of management skills in PCTs, exacerbated by constant NHS reorganisation, whilst the latter also identified weak clinical leadership and engagement in commissioning and a lack of data (Smith et al., Reference Smith, Curry, Mays and Dixon2010: 21, 25). Paton's commentary concurs, arguing that commissioning managers have failed sufficiently to engage with managers in providers in order to secure improved services, and that the former are generally of poorer quality than the latter (Paton, Reference Paton2010: 1979). The validity of these assertions about managerial capacity is doubtful, since no systematic evidence exists, and regulatory assessments of PCTs have been generally positive (Care Quality Commission, 2009). Qualitative evidence about the capacity of practice-based commissioners is somewhat mixed (Coleman et al., Reference Coleman, Checkland, Harrison and Dowswell2009b; Curry et al., Reference Curry, Goodwin, Naylor and Robertson2008) and one survey has suggested that a large proportion of such commissioners self-report as lacking necessary skills (Curry et al., Reference Curry, Goodwin, Naylor and Robertson2008). We are therefore unable to reach a conclusion as to the validity in its own terms of the second line of explanation.
The purpose of this paper is to propose an overarching explanation for the ostensible failure of commissioning that is both theoretically informed and based on detailed empirical research. Specifically, we draw on evidence from two recent studies of NHS commissioning that we have conducted in order to explore how far the perceived ‘failures’ of NHS commissioning can be seen as a lack of ‘fit’ between the concept of commissioning and the structures and processes of the NHS as it has developed since its inception.
What follows is divided into five sections, beginning with an account of so-called ‘new institutional theory’ and its applicability to the NHS. Our methods are then described, and results presented alongside a discussion of the relevant theoretical ideas. A final section examines the insights that institutional theory provides into the perceived failures of NHS commissioning, particularly in the light of on-going organisational changes.
New institutionalism
Institutionalism is less a single theory than it is a set of ideas that has been addressed from a number of different theoretical approaches and within a number of different disciplines. Hall and Taylor (Reference Hall and Taylor1996) identify three traditions within this broad heading. Firstly, they identify historical institutionalism, deriving largely from the disciplines of political science and history, and focusing upon the role of the formal or informal rules, procedures, routines and norms embedded in the organisational structure of the political economy in structuring interactions and determining outcomes, often through the phenomenon of path-dependence (Pierson, Reference Pierson2004). ‘Rational choice’ institutionalism, by contrast, sees individuals’ preferences as exogenous to the institutions that they inhabit, emphasising instrumental behaviour in pursuit of relatively stable and calculated goals. From this perspective, institutions are seen as a way of maximising the efficiency with which these goals are pursued (Lowndes, Reference Lowndes1996: 184). Finally, Hall and Taylor (Reference Hall and Taylor1996) identify sociological institutionalism, largely arising from organisational studies. In this tradition, institutions are seen as not simply rationally constituted in pursuit of goals, but as defined by cultural practices, including ‘cognitive scripts’ and ‘moral templates’ (1996: 947) that define not only what should happen in given circumstances but also set the limits of what can be considered as possible. Sociological institutionalists see a recursive relationship between actors and institutions by which each defines and sets limits upon the other (Giddens, Reference Giddens, Clark, Modgil and Modgil1990). The strength of this approach is that it is able to explain the persistence and duplication of institutions that might appear to be suboptimal in their operation, since institutional forms and processes are transmitted via cultural means.
In this paper, we focus upon this latter approach. Scott (Reference Scott2008b, Reference Scott2008a) argues that institutions rest upon ‘three pillars’. The regulative pillar encompasses formal structures and rules. The normative pillar denotes moral assumptions about what ought to occur. The cultural-cognitive pillar denotes constraints on what it is possible to envisage as meaningful, effective or appropriate action in any given situation, implying the imposition by the institution of a cognitive frame which excludes other ways of seeing. Di Maggio and Powell (Reference Di Maggio and Powell1983) argue that institutions can be transmitted or created by coercive means, using rules and formal structures, by normative appeals to what ‘should’ occur and by mimetic processes by which existing institutions are copied or seen as a template. In a similar vein, March and Olsen (Reference March and Olsen1984: 741) suggest that ‘political actors associate certain actions with certain situations by rules of appropriateness. What is appropriate for a particular person in a particular situation is defined by the political and social system and transmitted through socialisation.’ Thus, these scholars share a conception of institutionalism which rests upon the investigation of formal rules and regulations, norms of behaviour and cultural and cognitive constraints upon what it is possible to think or do in any given situation. These insights have been used, for example, to ask questions about the circumstances in which de-institutionalisation might occur (Oliver, Reference Oliver1992) or the ways in which institutions change over time (Christensen and Molin, Reference Christensen, Molin, Scott and Christensen1995).
Much writing in this tradition is vague as to what constitutes ‘an institution’ (Mohr and White, Reference Mohr and White2008: 421), with some authors treating institutions as synonymous with organisations (Peters, Reference Peters2005: 29). Scott defines institutions broadly, as containing ‘regulative, normative and cultural-cognitive elements’ (Scott, Reference Scott2008b: 48); this is less a definition of a ‘thing’ than it is a manifesto for a research programme (Scott and Christensen, Reference Scott and Christensen1995). However, investigating an organisation as large and as complex as the NHS requires the researcher to at least attempt to define in advance the organisational territory to be explored. Scott (Reference Scott2008a: 435) suggests that the social world can be usefully thought of as operating at a number of different levels. The world system is divided largely by national boundaries into a number of different societies. Within each society there exist a number of separate ‘organisational fields’, defined as a population of organisations concerned with a common enterprise or type of enterprise (Scott, Reference Scott2008b: 86). Thus, the English NHS could be said to be an organisational field concerned with health and health care. Within this ‘field’ there exist a number of different ‘populations’ of similar organisations, such as hospitals, Primary Care Trusts and general practices. Each one of these is itself an organisation, which may also contain organisational subsystems, such as operating theatres within a hospital. ‘Institutions’ may exist and have influence at each of these levels, and it is a matter for empirical investigation what they are and how they act. Thus, for example, one could postulate that the organisational field that is the NHS might be influenced by the institutions of the medical and nursing professions. The latter are governed by rules relating to occupational licensure, but also contain powerful norms and cultural-cognitive assumptions, which may be in conflict.
The NHS as an institution
It has been argued above that the NHS constitutes an organisational field. But is it also useful to think of it as a recognisable institution, characterised by Scott's regulative, normative and cultural-cognitive pillars? Whilst there have always been local variations in the delivery of care (Powell, Reference Powell1998), historically the notion of belonging to and working for the NHS has implied some kind of solidarity and commonality of purpose and experience (Sabin, Reference Sabin1998). In everyday speech, both NHS employees and commentators casually refer to ‘the NHS’ as a meaningful entity, even when many NHS services are delivered by private sector organisations. Documents such as the NHS Operating Framework (Department of Health, 2009b) and associated legislation could be said to constitute the regulative pillar of the institution. Furthermore, it is also commonplace to refer the ‘NHS family’ and to ‘NHS values’, with, for example, the 2010 NHS constitution explicitly setting out values which are said to underpin the NHS as a whole. This suggests that, in addition to a regulative framework, the NHS could be said to be underpinned by a codified set of normative values. The existence of a cultural-cognitive pillar is perhaps a matter for empirical enquiry, but the apparent resistance of the NHS to change in the face of continual structural upheaval has been cited as evidence for the existence of a recognisable ‘NHS way of doing things’, constrained by a cognitive frame as to what action can be conceived of as possible, which could be labelled a cultural-cognitive pillar (Plamping, Reference Plamping1998). Taken together, these notions suggest that it may be useful analytically to treat the NHS as an institution as defined by Scott. This is not to suggest that the NHS should be seen only as an institution in its own right; indeed, part of the strength of Scott's approach to institutions is that it makes it clear that any particular organisation could both contain institutional elements in its own right and be acted upon by other institutions, or, indeed, contain within it a number of different smaller institutions. The argument being made here is simply that treating the NHS as an institution opens up some interesting analytical possibilities. With this in mind, this study set out to address the following research questions:
• How do members of the different organisations that constitute the organisational field of the NHS interact within the context of the activity known as ‘commissioning, and is it possible to discern the impact of something recognisable as an ‘NHS institution’ within this activity?
• What does this tell us about commissioning as an approach to planning and managing healthcare, and is it possible to reach an institutional explanation for the ‘problems’ identified by commentators?
Methods
This paper utilises findings from two studies of different aspects of commissioning in the English NHS. The first study was carried out between January 2007 and June 2009, exploring the development and operation of Practice-based Commissioning (Department of Health, 2006). PBC was introduced in order to increase the involvement of clinicians in the commissioning process, and it entailed the devolution of notional budgets with which to purchase care to groups (‘consortia’) of GPs (Coleman et al., Reference Coleman, Checkland, Harrison and Dowswell2009b). Consortia had managerial support from PCT commissioning managers, and were entitled to reinvest any money saved in patient care. The detailed methods and results from this study have been reported elsewhere (Checkland et al., Reference Checkland, Coleman, Harrison and Hiroeh2008; Checkland et al., Reference Checkland, Coleman, Harrison and Hiroeh2009a; Coleman et al., Reference Coleman, Checkland, Harrison and Dowswell2009; Coleman et al., Reference Coleman, Checkland, Harrison and Hiroeh2010); this paper concentrates upon the commissioning work undertaken by the groups. Initial data collection in three ‘early adopter’ sites was followed by further detailed study of another six sites. A total of fourteen consortia in eight PCTs were studied. Data collection methods included 131 interviews with PCT managers, GPs and others involved with PBC in any capacity (e.g. practice managers and public health staff) and more than 300 hours of observation of a variety of meetings relating to PBC.
The second study arose from the finding that the actions and approaches of PCT middle-level managers were important determinants of PBC and its outcomes. However, the role of such managers in PCTs had been little studied; our second study therefore focused upon the roles and activities of PCT commissioning managers. Four case studies were undertaken in PCTs in the north of England between April 2009 and June 2010. Following initial meetings to identify relevant staff, thirty semi-structured interviews were carried out with middle grade managers (i.e. below board level), in which the roles, responsibilities and experiences of the commissioning managers were explored. In addition, we undertook 150 hours of observation in commissioning meetings and informally shadowing commissioning managers (Checkland et al., Reference Checkland, Snow, McDermott, Harrison and Coleman2011), both those involved with PBC and those charged with commissioning on behalf of the PCT. A wide variety of meetings relating to all aspects of commissioning were included; for example, internal meetings to discuss commissioning particular pathways, meetings to discuss aspects of the World Class Commissioning performance assessment initiative (Department of Health, 2007) and meetings with providers.
In both studies, detailed contemporaneous fieldnotes were kept, and analysed alongside interview transcripts, using the qualitative analysis software Atlas.ti. Initial coding of transcripts and fieldnotes using theoretical concepts identified from relevant literature was followed by more open coding, in which emergent themes and concepts were identified. Analytic memos (Miles and Huberman, Reference Miles and Huberman1994) were written and discussed. Emerging findings from the two studies were compared, with a specific focus upon the way that ‘commissioning’ played out in practice. This approach provides a detailed ‘bottom-up’ account of the realities of commissioning as it actually happens, providing information about the roles of managers, the interaction between managers and clinicians, and the interactions between commissioners and providers within the various commissioning forums. Both studies received relevant NHS ethical and research governance approvals.
NHS commissioning in practice: the evidence
In this section the results from the two studies are presented together, alongside a discussion of the relevant concepts and ideas from neo-institutional theory, exploring the roles of regulative, normative and cultural-cognitive elements. Quotations and fieldnote extracts are included in the text where they illustrate the points being made or are typical of the responses made. Whilst only a small number of illustrative examples can be provided, the themes and topics demonstrated here were found with remarkable consistency across studies, sites and contexts, in particular both PCT and PBC forums. In order to preserve anonymity, ID numbers and vague job titles are used throughout. The text identifies the quotation source as Study 1 or Study 2.
What does it mean to be ‘commissioned’?
The regulations associated with the separation of ‘commissioning’ and ‘providing’ care embody a contractual relationship based upon the notion of ‘customer’ and ‘service provider’, with commissioners specifying the service required, and providers tendering to provide it. We have already noted that other aspects of the regulatory pillar, such as transaction rules, tend to favour providers. But we also found that both commissioners and providers struggled with the more fundamental ideas underpinning commissioning. Thus, for example, it was common for providers to attend meetings to present the details of services that they wished to provide, rather than for commissioners to specify what they required:
Item 6 Community paediatrics bid
[Representatives from the local hospital attended the meeting to present a proposal for a new service run by hospital paediatric nurses, designed to avoid hospital admission.]
GP 4 – it makes a nonsense of us being commissioners if we are given something like this and expected to sign off by Wednesday.
GPs asked several questions about payment and tariff – how much would visits by these nurses cost, would children who went in for assessment cost tariff price etc. – consultant who came to present it answered with lots of aspirational stuff about ‘keeping children at home’ and ‘providing care in the community’, but with no evidence as to numbers.
. . . [Once the consultant and his manager had left there was general discussion around the table about how providers seemed to think that all they had to do was dream up a new service and PBC would fund it.] [Study 1 PBC meeting Fieldnote ID M13]
This problem was clearly demonstrated in Study 2 with regard to obstetric services. ‘Fetal Medicine’ involves the care of complicated pregnancies by specialist paediatricians and obstetricians. In one site, paediatric and obstetric services were provided by a local District General Hospital (DGH), with Fetal Medicine provided by a large Teaching Hospital nearby. The DGH appointed a new consultant with expertise in Fetal Medicine, who started caring for some patients locally, referring more complicated cases on as before. A meeting between the PCT/PBC commissioners and the DGH hospital managers was observed at which the DGH managers tried to renegotiate the price they received for this service. However, a significant proportion of patients would still require onward referral, resulting in the PCT paying twice for a service that previously attracted a single payment. The DGH had unilaterally changed the service that they provided and were retrospectively asking the PCT to commission it:
[PCT manager 1] clarified that she was understanding it properly – it was about providing as much care locally as possible? Whilst that seems reasonable, they need to understand if it will cost the PCT more. [PBC manager 1] reiterated that they need to understand what they are currently paying Teaching Hospital and what they are currently paying DGH, They would argue that this should be paid for at the normal tariff.
[PBC GP] said that the problem was that DGH are already doing it.
[DGH manager 1] asked why that was a problem?
[PBC GP] said ‘it is a problem from the money point of view’.
[DGH manager] became a bit defensive, saying that it was saving women trips to Teaching Hospital.
[PCT manager 1] tried to be soothing, saying that it made sense from a quality point of view, but they ‘need to understand the money. Is it a problem? If we haven't commissioned the service, we don't know the price.’
[DGH consultant obstetrician] joked about ‘stopping doing it’ as if this was an absurd suggestion.
[PCT manager 1] said ‘no, but we need to understand it’. [Fieldnote ID M33]
In institutional terms, these examples suggest a mismatch between the regulatory framework of commissioning, which mandates that commissioners specify the service that they require, and a longstanding shared norm that providers should be in the business of trying to improve their services. Changes were justified by appealing to a shared common moral norm that improving local services was by definition a good thing. In March and Olsen's (Reference March and Olsen1984, 2001) terms, the ‘logic of appropriateness’ in this situation is that it is inappropriate to impede the ability of the hospital to develop a new service (as evidenced by the joking reference to ‘stopping the service’).
Blurred boundaries: what is ‘the thing’ that is being commissioned?
Splitting commissioning from provision requires a clear definition of the things to be commissioned. Harrison (Reference Harrison2009) has called this ‘conceptual commodification’, arguing that the control of health care providers requires the explicit definition of units of activity that can be measured and controlled. The HRG system is an example of this, assigning patients to a price category. Other examples include the increasingly complex ‘patient pathways’ developed to plot all the possible routes that patients might take (Department of Health, 2009a). However, in the real world of commissioning we found blurred boundaries, with little agreement as to how particular services should be classified, or indeed whether or not such classification was useful.
These difficulties were evident in the division of commissioning work between teams. In all four of our study sites in Study 2 we found the allocation of responsibility for particular commissioning topics to be a source of ongoing debate and discussion. For example, in Site 1, teams of managers, supported by clinicians, were responsible for different commissioning areas, such as respiratory disease or urgent care. However, the boundaries between these categories are rarely clear. This manager expressed it thus:
[in answer to a question about how work is divided up]: There's diabetes, there's end of life care, there's unscheduled care, um, I think new neuro one has been set up as well . . . Stroke.
Q: Yeah. There's so many out there, they do overlap, don't they?
A: Right. They do . . . we've got a coronary heart disease group and we've got a stroke group and obviously there's a lot of overlap there. Yeah, they are quite arbitrary. [PCT manager ID 6]
These difficulties were also found elsewhere, with several of our sites having recently reorganised their commissioning groups, as they struggled to work out how the lines should be drawn or the ‘thing’ being commissioned defined.
A further example of this problem comes from attempts to introduce new ‘patient pathways’. These are an attempt to codify care so that defined packages can be ‘bought’ from the available providers (Jones, Reference Jones1999). However, in our studies we found that whenever such a ‘new pathway’ was presented to an audience of clinicians, debate about the details and boundaries of the pathway ensued. For example, in Study 2, Site 3, a new pathway for the management of carpal tunnel syndrome was introduced at a meeting. This is a relatively straightforward, uncomplicated condition, and yet a long and complicated discussion ensued. The same pathway was raised at another meeting between PCT Commissioners and the local hospital managers, resulting in further discussion and objection, with an eventual decision to take comments back to the team responsible. Thus, even for a simple pathway, there was little consensus and no eventual decision. Furthermore, where ‘new pathways’ were introduced, clinicians were often reluctant to use them, finding them too complex or unhelpful in the context of the complexities embodied by ‘real’ patients:
I keep stressing to people that if they want GPs to do things, it has to be kept simple. Um, and you got to know that the patient will fit into the pathway before you start going on with it. So, [patients] have this awful habit of not fitting in with all the pathways. [emphasis added. Study 1 PBC GP ID 7]
Thus, both studies demonstrated that one of the principal practical difficulties with commissioning is in defining exactly what the ‘thing’ to be commissioned is. Whilst the regulatory pillar associated with a split between commissioning and providing takes for granted the idea that a concrete entity can be defined, packaged and managed, in real life our participants were struggling to commodify (Harrison, Reference Harrison2009) care in this way. Furthermore, whilst many spoke eloquently about these difficulties, none was able to articulate the problem analytically. Indeed, the cycles of reorganisation that we encountered suggest that, urged on by the regulatory framework, our participants assumed that if only they could be smarter and divide the work up better, all would be well. Our institutional analysis, however, suggests that, rather than being a simple problem of allocation to appropriate categories, these findings demonstrate a fundamental mismatch between the demands of the new regulatory landscape of the NHS and the deeper, unconscious cultural-cognitive reality within which NHS staff are socialised and do their work. This deeper ‘institutional reality’ is one in which doctors caring for patients see the individual patient as the important ‘thing’ at issue rather than the illness or package of care required. Furthermore, these cultural-cognitive institutional pillars are deeply seated and unarticulated (Scott, Reference Scott2008a), representing fundamental constraints on how it is possible to think about a particular issue; this may explain why our participants found it hard to articulate the root causes of the difficulties that they faced, and why they continued to attempt to solve their problems by reorganising themselves.
The problem of institutional change: ‘we're all in this together’
Splitting ‘commissioning’ from ‘providing’ care is, in part, designed to encourage competition. However, both of our studies found a tendency for commissioners to default to negotiations with a trusted local provider of care. In Study 1, the problem was expressed thus:
Another one would be, that it might put our acute trust at risk . . . we are certainly conscious of it, that we don't want to put our main provider at risk, you know we are quite aware that they have quite a big PFI bill that's going to hit their door and that they have to start paying for, so you have to balance that with being able to provide services as another provider, local to patients . . . but you know . . . whose interests are we serving? You know, I suppose in the end it is the patients and then you have to think, what serves them best is having a trust on their doorstep that can manage its books, or having services with more locally provided. (PCT employee ID 32)
This tendency carries with it a powerful internal logic. It is clear from public reactions to planned service changes that local hospitals carry high levels of support amongst their local populations. In institutional terms, whilst the new regulatory framework suggests that commissioners should be proactive in moving their business around, entrenched institutional norms about the value of local hospitals militates against any significant changes in the pattern of service delivery. However, our data suggest that a yet deeper cultural-cognitive level of assumptions also exists. To many of our participants the notion of a ‘local health economy’ whose integrity must be protected is meaningful. For these clinicians and managers, whilst it might be possible for the PCT to extract a short-term gain by moving a service to another provider, this was seen overall as a zero sum game. One PCT manager expressed it thus:
I mean I've, you know, found it really difficult where I used to be director of contracting, across, three PCTs and I just found it really disheartening . . . I just, I realised that my job, if I did my job well, I would add to the hospital's deficit and reduce our deficit. And if I did my job badly, I would add to our deficit and reduce their deficit so I was just, you know . . . [Study 2 PCT manager ID 29]
In several sites in both studies we were told that high level negotiations took place in order to ensure that budgets were balanced. For example, this manager discussed ‘horse-trading’ between executives:
‘But when we were doing the work for the [financial group], for the dementia pathway, we discovered a cohort of patients that we'd been charged an extortionate amount for being admitted to [local FT] as acute patients . . . And when we challenged that, we were told . . . that's all part of the horse trading that's gone on between the two CEOs, end of.’ [Study 2 PCT manager ID 32]
As well as embodying shared normative values, this attitude also suggests a deeper cultural-cognitive assumption that the notion of ‘we’ includes all of those within the local NHS family rather than simply the participant's own organisation. Thus, acting to maximise the gain of one organisation within the ‘family’ at the expense of another remains an unthinkable step for many in the NHS; in institutional terms, the cultural-cognitive frame through which these actors saw the NHS prevented them from ‘seeing’ themselves as anything other than as part of a common enterprise. This is by no means universal, and in both studies we found isolated incidences of clinicians and managers trying to disrupt historical patterns of service delivery. However, this was generally only visible at the margins, affecting easily delimited services such as cataract operations, although it was clear that in some areas, well-established Foundation Trusts were adopting a more commercial approach that was felt by some to be beginning to change the terms of the debate.
Discussion
In summary, we have used data from two linked studies to explore the nature of commissioning in the English NHS. We contend that the lack of impact identified by commentators reflects not only inadequacies in the regulatory structure that could be fixed by modifying transaction rules (Smith et al., Reference Smith, Curry, Mays and Dixon2010), or (alleged) inadequacies in the performance of individual actors that might be fixed by better training or recruitment (Health Select Committee, 2010), but also deeper characteristics of the NHS institution. A combination of cultural-cognitive assumptions and beliefs of NHS commissioning staff have apparently served to undermine the ostensibly straightforward notion that commissioning organisations are able to make choices about what services should be purchased and from which providers. Whilst we did observe occasional examples of commissioners making such choices, it was notable that this only occurred at the margins, involving small volumes of straightforward elective care. What is striking is that these findings are strongly redolent of research findings from earlier phases of the NHS purchaser/provider split, in that formal changes in transaction rules are mediated through pre-existing social and institutional arrangements (Flynn et al., Reference Flynn, Williams and Pickard1996; Ferlie, Reference Ferlie, Bartlett, Propper, Wilson and Le Grand1994; Hughes et al., Reference Hughes, Griffiths and McHale1997; Mays et al., Reference Mays, Goodwin, Killoran and Malbon1998). In more abstract terms, commissioning has remained a relational (Powell, Reference Powell, Thompson, Frances, Levacic and Mitchell1991) rather than an economic phenomenon, embedded (Granovetter, Reference Granovetter1995) in existing social and institutional structures. Taken together with our findings, there is a clear implication that assumptions about the value of separating commissioning and providing may be somewhat simplistic; the impact of such policy prescriptions will vary in different social and institutional contexts.
In terms of our title (‘what's the problem?’), it is possible to argue that the institutional characteristics of the NHS are the problem, inhibiting the development and operation of what might be seen as a desirable shift towards greater separation between the commissioners and providers of health care. On the other hand, it is equally possible to argue that it is commissioning that is the problem, representing a forcibly introduced concept that is somewhat alien to a valued NHS institution. Any choice between perspectives here is clearly normative and political, and is not a focus of this paper. However, it is appropriate to explore how far, in the context of forthcoming NHS organisational changes, it is the concept of commissioning, or the institutional characteristics of the NHS, that is most likely to give.
Whatever final details emerge about the current NHS reorganisation, it seems clear that a split between ‘commissioning’ and ‘providing’ will remain the underlying organising principle, with groups of GPs assuming greater responsibility for commissioning care from competing publicly and privately owned providers. This implies a move away from a monolithic NHS, setting out a vision of a much looser structure of competing organisations which will be ‘accredited’ to provide NHS services. This might be described as a departure from the notion of an NHS ‘common enterprise’ towards an ‘NHS brand’ (symbolised by the familiar blue and white logo), with which accredited providers will ‘badge’ their services (Lister, Reference Lister2010). This is similar to changes which affected UK Local Government a decade previously (Barlow, Reference Barlow and Röber1996). In institutional terms, these changes represent a significant reconstruction of the regulatory pillars governing the NHS. It is unclear at present whether these regulatory changes will in turn generate sufficient change in the normative and cultural-cognitive institutional pillars to allow commissioning to function as its proponents claim it should.
In order to explore this further, it is worth examining the literature that addresses the question of how institutional change may occur. Zucker (Reference Zucker1977, Reference Zucker and Zucker1988) focuses on the ‘transmission’ of institutional elements and suggests, for example, that imperfect transmission of institutional rules may result in gradual change over time. Kraatz and Zajac (Reference Kraatz and Zajac1996) examine the changing nature of the traditional American liberal arts education, and highlight the gradual substitution of one institutional logic by another, given impetus by significant changes in the demands of the environment. Others have argued that changes in organisational practices can, over time, change the cultural-cognitive and normative pillars of institutions. Thus, for example, Scott (Reference Scott2008b: 196) argues that the gradual shift by US doctors from membership of the American Medical Association to individual speciality associations resulted in a significant loss of professional consensus. Yet others have argued that change arises from mismatch between the regulatory, normative and cultural-cognitive aspects of the institution. Caronna (Reference Caronna2004) examines the perceived failings of the US health system, and argues that moves towards a focus on managerialism, managed care and standardised treatment protocols has generated a significant tension between new regulatory and cognitive-cultural pillars and the traditional medical normative assumptions about autonomy and individualised care. In other words, she seems to be suggesting that, in this case, failure to bring about change in the normative pillar of the institution alongside the other changes generated a situation in which the institution as a whole became dysfunctional. Olivier (Reference Oliver1992) provides a list of ‘antecedents’ for ‘deinstitutionalisation’, including, for example, internal power reallocations, increasing workforce diversity and environmental competitive pressures. However, it is also clear from the wider literature on organisational stability that these factors will not necessarily generate significant institutional change in every case; what remains obscure are the contexts in which particular internal or external environmental changes affecting institutions will impact sufficiently upon the entrenched institutional pillars to allow desired change to occur (or, indeed, to precipitate undesired change).
Against this background, how might the lack of ‘fit’ that we have identified between NHS institutions and the notion of commissioning play out in practice? Firstly, we found some evidence that the presence of well-established Foundation Trust Hospitals, taking a more commercial approach, might begin to disrupt the collegiality associated with notions of a shared NHS family. The current new legislation mandates that all Trusts should become Foundation Trusts; it remains to be seen whether this will accelerate or dilute this effect, and whether it will in turn have knock-on effects upon the other institutional factors that make commissioning difficult. On the other hand, there has been a strong backlash from both professionals and the wider public against the idea of competition as a key organising principle for the NHS (O'Dowd, Reference O'Dowd2011). This suggests at the very least that our observation of deep-seated cultural-cognitive assumptions about the importance of integrating care around the individual rather than packaging it out to different providers is borne out more widely. This may in itself be enough to frustrate attempts to revitalise commissioning. Secondly, politicians and commentators have recently suggested that GPs should share their new commissioning responsibilities with hospital consultants and others (Health Select Committee, 2011). This suggests that the entrenched cultural-cognitive assumption that we observed that it is the legitimate role of providers to determine what care should be offered (rather than responding to the needs and demands of commissioners) holds even beyond the boundaries of the NHS, again acting to frustrate or limit change. Overall, it seems as if those aspects of the current proposals most likely to impact upon the normative and cultural-cognitive pillars of the NHS institution (for example the enforcement of competition) are those which have, so far, attracted the most effective opposition. Thus, it would seem that entrenched NHS institutional elements are acting to modify the proposals. However, it remains early days, with the legislation still not enacted, and it will be a number of years before we will know whether the lack of ‘fit’ that we have identified acts to generate significant institutional change or whether the policy itself founders.
Acknowledgements
We are grateful to all our participants, who were generous with their time and welcomed us in to their organisations. We are also grateful to George Dowswell, who undertook some of the fieldwork for Study 1. Study 1 was funded by the National Primary Care Research and Development Centre from its core Department of Health grant. Study 2 was funded by the NIHR Service Delivery and Organisation programme (grant number 08/1808/240).
The views and opinions expressed therein are those of the authors and do not necessarily reflect those of the NIHR SDO programme or the Department of Health.
