Article Text


Views of NHS commissioners on commissioning support provision. Evidence from a qualitative study examining the early development of clinical commissioning groups in England
  1. Christina Petsoulas1,
  2. Pauline Allen1,
  3. Kath Checkland2,
  4. Anna Coleman3,
  5. Julia Segar3,
  6. Stephen Peckham3,
  7. Imelda Mcdermott3
  1. 1Department of Health Services Research and Policy, London School of Hygiene and Tropical Medicine, London, UK
  2. 2Centre for Primary Care, University of Manchester, Manchester, UK
  3. 3Centre for Health Services Studies, University of Kent, Canterbury, Kent, UK
  1. Correspondence to Dr Christina Petsoulas; christina.petsoulas{at}


Objective The 2010 healthcare reform in England introduced primary care-led commissioning in the National Health Service (NHS) by establishing clinical commissioning groups (CCGs). A key factor for the success of the reform is the provision of excellent commissioning support services to CCGs. The Government's aim is to create a vibrant market of competing providers of such services (from both for-profit and not-for-profit sectors). Until this market develops, however, commissioning support units (CSUs) have been created from which CCGs are buying commissioning support functions. This study explored the attitudes of CCGs towards outsourcing commissioning support functions during the initial stage of the reform.

Design The research took place between September 2011 and June 2012. We used a case study research design in eight CCGs, conducting in-depth interviews, observation of meetings and analysis of policy documents.

Setting/participants We conducted 96 interviews and observed 146 meetings (a total of approximately 439 h).

Results Many CCGs were reluctant to outsource core commissioning support functions (such as contracting) for fear of losing local knowledge and trusted relationships. Others were disappointed by the absence of choice and saw CSUs as monopolies and a recreation of the abolished PCTs. Many expressed doubts about the expectation that outsourcing of commissioning support functions will result in lower administrative costs.

Conclusions Given the nature of healthcare commissioning, outsourcing vital commissioning support functions may not be the preferred option of CCGs. Considerations of high transaction costs, and the risk of fragmentation of services and loss of trusted relationships involved in short-term contracting, may lead most CCGs to decide to form long-term partnerships with commissioning support suppliers in the future. This option, however, limits competition by creating ‘network closure’ and calls into question the Government's intention to create a vibrant market of commissioning support provision.


This is an Open Access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See:

Statistics from

Strengths and limitations of this study

  • This is the first study which discusses attitudes of clinical commissioning groups (CCGs) towards commissioning support provision.

  • Evidence was gathered from eight case studies across England selected to provide geographical spread and a good mix of emerging models of CCGs.

  • Evidence from in-depth interviews was supplemented with observation of a variety of meetings and analysis of relevant documents.

  • Although the findings reported are from the early phase of clinical support provision, evidence from ongoing research on the development of clinical commissioning groups indicates that the arguments for and against outsourcing commissioning support functions are still relevant.


The National Health Service (NHS) was established in the UK in 1948 to provide universal healthcare coverage, free at the point of delivery and funded out of general taxation. Since then, it has been subject to many restructurings and varying levels of funding. In 2010, the Conservative-Liberal Democrat coalition government embarked on a large scale organisational overhaul of the NHS in England, intended to cut the costs of bureaucracy by more than 45% and achieve productivity improvements of £20 billion by 2015.1 This reorganisation took place only in England rather than the whole of the UK, as Scotland, Wales and Northern Ireland have devolved healthcare systems.

Since the early 1990s, public policy analysts have been drawing attention to the rise of the ‘contract state’, ‘government by contract’, or ‘contractual governance’.2–4 These terms are closely related to what is known in the literature as new public management (NPM), a shorthand description of the importation of management techniques from the private into the public sector and the use of market forces such as competition and contract as a way of increasing efficiency and choice in the delivery of public services.5–7 In the UK healthcare sector, NPM started with the introduction of ‘managerialism’ in hospitals in the early 1980s.8 This was followed in the early 1990s by the introduction of the ‘quasi’ or ‘internal’ market involving a split between purchaser and provider.9 ,10 Faithful to the NPM belief that markets rather than bureaucracies are more likely to achieve efficiency, the reorganisation of 2010 emphasised the use of more competition, private sector involvement and contracting.

One major focus of the 2010 restructuring is the function of commissioning, which includes assessing local health needs, prioritising the allocation of resources, procuring and purchasing healthcare services, and monitoring contracts. Until April 2013, the state allocated funds to primary care trusts (PCTs) which were charged with commissioning primary, community and secondary care services. Arguing that general practitioners (GPs), as the gatekeepers of healthcare provision, should be in charge of commissioning, the Coalition Government made GPs responsible for commissioning most non-specialist secondary care, community and mental health services. In April 2013 the managerially-led PCTs were replaced by clinically-led clinical commissioning groups (CCGs).11

Commissioning healthcare requires highly specialised skills and resources. The new Labour government (1997–2010) had encouraged PCTs to buy in necessary specialist external support from the private sector.12 ,13 A subsequent House of Commons Health Committee, however, raised concerns about the value for money these arrangements offered, especially in the context of the economic downturn.14

The Coalition Government announced that CCGs would have access to a wide range of commissioning support expertise, with the 2010 White Paper suggesting that CCGs would be free to decide for themselves how to obtain commissioning support, choosing suppliers from an emerging market.15 In reality, however, their options were heavily circumscribed. Contrary to policy rhetoric, a market in commissioning support provision was slow to emerge, which left CCGs with fewer choices.16 ,17 In addition, soon after the introduction of the reforms, many CCGs expressed uncertainty and unease about outsourcing commissioning support services, fearing that valuable local knowledge and trusted relationships would be lost.18 CCG reluctance to outsource commissioning support services meant that policymakers were forced to adapt policy plans, with new guidance issued in 2012.19 ,20 Of all the pieces of the health reform jigsaw puzzle, commissioning support arrangements would seem to have fitted least easily into place. This paper discusses the attitude of CCGs towards commissioning support provision by using data from a project which explored the early development of CCGs. The main aim of the study as a whole was to examine the early experiences and challenges faced by CCGs (including putting in place adequate commissioning support provision) as they set themselves up and moved towards authorisation. Commissioning support emerged as a key issue in all of our sites, and in this paper we focus on this evidence, seeking to answer the following questions:

  • How did CCGs react to the idea of buying in commissioning support?

  • What factors affected their choices?

  • What lessons can be learned for their future commissioning support needs?

Development and future of commissioning support units

The process and conduct of the 2010 NHS reforms have been described in detail elsewhere. For a summary, see the full project report.21 By April 2013, a total of 211 CCGs had been authorised and took over responsibility for commissioning.

Early guidance suggested that CCGs would be able to choose how to carry out their commissioning responsibilities, including buying in commissioning support from external providers such as local authorities, private companies and ‘third sector’ organisations.22 In late 2011, however, as it became apparent that a marketplace of commissioning support providers was slow to develop, the Government established commissioning support units (CSUs), staffed by former PCT commissioning managers and intended to remain within the public sector until April 2016.19 ,20 CSUs developed into large regional organisations, seeking to maximise economies of scale by providing services to many CCGs. The services they could potentially provide range from business support (eg, financial planning, Human Resources, IT) to support with the commissioning cycle (eg, health needs assessment, clinical pathway redesign, healthcare procurement, contract negotiation and management, data management, business intelligence) to clinical support (eg, medicines management, continuing care, complex case management). Although CCGs have been firmly in place since April 2013, commissioning support provision (including CSUs) is still in a state of flux. By January 2014, the 23 CSUs originally approved were reduced to 17 as a result of mergers. This number is likely to be reduced further as more mergers were announced at the time of writing.23

At the beginning of 2014, NHS England announced plans to ‘externalise’ or make CSUs independent from the NHS. ‘Based on feedback from extensive engagement with stakeholders,’ four potential organisational forms for CSUs were identified24:

  1. Social enterprise (taking the legal form of a community interest company limited by guarantee)

  2. Staff mutual, which will abide by the seven principles of the cooperative movement (taking the legal form of an industrial and provident society)

  3. Customer controlled social enterprise, which would be akin to the in-house department of the customers (taking the legal form of a community interest company limited by guarantee)

  4. Joint venture, which would be formed by bringing in other parties in order to raise the value for money and quality of the CSU's services (taking the legal form of a company limited by shares)

The option of retaining CSUs within the public sector was rejected, as was the option of transferring CSUs fully to private providers. According to NHS England, the approach taken about the future of CSUs represents the right balance between making them autonomous while at the same time protecting the public interest.24

Make, buy, ally

Whatever the form of CSUs, CCGs have, broadly speaking, three options when it comes to choosing commissioning support provision: ‘make, buy, or ally’.25 ,26 Specifically, one option is to keep most commissioning expertise in-house (make), buying in additional short-term support from external providers ad hoc. However, CCGs have been provided with a relatively small allowance to pay their running and managerial costs,27 so this is not a viable option for most groups. Alternatively, they can outsource most of the commissioning support functions, either through short-term contracts and frequent testing of the market (buy) or through long-term partnerships with commissioning support providers (ally). The decision will depend on factors such as the size of their organisation and the availability of market competition.28 ,29

Although the public sector has always bought products from the private sector through public procurement and supply chain management (SCM), the provision of public services through contracting out has intensified since the introduction of market reforms.28 ,30–34 SCM theories have been applied specifically to the healthcare sector but they have so far focused on the US model rather than the English NHS.35 The recent policy reforms, however, make the public procurement literature more relevant to NHS commissioning. Buying in external support will inevitably create a more complex picture, as CCGs deciding to outsource commissioning support will enter into contracts with commissioning support providers, who will in turn be managing contracts with healthcare providers on behalf of their clients (ie, CCGs). An additional layer of contract monitoring will therefore be required, since an additional ‘principal-agent’ relationship is being created. This new layer of contracting may magnify the problems already associated with contract monitoring in healthcare.36–39

Deciding whether to make or buy is not easy. If organisations provide services in-house (‘make’), transaction costs are thought to be kept low compared to contracting out, because there are fewer incentives for opportunistic behaviour and therefore less need for comprehensive (and expensive) monitoring of outcomes. On the other hand, there are some drawbacks, such as the possibilities of missing out on competitive prices, economies of scale, specialist expertise and opportunities for switching between suppliers in case of non-performance.40 Subcontracting through competitive tendering mitigates these risks by enabling choice from a wide variety of competing providers. However, in such contracts the incentives for purchaser and provider are not necessarily aligned, producing a potential loss of service quality. For example, the supplier may compromise the quality of the product in order to secure the contract by lowering costs. Mitigating this risk involves expensive contract monitoring processes.41

A possible solution to this problem would be to put in place longer-term contracts (eg, 3 years) characterised by relations of trust and aligned incentives. Such a model has been described in the literature as ‘supplier partnership’, ‘partnership sourcing’40 ,42 or ‘alliancing’.43 This model suggests forming alliances with a single supplier on the basis of a long-term relationship of loyalty, cooperation and trust. The ‘customer-controlled social enterprise’ and the ‘joint-venture’ options for CSUs can be seen as being different examples of this model. Within such partnerships the supplier works closely with the purchaser towards continuous product development.40 Advocates claim that such partnerships generate close working relationships within ‘networks’,44 which are claimed to be ‘“lighter on their feet” than hierarchies’, with gains available from the pooling of resources.45 Furthermore, such long-term partnerships reduce the risk of opportunistic behaviour by suppliers, as both parties invest in shared long term goals and develop trust.46 ,47 Many policy analysts, however, have also highlighted limitations of a network form of governance, suggesting such organisational forms are prone to power asymmetries, a leadership deficit and the potential onset of dependencies and rigidities.48–50 Others caution against the ‘dark side’ of supplier partnerships, warning that they may easily result in a lack of flexibility in adapting to changing environments, thus adversely affecting their performance.51 Too much trust on the part of the buyer may lead to lack of vigilance in monitoring performance, which can in turn result in complacency and lower standards. It seems, then, that successful supplier partnerships need both close collaboration between buyer and supplier and a frequent reassessment of their relationship. It is not easy to strike the right balance between frequently questioning the competence of the supplier and maintaining a trusting relationship.

Establishing long-term public-private partnerships has been a key theme of strategic public procurement in the UK, but research suggests that such partnerships have not been easy either to establish or to maintain, with problems arising from cultural differences between public and private sectors, budgetary constraints and differing attitudes to risk and innovation.31 ,34 Compared with contracts operating in commercial markets, public sector contracts in quasi markets such as the NHS are subject to greater levels of public control, operating within a hierarchical regulatory framework intended to ensure the promotion of distributive justice and safeguard the public interest in addition to increasing efficiency and choice.52

It is thus clear that the key assumption underpinning recent changes to commissioning in the English NHS—that a competitive market in outsourced commissioning support will by definition be more efficient and effective—cannot simply be taken for granted. It will be some time before we can definitively assess the new arrangements, but it is valuable to explore the emerging landscape at this early stage in order to provide evidence to inform the ongoing development of commissioning support services.


This paper uses evidence from a national study of emerging CCGs which was funded by the Policy Research Unit, Department of Health. The research took place between September 2011 and June 2012. Our methods are described in full in the project report.21 A case study research design with eight in-depth case studies was supplemented by descriptive information from two web surveys (one conducted in December 2011 and the other in April 2012). Case study sites were selected to provide a maximum variety sample across a number of domains, including: size; sociodemographic profile; presence/absence of formal federation between CCGs; the number of main providers with which the CCG interacted; the number of Local Authorities with which the CCG interacted; and how far the CCG represented the recreation of a previous administrative grouping.

We conducted 96 interviews with a variety of CCG staff (NHS managers, GPs, lay members and practice managers). Interviews were recorded and transcribed after obtaining written consent from participants. We observed 146 meetings (a total of approximately 439 h), including: governing body meetings; executive or operational group meetings; meetings of GP members; and locality meetings. Researchers recorded detailed field notes during meetings, and these were analysed alongside the interview data.

This approach enabled us to move beyond the personal viewpoints of our interviewees to also observe what actually happened in practice as the developing groups wrestled with the complex situation that they faced. We take the view that interview data alone are insufficient to illuminate complex questions such as this. We therefore combined exploration of the research questions in interviews with ethnographic observation of meetings at which commissioning support was discussed, and examined documents produced by our case study sites as well as national policy documents. While this could legitimately be called ‘triangulation’, we regard it as an opportunity to develop and refine our findings, rather than as a simplistic test of validity. Data from all sources were stored and managed with the assistance of Atlas.ti software, providing a medium through which research team members were able to work together on the analysis. Transcripts and fieldnotes were read repeatedly for familiarisation, and coded using a framework based on our research questions, our knowledge of the literature in this area and from our reading of relevant policy documents. In addition, inductive coding allowed us to capture unexpected themes. Frequent team meetings were held, at which coding definitions and emerging theoretical ideas were discussed and refined. In addition, the team produced and updated ongoing summaries of the case study sites under headings derived from the developing analytical framework, and these were discussed at the team meetings. This allowed us to maintain ongoing cross-case comparisons, and aided in the handling of such large amounts of data. Emerging analytical ideas were set out in written memos, and these were tested among the research team members and refined. Coded data were then further read and analysed by a number of team members in order to ensure consistency of approach, and the PI repeatedly read the whole data set in order to further refine and develop the emerging analysis.


For CCGs, the most pressing question was deciding whether to ‘make or buy’ commissioning support functions. It is important to note here that the option of ‘ally’ was not considered by CCGs at the time of the research, since the uncertainty surrounding the future of CSUs made difficult the negotiation of longer term contracts. In addition to the ‘make or buy’ options, CCGs could establish federated structures by pooling managerial resources and sharing commissioning support (CS) provision.

Some of the dilemmas CCGs faced at this time were due to uncertainty about policy expectations and the sheer novelty of the reforms. As already mentioned, while CCGs were preparing their applications for ‘authorisation’, CSUs were also going through a process of assurance. One of the problems facing CCGs was the fact that, as new organisations, they had not had time to become ‘informed customers’ and they also lacked detailed knowledge of their requirements. Since neither CCGs nor CSUs had established themselves as organisations, a ‘chicken and egg’ problem was apparent: CSUs were expecting CCGs to tell them what functions they wanted to commission, while CCGs were expecting the CSUs to tell them what services they offered. One participant summarised the uncertain situation in which CCG found themselves: We are nowhere ready to go out to some kind of competitive procurement of these services, because we can't even articulate what we want, never mind how much it should cost, or who else is out there who can supply it. Because the other thing which is true of course is there is no market place. We couldn't buy this if we wanted to because it doesn't exist. [Manager ID 244]

Despite initial government rhetoric and enthusiasm about the choices opened to CCGs by a competitive market in commissioning support, CCGs soon discovered that the only game in town was some version of amalgamation of former PCTs.

Make, buy or share?

A number of issues highlighted by participants were more general and related not so much to the timing as to the nature of the changes. Many discussions across the study sites focused on deciding which commissioning support functions should be retained in-house, which should be contracted out and which should be shared with other CCGs.


CCGs have been provided with a limited managerial budget of £25 per capita (which totals significantly less than the average running cost of their predecessor organisations). Many participants told us that their initial intention was to employ most of the commissioning staff in-house, but this decision had to be revisited once the managerial budgets were announced. …part of the reason why we still haven't worked all this out yet, is because we thought that as a large CCG we'd be doing a lot of this in-house anyway, so it came to us fairly late in the day that we would have to start thinking about externalisation of, of a lot of this. [Manager, ID 171]

One of our larger sites decided to keep the CS functions in-house. One reason for this was the absence of a developed market in CS services but additional arguments were also important. First, members of the CCG were concerned that they would lose trusted local managerial personnel. Participants highlighted the importance of relationships and the sense of ‘knowing who you are dealing with’, being in control rather than having to interact with a remote regional provider. One participant at a different site expressed concerns about this: ‘It's going to become more and more difficult, it's going to be more and more lengthy to get things done, I think. Because you don't have that shorthand. [you can't just pick up a phone and say] ‘Hi, so and so, you know you did that?’ ‘Can I have a word’ You know? It's not going to be like that, it's going to be fill in a form, or you speak to somebody you might even never have met. [Manager ID 122]

A second reason, repeated by several CCGs, was the importance of detailed knowledge about unique local problems and the difficulty of interpreting generic data produced by a distant CS provider in the absence of local knowledge. If we buy in a performance system from...could be anywhere...they don't have the ability to interpret what that intelligence says, and I think if you don't have that local interpretation...I think you're gonna lose the richness of the information, however slick it is.’ [Manager ID 196]

But even the CCG that decided to retain most of the commissioning support functions in-house made it clear that they would keep an open mind and review the provision of those functions on a regular basis retaining the flexibility of outsourcing them at a later stage. But the caveat that the measures envisaged would include is, say, that isn't a licence that there'll be no further changes, and that there will be an opportunity to review on an annual basis the appropriate functionality of the commissioning support of the CCG.’ [Manager ID 63]

As the literature shows, however, constantly checking for alternative providers can create feelings of uncertainty and anxiety in existing personnel. One participant mentioned the need to strike the right balance: Because if every year they [CCGs] market test 20% of their services and outsource 10% of it, as soon as you tell the staff that we're market testing your area they know that there's a one in two, or 50–50, whether or not they'll have a job after it. So you introduce ambiguity and anxiety when you've just gone through all of that...There is a balance to be struck's incumbent on the commissioning support staff to deliver the best for the CCG, because in that lies the security of continuing employment. [Manager ID 63]

This suggests that the fear of losing out to a competitor may be sufficient to prevent complacency on the part of in-house providers.


Deciding to provide all commissioning support in-house was the exception in our case studies, and in the country as a whole. Large CCGs had the resources to have CS functions in-house but for smaller CCGs outsourcing was the only option, although many also planned to employ a small core staff in-house. In the absence of a market including private providers, CCGs looked to the emerging regional CSUs. Participants expressed a mixture of views on this development. Many were reluctant to outsource all CS functions for fear of losing control and ownership of the process as well as losing ‘collective memory and experience’. Others felt that they were being forced to outsource by top-down policy expectations. We certainly planned to have a lot of our staff employed in the CCG, and, that's not going to be the case…We have spent a lot of time building up local relationships in the localities, and bringing people in from elsewhere just isn't the same …That's the policy, we have to work with it and make the best of it, but it would have been better if we could have had more staff …working with us directly. [Manager ID 193]

Yet others were excited at the prospect of being able to shop around for CS providers rather than being dependent on the limited skills offered by PCTs. Instead of being ‘stuck’ with a CSU which might look like the old PCT, some participants welcomed the freedom to ‘pick and choose’ rather than buy all their CS functions from the same provider. Some of the conversations I have with my colleagues, it's like...some of them can't wait to free themselves from the PCT and I say, but surely, this is an opportunity to make this work, not go into it thinking it's going to fail and I can't wait to go in contract with McKinseys or something. [Manager ID 114]

Some participants were unhappy with the creation of regional CSUs because they thought it created monopolies and restricted choice. We're now back to a place where in fact what we're trying to do is to build massive shared service organisations involving hundreds of people covering very large areas in a way which won't give the customers any choice at all, they're monopolies. [Manager ID 244]

The CCGs that did outsource to CSUs were careful to avoid long-term contracts in order to maintain flexibility to switch provider when the market matured.

Most participants were willing to unbundle some CS functions but they also insisted on keeping in-house what they considered to be core functions. There was, however, a great variety of opinions about which CS functions should be considered as ‘core’ and which were appropriate for outsourcing. Some participants, for example, thought that financial management needed to be provided in-house but others believed it could be outsourced. Some CCGs deemed that functions like negotiating and monitoring contracts with heath care providers could be easily outsourced. Others thought that, because of the complexity and the importance of building good relationships with local providers, contract negotiation and monitoring should be kept in-house. Acute contracts are notoriously complex, and I think at the moment …how people perceive contracting, they think it's quite an automated process. That it's like you go to the supermarket, you buy a tin of beans … you put it through the till and out pops the bill sort of thing. It isn't like that in the NHS …I'm willing to work with let's have it at an arm's length in terms of a CSU, and there might be some economies of scale …, but if we start losing any expertise or that closeness, because you need to work very closely with people in contracting, then I'll have to say no, it isn't going to happen. [Manager ID 287]

A distinction was also made between ‘relational’ and ‘transactional’ types of CS functions. The data analysis that supports contracting was cited as an example of a transactional function which could be outsourced easily, while contract negotiation was cited as a relational function which had to be kept in-house. Some participants believed that the design of pathways is ‘fairly transactional’ and could therefore be outsourced but others had serious misgivings about the ability of remote external suppliers to redesign clinical pathways in the absence of local knowledge. Others again thought that generic clinical pathways could be designed but then they would have to be adapted to local conditions. There was widespread agreement among participants that ‘back office’ support functions could be outsourced, but such functions had already been outsourced by many PCTs.

Owing to CCGs being left with small managerial teams, participants stressed the need for CS providers to produce CCG-specific information.

Many CCGs in our case studies understood the rationale for creating economies of scale but emphasised the importance of maintaining good relationships and clear lines of accountability between CS providers and CCGs. They also stressed the need for an additional layer of monitoring the performance of CS providers and the associated increase in transaction costs. You have to know that what you're getting from any commissioning support organisation is robust and correct. So that you end up pulling a whole bunch of people to check what they're doing is the right thing, or do you just employ the people and check them yourself? I think we're struggling a bit with that at the moment …Who's checking how many knee operations the provider is charging us for? And you have somebody checking the person who's checking how many. [GP ID 283]

Many CCGs questioned the wisdom of the reforms and in particular the idea that they would result in efficiency gains. Several participants doubted that outsourcing CS functions to centralised and remote CSUs would necessarily save money, because of the associated duplication of work. I know the push from on high is sort of, you know, buy it in from a large national organisation but actually it's not always cheaper to do it that way. I mean, if it's too remote, you end up reinventing it within your locality because you don't trust the organisations providing … So we're employing someone to check what they're doing is right and you don't always get the economies of scale if you make it too remote. [GP ID 282]

Many CCGs feared that the new structures would not be affordable, given the limited financial resources allocated to CCGs for management support.


The issue of limited resources led smaller CCGs to collaboration. Instead of contracting out most CS functions to a regional provider, smaller CCGs selected the sharing option for the delivery of core commissioning support functions. Keen to keep vital CS functions in-house, a few CCGs in geographic proximity to each other agreed to share key commissioning support personnel, reducing thereby their overall management costs. Some CCGs were also exploring the option of sharing commissioning support with their local authority. This meant that, We get the local knowledge but we're reducing the overheads so it actually may be a more cost effective way to share with the local authority and other CCGs than go to a national or regional model. [GP ID 282]

This fits in with the national direction of creating economies of scale but it also allows the cooperating CCGs to keep the functions local and maintain control.


The picture of commissioning support provision is still developing. The preferred model so far has been a ‘hub and spoke’ formation in which large regional CSUs provide services to a number of CCGs. The policy plan is that by the end of 2016 CSUs will become autonomous entities in order to be able to compete in a developing market for CS provision. It is likely that CSUs will become even larger entities as a way of securing their viability. A wave of mergers is already underway.53 At the same time, there are signs that CCGs are becoming increasingly unhappy about contracting out vital CS functions to CSUs and are transferring them back in-house.54 ,55 Fragmentation of services and lack of quality and value for money are reported to be some of the reasons for this move.

The new structures are meant to increase cost effectiveness by creating economies of scale. Whether they really are more cost effective than the old arrangements, however, is not yet known and the issue will need to be subjected to economic evaluation, which is beyond the scope of this paper. NHS England are currently undertaking an impact assessment of CSU autonomisation which is expected to be published in late 2014.56 Recently, public policy analysts have questioned claims that the adoption of NPM approaches brings about significant administrative savings.57 Transaction cost theorists draw attention to high costs of contract monitoring.58 Models which rest on short-term competitive tendering may be less efficient (because of high transaction costs) than models promoting social capital and relationships of trust and reciprocity. On the other hand, integrationist models promoting social capital and trust may prove equally, if not more, inefficient because they result in monopolies, higher prices and a lack of quality improvement.

Respondents in our case studies expressed both these views. Some seemed happy at the prospect of choosing commissioning support functions from a range of competing providers and were disappointed when they discovered that, at least for the first few years, they had to obtain commissioning support from regional CSUs. They felt that CSUs were a form of monopoly. On the other hand, many sites were worried about opening commissioning support functions to competition and saw this as part of a Government plan to privatise the NHS. Several participants also feared that remote commissioning support providers would find it difficult to understand distinct local issues, especially in large rural areas. The majority of participants, however, adopted a more balanced view, according to which there was room for a market in commissioning support functions as long as it allowed CCGs to make their own choices about which functions to externalise. Most participants felt that a number of ‘core’ functions had to be provided in-house. There was, however, lack of agreement about which functions were ‘core’ and which could be externalised.

Importantly, some participants drew a distinction between ‘transactional’ and ‘relational’ commissioning functions and expressed a variety of views either about their definition or about their suitability for outsourcing. NHS England draws the distinction between ‘transformational’ (eg, service or pathway re-design) and ‘transactional’ (eg, market management, healthcare procurement, contract negotiation and monitoring, information analysis) commissioning functions, but emphasises that commissioning support providers ought to be used by CCGs to drive strategic or transformational change rather than merely to deliver the transactional functions.19 Some participants, as we saw, were happy with the creation of national pathways but feared that important local knowledge might be lost. For reasons predicted by the outsourcing literature, many participants felt uncomfortable about outsourcing core commissioning support functions (eg, contract negotiation and monitoring) but were at ease with the idea of outsourcing the transactional type of commissioning functions, like business intelligence and data management.

In addition, CCGs expressed concerns about outsourcing commissioning support functions in general. One important issue is the monitoring of commissioning support providers. At the time of writing, NHS England is overseeing the new commissioning structures, including the development of the procurement skills of commissioners and a ‘lead provider framework’ of accredited CS providers (including CSUs). Even now, however, it is not clear who is responsible for monitoring the work of CSUs. It is even less clear who will be monitoring the commissioning support providers after 2016 when the market is opened up to the private and voluntary sectors. One of the stated aims of the reforms is the creation of more transparency and accountability in the NHS.15 Under the new structures, CCGs are accountable to NHS England but the lines of accountability between CCGs and other organisations, including commissioning support providers, are far less clear.59 Commissioning support providers will be ‘contractually’ accountable to CCGs. But if CCGs lack the resources or skills to monitor them properly then the sense in which they are really accountable to CCGs becomes less obvious. A further danger is that if CCGs lack the means to hold commissioning support providers properly to account, they may become dependent and led by them.

Another issue is whether CCGs will have the resources or skills to scan the market regularly and keep up to date with developments in commissioning support provision. Moreover, CCGs will face the dilemma of monitoring the market at the risk of alienating their current providers. It may not be easy to maintain a good working relationship with commissioning support providers under these conditions. In order to develop truly collaborative relationships, CCGs will have to put in place longer term contracts characteristic of the partnership type of outsourcing. Our discussion of the literature shows that there are advantages to be had in selecting this model. As we saw, however, this model is not free from dangers, the most significant of which is the creation of exclusive networks or ‘network closure’. The creation of long-term partnerships (rather than frequent competitive contract tendering) may also call into question the Government's intention to create a vibrant market for commissioning support provision. Externalising commissioning support provision may yet develop into one further NPM paradox.60 There is a danger that elements of commissioning support which might be delivered better through vertical integration or regional collaboration (eg, contract management or pathway developments), may in the end be externalised, when this could entail the loss of vital relations and local focus, damage to services, and a waste of resources.


This paper discussed the attitude of CCGs towards commissioning support provision during the initial stage of their development. We witnessed a mixed reaction with arguments for and against buying commissioning support from external providers. Provision of commissioning support is still very much an evolving theme in the English NHS and further research needs to be conducted on the issue. Our research highlighted a number of questions that need to be addressed:

  • Does buying commissioning support represent value for money for CCGs?

  • Given the limited resources of CCGs will they be able to hold external commissioning support providers to account?

  • Will CCGs which externalise commissioning support provision be able at the same time to maintain close relationships with healthcare providers?

It remains to be seen whether more CCGs will opt to keep commissioning support provision in-house or form customer controlled social enterprises with CSUs.


The study formed part of the programme of the Policy Research Unit on Commissioning and the Healthcare System. The authors are grateful to their participants for allowing access to their organisations at a time of considerable turmoil and change. The authors wish to thank Dr Andrew Wallace and Ms Ros Miller for undertaking part of the fieldwork and contributing to the analysis of data and final report. Employees of the Department of Health were members of an advisory group who supported the conduct of the research, and commented on an initial draft of the study final report. The findings and views expressed here are those of the authors.


View Abstract


  • Contributors CP was involved in data collection and analysis, produced the first draft and finalised the paper. PA was involved in data analysis and made substantial critical comments on all versions of the paper. KC and AC conceived and designed the project, were involved in data collection and analysis and made substantial comments on all versions of the paper. KC was PI for the project. JS and IM were involved in data collection and analysis and made substantial comments on all versions of the paper. SP was involved in the initial design of the project and made substantial comments on all versions of the paper.

  • Funding The Policy Research Unit in Commissioning and the Healthcare System (PRUComm), is funded by the Department of Health Policy Research Programme.

  • Competing interests None

  • Ethics approval Ethical approval was obtained from NHS Research Ethics Service (study number: 11/NW/0375) and research governance approval was obtained for each case study site.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data sharing statement No additional data are available.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.