NHS England’s decision to consult on new contracts for accountable care organisations (ACOs), alongside the two legal challenges that have been mounted on the status of these potential organisations, have brought the somewhat niche world of NHS commissioning firmly into the public eye.
While the focus of the debate has remained on whether or not these new organisational arrangements will result in increased privatisation within the NHS, there has been less attention on the impact they may have on the nuts and bolts of NHS commissioning.
Yet ironically, given concerns about privatisation, the development of place-based integrated care arrangements – in the varying forms of accountable care systems and organisations – has been taken to signal a shift away from the purchaser/provider split policy that dates to the early 1990s. This raises some fundamental questions about the future of commissioning and where decision-making will sit in a reformed system.
A recent review of strategic (health care) purchasing in 10 European countries suggests it might be time to take a different approach to policy than investing more effort to develop strategic purchasers. The paper does, however, acknowledge that there are still some elements in the role that have value.
In talking about the development of organisations and systems providing integrated care, NHS England chief executive Simon Stevens has suggested that their development could lead to the end of the purchaser/provider split. But the reality is more complex than that, as we will see in this long read. As their name suggests, ACOs will need to be accountable to some kind of body with strategic oversight, so it is unlikely that the NHS at the end of this process will be as it was in 1988, as some people might like to imagine.
What is commissioning?
The term encompasses a large number of different activities: from detailed transactional activities associated with service planning and contracting, right through to strategic oversight and system leadership. Within each of those broad descriptions lie many separate functions undertaken at different levels in the system. Regular reorganisations of commissioning in the NHS have added to the lack of clarity about what the functions are and where in the system they sit.
What are ACSs and ACOs?
The complexity of NHS commissioning has recently been amplified with the implementation of place-based accountable care systems. Two main models seem to be developing, although within them there is huge variation.
In broad terms, an accountable care system (ACS) is a group of organisations working closely together in an area. Working arrangements might be underpinned by a memorandum of understanding, but are largely based on partnerships and agreements between different statutory organisations, each of which will hold individual contracts with the commissioner. An accountable care organisation (ACO) is a more formal structure than an ACS, and is likely to involve several organisations merging to form a single care provider.
In the refreshed planning guidance for 2018/19, NHS England and NHS Improvement state that they “are now using the term ‘integrated care system’ as a collective term for both devolved health and care systems and for those areas previously designated as ‘shadow accountable care systems’. An integrated care system is where health and care organisations voluntarily come together to provide integrated services for a defined population.” In this long read, I will use the term integrated care system (ICS) unless I am referring to an issue that is specifically relevant to ACOs.
Although several forms of these arrangements are developing, they essentially all seek to give greater powers to care providers to allocate resources and design services that deliver maximum value for the population in the geographical area they cover. Associated with this is some level of risk transfer, which places responsibility on providers like hospitals and collections of GP surgeries for ensuring that care for the local population is provided within the funds available.
In this long read, I explore the risks and opportunities for commissioning that are likely to arise as we move towards integrated care – drawing on my own observations of the near 30-year history of commissioning, discussion from the Nuffield Trust’s learning group for clinical commissioning group leaders, and the available evidence on how other countries have fared.
NHS commissioning to date: a complex and imperfect arrangement
My first observation is that, over the years, commissioners have been told by national bodies that they are expected to be the system leaders. This has often meant they have replicated the roles of regulators and providers, and have for example often involved themselves in the detailed issues within hospitals from performance management to infection control. Indeed it’s often appeared that commissioning has just continued old-fashioned hierarchical control by other means.
This approach may partly explain why, compared to similar strategic purchasing functions in other countries, NHS commissioners have been overly concerned with the detailed management of care providers. This has, it seems, meant they have often become too involved in defining pathways, care processes and matters of detail that they generally knew less about than the providers.
The limited effectiveness of the market
While the tools of market-based systems – including payment systems, contracts and specifications – promised to streamline processes and drive up efficiency, their effectiveness seems to have been limited. They can also be costly, with little evidence of a return on investment for the effort expended. In some areas it has appeared to me that the tools of market-based systems are being used to enforce changes such as price reductions, provider mergers or other improvements that could be achieved more effectively, quickly and cheaply through more direct management methods.
Multiple payment incentives and large numbers of complex contracts run the risk of diluting the intended impact – the delivery of improved outcomes for the population. This complexity can create significant and risk diverting people’s efforts. Complete contracts (i.e. those that specify in detail how all eventualities are dealt with) are difficult to agree in complex and dynamic areas such as health care, and so relationships matter more than they would in simpler environments. I fear that traditional commissioning has sometimes focused too much on the need for certainty, rather than on the importance of relationships and trust to allow uncertainty to be managed.
A further issue I have observed is that the capacity of commissioners to use contractual tools effectively (and of providers to respond) has also often been constrained, due to the difficulty of letting providers fail and of dealing with stranded costs and the requirements of a balanced annual budget. This has made big shifts in activity difficult for NHS commissioners – meaning they have been more exposed to risk than is justified by their financial resources. Large payers in other countries can often afford to be less concerned with the viability of cost problems of individual providers.
Beware losing the grit in the oyster
We have, however, learned from history and other systems that the absence of any sort of purchaser/provider split runs the risk of losing sight of population health goals, or at the very least of being captured by the providers and the short termism that tends to infect many health systems.
It is worth reminding ourselves that the purchaser/provider split was first set up partially in response to a system becoming captured by providers and failing to focus on the needs of the population. While there are compelling arguments in moving further from the way the current purchaser/provider split operates, not least that we have not in fact succeeded in addressing that risk of short termism, it would seem unwise to completely do away with a population perspective that is separate from that of the providers’.
Commissioning in integrated care systems: greater freedom to plan strategically?
The big challenge and opportunity that comes from the development of ICSs, and in particular ACOs, is that many of the transactional activities that commissioners currently carry out may be more logically located with integrated care providers. These could include: the detailed specification of pathways, the monitoring of day-to-day activity, most provider contracting and procurement, the operation of detailed payment systems, the management of volume risk and a significant amount of the planning of the provider network.
Importantly, there is potential under an ICS model for commissioners to be freed up from the annual accounting cycle to look strategically over the longer term. They will still play an important role in holding providers to account but, instead of the multitude of process measures used in the past, there is an opportunity for a move towards outcome measures that ensure the vision is realised.
By creating a system where the detailed transactional activities are shifted towards provider entities, there is an opportunity for commissioners to put greater focus on setting a vision for population health.
Holding providers to account: setting standards
ICSs are likely to operate with long-term contracts – in line with practice in other countries that also reduces the amount of transactional work required of commissioners and allows time for investments in change to pay off. The length of these contracts is a key concern for the individuals and organisations bringing legal challenges to the development of the ACO contract, as in the ACO model the majority of provider eggs will be in one basket – increasing the potential risk should that provider fail.
It seems likely, regardless of the length of contracts, that commissioners will still have a key role in developing a detailed understanding of current and future needs in the population they serve, and for setting the outcome measures they want to deliver. There may be some process measures that are also important, but they should use these sparingly. They are also important as a mechanism for holding the systems to account for their performance.
It seems to me that within an ICS, commissioners will need to set some system rules about behaviours and how the local system operates, and go beyond the minimum standards set by regulators, whose work they should not duplicate.
There may be standards for some services, such as:
- minimum procedure numbers for centres and operators, such as for major surgery
- clinical back up to support some specialist care.
System rules that commissioners might need to set may also need to cover elements of provider structure and behaviour. The development of place-based integrated care organisations carries a significant risk that ‘make-or-buy’ decisions are distorted by provider rather than patient considerations, leading to effective monopolies in local areas. We are seeing a bit of this in the way that some STPs are repatriating work from other areas or elective surgery from independent sector providers. This has nothing to do with patient choice and a lot to do with covering local provider overheads.
This monopolistic behaviour may be rational but it is short-sighted – in elective surgery, diagnostic imaging, talking therapies, physiotherapy and a range of other services, it is often beneficial to have patient choice and a diverse ecosystem of providers. Strategic commissioners working in an ICS model should seek to set some system rules to cover this and other potentially adverse effects of a local geographical monopoly.
This also means that much of the dialogue with the public about operational choices – how services are organised, where they are located, how they operate – will need to be conducted by providers in the ICS. This dialogue may be backed up by the commissioners, as they remain accountable for the decisions about which providers should be commissioned to deliver their care. In some cases it may need to be led by commissioners to deal with the legal maze the 2012 Act created (due to the duty to consult resting with the clinical commissioning group and local HealthWatch).
But commissioners’ conversations with the public should be more about understanding need, and defining outcomes and values that show how that need is being met. Without more explicit role clarity, there is a risk that the already limited and unclear accountability of commissioners to the public is further weakened.
Small isn’t always beautiful
Generally it seems that, under the current system, smaller commissioners are a better size to develop relationships with primary care than bodies covering large populations. But they have difficulties matching the power of large hospitals and may lack the analytical firepower to manage population health effectively, not least because of the cost ceiling they have been given by NHS England.
The structure has also led to a replication of effort and decision-making that is costly and difficult to defend. For instance, it is not clear why each CCG needs to develop their own policies on BMI (body mass index) and smoking status for elective surgery. Local sensitivity and priorities are important but can be overstated in justifying variation.
The lesson from other health systems is that there are significant economies of scope and scale available to strategic commissioners. The analytics, actuarial methods for budget setting and other functions are scalable.
There has been very substantial consolidation of payers in the USA, Germany and the Netherlands, and there has also been a move to larger administrative units in other countries (such as Denmark, Finland and Norway). Others like Spain, Portugal, Canada, Australia and New Zealand operate at larger levels of aggregation than the NHS does, but with a less powerful national layer, and even Sweden with its long history of county-level governance has created a regional level. It is, however, worth noting that generally one to five million people seems, in my view, to be the scale at which relationships can be developed and sustained, and people held effectively to account.
As the NHS develops the ICS model, consideration should be given to the most appropriate scale of organisations undertaking strategic commissioning. There will be less need for them to be so rooted in local areas, so there is potential for them to become larger organisations that operate at a higher level to existing CCGs.
There is a risk too that these provider bodies become significantly more powerful than the commissioners who will be tasked with setting a strategic vision and holding them to account for it. Given how commissioners have struggled in the past to overcome this imbalance of power, it will be important that the commissioning function in the newly evolved system has sufficient capacity and capability to hold provider groups to account. Creating larger-scale organisations may go some way to redressing the historical imbalance of power.
Effective strategic commissioning is still very much dependent on high-quality relationships, regardless of area size. One of the striking features of some of the small countries or large regions I have worked with is that it is possible to get all the key senior decision-makers around a reasonable-sized table. Getting these relationships operating at the most senior level, rather than being mediated through middle-level managers or finance specialists, removes some of the transactional nature of the relationships and could allow a higher quality and more strategic conversation.
This does not mean that the answer is another top-down reorganisation. It has been pretty much demonstrated that one-size-fits-all organisational redesign and big bang change is at best not very effective in the NHS, and at worst is destructive.
It is going to take time for the ICSs to develop, and some of them may not work at all. In some cases, there will be opportunities to experiment with forms that create joint strategic commissioning with local authorities. In others, different approaches will emerge. For example, I was at a recent conference hearing about place-based integrated models being developed by CCGs, local government and providers, which were very interesting but looked a lot like planning-based approaches found in New Zealand or Scandinavia.
In other places, very transactional and adversarial approaches remain with a lot of effort and cost going into shifting the financial problem from one part of a system to another through procurement, commissioning coding reviews, tearing up elements of contracts and other transactional tactics. I have not yet seen a convincing argument that using contractual mechanisms is a cost-effective way of addressing these issues.
CCGs adopting these practices, and providers using their equivalent methods, run the risk of destroying value and relationships, and in doing so jeopardising the long-term success of their local system. However, given that a transactional purchaser/provider split has governed relationships in the NHS for nearly 30 years, it is important to acknowledge that it may take time for mindsets of both providers and commissioners to make the shift away from an adversarial approach.
A more evolutionary approach seems to be required, but there will be some difficult tasks involved in this. There is a lot to do to identify the indicators the new integrated providers need to demonstrate they are delivering population health, and how these will be used by the commissioners to hold them to account. It’s easy to talk glibly about outcomes or value-based purchasing, but actually turning these ideas into concrete and useable instruments has not proved straightforward.
Another task is to understand what functions are required, how these will work and where they will sit. Even more difficult will be to develop the relationships, skills and ways of working that will create a new approach to how systems work together, not to mention the technical issues of how to contract for or procure services.
How to create some space to do this when the whole working week is often consumed by immediate issues and a profusion of meetings is a very significant challenge. The NHS can be surprisingly quick but it does seem likely this may take longer, and be significantly more difficult than the policy-makers assume. As with previous changes, they are in danger of repeating the pattern of spending most of the intellectual effort and money in supporting the provider development required to create ICSs, but without matching investment to create a capable commissioning function to match.
The next few months will see considerable public debate about the move towards ICSs in England. An underexplored issue will be how to ensure the parts of commissioning that are about creating improvements in population health, as well as involving the public and ensuring health services are fit for the future, can function effectively.
Edwards N (2018) "Integrated care: what does it mean for commissioning?", Nuffield Trust long read. https://www.nuffieldtrust.org.uk/news-item/integrated-care-what-does-it-mean-for-commissioning