Method

Over the course of 6 months between November 2016 and April 2017, the Health Foundation and NHS Providers sought the views and experience of individuals across all types of NHS trusts that are directly involved in, or affected by, the current payment system. This took the form of three workshops and 18 in-depth telephone interviews with people working in the provider sector. In total, 73 people were involved: 12 clinicians, 14 staff in strategic and operational roles, and 47 staff working in contracting or finance. They represented 39 acute trusts, 12 mental health trusts, four community trusts, and two ambulance trusts. Many of the individuals brought years of experience through careers at multiple organisations.

Analysis of discussions from these workshops and interviews was performed using the web-based qualitative analysis software Dedoose. It resulted in a preliminary set of principles for designing and implementing payment systems. These were then tested with an additional 20 NHS staff in a scenario-based workshop, at which the draft principles were mapped against the strategic aims of the provider sector, to ensure they were relevant and applicable. Further refinement following this workshop led to the development of the final eight principles outlined in this report.

Scope

The eight principles reflect the expressed needs of those working in the NHS provider sector. They were asked what a reformed payment system should take into account, and how it could support them in providing quality care in appropriate settings. They were not asked what it should look like – this report does not attempt to outline an ideal payment system.

Acute trusts were slightly over-represented in the sample (39 out of 57 trusts were acute). They made up 68% of the sample, whereas 58% of all NHS trusts in England are acute trusts. However, the principles were tested with – and found to be applicable to – staff from across the range of trust types. The principles underpinning a payment system – unlike the specific payment mechanisms employed – should not vary depending on the type of care delivered.

This report does not capture the views of the whole system: it focuses on the views of NHS trusts providing services to patients. It reflects these views for the benefit of NHS England and NHS Improvement – which share responsibility for the current system – as they consider and reform the payment system for the benefit of providers, commissioners and patients.

The principles are the result of rigorous qualitative analysis, and are closely interlinked. The authors recognise that, consequently, some themes recur across them.

Payment mechanisms are used in health services across England, and further afield. Each mechanism has its own advantages and disadvantages, but an effective payment system will combine complementary payment and non-payment levers, with the optimal mix dependent on the priorities of the system. Payment mechanisms differ in the extent to which they bundle payments for services, and include:

• block budgets, in which payments for all services provided are bundled together, with a lump sum paid to providers at intervals, independent of the level of activity
• capitation, under which bundled payments are made per patient
• case-based payments, in which providers receive a prospective fixed sum for an episode – rather than single instance – of care
• fee-for-service, where providers are paid retrospectively per unit of activity undertaken.

No single mechanism is perfect (the pros and cons of each payment mechanism are discussed in more detail elsewhere). It is common to apply a combination of different mechanisms within the overall health care payment system. This is true for the NHS, where the payment system consists of a blend of methods used across services, incorporating block budgets, capitation and case-based models.

The acute sector: payment by results

Payment by results (PbR) covers the largest segment of NHS spend. Introduced in 2003/04 to cover a small proportion of elective hospital care, it accounted for 60% of the total income received by all NHS trusts and 67% of acute income by 2014/15. PbR is an case-based payment system with nationally set prices for units of care that apply across providers. It therefore supports patient choice by allowing funding to follow the patient to wherever they choose to be treated, within a range of available options. PbR is also applied to emergency care to provide ‘yardstick’ competition, incentivising providers to improve efficiency where patient choice is not possible.

This shift from block to activity based payment in the NHS acute sector at a time of long waiting lists – accompanied by other complementary incentives such as waiting time targets – had the intended impact of increasing activity levels, with a rise in elective spells leading to a reduction in waiting times. Resulting resource savings were estimated to be between 1% and 3% over a 5-year period following the introduction of PbR, with no evidence of a coincident deterioration in quality. This suggests that the introduction of PbR led to improved efficiency.

These findings are consistent with those from other countries that have moved away from block payments to activity based systems. Internationally, activity based payment systems similar to PbR have been associated with increases in life expectancy. However, they do tend to be more complex than some other systems, are costly to implement and run, and make financial control difficult as increased activity is incentivised, which can encourage supply induced demand. They can also discourage joint working with other providers, or investment in initiatives to prevent future ill health – both of which are now national policy objectives.

Through this sort of payment system, providers are incentivised to improve their efficiency – reducing costs while maintaining or improving quality – in the following ways.

• Setting national prices based on average cost means higher-cost providers are incentivised to improve efficiency to reduce cost through yardstick competition. Providers with below-average costs are incentivised to keep them below the average as they will retain the marginal difference.
• The fixed price means that providers must compete based on quality of their service rather than price, thus incentivising cost reductions through improved efficiency rather than reduced quality.
• A national efficiency factor is incorporated preventing the price paid from rising at the same rate as costs, so providers must continuously improve efficiency.

Initially the national efficiency factor was set at a level to encourage efficiency improvements, but also to allow the overall price paid to rise above inflation. Following national austerity measures introduced in 2010/11, the efficiency factor has been set much higher, creating annual net unit price reductions, with the aim of driving greater efficiency savings.

The system has continued to evolve, with various pay-for-performance schemes, such as best practice tariffs (BPT) and Commissioning for Quality and Innovation (CQUIN) payments, introduced alongside PbR to further encourage quality improvement. These have had variable impact, with the level of clinical engagement appearing to be a key deciding factor of their success. The maternity pathway tariff has been introduced to reduce variation in care pathways in different hospitals. Additional cost-saving measures have been introduced to encourage providers to reduce their emergency activity, such as the marginal rate emergency tariff (MRET).

Community and mental health services: block contracts

In contrast, the predominant payment systems in community and mental health services are block contracts. While there has been some development of patient-based payment systems, including mental health clusters, these sectors have seen much less innovation in their payment systems. Block contracts can be more straightforward, resulting in lower transaction costs. They may allow more flexibility for innovation by providers as a result. They also make expenditure predictable and budgets easier to control. But this can be at the expense of the efficiency of the service and can mean a lack of transparency. It can also incentivise inappropriate care settings, with providers potentially avoiding more complex patients.

Block contracts may therefore lead to lower responsiveness as increases in activity are discouraged. CQUIN payments and Any Qualified Provider schemes have been extended into mental health and community services. This is to counteract the potential limitations of block contracts on quality and patient choice, but the lack of efficiency incentives remains a fundamental challenge.

Primary care: capitated payments

General practice is beyond the scope of this research, but provides an example of capitated payment systems, with risk-adjusted per-patient payments in operation alongside the Quality Outcomes Framework, which has been used – successfully to a certain point – to incentivise improvements in the quality of care processes for chronic conditions.

Direction of travel: towards whole-population budgets

Although a combination of methods is likely to be appropriate in most instances, the current combination of a case-based system for most acute care and block budgets in out-of-hospital services has provided a balance of incentives that are counter to the national ambition to provide more care out of hospitals and to treat mental and physical health services with parity. Equally they do not incentivise prevention or early intervention.

New payment models are being developed and tested in local areas in line with the development of the various new models of delivering care. As one example of this, a version of capitation-based payment known as ‘whole-population budgets’ has recently been suggested to support these new models of care. However, arrangements for ongoing evaluation of these new payment systems and spreading of best practice are not currently clear, and must be developed and shared.

First and foremost, an effective payment system needs a clear primary purpose. Whether this is equitably allocating resources, improving care pathways, driving efficiency, improving population health – or something else – this purpose must be clearly defined and understood by all parts of the system.

The current situation

The current payment system lacks a clear overarching purpose, so the rationale behind changes can be difficult to interpret. In 2013, NHS England suggested the payment system should support sustainably delivered continuous quality improvement, and the appropriate allocation and management of risk. This should be done by pursuing four objectives.

• Reimburse providers for delivering specified outcomes for patients, rather than treatments or inputs.
• Promote the long term, sustainable wellbeing of the whole person.
• Allow for different payment approaches where people’s care needs differ, with room for local flexibility bounded by a clear structure of rules.
• Signal clearly to commissioners and providers the choices available to them that will promote sustainably better outcomes for patients.

However, many providers perceived the existing payment system to have a multitude of objectives – many of which were unclear – and this may be contributing to a loss of focus during commissioner–provider negotiations. This is supported by a paper that attributed 12 policy objectives to the current NHS payment system, which contrasts against the three to five objectives pursued in other comparable European health systems.

For example, it is often unclear whether innovations in a payment system are intended to control expenditure, increase transparency, improve quality, or a combination of the above (or, indeed, achieve other objectives). The reality is that no single optimal payment system would be able to improve all of these objectives.

A system that prioritises expenditure control will therefore be different from one that intends to increase activity (Table 1). While each objective may be worthy in and of itself, expecting any payment system to achieve multiple objectives can lead to complexity and opacity, leaving rules and guidance open to conflicting interpretations. This is not to say that the payment system cannot have a role in multiple objectives, and many systems are able to meet more than one goal. But a primary objective should be agreed, rather than giving all objectives equal weighting. Otherwise the risk is that no objective is fully achieved.

This report’s research participants emphasised how difficult it can be to make decisions regarding the payment system – from developing implementation strategies to negotiating local prices – without a clearly defined purpose. They desired an NHS-wide purpose – one that would clarify what the payment system expects of them at a strategic level.

Whatever this primary objective is, it must be clearly stated and prioritised. There may be secondary objectives in support of other priorities, but there must be clear agreement on the primary purpose. Changes or additions for any secondary purpose can then only occur if they do not hinder the primary purpose.

Evolution

When the primary objective changes, which is to be expected as the NHS encounters new challenges, the new objective must be clearly defined before any part of the payment system is redesigned. This will make sure that changes to the payment system harmonise with the vision of the wider health care system. It will also avoid confusion resulting from multiple interpretations of what the payment system should deliver.

The current tariff system was designed during a period of major investment in the NHS aimed at reducing waiting times. In an era of austerity, with the focus on improving efficiency, the primary objective for the payment system has arguably changed – yet the payment system itself has not. While stability is an important feature of a payment system (see principle 8), priorities can and do change over time. When that happens, it becomes necessary to re-examine and reformulate the payment system’s underlying purpose.

Conclusion

Many of those spoken to for this research understood clearly that the principle objective of PbR, when it was introduced in the early 2000s, had been the reduction of waiting times. The stated reason when PbR initially introduced was ‘to incentivise expansion of elective surgery so that waiting times fall’. This outcome was easily measurable and understood across commissioners and providers. The system – together with complementary measures and increased funding – was effective in achieving its aim.

This is not to say that PbR was only designed to meet this one purpose. The complexity of the design meant that other objectives could be supported, such as driving efficiency through yardstick competition, incentivising quality by ensuring providers did not compete on price, and facilitating patient choice with money following the patient. The system evolved over time as the purpose developed, and PbR was expanded to cover non-elective and A&E care by 2006/07.

However, the continued expansion of the payment system’s priorities in later years – evident from the literature and perspectives gathered through research – has muddied the waters. Many participants in the research said there is no longer a clear purpose for the current payment system. Trade-offs occur when a large number of objectives are pursued. Any change or reform must have a clear, strongly signalled purpose that can be easily interpreted and implemented. This enables organisations to more easily translate the payment system into local action.

Such clarity of purpose underpins the other seven principles outlined in this report: without it, reform will be undermined.

The payment system can play an important – although limited – role in improving the quality and efficiency of services, but it cannot by itself overcome the many challenges that characterise complex care systems. Where payment mechanisms have improved quality and efficiency, the effect tends to be small. Their impact is also very dependent on the wider policy and delivery context.

A number of factors (eg organisational culture, relationships between organisations, and system-wide funding and demand pressures) can either undermine or enhance the impact of a payment incentive and must be considered. Payment rules are just one lever among a range of strategies – a system designed without consideration of the goal and methods of other strategies will be less effective.

The importance of culture

The culture of an organisation and its wider health system will influence the success of any payment method. The complex culture in which the payment system operates is evident in relationships between those who work directly with the payment system and those who deliver care. The quality of communication between these groups can affect whether the payment system achieves its intended impact.

Highlighting the benefits of the payment system to clinicians can improve this dialogue. For example, when speaking about best practice tariffs (BPTs), one interviewee observed:

It really does incentivise people… we go out there and talk to clinicians and say, ‘Do you realise, if you do this, we get some extra payment and it really does make a difference?’ NHS contract manager

The degree of clinician engagement varies, and incentives targeting clinicians’ behaviour will be less effective when clinicians are less engaged. Where this culture does not exist at all, relying on a payment system to drive change without a complementary strategy to improve communication will limit the impact.

Relationships between organisations

Relationships between organisations are also important. Financial restrictions on both commissioners and providers mean there will be certain challenges that a payment system alone is incapable of solving, and behaviours it will be unable to incentivise. Indeed, both the payment system and the health service culture more broadly can cause unhelpful degrees of complexity and drive problematic practices. The dominant role of organisational culture can counter (or indeed support) the incentives for behaviour change and limit (or boost) the effectiveness of a payment system.

I don’t think [models] are the sole driver of [contractual constraints]. It’s a complicated interplay, related to relationships, different leadership, and a variety of things. NHS finance professional

Barriers to impact

Barriers to the intended impact of a payment system can also be observed at all levels. Participants of research for this report noted that day-to-day operational challenges – also identified in literature, including limited capacity, spiralling demand and workforce shortages – can inhibit the effectiveness of payment levers.

While there are examples of payment systems facilitating change, policymakers must be realistic about what can be achieved with this tool alone. The powerful nature of established organisational culture and cross-organisational barriers must be considered in the design of a reformed payment system.

Conclusion

No payment system should be seen as a ‘silver bullet’ for the challenges facing the health care system. It is just one incentive tool among an armoury that can be used to influence provider behaviour. Local and national health care leaders must be pragmatic about the degree of change and benefit that payment systems can deliver, particularly given current pressures on finances, workforce and rising demand. A payment system cannot change or incentivise behaviour in isolation – other factors may render it ineffective. The objectives set by a payment system should be shared with other aligned policy initiatives and levers, such as clinical governance, or guidance from the Care Quality Commission. While a well-designed payment system can drive improvement, ultimately the scale and size of change that can be achieved are limited.

A payment system requires a consistent national framework to support a primary objective, but with flexibility to acknowledge the distinctive needs of different regions and provider types.

A system-wide perspective

The research for this report highlighted a desire among providers for national consistency across services, designed around a clear purpose (see principle 1). The system-wide view must account for the impact of incentives and disincentives across all NHS settings. For example, continual changes to the payment system for acute care, without an assessment of their impact on mental health services, may have a detrimental effect on parity of esteem for access and quality. Mental health patients are three times more likely to attend A&E, yet in most local health economies the payment systems for the acute provider and mental health provider are not in sync. Systems across services need not be identical, but must be complementary and interoperable. They must also support the equitable distribution of resources between regions.

We don’t want a payment system that pumps all the money into London or all the money to Newcastle; it has to distribute the money as fairly as possible. NHS finance director

Local implementation, with national support

While national consistency is important, participants were also clear that local health systems and providers should still have some autonomy to operate local pathways as efficiently and effectively as possible. Provision for local variation is set out in the 2012 Health and Social Care Act and is already happening to a certain extent in the current system with the development of integrated budgets. It is also an objective supported by NHS England. However, local arrangements must always be transparent and based on national guidelines, and their system-wide impact must be considered. Crucially, they must be made in accordance with the national priority of the payment system, or in a way that does not impede this priority.

The research for this report found that fragmented decision making, both centrally and locally, generates results that are inconsistent and contradictory to the overall ambitions of national and local health systems. This causes challenges for many providers, with financial sustainability currently achieved through local price negotiations with commissioners on an annual basis. These take up considerable time and resource.

Some participants called for an end to perceived ‘special arrangements’ agreed with commissioners during contract negotiations. But a more pragmatic solution may be to facilitate greater transparency of arrangements made at any level – local, regional or national. Currently, commissioners are responsible for publishing local variations and submitting them to NHS Improvement.

A national scheme for publishing these variations could address concerns regarding what some see as inequity in the system. Some providers also called for a federalised payment structure to allow system design to be led at either local or regional level depending on local needs. Others suggested a national payment systems toolkit, offering a range of solutions that can be selected to suit local health economies.

Participants expressed the need for a central repository of information about what has and has not worked when implementing local tariffs and new services, so providers can see how to innovate without worrying about not receiving payment.

It is hard to get [information from] a central repository. For example, when we were setting up the Hospital at Home and wanted to find out how many people were charging for Hospital at Home, we [had to be…] on the telephone ringing round. NHS contract manager

In the period between interviews and the publication of this report, NHS England published an accountable care organisation (ACO) contract package, with supporting documents including guidance to local areas for establishing integrated budgets for whole-population models of provision. This handbook is informed by learning from NHS England and NHS Improvement’s work with a number of the vanguard sites to develop whole-population budgets – it should therefore meet some of the needs expressed during the research for this report. It will be important to evaluate how useful it is to local areas in informing the development of whole-population budgets.

Crossing geographical boundaries

A national perspective is also important when patients move between geographical areas, as is often the case with specialised services. Respondents described the difficulties in applying national identification rules for specialised services by regional teams. This results in a wide variation of locally agreed prices. Sometimes service provision itself varies greatly across the country. Complicating the matter further, referrals to specialist services often come from another region, where commissioners and providers interpret identification rules differently.

Misattribution between clinical commissioning groups [for] specialised [services] has got such a [variety of] ways in which they can be misunderstood… at the moment they are going through this vast exercise to try and change the rules from what’s specialised and what’s not specialised. NHS contract manager

Other participants stressed the need for national pricing systems to acknowledge the demands on local providers who carry out complex care and treatment.

Our patients come from the entire country, so some means of sharing the risk for the outliers needs to be created in any payment system. NHS medical director

It will be difficult to strike a balance between a consistent national approach and local flexibility. In discussions with providers, the issue of the inequitable application of national rules frequently arose. But many also desired greater flexibility to build bespoke solutions suitable for local systems. Ultimately, a rigid national payment system will not promote efficiency and will undermine the financial sustainability of the whole system. But if a national payment system is to flexibly support local variability, such variability must equally support the key objective of that payment system.

Conclusion

The recent developments of accountable care systems (ACSs), ACOs and STPs indicate a willingness among national policymakers to allow areas to develop their own care systems and appropriate payment systems for these. Research for this report found that flexibility to reflect local needs and ambitions is valuable to providers. However, organisations and regions should be treated equitably, and such flexibility should be within the limits of a national framework, with national support. A fine balance exists between consistency and flexibility – transparency is crucial to finding and maintaining this balance.

Incentives in the payment system must be designed to encourage all parties to work towards the same, or aligned, objectives. Shared objectives will foster effective relationships between providers and commissioners. Payment incentives should also be carefully aligned with non-payment measures, such as performance metrics and regulatory frameworks.

Targeting the right actors

Several participants thought that contracting and finance departments were normally aware of payment system incentives, but that a lot of work was required to engage clinicians to fully align the incentives and make them effective. Sometimes the placement of an incentive is a step removed from those whose behaviour it is intended to alter.

It’s clinicians generating this informationand it doesn’t make a difference to anything that they do, other than we don’t harass them if they create data. If they don’t do it, then there’s not really any consequence. NHS contract manager

This is consistent with findings by the Nuffield Trust, which stated that: ‘Pay-for-performance schemes in secondary care in England have not translated into incentives for individual staff, only for hospitals as institutions.’

This is not to say that best practice should not be incentivised. Rather, if the payment system is the tool used for incentives, it will work better if these directly affect those who are making clinical decisions that determine both costs and patient outcomes. Otherwise, other incentive tools may be more appropriate.

Payment mechanisms must consider that NHS providers are obliged to treat patients presenting to them. Providers agreed that patient care should always come before the payment system.

[But incentives can] potentially stop people taking the steps to doing the right thing. NHS contract manager

For example, basic block contracts can create a perverse incentive to undertake less activity, and undermine choice where other providers are paid on an activity basis for the same services. Furthermore, evidence suggests they may not incentivise quality improvements as successfully as other payment mechanisms.

Problems arise when payment incentives are not correctly targeted to the right part of the system. For example:

There is a penalty for providers not receiving electronic referrals. It’s the GPs who send the paper and yet [providers will] be the ones who get penalised. NHS finance professional

It is important that payment systems within the secondary care sector are not designed in isolation and are aligned with the broader health and social care sector, and that the financial incentives are targeted at the part of the system that can influence the outcome. However, this is challenging, particularly given the move to more integrated systems, and the frequent lack of integrated accountability frameworks.

Making financial sense

Financial risk needs to be balanced and incentives or disincentives targeted at those directly able to improve performance. For example, participants expressed concerns that block contracts burdened trusts with disproportionate degrees of risk; higher levels of activity, or rising patient acuity, are typically not funded in year despite providers experiencing higher costs.

Under block [contracts] the risk seems to be on the provider side, because things come on board like a new drug, which is very expensive. So it disincentivises development in a sense that when new business cases, new initiatives or innovation come up, it costs the providers money… that is bad for the patient. NHS finance professional

Some interviewees reported that PbR can hinder innovation and that, being based on average cost, the tariff does not incentivise best practice or quality improvement.

You get paid for doing the work whether you’re doing the worst hip replacement in the country or the best… you get paid the same. NHS board director

There is often a disconnect between the size of an incentive and the cost of achieving it. Challenging yet achievable targets should be set, which take into account the cost of delivery. But if the cost of achieving these targets is greater than the reward, they will not incentivise improvement. Best practice tariffs (BPTs) aim to improve quality for certain procedures, but their impact is not universal, in part due to the associated costs of improvement and/or set up, along with opportunity costs. The size of the trust and current activity or practice may determine this.

For a small trust… BPT arrangements make no sense at all because you’ve got to set up infrastructure just to count and monitor. NHS contract manager

Alignment across the system

Payment mechanisms do not always align with national rules or ambitions, or other non-payment mechanisms. For example, a community trust funded through a block contract was compelled to enforce a national agency nurse cap, and as a result had to close beds due to weekend staff shortages. The block contract meant that commissioners were unable to repurpose the funding to commission alternative services. Following the bed closures, the activity instead flowed towards the local acute provider and was funded by commissioners via PbR.

As a result, costs to the local health economy increased, and it is possible that some patients were treated in inappropriate care settings. In this case, the national agency cap distorted the incentives of the existing payment system, demonstrating some of the problems that arise when providers use different payment mechanisms with conflicting objectives.

The fragmented nature of health services in recent years has led to an increase in local negotiations and contractual arrangements. One provider described developing a block contract with their CCG over a 3- to 4-year period. Another provider in the same area had signed a cost and volume arrangement, which meant that if patients didn’t get ‘the answer they wanted’, they were referred to the other provider, which was willing to investigate patients because they were incentivised to do so by PbR. Unnecessary and avoidable clinical activity was thus being driven by conflicting payment mechanisms within one local health economy.

Even when respondents were able to interpret the purpose of a component of the payment system, it did not always align with the prevailing national or local strategies of the health system. One participant highlighted this problem in the context of STPs, saying that PbR is hindering the drive towards co-planning and whole-population management.

Incentives and risk must be balanced between providers and commissioners. For example, with the marginal rate emergency tariff (MRET), the financial penalty falls disproportionately on providers.

[A]ll the pressures, all the penalties, MRET, everything all sits with the acute trust and there is absolutely no way you can fight it – there are no levers in the contract, you can’t respond. NHS contract manager

Commissioners and the wider system are arguably better able to reduce the volume of patients attending A&E by investing in preventive services and alternatives to A&E where appropriate, and should be incentivised to do so. Currently, acute trusts are experiencing rising A&E attendances, yet are receiving reduced payment for this additional activity.

The financial incentive or disincentive must be targeted at those who have influence over the outcome. This is clearly an ongoing challenge.

It’s quite unhelpful, PbR, in terms of the STP conversations we’re having… The whole focus on having to pull costs out of our local health economy should be based on avoided cost. And a lot of the planning is on avoided commissioner payment, which is income, and that’s not necessarily what providers can pull out in cost terms. NHS contract manager

Any development of the payment system must ensure that incentives target those who are able to influence the desired outcomes, and must anticipate the potential consequences of any change on other parts of the system.

Conclusion

While it is widely accepted that payment systems should be used to incentivise or deter certain actions, these systems must be coordinated with non-monetary incentives such as medical education and clinical guidelines.

During research for this report, it was not unanimous what should be incentivised and how. Some participants felt existing incentives inhibited their ability to make appropriate decisions for high quality, efficient care provision. Others felt incentives are targeted at parts of the system unable to influence outcomes. Accountability must be agreed across the system, and payments aligned to this.

Any payment system requires good quality data. Designing a payment system without ensuring that sufficiently accurate data are available will greatly reduce its effectiveness.

Driving data quality

There was a general consensus among participants that PbR contracts ensured trusts and commissioners collected and used better quality data compared with block contracts. Whatever the benefits or disadvantages of the payment system, providers believed that switching to PbR helped improve their data processes. This in turn increased PbR’s effectiveness as a payment system.

Since PbR has come in, we have got an awful lot better – infinitely better – at counting activity and coding it and classifying it. NHS finance director

However, improving data is not currently the primary purpose of the payment system. Good quality data are important for all aspects of NHS service delivery, so the quality and use of data should be driven independently of the payment system. Health care resource groups (HRGs), which currently underpin units paid for through PbR, have been in use since 1992 – prior to the introduction of the existing payment system. Their use extends beyond payment, to benchmarking, resource planning and monitoring. So linking payments to HRGs should only be done if such a process supports the primary objective, not as a means to drive data quality.

Relying solely on the payment system to improve data may lead to gaming, which ultimately reduces data quality. In recent years, this has led to a lack of trust between commissioners and providers. Disputes arising during the data challenges submitted as part of the monthly PbR reconciliation process can be hard to resolve – this can result in a large number of outstanding challenges at the end of the year.

The impact and burden of counting and coding disputes between commissioners and providers are seemingly unreported. Evidence of the extent of the administration costs associated with this issue is limited. A single high-quality data set, available to all, could be one solution to this.

Wider uses of data

Many participants noted that the vast amount of existing data is not being exploited to its full potential, as it is mainly used at a local transactional level, rather than a national strategic level to help shape policy decisions regarding payment. The Getting It Right First Time (GIRFT) programme, which aims to improve quality of care by reducing unwarranted variations and improving efficiency and patient outcomes, is beginning to use some of these data in a more strategic way. Data quality, an improved understanding of providers’ core costs and an accurate record of treatment outcomes were all seen as important for new payment models.

The data ought to be absolutely visible for everyone because it’s a publicly owned service, and by that you will drive up both the quality of the debate and the quality of the decision making [around payment systems]. NHS medical director

However, many participants of this research viewed the prospect of developing data of this quality for the NHS as a long way off.

I could put forward a model but it’s a completely unworkable one [at the moment…] if purchasers were not paying us for what we did but for the data associated with it, the ability to make rational policy decisions would be greatly altered. NHS medical director

For a number of years Monitor, now part of NHS Improvement, has sought to improve data quality through the patient-level information and costing systems (PLICS) programme. The aim is that, by 2020, costing information will more accurately reflect a patient’s treatment and will be produced more consistently across the provider sector. This programme may significantly improve the data underpinning future payment systems.

Conclusion

Payment systems need to be underpinned by accurate data to ensure that providers’ costs are properly covered, and that any incentives or penalties have the correct balance of risk associated with them. However, the purpose of the payment system is not to improve the quality of data, especially where this can conflict with the primary purpose and could lead to inappropriate recording of data. Ultimately, the payment system should be a beneficiary of quality data, not the key driver for it.

A consistent theme in the research was the highly complex nature of payment systems. In part, this reflects the complicated task of delivering health care, so a degree of complexity may be unavoidable. But care should be taken to ensure that the level of complexity is proportionate to the primary purpose of the payment system.

Increasing layers of complexity

Piecemeal evolution of the current system has meant that new rules or mechanisms are put in place to address specific concerns, without full assessment of their impact. More could be done to test whether the resulting complexity creates bigger challenges than the ones it was intended to fix.

PbR has got more and more complicated and we increased the number of health care resource groups. NHS contract manager

Clarity of purpose for the system (see principle 1) can help determine whether or not to implement a change. A principle that requires payment to influence behaviour will require a more complex system than one that purely facilitates financial transactions. Any added complexity should directly support, or at least not conflict with, the primary purpose of the system.

Support for operating in a complex system

Where complexity is necessary, adequate national support should be provided to reduce the impact on those operating systems at local and regional levels. For example, making the software code for applying payment rules publicly available would allow it to be applied and adjusted as appropriate. This would avoid duplication of effort across providers and commissioners, as well as researchers, while improving consistency and accuracy.

The terminology and mechanisms associated with payment systems can be burdensome and even incomprehensible.

Sometimes we just take for granted that most people understand this, and you’d be surprised how many people still are not savvy on finance structures. NHS medical director

This is a barrier to engaging staff on payment issues, especially clinicians, whose engagement is important for the success of payment incentives. It may also prevent providers from mounting an effective challenge when faced with difficulties in payment systems.

There is a lack of knowledge [of PbR within the trust]… I think that lack of knowledge… creates confusion. NHS contract manager

It can also create challenges when making service changes that require flexibility in the payment system.

Even agreeing relatively modest service changes in the current environment, where you have to agree a [new] price, is incredibly resource-intensive. The best example we’ve got is a fracture clinic, which has been going on for years. It’s a way of providing a service so [patients] come in as outpatients… it should be easy [to set up], but because we’ve overlaid everything [with tariff rules and prices] we have this huge complexity… No one feels empowered enough to just say, ‘This is actually a big improvement for patients.’ Yes, we need to collaborate [with commissioners] on how we deliver the service and the pathway, and this is how it will work, but how long has this been going on? NHS contract manager

Some level of complexity may be unavoidable given the nature of the NHS. But care should be taken to ensure that the level is appropriate to meeting the primary objective, avoiding excessive complexity in order to meet competing secondary objectives. Furthermore, adequate support must be provided to staff implementing complex systems.

System designers should view complexity as a design challenge, while ensuring that execution remains manageable – just as a car is complex in design, but relatively simple to operate with basic training. The current payment system provides the components and instructions – but not a consistently functional and effective product.

Conclusion

Many participants were uncertain about the level and breadth of understanding of the payment system(s) within their organisation, particularly outside finance and contracting departments. Evidence suggests that new payment systems, and PbR in particular, require technical skills to understand and negotiate. This can create extra work and requires a higher level of expertise, which may consequently add to costs. When designing a new system, policymakers must consider the resource burden of implementation.

Clear and independent oversight is necessary to ensure the payment system is delivering in terms of its key purpose, and to help resolve local issues and interpretations.

What oversight and support do providers say they need?

Interviewees reported a need for clear, high quality national guidance and support, and that an inability to access this can hinder both national and individual trust strategies. Payment for health care services needs to be properly regulated, so independent, national support for arbitration and for clarifying rules will be important to those working with the payment system at a local level. Such support could come from regional teams within arm’s-length bodies, who may have a better understanding of local issues and relationships.

However, it is important that oversight remains independent, with a national steer to ensure equity across the country. For example, the governance arrangements surrounding NHS England’s specialised commissioning have been criticised in the past, and a more transparent approach to decision making has been called for.

Providers felt they would benefit from a more robust role for the national bodies in maintaining and setting standards regarding payment systems.

The role of the national bodies [should] be defining standards, maintaining standards, maintaining pathways. So you get a truly national service… with the current system, a clinical service in one part of the country can be completely different to another. NHS finance director

National support should be available from a truly independent body. NHS England is both a commissioner and a national body involved in overseeing the payment system, which some saw as representing a potential conflict of interest. Equally, with NHS Improvement’s role in reducing trust deficits, they have a clearer focus on providers.

Streamlining the process

Participants described the burdensome process they, together with commissioners and central bodies, undergo each year to alter contracts and locally agreed tariffs.

[It is] a perennial process, year on year, where we produce the evidence, the tariff is produced, we dispute it, we produce evidence and then we come back to local prices and negotiate those. NHS board director

Many felt these challenges could be avoided with improved national support and guidance, and considered a consistent and joined-up message from the centre to be essential.

The current process of refining tariff payments and pricing occurs in parallel with the national strategic work regarding new payment and contracting models, as well as the national strategy for the NHS more generally. This leads to design lag, in which a new payment system follows changes in national policy. For example, integrated care has been an ambition of the NHS for a number of years, but it is only recently that the removal of PbR has been discussed nationally. Instead, the tariff development process and associated guidance should be fully aligned and should cross-reference these work streams.

Conclusion

For any payment system to succeed, truly independent oversight and support is needed from central bodies. There can be no conflict of interest, and messaging must be transparent and consistent. The centre needs to remain informed and impartial, assisting both providers and commissioners. There is a lack of available evidence that explores the impact of the competing roles of the arm’s-length bodies, but many providers feel the current relationships are affecting the way the payment system is overseen.

Any significant change introduced into the payment system needs to be allowed a period of uninterrupted implementation and evaluation. Educating staff, evolving culture and refining technique all take time. Not only will this produce a better assessment of a new policy, but it is also more likely to lead to successful outcomes.

The need for stability

Stability leads to a consistent application of policy, after which processes are simplified and become less onerous. But many processes currently lack stability.

With every new contracting model, there is the additional burden of having to record new data… Every change requires new reporting requirements on top of everything you’re doing. NHS contract manager

Constant changes to the payment system make it difficult to implement those changes fully and effectively and complicate engagement with the workforce. This all leads to increased implementation costs.

Significant, sustainable change takes time to embed – stability provides time for the system to be adjusted and refined, and facilitates local innovation. One participant reported that, while moving to a 2-year contracting cycle has been helpful, it was in conversation with commissioners around a 5-year contract for its musculoskeletal services. Longer-term contracts provide greater consistency and stability. Conversely, instability undoes progress and creates problems.

Another participant commented on the effectiveness of creating stability over time.

We’ve been doing [PbR] for the last 10 to 15 years and we’ve got commissioners who understand what they are working for in terms of income to the organisation. And [this in turn] helps stimulate the development and discussion and drive to improve services to do things differently. NHS contract manager

The need for evaluation

As with any improvement project, for a payment system to succeed it should undergo continual evaluation, with a manageable number of measurable outputs. This allows for effective feedback to those involved on how well incentives are working, identifies sooner issues to be resolved, and offers assurances of success where appropriate.

The current mechanism for looking at payment systems does not look strong, and a transparent, ongoing evaluation function is needed. This could be a role for NHS England and NHS Improvement, which have responsibility for the payment system, or could be delegated to a single independent centre (for example, a centre funded by the National Institute for Health Research) that can build up intelligence about the payment system.

Stability versus evolution

The need for stability must be balanced against the first principle of a payment system – a clear primary purpose. Where the primary purpose changes, the need for a change to the system can be destabilising and burdensome. Thus, wherever possible, aspects of the system that support a new primary purpose should be maintained. Equally, the need for stability should influence any decision to change the system’s primary purpose, to ensure that such change happens only when absolutely necessary.

It takes time for those working on payment systems to learn the ins and outs of each system. Sufficient time for embedding new systems enables staff to evaluate what does and does not work. Evaluation at both system and organisation levels is a crucial element for success.

Conclusion

Stability, balanced with the need for evolution of the priorities of the payment system, is beneficial in terms of operating efficiency. It also gives providers and commissioners certainty and allows them to plan for their populations in the longer term.

Evaluation is important to central bodies, as well as those innovating locally, in system reform and spreading good practice from areas with new models of delivering care. The ability to examine the effectiveness of current payment systems has been hampered by poor evaluation design and methods that have provided insufficient conclusions. Many participants spoken to for this report strove for continuous improvement and wanted to progress the payment system – there is widespread demand for robust system-wide evaluation.

There is broad consensus that a well-designed system can support positive change for the NHS. While no one payment system can ever perfectly suit all purposes across a system as complex as the NHS, a well-designed system of complementary mechanisms has a role to play in supporting improvements in the quality and efficiency of care and outcomes.

Parts of the system work well, yet others are not fit for purpose

However, this report, which supports findings in previous literature,, shows that the current NHS payment system is not fit for purpose. The design and implementation of certain aspects lead to inefficiency and can adversely affect patient care in unintended ways. This was the common theme running through the research for this report, and has been recognised by those with overall responsibility for the system., To support this, payment mechanisms should work well beyond organisational boundaries.

But there are also aspects of the system that work well. Many participants of the research for this report agreed that the current payment system has improved aspects of patient care, for example by reducing waiting times and improving efficiency.

It is hard to identify the best aspects of the current system without a clear priority of what the system should achieve. But a complete overhaul of the system may not be required. Focused improvements to some areas may be sufficient, and may be in the best interests of provider stability.

Multiple objectives can create conflicting priorities

However, it is impossible to know the scale of change required without a clear understanding and agreement of the primary purpose of the payment system. As the maxim goes, a system with too many priorities ultimately has none. In a briefing produced by NHS Confederation, 12 priorities were identified for the tariff system alone – more than double the number identified for payment systems of other countries listed. These included improving patient satisfaction and choice, driving efficiency, and controlling costs. The list did not include driving improvements to data, although this is another common requirement attributed to the current system.

Each of these objectives is worthy of attention. But trying to use the payment system to meet them all will inevitably lead to an overly complex system that is ultimately unable to deliver on any of them. There will always be conflict between different objectives and, without a clear priority, decisions become more difficult, if not impossible. It is important to remember the tariff is only one part of the whole system.

This report has not attempted to set out what the current priority for the payment system should be. NHS Improvement and NHS England share the responsibility for NHS payment systems, so they – in consultation with commissioners, providers and patients – should set the primary objective. This must be done by thorough engagement across the system, and be matched to the national priorities of the NHS, coupled with a good understanding of other options available to drive national priorities.

A primary objective does not preclude secondary objectives, indeed these are likely to be important. For example, a system that prioritises cost-containment fully at the expense of quality, or vice versa, would not be appropriate. It is therefore possible that the four existing objectives (see principle 1) are maintained. But an order of priority is essential to enable appropriate design and decisions to come from the top down. Naturally, the primary objective must be for the benefit of patients, whether directly through improved satisfaction and outcomes, or indirectly such as through improving efficiency.

Clarity of purpose must be the core principle for any payment system. Only once this primary purpose is agreed can the current system be assessed against it and areas for improvement identified. The extent of change required will only be understood once this is established. The other seven principles in this report provide guidance on what is important to those delivering care to people.

A mix of consistency and flexibility is key to local success

Consistency of approach across the system, accompanied by realistic expectations and appropriate incentives, will help all parties work towards the agreed primary objective. The system need not be identical across all settings – a mix of approaches is likely to be more effective to meet some priorities. But the different approaches must all support the same purpose, with complementary incentives.

Geographic variability in population health and health service provision means that some level of local flexibility will always be necessary and desirable to allow local or regional organisations to adjust national rules to meet local needs.

Complexity will vary depending on the objective

The level of complexity required in the payment system will vary depending on the primary purpose. Some objectives may be met with a simpler system, while others may be better met with a more detailed one. But where complexity is unavoidable, it must still support the key purpose. Efforts should be made to ensure that complex systems are easy to understand and operate by multiple actors. Any system will need high quality data, but data quality should be seen as a required input, not an objective of the system.

Oversight and support must be transparent and independent

This report has shown that systems will have conflicting interpretations, so oversight and support must be available from an independent source. Currently, NHS Improvement and NHS England provide this oversight, but also have other regulatory and commissioning responsibilities. Whether these affect their ability to offer independent oversight depends on whether the agreed primary purpose conflicts with their other responsibilities. Where there is potential conflict, the oversight arrangements may need to be reviewed.

Thorough evaluation can support and spread change

Finally, thorough evaluation will always be essential to the system delivering fully against the primary objective. It can identify areas for change where this is not the case or when the primary purpose changes. It also enables the spread of good practice as new models of delivering care – requiring new ways of paying for care – are extended across the NHS.

Taking the principles in this report into account when designing and reforming payment systems – at both national and local levels, and however large or small the reforms – will promote the development of a payment system that is fit for purpose. This will support the efficient delivery of high quality care in appropriate settings, and collaborative as well as integrated ways of working.

There are many positive aspects of the current NHS payment system, but it is far from perfect. This report shows that those providing care to patients would benefit from a re-alignment of the aims of the current payment system.

The timing is right for a review of those aims. The Five year forward view established the need for innovation, through new models of care and the development of STPs. Additionally, fixing the national PbR tariff for 2 years to 2019 has allowed for stability in the system. It has also provided a period to take stock and consider the future payment system and moves towards implementing this, with minimal disruption to patient care.

The responsibility for the national payment system lies with NHS England and NHS Improvement, so the review should be initiated and overseen by them. But it must be done in full consultation with providers and commissioners of patient care, as well as patients themselves.

The principles presented here, which reflect the needs of the provider sector in delivering care, act as a guide for this review. There will not be one simple or ‘one-size-fits-all’ system for the NHS, and the optimal solution is likely to be a mixture of mechanisms. But a system designed with these principles in mind will help providers deliver high quality care to patients.

It is important to remember, however, that the payment system is just one of many tools that can enable change, and is not an end in itself.

1. Department of Health. NHS Constitution for England. 2015. Available from: www.gov.uk/government/publications/the-nhs-constitution-for-england
2. NHS England. Five year forward view. 2014. Available from: www.england.nhs.uk/wp-content/uploads/2014/10/5yfv-web.pdf
3. NHS England. Vanguards. Available from: www.england.nhs.uk/ourwork/new-care-models/vanguards
4. NHS England. Sustainability and transformation partnerships. Available from: www.england.nhs.uk/stps
5. Organisation for Economic Co-operation and Development. Better ways to pay for health care. 2016. Available from: www.oecd.org/health/better-ways-to-pay-for-health-care-9789264258211-en.htm
6. O’Reilly J, Busse R, Häkkinen U, Or Z, Street A, Wiley M. Paying for hospital care: the experience with implementing activity-based funding in five European countries. Health Economics, Policy and Law. 2012;7(1):73–101. Available from: www.ncbi.nlm.nih.gov/pubmed/22221929
7. NHS Improvement. The sustainability and transformation fund and financial control totals for 2017/18 and 2018/19: guidance (Indicative). 2016. Available from: https://improvement.nhs.uk/uploads/documents/STF_
and_Financial_CT_1718_1819_Guidance_Indicative.pdf
8. Monitor. Reforming the payment system for NHS services: supporting the Five year forward view. NHS England; 2014. Available from: www.gov.uk/government/uploads/system/uploads/attachment_data/file/381637/ReformingPaymentSystem_NHSEMonitor.pdf
9. Dedoose. Great Research Made Easy. Available from: www.dedoose.com/about/history
10. Charlesworth A, Hawkins L, Marshall L. NHS payment reform: lessons from the past and directions for the future. Nuffield Trust; 2014. Available from: www.nuffieldtrust.org.uk/files/2017-01/2014-payment-reform-policy-response-web-final.pdf
11. Marshall L, Charlesworth A, Hurst J. The NHS payment system: evolving policy and emerging evidence. Nuffield Trust; 2014. Available from: www.nuffieldtrust.org.uk/files/2017-01/2014-nhs-payment-research-report-web-final.pdf
12. Lafond S, Charlesworth A, Roberts A. A Year of Plenty? An analysis of NHS finances and consultant productivity. Health Foundation; 2017. Available from: www.health.org.uk/sites/health/files/YearOfPlenty.pdf
13. Farrar S, Chalkley M, Yi D, Ma A. Payment by results – consequences for key outcomes measures and variations across HRGs, providers and patients. University of Aberdeen Health Economics Research Unit; 2010. Available from: https://abdn.pure.elsevier.com/en/publications/payment-by-results-consequences-for-key-outcomes-measures-and-var
14. Marshall L, Charlesworth A, Hurst J. The NHS payment system: evolving policy and emerging evidence. Nuffield Trust; 2014. Available from: www.nuffieldtrust.org.uk/files/2017-01/2014-nhs-payment-research-report-web-final.pdf
15. Charlesworth A, Davies A, Dixon J. Reforming payment for health care in Europe to achieve better value. Nuffield Trust; 2012. Available from: www.nuffieldtrust.org.uk/files/2017-01/reforming-payment-for-health-care-in-europe-web-final.pdf
16. Wubulihasimu P, Brouwer W, van Baal P. The impact of hospital payment schemes on healthcare and mortality: Evidence from hospital payment reforms in OECD countries. Health Economics. 2016;25(8):1005–19. Available from doi: 10.1002/hec.3205
17. Monitor. The maternity pathway payment system: Supplementary guidance. NHS England; 2014. Available from: www.gov.uk/government/uploads/system/uploads/attachment_data/file/342522/maternity_payment_pathway_system_supplementary_guidance.pdf
18. NHS England and NHS Improvement. 2017/18 and 2018/19 National tariff payment system, Annex C: Technical guidance for mental health clusters. NHS England; 2016. Available from: https://improvement.nhs.uk/uploads/documents/Annex_C_-_Mental_Health_Clustering_bookletl.pdf
19. NHS Digital. Quality and Outcomes Framework. Available from: http://content.digital.nhs.uk/qof
20. NHS England. Whole population models of provision: establishing integrated budgets. 2017. Available from: www.england.nhs.uk/publication/whole-population-models-of-provision-establishing-integrated-budgets-document-7b
21. Monitor. How can the NHS payment system do more for patients? A discussion paper. NHS England; 2013. Available from: www.gov.uk/government/uploads/system/uploads/attachment_data/file/302433/How_can_the_NHS_payment_system_do_more_for_patients_0.pdf
22. Geissler A, Quentin W, Scheller-Kreinsen D, Busse R. Introduction to DRGs in Europe: common objectives across different hospital systems. In: Busse R, Geissler A, Quentin W, Wiley M. (eds.) Diagnosis-related groups in Europe: moving towards transparency, efficiency and quality in hospitals. Open University Press; 2011. Available from: www.euro.who.int/__data/assets/pdf_file/0004/162265/e96538.pdf
23. NHS England. Delivering the NHS plan: next steps on investment, next stpes on reform. The Stationery Office; 2002. Available from: www.nhshistory.net/deliveringthenhsplan.pdf
24. Naylor R. NHS property and estates: Why the estate matters for patients. Department of Health; 2017. Available from: www.gov.uk/government/publications/nhs-property-and-estates-naylor-review
25. Maguire D, Dunn P, McKenna H. How hospital activity in the NHS in England has changed over time. The King’s Fund; 2016. Available from: www.kingsfund.org.uk/publications/hospital-activity-funding-changes
26. Gershlick B, Roberts A, Charlesworth A, Thorlby R, Jones H. A sustainable workforce – the lifeblood of the NHS and social care. Health Foundation; 2017. Available from: www.health.org.uk/publication/election-briefing-sustainable-workforce
27. Appleby J, Harrison T, Hawkins L, Dixon A. Payment by Results: How can payment systems help to deliver better care? The King’s Fund; 2012. Available from: www.kingsfund.org.uk/sites/default/files/field/field_publication_file/payment-by-results-the-kings-fund-nov-2012.pdf
28. Mental Health Taskforce. The five year forward view for mental health. NHS England; 2016. Available from: www.england.nhs.uk/wp-content/uploads/2016/02/Mental-Health-Taskforce-FYFV-final.pdf
29. NHS England and NHS Improvement. 2017/18 and 2018/19 National tariff payment system. NHS Improvement; 2016. Available from: https://improvement.nhs.uk/uploads/documents/2017-18_and_2018-19_National_Tariff_Payment_System.pdf#page=88
30. Ham C, Alderwick H, Dunn P, McKenna H. Delivering sustainability and transformation plans: from ambitious proposals to credible plans. The King’s Fund; 2017. Available from: www.kingsfund.org.uk/sites/default/files/field/field_publication_file/STPs_proposals_to_plans_Kings_Fund_Feb_2017_0.pdf
31. NHS Partners Network. Rise in block contracts risks patient choice. NHS Confederation Media Centre; 2016. Available from: www.nhsconfed.org/media-centre/2016/02/rise-in-block-contracts-risks-patient-choice
32. Independent Mental Health Services Alliance. New research shows impact of poor commissioning in mental health. 2015. Available from: www.imhsa.org.uk/upload/files/IMHSA_policypapers_31_146215583.pdf
33. Hudson A. Simpler, clearer, more stable: integrated accountability for integrated care. Health Foundation; 2016. Available from: www.health.org.uk/sites/health/files/SimplerClearerMoreStable.pdf
34. Park M, Braun T, Carrin G, Evans D. Provider payments and cost-containment lessons from OECD countries (Technical briefs for policy-makers, Number 2). World Health Organization; 2007. Available from: www.who.int/health_financing/documents/pb_e_07_2-provider_payments.pdf
35. Grašic K, Mason AR, Street A. Paying for the quantity and quality of hospital care: the foundations and evolution of payment policy in England. Health Economics Review. 2015;5(15). Available from: https://healtheconomicsreview.springeropen.com/articles/10.1186/s13561-015-0050-x
36. Shaw J, Taylor R, Dix K. Uses & abuses of performance data in healthcare. Dr Foster Intelligence; 2015. Available from: www.drfoster.com/updates/recent-publications/uses-and-abuses-of-performance-data-in-healthcare
37. Getting it right first time (GIRFT) programme. Available from: http://gettingitrightfirsttime.co.uk
38. Mason P, Fullwood Y, Singh K, Battye F. Payment by results: learning from the literature. A review prepared for the National Audit Office. ICF International; 2015. Available from: www.nao.org.uk/wp-content/uploads/2015/06/Payment-by-Results-Learning-from-the-Literature.pdf
39. National Audit Office. The commissioning of specialised services in the NHS. 2016. Available from: www.nao.org.uk/wp-content/uploads/2016/04/The-commissioning-of-specialised-services-in-the-NHS.pdf
40. NHS Improvement. NHS Improvement: 2016/17 highlights. Annual report 2016/17. 2017. Available from: https://improvement.nhs.uk/uploads/documents/1273_NHS_Improvement_201617_Ann_Report_FINAL_press.pdf
41. The Health Foundation. Evaluation: what to consider. Commonly asked questions about how to approach evaluation of quality improvement in healthcare. 2015. Available from: www.health.org.uk/publication/evaluation-what-consider
42. NHS England. The incentives framework for ACOs: Accountable care organisation contract package – supporting document. 2017. Available from: www.england.nhs.uk/wp-content/uploads/2017/08/1693_DraftMCP-10_A.pdf
43. Sagan A. Briefing 16. Lessons from Europe: the value of tariff redesign. Health Services Research Network, NHS Confederation, LSE Health, European Observatory on Health Systems and Policies; 2013. Available from: www.nhsconfed.org/~/media/Confederation/Files/Publications/Documents/Lessons-Europe-value-tariff-redesign.pdf
44. NHS England. NHS payment system. Available from: www.england.nhs.uk/resources/pay-syst