National tariff: the right blend

The NHS long-term plan committed the NHS to reforming the payment system. The longer term aim is to move away from activity-based payments to ensure a majority of funding is population-based, potentially using capitated outcomes-based budgets to cover a broad range of integrated care services.

But the plan also set out the goal of moving to a blended payment model for all services. Having started the move this year, NHS England and NHS Improvement are now set to expand the approach in 2020/21.

Blended payments may not deliver on the population-based payment commitment – but they are seen as a stepping stone towards it. They are viewed as a better way of supporting more integrated care and the development of new patient pathways or models of care. While some fixed payments may initially be derived based on historical activity and current prices, the approach would allow for different mechanisms to be used in the future – such as linking payments to population-based information – once the data has been made available to support this.

Under traditional episodic payment approaches (and without further controls in place), increases in activity above plan might typically lead to increased cost for the commissioner and increased income for the provider. Shortfalls in activity might reduce income for a provider despite the fact that fixed costs remain in place. The financial incentives were not seen as reinforcing a collaborative approach to reducing avoidable demand or meeting that demand in a different way.

Blended payment aims to address these issues. It is described as a framework rather than a single approach and is inherently permissive, with the intention that it should be adapted to local requirements. In fact, it is suggested that there could be multiple blended payment approaches within one local health system, tailored to the needs of different patient groups or services.

The approach uses an ‘intelligent’ fixed element. It is ‘intelligent’ in the sense that it is based on forward-looking forecasts of activity and best available cost data. This is a pronounced change on the approach used with the old marginal rate emergency tariff (MRET), where the activity was set at a baseline level from a much earlier year and often had little in common with current levels of demand. This fixed payment can then be combined with one or more of variable, risk-sharing and outcome-based elements.

The flexibility continues with the ability to assign different proportions of contract value to each of the elements, depending on the objectives that local areas are trying to achieve. It is even possible that the core, fixed element isn’t the largest proportion by value.

And it is not being forced on local health systems. If systems have risk sharing mechanisms in place that work for the local system, there is no requirement to replace this with blended payments, although variations to national tariff approaches should be reported.

Emergency services

The move to blended payments began with the 2019/20 national tariff, when blended payment was established as the default approach for emergency care and adult mental health services. For emergency care, the blend included a fixed payment calculated as the value of the forecast activity at national tariff prices – reduced by the 2017/18 value of the MRET and 30-day readmission rules.

The fixed payment should allow for excess bed days. In principle, this creates an opportunity for providers and commissioners both to agree targets for reducing excess bed days, but also to agree what needs to be in place to enable the provider to move to earlier discharge. Ideally, this would involve other providers and commissioners with a role in supporting discharge or improving demand management. However, it inevitably becomes more complex the more partners that are involved.

In effect, there is then a 20% marginal rate for activity above or below this forecast, although this is actually calculated based on overall values. Where the value of actual priced activity is higher than the value of forecast activity (before MRET reductions), the provider would receive 20% of the difference. And if activity is below the forecast level of priced activity, the provider would retain 80% of the difference.

Lee Rowlands, contracts director at Manchester University NHS Foundation Trust and chair of the HFMA’s payment systems and specialised services committee, says it is too early to call how the new approach has bedded-in. ‘Some areas will have tried to run variations on it and some areas will already have had similar approaches in place prior to the national policy,’ he says.

Within Greater Manchester, for example, a mixture of arrangements were already in place to share risk between providers and commissioners and across a wider range of services than urgent and emergency care. These were set up under local agreements using different models including fixed value contracts, caps and collars and aligned incentives.

NHS England and NHS Improvement publish details of local variations from national tariff rules. This is unlikely to be comprehensive, but shows a number of commissioners have adopted different payment arrangements with providers.

These variations range from remaining on payment by results activity-based contracts to block contracts with caps and different marginal rates. In one case the marginal rate ranges from 20% to 100% depending on the percentage of actual activity value compared with the value of planned activity.

Although there has been no official confirmation, it is believed that the majority of commissioners/providers have adopted something that is following the spirit of the policy. This could involve aligned incentive arrangements across a broader range of services, the blended payment arrangement as described in guidance, or a variation on it.

In previous years, there has often been a gap between the activity included in commissioner plans and that assumed by providers. This undermines attempts to deliver system-wide
balance, as at least one of the plans must be based on wrong assumptions. This has continued despite calls to ensure agreement and was arguably reinforced by an MRET system that set an unrealistic baseline based on an old activity level. There is no information yet on whether the activity levels agreed as part of the new payment approach have more accurately tracked actual activity as the year has progressed.

Rather than arguing over activity levels, the hope is that with a robust and agreed baseline for activity, organisations can then share any financial and activity risks across all system players. And they can spend their time working together to address increases in activity and avoid unnecessary admissions.

Mental health

The context for mental health was different. The introduction of blended payment follows years of trying to move services away from the use of simple block contracts, which often fail to recognise rising demand and changing complexity, and to support the mental health investment standard increase in mental health funding.

In addition to fixed and variable elements, for mental health there is also an element linked to locally agreed outcomes. This builds on work in recent years encouraging local systems to link payment for mental health services to outcomes – arguably putting the sector ahead of acute services in this area.

For mental health, the fixed payment takes account of historic activity and unit costs. With no comparable national tariff in place, this is not as straightforward as multiplying forecast activity by published tariff prices.

However, it should enable providers to put issues such as rising complexity and increasing quality demands on the table. Providers have complained that under block contracts, the increased costs associated with rising complexity are normally overlooked. According to guidance, the variable price should then reflect the ‘best possible estimate of the incremental costs of activity increasing or decreasing’.

An outcomes element – linked to outcomes such as access to cognitive behavioural therapy and the provision of crisis plans – was recommended to be worth a minimum of 2% of the total contract value. These approaches to emergency care and mental health services payment will be rolled forward
into 2020/21.

Outpatients

But under new proposals set out in the tariff engagement document Key areas of work for the 2020 national tariff – consulted on in November – the approach will be expanded to a number of new areas. For a start, blended payments will be used to support the long-term plan promise to redesign outpatient services, which aims to remove up to 30 million attendances a year over the next five years. 

As an initial step, the proposal is simply to reimburse all outpatient attendances next year using blended payments. Then from the following year, activity would start to be grouped and paid for by specialty. It is believed that this second stage could allow outpatient attendances to be grouped, for example, into those related to an elective care episode and those supporting patients with long-term conditions.

With the right data collection to support this, this might then open up the possibility of payment being bundled into the elective care pathway rate or supporting a year of care for long-term condition patients.

The largest element of the blended payment would be fixed, calculated on locally agreed activity volumes and national prices, where these exist, and agreed local prices for other activity. It should also include the set up and running costs of advice and guidance services.

This would then be complemented by a risk sharing agreement. The proposal is that this risk share should not reimburse additional activity as a pure activity-based variable element.

Instead it would ‘recognise that there may be either cost or performance implications if service demand increases outside of the direct control of the provider’. So, the risk share might distinguish between increased activity arising from GP or from consultant-to-consultant referrals. An optional outcomes-based element could also be incorporated.

In its response to the engagement document, the HFMA calls for further guidance to avoid difficult local discussions to set values. There were also concerns about extending the model to NHS England’s specialised services.

Maternity

The engagement document also suggests extending blended payments to maternity services with the relative stability of birth numbers and the importance of having capacity in place making it ideal territory.

Currently services are paid for using a three-part pathway approach. Providers receive one single prospective payment to cover all of a woman’s antenatal activity – with three different levels of payment based on complexity – and then another to cover postnatal activity. Births are paid for on a birth-by-birth basis – with six different prices paid retrospectively and a single price for home births.

Under the engagement document proposals, this would switch to blended payments from next April – a separate blended payment for antenatal care, births and postnatal care.

The approach helps to tackle one of the current problems with the current pathway system – provider-to-provider payments triggered when a non-lead-provider undertakes antenatal or postnatal activity.

Matthew Jolly, national clinical director for maternity at NHS England and NHS Improvement described this as ‘competitive cross charging chaos’, when he spoke during an NHS England and NHS Improvement tariff webinar during the summer.

‘There is currently a lot of administrative effort going into getting a small amount of extra money,’ he said. ‘Instead all of the financial reconciliation between units would happen at CCG level,’ said Mr Jolly, ‘reducing a huge amount of time, burden and stress.’

So for maternity services, the fixed payment would take into account planned activity at national prices plus any system investment. For example, this could include any recognition of transition costs that some areas may want to factor in when moving to a continuity of care model.
There would then be adjustments to provider payments based on the historic net income flows between providers in the local maternity system. 

A risk share element would share the financial impact of risks – such as a different casemix than anticipated or other issues. ‘For example, if there is an estimate on how much continuity of carer could cost the system included in the plan, there may be an estimating error and the system could agree to share that risk,’ Gary Andrews, the new care models pricing manager at NHS England and NHS Improvement, told the same summer webinar.

There are also plans to pilot blended payments for adult critical care services. It is clear that NHS England and NHS Improvement see blended payments as the future – or at least a stepping stone to future payment systems. They are designed to support greater collaboration and system working to deliver more integrated care. But they accept that this will not happen overnight.

In fact, Mr Andrews told the webinar that the move to blended payments should be seen as an iterative process. While blended payments are likely to work best in mature systems, they should also help systems to develop better relationships. ‘By moving from a transactional, confrontational approach to something more collaborative, relationships can shift quite quickly with a focus away from income chasing and arguments towards the management of system costs,’ he said. ‘So blended payment should improve relationships, which will in turn lead to better designed payments.’

On that basis, it is understandable that NHS England and NHS Improvement are keen for systems to adopt the new payment approach as soon as possible.