Costing questions

Policy and financial changes in the NHS mean the case for costing at the patient level is stronger than ever and there are increasing pressures for granular information to be produced on a more regular basis.

This was one of the key messages from NHS England and NHS Improvement director of pricing and costing Chris Walters when he addressed the annual HFMA Healthcare Costing for Value Institute costing conference in November. The conference was postponed from its normal timing in April because of the Covid-19 pandemic.

The pandemic has created challenges for costing practitioners and for the Costing Transformation Programme. Questions remain over the usefulness of nationally submitted costing data produced for 2020/21 (and submitted next year) because of the severe disruption to services.

But Mr Walters said the need for detailed, reliable cost data was inarguable. ‘My priorities for costing are framed by what it means to meet the commitments of the NHS long-term plan,’ he said. ‘That means two things. The first is supporting the many commitments – some clinical, some operational and some financial – that rely on reducing unwarranted variation across the NHS.

‘It also means supporting the three commitments for reform of the payment system – namely, to take better account of the cost of delivering efficient services locally, to move funding away from activity-based payment and to move to a blended payment model. It is self-evident how important patient-level costing is to both.’

Mr Walters added that future changes to legislation and the financial landscape would also create a greater reliance on patient-level cost data.

In particular, he highlighted the accelerated move to blended payment and moves towards system working. System working would be ‘mainstreamed’ in the next few years, he said, building on collaboration during the Covid-19 response.

‘Successful system working relies on integrated care systems having a common and shared understanding of system cost and value,’ he said. ‘Continuing to expand the reach of patient-level costing and improve the quality of information it provides will be essential to this on the provider side, as will rebooting our work on programme budgeting on the commissioner side.’

The plans for using blended payments suggest that payments should be based on the local costs of delivering activity in system plans. For most areas initially, the best estimate of local costs will be this year’s contract value. But over time the proposals envisage a greater level of cost-reflectiveness, which will rely heavily on robust local cost data.

And as competition is replaced by collaboration, systems will look to patient-level cost data to analyse and redesign healthcare pathways that run across organisational boundaries.

‘All three developments should present us with even more of a platform to showcase the undoubted benefits of value-added costing,’ said Mr Walters (pictured).

There are already significant challenges for costing practitioners in meeting the demands of the Costing Transformation Programme, which sees mental health and ambulance service trusts join acute trusts in making a mandatory patient-level submission for 2019/20 costs.

There are concerns among finance practitioners that the workload required to meet prescriptive and detailed costing standards and specific collection criteria leaves little time to actually use the cost data to drive improvement with clinical teams.

Despite this, Mr Walters said that more regular submissions were likely to be required in future. ‘The pandemic has thrown up an additional future priority – the need to collect patient-level costs more frequently and with less of a delay than at present,’ he said.

The daily briefings from 10 Downing Street during the pandemic – with daily updates on new Covid cases, admissions and deaths – had raised expectations among the public and politicians, he added. ‘As the costing community, we must be prepared to respond.’

Mr Walters told the conference ‘Knowing activity, casemix and patient-level costs on a daily or weekly basis won’t be possible for a very long time, if ever. But moving to submitting and collecting patient-level costs on a quarterly basis is realistic and something we should aspire to.’

Exceptional collection

A number of acute providers have already volunteered for an exceptional quarterly collection (EQC) this year, specifically aimed at understanding the costs of delivering care to Covid patients and the impact the pandemic has had on costs of non-Covid patients. This has been based on reference cost-style average cost collections, rather than patient-level costs and will continue with volunteer organisations for the rest of financial year.

Mandating a quarterly National Cost Collection (NCC) based on patient-level costs would move things to a completely different scale. It is understood that this would need to go through a mandation process involving an impact assessment and consultation. However, discovery and scoping work will begin in January.

Mr Walters said that understanding how this could avoid increasing the burden on costing teams would be an important part of the work.

This will be key to keeping costing practitioners on side. Many practitioners spoken to by Healthcare Finance see the national costing approach as too detailed. They want the focus put firmly on providers’ major costs by raising the level of materiality below which detailed costing data is not required.

Last year, the HFMA called for the approach to be streamlined, so that the resources needed to produce the cost data are more proportionate to the benefits. As part of this, it believes the resource/activity matrix that needs to be compiled for each individual patient should be reduced. Practitioners point out that while costs have to be broken down into hundreds of resources and activities, 80% of costs might typically be covered by just 20 to 30 resource groups.

In a response to the HFMA’s recommendations, NHS England and NHS Improvement said the 2020 standards reduced the number of resources required, though collection resources (the smaller number of resource categories that must be submitted as part of the cost collection) were increased. While some activities have also been reduced, there will be further rationalisation this year.

The central costing bodies have also retained the requirement to map the general ledger to a cost ledger as the first step in the costing process. Some practitioners claim there is little value in the exercise other than forcing organisations to look at their general ledger in more detail. And some costing software does not support the cost ledger function. In response to these concerns, a two-level mapping approach was introduced this year, allowing trusts to map directly to collection resources.

Practitioners remain concerned about the time taken to submit costs centrally compared with the time this leaves to support their organisations to use patient-level cost data to drive improvement. The move to a quarterly patient-level cost collection would exacerbate this.

‘The danger is that we would spend the whole year producing information for the national collection, rather than supporting clinicians and service managers to use the data,’ says Chris Marshall, costing manager at the Royal Marsden NHS Foundation Trust. ‘And it would exacerbate problems with recruitment and retention.’

There has long been a recognition that the move to patient-level costing would require an expansion of costing teams. However, this hasn’t materialised and costing practitioners insist that it is using costing data that makes the job attractive, rather than the nuts and bolts of deriving robust cost data. ‘And using the data is the best way to improve the quality of the data,’ adds Mr Marshall.

It is not the frequency of collection per se that would be the problem. Many trusts already report their own patient-level cost data on a quarterly basis, some even monthly. In fact, this is the data they use internally to support improvement work, not the national cost data, which is collected after the year-end and is not played back to providers for several months.

Many acknowledge that the national costing standards have helped to improve their costing processes and provide a common approach that makes data more comparable.

But there are key differences between the data generated and used locally and that required by the centre, with the National Cost Collection wanting certain costs excluded and far greater levels of detail.

So while one trust might be able to identify the pathology costs in a patient episode, under the National Cost Collection process you would be able to see how this pathology cost was made up of consultant and clinical time as well as consumables and other costs. The process around unbundling services such as chemotherapy, radiology, direct access and diagnostic imaging is one of the key differences between national and local approaches.

One of the big challenges is simply keeping on top of the guidance. The old reference cost guidance was at one time a booklet of between 100 and 150 pages. The guidance for acute trusts is now volumes one, two, three and seven of the collection guidance, separate standards for information requirements, costing methods and costing processes, an extract document and technical document – not to mention staying on top of FAQs, which override the guidance.

‘There is some good stuff in there that internally has changed and pushed the way we do our business-as-usual patient-level costs, which we produce monthly,’ says Dave Tunstall, senior costing accountant at The Leeds Teaching Hospitals NHS Trust.

‘But the volume of material is an issue. It not only takes a long time to work through, but it can also be difficult to find what is relevant. Especially when there are inconsistencies between volumes, which mean you end up having to contact NHS England and NHS Improvement and keep checking the FAQs for the latest clarifications and corrections.’

So quarterly patient-level cost submissions would not be as simple as hitting a computer key each quarter and producing a data file for NHS England and NHS Improvement.

Practitioners, accustomed to regular local reporting, say that the first three months of the financial year can be taken up with the national submission. To move the quarterly patient-level cost submission idea forward, some practitioners suggest it would need to become more of a simple local system output, perhaps with new resource/activity codes enabling NHS England and NHS Improvement to exclude unwanted costs centrally.

Covid challenges

The Covid-19 outbreak created challenges for the 2019/20 submission. Some practitioners were redeployed to other duties as part of the NHS response to the first wave. And typically they didn’t have the usual access to service managers and clinicians to validate data.

But Mr Walters said that a week before the mid-November deadline for acute and community providers, submissions were ‘right on track’, including dry-run submissions from mental health providers. The window for submissions for ambulance, mental health and IAPT (improving access to psychological therapies) costs opens in January.

Looking ahead, 2020/21 is likely to be far more challenging, and perhaps the biggest problem will be understanding exactly what staff did and where they did it, such was the level of redeployment. Consultants and other clinical staff changed routines and rotas to support the frontline effort. But changes in how staff were used during Covid has not been properly captured in activity systems.

There are lots of other issues too. Trusts leant heavily on private hospitals during the outbreak. In some cases, this was centrally funded and the private providers undertook all the care, while in other cases NHS consultants would simply be using these private facilities – leading to under-reporting of costs for this activity.

In the 2019/20 collection, exceptional Covid costs – those reported to NHS England and NHS Improvement and remunerated via a top-up process – were excluded from the cost quantum. If this approach is taken again for 2020/21, then this will lead to an under-reporting of costs for some patient episodes.

A number of practitioners are struggling to see what the 2020/21 data could be used for. It wouldn’t be comparable to other years and it would be difficult to see how it could be used to inform tariffs, whether they are produced locally or nationally.

Support for patient-level costing among costing practitioners remains high. They don’t need to be convinced that granular cost data could play a big role in driving improvement and supporting both system working and transformation. But there is still some distance to go before there is complete agreement on how this goal should be delivered.