Feature / Count the cost, feel the quality

The Department of Health believes NHS reference costs and costing in general have a major role to play in the push for better services and improved efficiency. And it is making this point more vocally. Steve Brown reports

If finance managers wanted evidence about the value attached to costing by the Department of Health, they should look no further than the reference costs publications in November.

Ministers and officials believe robust costing data has a pivotal role in supporting the elimination of waste and the delivery of better, more cost-effective services for patients. And both the 2010/11 cost data and draft collection guidance for 2011/12 underline this view.

There is a message in the very act of publication. Check your iPhones, Blackberries, wall planners – the 2010/11 national reference costs have been published in November – just four months after NHS bodies’ deadline for submitting local costs. That’s a good two months earlier than previous years.

The Department says the earlier publication is to ‘support effective analysis and use by the NHS locally’. But – along with the provision of greater analysis and a higher profile for this year’s reference costs publication – it also speaks of greater confidence in the costs produced. There has been a tendency within the finance community to constantly point to the known flaws within reference costs and a frequent assumption that the reference costs are only used to provide the starting point for the national tariff.

But reviews by the Audit Commission have highlighted both that there are significant other uses for reference costs beyond tariff construction and that much of the costing data is broadly accurate.

Aside from internal management purposes within NHS bodies, uses include holding the Department to account for use of NHS resources, assessing trusts readiness for FT status, supporting programme budgeting, informing productivity assessments and feeding academic research.

In terms of accuracy , the Audit Commission’s audit of reference costs last year found that most trusts’ reference costs submissions were ‘accurate in total’ and ‘unit costs for items covered by the tariff were usually accurate’.

However, the Department accepts there is room for improvement. For the first time this year it is excluding some trusts’ reference costs from the tariff calculation (where data was assessed as poor).

In fact, the Department is leading the call for big improvements in costing – partly to reinforce its commitment to data transparency, partly because robust cost data is seen as the way to unlocking greater efficiency, helping to identify unnecessary and costly variation in practice and informing the redesign of new pathways. Its new draft reference costs guidance states simply: ‘The need for high-quality reference costs cannot be overestimated.’

It directs managers to the Audit Commission’s 10-point checklist for senior managers to use to improve reference cost quality. The list was included in the watchdog’s Improving coding, costing and commissioning report this autumn (see Healthcare Finance, October, page 13).

And it accepts the limitations of average costs created through a top-down process. A keen supporter of patient level costing (see box above), the Department does not subscribe to the view that organisations have to choose between putting their energies into either patient-level or reference costing. It believes patient level costing can lead to more accurate average costs and has produced best practice guidance on how to use patient level costs for reference costs returns.

Another possible step-up in the Department’s drive on costing is a proposal to require medical directors to sign off the reference costs return – confirming that a minimum number of specialities have been reviewed by clinicians. Subject to confirmation in the final guidance – which is due to be published in January – this would drive greater engagement of clinicians in the accuracy of costing and service line profitability. Clinicians would not be able to wash their hands of local cost data on the grounds that they were nothing to do with them.

This would also tie in neatly with calls by NHS deputy chief executive David Flory for greater evidence that clinicians buy into and understand the implications of local cost improvement plans (see Mission critical, page 27).

There are some other big changes for next year’s collection – reflecting changes in NHS structures, to support developments with national tariff and in response to Audit Commission calls to address weak areas on national guidance.

PCTs will no longer take part in the exercise. Largely a natural consequence of them giving up community services provision under Transforming community services, the Department has also decided to remove the requirement for PCTs to submit details of services commissioned directly from the independent sector. It says past data has proven to be of poor quality and of limited usefulness.

Similarly PCTs will no longer submit a return for their personal medical services plus (PMS+) pilots. Again, this reflects the fact that many GPs provide similar services under general medical services or alternative provider medical services, and these services don’t form part of reference costs.

So reference costs will become a trust- and foundation trust-exclusive zone. Mental health trusts will submit a single cluster-based reference costs return following this year’s pilot exercise, which was run in addition to the standard return.

Spell progress

The voluntary reporting of spells costs alongside the mandatory reporting of finished consultant episode costs will also be ‘mainstreamed’ – although this too is subject to final confirmation in final guidance.

With the tariff paying on the basis of spells, and reference costs collected in finished consultant episodes and requiring a complicated conversion to inform the creation of tariffs, collecting on the basis of spells would be regarded as a major step forward in demystifying the whole costs-to-tariff calculation.

The Department of Health is passing responsibility for tariff setting to Monitor and the Commissioning Board as part of the current proposed changes.

Under the new arrangements, it will be Monitor that dictates which costs are used to inform the national tariff. But the importance attached to costing by the Department – and the broader local management and national uses for the data – mean it is determined to make as much progress on costing improvement as possible.

Perhaps to underline this point, the Department’s draft published timetable says that finance managers can look forward to reference costs being published even earlier next year – in October.

Reference costs 2011/11

A total of £53bn of NHS expenditure is included within the 2010/11 reference costs – a rise of £1.7bn on 2009/10. This represents about 50% of all NHS spending and covers more than 1,500 different treatments and procedures and more than 15 million episodes within admitted care alone. High level costs suggest the average cost of a day case is £670, an elective inpatient stay (excluding excess bed days) is £3,095 and a non-elective inpatient stay is £1,400. Each excess bed day costs an average of £260, while outpatient or accident and emergency attendances cost £100.

Of the £22bn costs incurred within inpatient and day case activity, the greatest expenditure was within the musculoskeletal system chapter of healthcare resource groups (chapter H), which accounted for £3.9bn. The digestive system (chapter F) consumed a further £2.7bn, while cardiac surgery and primary cardiac conditions pumped up another £2bn.

The full range for reference costs index (RCI), where 100 equals national average costs, stretches from 81 up to 134. However, this includes community and mental health trusts – where activity counting is known to be particularly problematic – and ambulance trusts. Looking at just general acute organisations (stripping out single specialty organisations as well), the range stretches from Gloucestershire Hospitals NHS FT at 86 to Weston Area Health NHS Trust at 117. This 31 percentage point difference represents a slight tightening of the range – it was 80 to 114 in 2009/10.

As always, RCI scores are a starting point for understanding an organisation’s costs, rather than an absolute indictor of relative costs. For example, while HRGs do reflect casemix differences to an extent, a trust with extensive specialised activity and treating more complex patients within some HRG groupings could exhibit a higher RCI than a trust with a less complex case load. Equally, differences in RCI could reflect problems with data collection or costing. Changes in the market forces factor – used to compensate for unavoidable cost differences faced by different organisations – can also influence an organisation’s RCI.

The reference costs schedules provide massive detail on the range of costs and averages for delivering procedures and treatments within specific HRGs. For example, there were 37,800 major hip procedures undertaken in 2010/11, where the patient was categorised as being in non-trauma category 1 and had no comorbidities or complications (HRG HB12C). The average cost of these procedures was £6,009, with the interquartile range extending from £5,078 to £6,618 (a 30% difference in costs). Average length of stay was just under five days.

Another high-volume procedure involved nearly 15,000 inguinal umbilical or femoral hernia repairs undertaken on patients aged more than 19 years old, again where there were no complications (FZ18C). The average cost was £1,875, with an interquartile range of £1,475 to £2,234 – a difference of more than 50%. These 2010/11 reference costs will inform the 2013/14 tariff.

Patient level costing survey

The Department of Health undertook its regular annual survey alongside the reference costs collection, looking at implementation of patient level costing. More than 90% of acute and mental health trusts took part.

While implementing patient level costing systems is not mandatory for NHS bodies, the Department supports its implementation. In 2010, it commissioned the HFMA to develop the clinical costing standards. Originally developed by the Department, the HFMA published revised standards in 2011 and is working on further revisions for next year.

The survey revealed that 126 organisations have either implemented a patient level costing system (81) or are in the process of doing so (45). These figures exclude PCTs and compare to 103 last year. A further 41 organisations plan to implement patient level costing – most within three years. Of the 81 already using patient level costing systems, 71 have used the data to underpin their reference costs returns. That means the 2010/11 reference cost data for more than a third of all acute providers have been informed by patient level costs.

An increasing number of mental health providers (10) have also implemented systems or are in the process of doing so.  A further 17 are planning to do so.