29 August 2008
An Audit Commission study revealing significant levels of error in the coding that underpins the payment-by-results system should be a call to action for both providers and commissioners, NHS leaders have said.
Results of the commission's first clinical coding audit programme revealed an average 9.4% error rate on Healthcare Resource Group assignments – the system under which diagnostic and treatment procedures
are grouped and which determines the tariff. The average error rate for coding the individual procedures was even higher, at 16.5%.
The August 28 report also uncovered huge variations in trusts' performance. While some recorded an HRG error rate of 0.3%, one trust's error rate was 52%.
Variation in the error rates identified at procedure coding level was wider still, with trusts ranging from close to zero to more than 70%.
Nigel Edwards, policy director at the NHS Confederation, told Public Finance that the results justified the need for the study.
'This is an important report that needs attention paying to it,' he said. 'We should get to a point where it shouldn't be needed every year, [but] commissioners and providers are going to have to pay some quite serious attention to documentation, systems, training of the coders and the internal quality control that they operate because there are significant sums of money that swing on this.'
The Audit Commission checked the coding quality of 300 'consultant episodes' (periods of continuous treatment under a consultant) from every acute trust. It found the errors contributed to a gross financial error of around £3.5m, approximately 5% of the price of the sampled treatments.
Andy McKeon, managing director for health at the Audit Commission, said: 'In most cases, not in every case, the positives and negatives cancelled each other out, so the net error was close to zero.
'We say, with reasonable confidence, that there's no evidence that there is systematic gaming or manipulation of the data to secure higher payments.'
But Edwards said this risk of 'gaming' was dwarfed by the more significant problem of inaccurate coding.
'You can't game it if you can't get it right,' he said. 'There isn't a sufficiently high level of accuracy that would be acceptable in payroll. If you imagine this level of error in payroll you'd be changing the contractor.'
The Audit Commission found the most common factor affecting accuracy was poor source documentation. Information passed on to coders by clinicians was not robust enough, it said. In those trusts where clinicians and coders met regularly to discuss problems, accuracy rates were found to be much higher.
Training should also be a priority. 'It's not the number of coders that affect accuracy, it's the adequacy of their training and development,' McKeon told PF. He also suggested that PbR might not be meeting the needs of specialist trusts, where error rates were generally found to be higher.
'That's partly because the classification system possibly doesn't really suit what they actually do,' McKeon said.
How well PbR works for specialist trusts was an issue flagged up in a Department of Health-commissioned report whose findings emerged this week. The PA Consulting report on PbR and the delivery of cancer services found a perception among specialist cancer providers that the highly complex procedures they carried out were not reflected in the tariff, leading to a significant loss of revenue.
The report stated: 'For example, a complex renal cancer procedure, estimated to cost approx £25k, can only be coded as a kidney removal, valued at £4k. Therefore the trust makes a loss of £21k for every operation they undertake.'
The report also noted that, with 200 different forms of cancer, the system did not reflect the necessary complexity of care.
These findings were seized on by the Liberal Democrats, who accused the government of suppressing the report. Although it was finalised in May, the report was not published on the DoH's website until August 8, the day the Olympic Games in Beijing began, the party said.
Health spokesman Norman Lamb wrote to Health Secretary Alan Johnson requesting an explanation for the delay, as well as an urgent review of PbR and its impact on cancer services.
Lamb said: 'This report reveals the very serious unintended consequences of how treatments are paid for. The government is effectively penalising hospitals for providing specialist care. The government must act on this report now. Its own cancer strategy is being fatally undermined by this flawed funding system.'
The Royal Marsden NHS Foundation Trust, a specialist cancer trust and a major contributor to the PA Consulting report, would not comment on Lamb's letter. A spokeswoman said: 'The trust has been working closely with Department of Health officials to ensure that payment by results appropriately reflects the costs of providing specialist cancer care, and it is broadly supportive of the report's recommendations, which we believe will help to deliver that goal.'
PFaug2008
