The quality of data being used by the NHS Payment by Results (PbR) system has improved since last year, but the performance of individual NHS Acute Trusts varies each year, according to the Audit Commission.

There has been a gradual improvement in clinical coding overall since PbR began in 2007-08, the Commission also finds, in its latest annual review of the quality of the data underpinning payments by PbR, the tariff that aims for fair hospital funding.

Performance on clinical coding has improved since the assurance framework started, with average error rates dropping each year and a narrower range of error, it says.

But there is still room for improvement, with many Trusts struggling consistently to achieve good levels of accuracy, and some have performed poorly, it says.

PbR "can only ever be as good as the data that underpins it. We have found the overall picture to be one of gradually improving quality and accuracy," said Andy McKeon, the Commission's managing director for health.

However, “health service providers can still do more to improve data quality. For example, it was disappointing that only a half of Trusts had made good progress in implementing previous audit recommendations," he added.

Work carried out in 2011-12 by the Commission's PbR assurance programme included auditing 33,373 episodes of care, following-up previous audit recommendations at individual Trusts to see how well they were being delivered, and reviewing Primary Care Trust (PCT)s' arrangements to secure quality data.