NICE's vascular risk advice based on 'severely misleading' data
By Nigel Praities
GPs are set for fundamental changes to how they assess cardiovascular risk after a Department of Health report dismissed NICE's current recommendations as based on ‘incorrect and severely misleading' data.
The report looks likely to strike a decisive blow in the wrangling over which risk score GPs should be using, after it heavily criticised the institute's use of unpublished data to justify recommending a modified version of Framingham.
The DH report was produced by epidemiological experts at the University of Oxford and found QRISK was a better predictor of cardiovascular risk in the UK than the Framingham score.
Some of the team's results were published in the BMJ last month, but the complete report on the DH website is much franker in its criticism of the evidence used by NICE.
It concluded unpublished findings from the University of East Anglia – which were influential in NICE choosing Framingham in its lipid modification guidelines last year – were based on flawed calculations.
‘We are unable to support claims and conclusions reported in the UEA paper. We are confident claims made in that paper, which has not been peer-reviewed or published, are incorrect, misleading and based on incorrect information,' the report concludes.
NICE is now looking at whether its guidelines should be revised on the back of the new report, with fears GPs will have to rerun thousands of risk calculations for the Government's vascular screening programme, which is currently being rolled out.
Some patients may become ineligible for statin prescriptions that they are already taking, while others may have been missed in the risk evaluations undertaken so far.
Phil Ranson, associate director at NICE, confirmed the institute was looking at whether its guidelines needed rewriting and defended use of data from UEA.
‘UEA were registered stakeholders at the time of consultation and its report was accepted as a comment on the guideline development group's interpretation of evidence on QRISK.
‘We welcome this new validation of QRISK that appears to show it provides a more accurate estimate of CVD risk in the English population than other calculators,' he said.
Dr Stewart Findlay, a GP in Bishop Auckland, County Durham and treasurer of the Primary Care Cardiovascular Society, warned GPs to expect disruption should the risk score change, but urged for clarity on what score they should use.
‘Why can't NICE and the DH make a decision and then we can all get on with the assessments? It is crazy, confusing and rather daft dilemma we have got ourselves into, with people arguing over the minutiae of data between scores,' he said.
Professor Julia Hippisley-Cox, QRISK researcher and professor of primary care at the University of Nottingham, said: ‘Many practices and PCT are using QRISK as their risk engine of choice as it better predicts risks for individual patients taking into account the medical profile and ethnicity. We look forward to the review by NICE.'What will be the impact?
This DH report could be the final nail in the coffin of Framingham, with NICE to review the advice it gave on assessing patients for statin treatment.
NICE recommended GPs use a modified Framingham score last year, but the DH has been pushing for QRISK to be used for some time as it includes measures of deprivation and is thought to be the most accurate predictive score for the UK.
Department of Health standard operating procedures on the NHS Health Check scheme already provide advice on what data is needed to use both Framingham and QRISK. Some areas are already using QRISK routinely and it looks more likely the whole country could follow in the light of this new evidence.