NICE under pressure to axe Framingham
By Nigel Praities
New research shows UK-based QRISK is better at predicting CVD risk
The Framingham risk score may now be living on borrowed time after NICE came under strong pressure to advise use of a new UK-based system instead.
A study by a leading team of GP researchers revealed that their QRISK algorithm did a better job than Framingham of predicting cardiovascular risk.
Study author Dr Peter Brindle, NICE's top adviser on cardiovascular risk assessment, urged the institute to adopt QRISK in its lipid modification guidance due in January.
Draft guidance from NICE, released in July, relied on a modified version of the Framingham risk score, which is known to overpredict cardiovascular risk overall but underestimate risk in women, ethnic minorities and deprived patients.
Dr Brindle said replacing Framingham was ‘almost a mor-al issue' because it led to people from deprived backgrounds missing out on treatment.
‘If we have Framingham as a national screening tool, advocated by NICE, it could exacerbate health inequalities on a large scale,' he warned, saying a replacement was long overdue.
Published online by Heart, the research validated the predictive power of QRISK in a
new population of primary care patients, without cardiovascular disease or diabetes.
In two cohorts comprising a total of 1.6 million patients from 813 UK practices, QRISK underpredicted cardiovascular risk overall by 1-12% compared with overpredictions of 23-36% using Framingham.
About half the patients described as high-risk by Framingham were described as low-risk by QRISK in both cohorts, compared with 24% vice versa.
Study leader Professor Julia Hippisley-Cox, a GP in Nottingham and professor of general practice at the University of Nottingham, said QRISK included important factors that were missed by Framingham. ‘Framingham doesn't include BMI, family history of heart disease or blood pressure medication, and it doesn't have weighting for social deprivation,' she said.
But Professor Ian Graham, chair of the joint task force that sets European CVD guidelines and consultant cardiologist at the National Children's Hospital in Ireland, said QRISK had major problems. ‘There is so much missing information that they had to make extrapolations. For example, cholesterol isn't a major risk factor in it which is clearly crazy,' he said.
A NICE spokesperson said it was aware of QRISK and was working to ensure as much new evidence as possible was considered before publishing its final recommendations.
QRISK vs Framingham
• Validated with data from UK primary care on nearly 3 million patients over several studies
• Risk factors include age, gender, smoking status, systolic blood pressure, total cholesterol to
HDL-cholesterol ratio, BMI, family history of CVD in first-degree relative aged under 60, area measure of deprivation and current prescription of at least one antihypertensive
• NICE considering evidence
• Based on a cohort study of 5,573 affluent white patients in Boston, USA
• Risk factors include age, gender, smoking status,
systolic blood pressure, total cholesterol to HDL-cholesterol ratio and left ventricular hypertrophy
• Currently included in the draft NICE guidelines for CVD