Is there overdiagnosis of CVD risk?

A new CVD risk tool could change statin prescribing, writes Rachel Liddle.

What is the story?
Thousands of people are being prescribed statins unnecessarily, the press has reported.

The traditional method of assessing cardiovascular disease (CVD) risk is predicting up to 35 per cent (1.5 million) too many cases, suggest UK research findings.

As a result many people in the UK could be needlessly being prescribed statins, adding to the NHS annual £2 billion statin bill.

Additionally, the study found the risk of CVD is lower in white middle-aged men than previously thought, and higher in women from poor backgrounds.

In women over 60, one in three are at risk of CVD, compared with previous estimates of one in four.

What is the research?
The story follows publication of a new CVD risk assessment tool called QRISK, which has been developed specifically for use in UK primary care settings.

This was derived from QRESEARCH, which contains EMIS records from 529 general practices, encompassing data on 10 million patients over 17 years.

For the first part of the study, 12 years of medical records for 1.28 million healthy men and women, aged 35 to 74, were analysed to derive the QRISK tool.

During this time, 65,671 cases of CVD, including MI, CHD, stroke and TIA, were recorded. This is roughly equivalent to an incidence rate of 7.96 per 1,000 person-years.

Analysis of risk factors and outcomes led to QRISK, which incorporates the logarithm of age, ratio of serum total cholesterol to HDL cholesterol, systolic BP, BMI, family history of premature CVD, smoking status, Townsend deprivation score and use of one or more BP treatments.

For the second part of the study, the researchers applied QRISK to 610,000 patients in QRESEARCH to validate the risk tool. This cohort was also assessed using the Framingham algorithm - which has been recommended in draft NICE guidance on CVD screening - and the ASSIGN tool recently developed for Scotland.

The observed 10-year risk of a cardiovascular event was 6.1 per cent in women and 9.28 per cent in men. But the Framingham tool over-predicted CVD risk by 35 per cent and ASSIGN by 36 per cent. In contrast, QRISK over-predicted by just 0.4 per cent.

Applied to all under-75s in the UK in 2005, QRISK identified 3.2 million at high risk compared with 4.7 million with Framingham and 5.1 million with ASSIGN.

Additionally, QRISK estimated that 34 per cent of women aged 64-75 would be at high risk, compared with 24 per cent under Framingham.

For men of the same age, QRISK put 73 per cent in the high-risk category and Framingham 86 per cent.

What do researchers say?
Dr Peter Brindle, GP and research and development lead for Bristol PCT, said: 'The difference between QRISK and Framingham is deprivation, BMI and looking at whether the patient is on BP treatment.'

Importantly, QRISK is well suited to assessment in UK primary care because it's derived from a GP database, he added.

Dr Brindle hopes NICE will consider adding QRISK to its guideline on cardiovascular assessment. The draft form published earlier this month centred on the use of Framingham.

'There will be less men being given statins and more women and people from deprived areas given them,' he said.

What do other experts say?
North Yorkshire GP Dr Terry McCormack, chairman of the Primary Care Cardiovascular Society, said: 'It is more accurate than Framingham for the UK population.'

Dr Rubin Minhas, CHD lead for Medway PCT in Kent, said: 'Framingham is showing its age despite all the tweaks. QRISK can be updated in real time using GP databases.'

BMJ Online First

Informing patients

  • QRISK is a CVD risk assessment tool that has been developed from UK GP records.
  • QRISK is calculated using age, total:HDL cholesterol, systolic BP, BMI, family history, smoking status, deprivation and BP treatments.
  • Using the Framingham assessment may overestimate number of people at high risk by 35 per cent.
Comment below and tell us what you think

Have you registered with us yet?

Register now to enjoy more articles and free email bulletins

Register

Already registered?

Sign in

Before commenting please read our rules for commenting on articles.

If you see a comment you find offensive, you can flag it as inappropriate. In the top right-hand corner of an individual comment, you will see 'flag as inappropriate'. Clicking this prompts us to review the comment. For further information see our rules for commenting on articles.

comments powered by Disqus