Six in ten patients on statins 'not at high risk'
Around six in 10 patients prescribed statins in the UK are not at high risk for cardiovascular disease, finds a new study.
The findings, from an analysis of primary care records for over 350,000 patients at 421 UK general practices, indicate most patients prescribed statins are not strictly eligible for the treatment based on their global cardiovascular risk.
The study’s authors, from the University of Birmingham, concluded statin prescribing is ‘very unfocused’ and based too much on specific risk factors.
They also suggested that less than a third of patients who should be taking a statin are prescribed them. GP experts put this down to confusion over CVD risk scores, as highlighted in another recent report on statin prescribing, and an over-emphasis on specific conditions in the QOF.
The research team used data from The Health Improvement Network to identify 365,718 patients aged 30 to 74 years who were free of CVD and not taking lipid-lowering drugs in May 2008, and who had complete records on risk factors.
In all, 50,558 (13.8%) patients were prescribed lipid lowering drugs – predominantly statins – over the next two years.
These included just 28.5% of the 74,137 patients who were considered eligible for treatment based on their predicted 10-year CVD risk, which was assessed using the Framingham-based risk score advocated by the 2005 Joint British Societies 2 guidelines.
The remainder constituted 10.1% of the 291,581 patients who were ineligible for treatment based on this 10-year predicted risk.
Overall, less than half – 41.7% – of patients prescribed lipid-lowering drugs were eligible for treatment.
People aged 65 and older, those with diabetes and those with four or more blood pressure measurements in the past year were four times as likely to be prescribed lipid lowering therapy as younger patients, those without diabetes and those with fewer blood pressure recordings. Patients with TC levels of 7 mmol/L or more were twice as likely to be prescribed statins as those with lower cholesterol levels.
Lead author Dr Tom Marshall told Pulse the analysis suggests many GPs are prescribing according to specific risk factors rather than global risk scores used in guidelines.
Dr Marshall said: ‘We found that prescribing was strongly influenced by specific risk factors. That is why having diabetes or a total cholesterol level above 7 mmol/L are very strongly linked to prescribing statins.
‘But there are patients with normal cholesterol levels who are high risk – usually because they are older and have other risk factors – and there are patients with total cholesterol level above 7 mmol/L who are not high risk. So this is not a good substitute for the risk score.’
Dr Christine A’Court, clinical researcher at the University of Oxford and a GP in Carterton, Oxfordshire, said it was possible some eligible patients who were treated ended up excluded from the study because they developed CVD during follow-up, but added that overall the findings were ‘depressing’.
Dr A’Court said GPs’ ability to carry out CVD primary prevention has been made more difficult by repeated changes in recommended risk scores.
‘These changes by degree, their delayed incorporation into GP computer systems, together with the fact that many of the the risk calculators have a restricted age range, may have undermined GPs confidence in their usage,’ she said.
Dr A’Court added: ‘Contrary to the assumption of the authors, patient preference does impact on decision making in primary prevention. Patients frequently express concerns and misconceptions about statins, fuelled by misinformation from the Internet and certain newspapers.’
In addition, Dr A’Court suggested QOF may have influenced GPs’ priorities ‘towards diagnosis and management of hypertension, diabetes and established CVD, and away from risk assessment’.
Dr Alan Begg, member of the SIGN guidelines steering committee and a GP in Montrose, said the findings were not surprising and that GPs ‘get mixed messages on what score to use and confusion leads to lack of use’.
Dr Begg agreed that QOF could also be contributing to the lack of proper risk assessments.
He said: ‘The use of risk assessment as it appears in QOF does not help. Instead of measuring global risk and using it to determine management, it encourages making the diagnosis (hypertension) and then ensuring a risk assessment is carried out – not my understanding of how risk assessment should work.’
Dr Begg said there is also too much emphasis in guidance on single cholesterol levels and treating a high TC/HDL ratio, whereas with diabetes the global risk is assumed to be high and a statin is recommended irrespective of baseline cholesterol.
Note: The headline of this article was corrected after publication from ‘one in ten’ patients on statins to ‘six in ten’ not at high risk for cardiovascular disease