Primary prevention of cardiovascular disease

Two linked analysis articles (doi:10.1136/bmj.c6244; doi:10.1136/bmj.c6312) and one research study (doi:10.1136/bmj.d108) on aspects of prevention of cardiovascular disease raise concerns as to what measures are the most efficient and cost effective.1 2 3 Despite major initiatives and improvements in the prevention of cardiovascular disease, increases in obesity and diabetes have the potential to worsen rather than improve future outcomes.

Severe hypertension and familial hypercholesterolaemia are single risk factors for cardiovascular disease, but blood pressure and cholesterol are measured on a continuous scale and risk of cardiovascular disease is multifactorial. Guidelines from the United States and United Kingdom cover the identification of people who are at high risk from multifactorial causes, but these guidelines are due for revision shortly⇓.4 5 6

• Download figure
• Open in new tab
• Download powerpoint

Jim Varney/SPL

Most events occur in people with modest values of cholesterol (or low density lipoprotein-cholesterol) that overlap with those seen in people without cardiovascular disease, and Hingorani and Hemingway debate whether to target high risk people or screen whole populations.1 They favour wider eligibility for treatment with inexpensive generic statins and a wider population effort to reduce cardiovascular disease. However, a previous editorial in the BMJ concluded that although universal screening was cost effective compared with no screening, small gains did not justify the extra cost compared with targeted screening.7

Secondary prevention of cardiovascular disease should be straightforward, but not all people with the disease are treated and many are not treated optimally. Even people at very high risk are often not identified for primary prevention, let alone treatment. Only 15% of 120 000 people with familial hypercholesterolaemia in the UK are identified and, when one family member has had an event, family cascade screening is being implemented only patchily.8 People at high risk may be identified at casual consultations, but the National Institute for Health and Clinical Excellence advised a formal process.6 Primary care should undertake “virtual” patient assessments from practice computer records, impute missing values, and call people who might be at higher risk for full profiling and management.

Khunti and colleagues discuss the NHS health checks programme,9 which extended screening to everyone aged 40-75 years for cardiovascular disease and metabolic factors.2 This major programme has been modelled but not fully implemented, and pilot schemes have shown variable results. Screening may do no physical harm but some people find it emotionally distressing. Apart from drug treatment, can we provide appropriate and effective lifestyle changes for so many people when adherence is known to be poor, especially in people who are difficult to reach? Geographical variation of implementation and appropriate outcome audit are important factors in determining practicality and cost effectiveness and in tackling inequalities.

With the present financial pressures will there be enough local uniform and appropriate funding? The current quality and outcomes framework in primary care does not fully incentivise screening for cardiovascular disease risk. Its targets are too conservative compared with the extensive evidence base, which shows benefit from additional lowering of low density lipoprotein-cholesterol in the population at risk (down to low density lipoprotein-cholesterol values below 1.8 mmol/L in those at risk).10 Serumaga and colleagues found that these payments to primary care had no effect on controlling hypertension.3 Health Service for England 1998-2008 data also show limited improvement in the management of blood pressure in patients with hypertension. However, for lipids, treatment has increased over this period for men with coronary heart disease (22% to 76%), stroke (6% to 47%), and diabetes (7% to 61%) (data from www.ic.nhs.uk/statistics-and-data-collections). EuroAspire-III data (comparing 1995, 2000, and 2007) show that the UK is in the top three of 22 European countries for achieving blood pressure and cholesterol targets for secondary prevention.11 For primary prevention, the UK was second best of 12 European countries for non-smoking rates, and for achieving blood pressure targets it was fourth for people without diabetes and second for those with diabetes. The UK was best (by at least 20% more than other countries) at reaching targets for cholesterol.12 However, much more could be done to achieve effective prevention in all high risk patients.

Targeted identification is more cost effective but identifies fewer people than NHS population screening health checks. All those identified will need dietary and lifestyle advice, and large numbers of “well” people will be identified for drug treatment. Perhaps half of asymptomatic men would be eligible for statins in their last 25 years of life, which raises concerns about “medicalising” such large numbers of people. Equality of access and treatment requires particular efforts to include those from deprived backgrounds, who are often at higher risk, and to improve concordance of those who are normally poor medical attendees. NHS cardiovascular disease health checks will need to involve these groups and not be over-represented by the “worried well” or “semi-worried, semi-well.” Financial pressures may limit screening implementation and lead to patchy uptake.

The goal is to identify, treat, and treat to target high risk individuals. Adoption of one method should not preclude the use of others, and the NHS health checks programme must not prevent other pathways. Practitioners should target patients with cardiovascular disease and those in known risk groups, identifying some opportunistically. Cascade screening is needed for families affected by premature cardiovascular disease (such as familial hypercholesterolaemia8). An arbitrary treatment threshold of 20% risk over 10 years will limit numbers, although treatment is clinically effective and cost effective at a 10% risk over 10 year threshold if low priced statins are used. An individual who is a little below the 20% risk over 10 year threshold who wants treatment after full counselling should not be denied, and perhaps lifetime rather than 10 year risk thresholds should be used. The NHS health checks programme is not the only answer and not the only way to go.