Intended for healthcare professionals

Head To Head

Will screening individuals at high risk of cardiovascular events deliver large benefits? Yes

BMJ 2008; 337 doi: https://doi.org/10.1136/bmj.a1371 (Published 28 August 2008) Cite this as: BMJ 2008;337:a1371
  1. Rod Jackson, professor of epidemiology,
  2. Sue Wells, senior lecturer in clinical epidemiology,
  3. Anthony Rodgers, professor of epidemiology
  1. 1School of Population Health, University of Auckland, New Zealand
  1. Correspondence to: R Jackson rt.jackson{at}auckland.ac.nz

    The Department of Health is planning to identify and treat all adults over 40 at high risk of a cardiovascular event. Rod Jackson and colleagues argue that a well targeted programme will save many lives, but Simon Capewell (doi: 10.1136/bmj.a1395) thinks whole population approaches would be more cost effective

    Provisional modelling by the Department of Health suggests that up to 9500 heart attacks and strokes and 2000 deaths could be prevented each year by its plan to screen for and manage vascular risk in people aged between 40 and 74.1 Checks would comprise a brief history, examination, and blood test and could take place in general practices or pharmacies. The programme will be implemented in 2009-10 at an annual cost of about £250m (€315m; $465m).1 This strategy assumes that the high risk approach to preventing cardiovascular disease is effective and cost effective, which we believe to be true.

    A large World Health Organization review comparing high risk and population-wide interventions to prevent cardiovascular disease globally showed that treating high risk patients with aspirin, off-patent statins, and blood pressure lowering drugs was not only cost effective but would avert more disability adjusted life years (DALYs) worldwide than population based interventions to reduce salt intake, obesity, and cholesterol concentrations.1 2 Furthermore the high risk approach was much more effective and cost effective than approaches targeting high risk factors (such as treatment of hypertension). Rose was right to point out the disadvantages of high risk factor approaches, but he didn’t assess the high predicted cardiovascular risk approaches.3 The key questions to be resolved are how best to define high risk and the optimal mix of high risk personal strategies with other population strategies.4 5

    Potential impact

    Myriam Hunink’s 1997 disease modelling study suggested that over two thirds of the fall in deaths from coronary heart disease in the US between 1980 and 1990 occurred among the 6% of 35-84 year old Americans with prior coronary disease.6 Her hypothesis is even more relevant today with the increasing numbers of people with coronary disease7 and the widespread availability of affordable, effective drugs. Between one third and one half of all major coronary disease events in 35-74 year olds now occur in the 5-6% of the population with a previous cardiovascular disease event,8 9 and triple therapy with aspirin, statins, and blood pressure lowering drugs can reduce event rates by two thirds or more.10 11 If only half these very high risk patients were adherent, this combination of drugs alone would be responsible for a 10% fall in national coronary disease event rates in less than 10 years.

    To achieve a similar national reduction in coronary disease events, a primary prevention strategy would need to lower the average coronary risk in the rest of the population by about 20%—a huge challenge, now that much of the low hanging fruit receptive to population-wide strategies has been picked.12

    Targeting very high risk patients

    The most cost effective strategy for preventing cardiovascular events therefore starts by targeting patients who have had a cardiovascular disease event. They are at the highest risk (10 year risk is usually >40%), there are relatively few of them, they are easy to identify, and they are likely to be more motivated than patients without symptoms. However, the NHS intends to provide drug based management for everyone aged between 40 and 74 years with a 20% cardiovascular risk over 10 years.1 If, for example, the QRISK2 cardiovascular disease risk prediction equation13 was used to identify patients meeting the NHS treatment thresholds, then about 10% of Britons aged 35-75 years would be recommended for drug treatment, in addition to the 5-6% with a history of cardiovascular disease.

    However, about twice the numbers of asymptomatic patients would need to be treated to prevent half the number of events achieved by treating those with prior cardiovascular disease.

    Currently, most patients with established cardiovascular disease are not taking triple therapy. The treatment gap is closing very slowly, but the number of low risk people treated far exceeds the number of high risk people treated.14 So priority must be given to ensuring general practitioners are able to manage patients with prior cardiovascular disease before widening the net too far to lower risk patients.

    New equations to predict risk of cardiovascular disease are better calibrated to UK populations,13 15 and their accuracy is improving with the addition of measures of social deprivation and ethnicity, providing an opportunity to target primary prevention more accurately while simultaneously reducing social inequities. However, as risk thresholds for treatment are lowered, large numbers of patients will be eligible for drugs. To cope with the increased workload and cost, simplified primary prevention treatment regimens are required. The recent NICE lipid guidelines are a good start, recommending 40 mg of simvastatin for all asymptomatic high risk patients, without any dose tailoring or systematic follow-up.16 Similar recommendations are required for blood pressure lowering drugs, and the end game may turn out to be a single daily combination pill.

    Getting the right balance

    Too often debate about prevention of cardiovascular disease is polarised between opposing evangelists for the high risk and population-wide approaches. An agnostic approach based on effectiveness and cost effectiveness is preferable. Cost effective high risk strategies must be complemented by cost effective population-wide interventions. There is no longer any justification for questioning the role of well targeted treatment with safe, inexpensive, and effective drugs for patients at high risk of cardiovascular disease as a key component of any global cardiovascular disease prevention programme.

    Notes

    Cite this as: BMJ 2008;337:a1371.

    Footnotes

    • Competing interests: RJ and AR are investigators on several randomised controlled trials investigating the effectiveness of polypills for managing risk of cardiovascular disease.

    References