Comments of Prevention and health promotion

Primary prevention: Treating populations or individuals?

Primordial prevention focuses on population health, but, once we move to primary prevention, then individuals and their families can be targetted. Most health workers in contemporary society are concerned with treating individuals and their families.

People with established disease are at high risk by definition, but risk can also be measured in people who have no established disease. There is controversy over how best to measure risk and at what level to treat people. The WHO recommends measuring cardiovascular risk by using charts that combine risk factors including age, smoking status, whether or not people have diabetes and systolic blood pressure (17). Charts for well resourced countries also include blood cholesterol levels, but there are charts that exclude cholesterol for places where it is impossible or prohibitively expensive to access laboratories to measure cholesterol. The point of using these charts is that they give a much more accurate estimate of risk than using any one factor alone, although some argue that age is such a powerful determinant of risk that it can be used alone (Nick Wald, personal communication; publication pending).

These charts are developed using data from the famous studies in Framingham in the USA, where a large population was followed up for years. Some experts argue that it is inadequate to use the Framingham data for other countries, where the makeup of the population may be very different. The United Kingdom, for example, which probably has a population which is less different from that of Framingham than many other countries, has used electronic records to generate a new risk assessment tool called QRISK, which has been shown to be a better predictor for the UK than the Framingham tool (18, 19).

Figure 2.  Overlap among Women and Men who will Experience a Cardiovascular Event in the next 10 Years and who are Predicted to Do so by the QRISK and Framingham Risk Assessments

Figure 2 shows, however, that neither tool is very good at measuring risk at the population level. QRISK identifies 10% of men as «high risk» (having a 20% chance of having a cardiovascular event within the next 10 years) but only 30% of cardiovascular events will occur in those men (18). In other words, 70% of cardiovascular events will occur in men defined as being at low risk because they make up 90% of the population.

For women it is worse: QRISK identifies 4% of women as being at high risk, but only 18% of cardiovascular events occur in this group (19).

The WHO recommends lifestyle improvements for people at all levels of risk, as well as regular monitoring for those with a 10-20% risk, and pharmacological treatment for patients with a risk above 20%. The National Institute of Health and Clinical Excellence in England and Wales recommends the same measures (20). The American Heart Association recommends low dose aspirin for patients who have more than a 10% chance of a major cardiovascular event in the next 10 years (21). A recent systematic review suggests that this advice may be misguided (22).

But there is an argument that a 20% chance of a cardiovascular event in the next 10 years is an unacceptably high risk of something that might well result in death or severe disability. People spend large amounts of money every year to insure their houses, which pose nothing like a 20% chance of being burnt down or suffering severe damage in the next 10 years. The risk of potential harm must, of course, be measured against the risk carried by treatment and that is why the authors of the recent systematic review argued against the use of aspirin in people at low risk (22): aspirin will undoubtedly reduce the chances of a thrombosis leading to a heart attack or stroke, but it also increases the risk of a gastrointestinal or cerebral bleed, with the risk of treatment cancelling out any potential benefit.