Balancing risks and benefits in primary prevention. Development of an individualised approach.

In the Netherlands about 30,000 Dutch people suffer a stroke each year and in 2000 there were approximately 140,000 people who had suffered a stroke in their lifetime. Stroke is the second most common cause of death worldwide. However, mortality data underestimate the true burden of stroke, since in contrast to coronary heart disease and cancer, the major burden of stroke is chronic disability rather than death. Stroke therefore is a major burden for public health and also a major disease in terms of health care expenditure. Primary prevention is the only means by which a fatal stroke or disability from stroke can be prevented. Therefore the development of effective strategies for primary prevention of stroke is extremely important. When studying differences in risk factors between ischaemic events and intracerebral haemorrhage (ICH), differences are found for hypercholesterolemia: It seems to lower the risk of ICH but clearly increases the risk of ischaemic stroke. Furthermore aspirin and other antithrombotics are known to prevent ischaemic stroke and also myocardial infarction, but increase the risk of intra- and extracranial haemorrhage. If it is possible to select those subjects who will benefit (benefits outweighing risks) from either cholesterol lowering therapy or from aspirin therapy, than it is possible to improve prevention by giving the therapy only to the subjects who will benefit. With the current project we propose to evaluate in which individuals there is a positive overall balance of benefits and risks for three different preventive strategies. 1. Aspirin for prevention of stroke and other ischaemic events. 2. Cholesterol lowering drugs for prevention of stroke and other ischaemic events. 3. Preventive surgery in patients with an unruptured intracranial aneurysm. We aim to identify individuals in whom the benefits of the preventive strategy outweigh the risks of that preventive strategy. First step: To identify these subjects Cox proportional hazards models will be developed to predict ischaemic and haemorrhagic events. With this characteristics can be identified that can be used to make subsets of individuals with and without certain characteristics. Second step: A Markov model will be developed for each preventive strategy and will be based on previous developed models. For each subset of individuals with Markov modelling benefits and risks will be balanced (in terms of quality adjusted life years) to evaluate in which subset of individuals benefits outweigh the risks. Furthermore the cost-effectiveness of these strategies will be evaluated. The ultimate aim is to provide recommendations, which subjects will benefit for use in daily clinical practice.