[National guidelines for clinical decisions, a methodological view]

Medical guidelines used to be based mostly on expertise and experience. Since the eighties they are increasingly scientifically founded. In recent years results of possible treatments have been estimated quantitatively and compared. In drawing up the consensus text 'Antithrombotic prophylaxis of vascular events in patients with manifest atherosclerotic vasculopathy' the preparatory committee, after systematic arrangement of the relevant literature, made maximal use of the results of randomized prospective clinical trials of good quality and sufficient magnitude, published in peer-reviewed journals. For most indications, the pathophysiological reasoning and the study results were in agreement. A demonstrated effect of a treatment was evaluated on the basis of its magnitude and related to the associated costs and efforts. For the consensus 'Treatment and prevention of coronary heart disease by lowering the serum cholesterol level' use was made of estimates of effects of treatment with statins versus placebo. For prevention of total mortality and non-fatal myocardial infarctions and strokes, the relative risk reduction was 30-35%. The decrease of the absolute risk depended on the initial risk. The committee was of the opinion that treatment would be useful given an absolute risk of 25% of a (subsequent) manifestation of cardiovascular disease within 10 years. This would cost Dfl. 40,000.-per year of life gained, which the committee considered acceptable. In the argumentation of guidelines there is a trend to systematic evaluation and quantitative application of the research data. Subjective assessments remain necessary, particularly the evaluation of clinical relevance of observed or assumed effects of treatments.