Methodological differences and variations in health care regulations among countries often preclude direct comparisons of cost-effectiveness studies. A projected risk model was applied, designed to determine the economic value in the United States of pravastatin in the secondary prevention of coronary heart disease (CHD), to Belgium using local health care costs. A Markov process was used to model the effectiveness of treatment for 3 years with pravastatin versus placebo in 1000 male CHD patients aged 60 years and clinically similar to those in the pravastatin limitation of atherosclerosis in the coronary arteries (PLAC I) and pravastatin, lipids and atherosclerosis in the carotid arteries (PLAC II) studies. The PLAC I and II trials have shown that pravastatin treatment for 3 years at a weighted mean dose of 36.64 mg daily significantly reduced the incidence of non-fatal myocardial infarction in patients with CHD. Framingham data were used to project the risk of mortality 10 years post-myocardial infarction. The incremental cost per life year gained (LYG), after discounting costs and benefits by 5% annually, in the setting of Belgian health care regulations, was Belgian francs (BEF) 720794 (US$ 24359) for CHD patients with one additional risk factor; BEF 526464 (US$ 17792) for those with two additional risk factors; and BEF 392765 (US$ 13274) for those with three or more additional risk factors. The cost per LYG in Belgium appeared to be more sensitive to drug acquisition cost than to costs of medical interventions. The cost-effectiveness ratios of pravastatin monotherapy for 3 years in secondary prevention of CHD, obtained with the same projected risk model, are from 86 to 92% higher in Belgium than in the United States, due to differences in medical patterns of practice and in intervention costs.