Evidence-based treatment of hypercholesterolemia currently recommended for rationalizing drug prescription requires justification of treatment by randomized trials, such as the West of Scotland Coronary Prevention Study (WOSCOPS) or the Scandinavian Simvastatin Survival Study (4S), and evaluation of its benefit from the estimation of the coronary risk of each patient. The latest European guidelines and Sheffield tables apply these principles and justify the decision to treat hypercholesterolemia if the Framingham coronary multivariate risk estimate is high enough, ie, >20% risk of coronary event at 10 years in the former and >1.5% risk of coronary death per year in the latter. Nevertheless, the practice of these two recent guidelines results in discrepancies in the decision to treat, because coronary morbidity was considered in one but mortality was considered in the other, and the risk required for treating may be extrapolated from different trials (4S or WOSCOPS). Although the principle of targeting lipid-lowering treatment to high-risk subjects is unquestioned, further studies are needed to demonstrate that the Framingham risk profile is useful in selecting persons who are likely to benefit and to determine the place of newer risk factors and that of early noninvasive detection of atheroma in the risk estimation-based treatment.