The distribution of serum cholesterol in the population depends on sex and age. By analyzing data for men and women aged 20-69 from the Copenhagen City Heart Study 1981-1983, it is shown that this dependence is of major importance for the sex and age distribution of subjects classified as hypercholesterolaemic, when single threshold values are used to define the condition (7 mmol/l; 6 mmol/l for subjects less than 30 years): 56% are women, and 38% of these are women aged 50-59. Men between 30 and 59 years of age, for whom the association between hypercholesterolemia and ischaemic heart disease is most pronounced, only constitute 28%. As an alternative to single threshold values, one can use sex- and age-specific percentiles for cholesterol. The advantage appears from the explained cumulative percentage distribution curves in Figures 1 and 2. With the 90-percentiles, it is possible to identify the "cholesterol top-10" of the entire population, and although the use may seem more complicated, it has several advantages, notably the possibility of identifying younger hypercholesterolaemic subjects and making more appropriate use of the available resources for treatment.