We have studied the influence of oral contraceptive use on lipoprotein (a) levels in a cohort of women aged 18, 21 and 24 years (n = 559). Data was available on oral contraceptive formulation and dosage, anthropometric variables, exercise, serum lipoprotein (a), insulin, lipid and apolipoprotein levels. Lipoprotein (a) was determined by radioimmunoassay. Differences were assessed with non-parametric statistical methods. Forty per cent of the study women used oral contraceptives. The use of desogestrel-containing monophasic preparations was associated with lower levels of lipoprotein (a) compared to triphasic/levonorgestrel formulations or to non-users (P = 0.005). This effect was seen only in non-smoking women. Oral contraceptive users had higher levels of serum apolipoprotein B, HDL3-cholesterol, apolipoprotein A-I, triglycerides and systolic blood pressure, and lower serum lecithin:cholesterol acyltransferase activity. Lifestyle factors (smoking, exercise) showed no significant influence on lipoprotein levels. In conclusion, the use of desogestrel-containing oral contraceptives has a marked lowering effect on lipoprotein levels. Prospective studies will be needed to assess the net influence of oral contraceptive use on cardiovascular health.