Data from a previous study, designed to compare metabolic risk markers for cardiovascular disease in non-users and oral contraceptive (OC) users, were analysed to evaluate the influence of OC composition on blood pressure. Healthy, female volunteers (1189 women) either not using OC (non-users) or currently using one of six different combined formulations (users) were compared. Combinations studied contained 30-40 mu g ethinyl estradiol combined with the progestins levonorgestrel, norethindrone (at two and three different doses, respectively) or desogestrel. After statistical standardisation to account for the significantly greater age of the non-users and longer duration of OC use amongst the levonorgestrel combination users, mean blood pressure was higher, compared with non-users, in users of monophasic or triphasic levonorgestrel combinations (systolic: + 4.3 mmHg (p < 0.001) and + 2.7 mmHg (p < 0.0001, respectively; diastolic: + 2.6 mmHg (p < 0.001) and + 2.3 mmHg (p < 0.05), respectively. Blood pressures in users of monophasic norethindrone and desogestrel combinations were not significantly raised and there was no increase in the proportion of women with abnormal values. Diastolic and systolic blood pressures were positively associated with oral glucose tolerance test insulin response (r = 0.11 (p < 0.01) and r = 0.15 (p < 0.001), respectively) in users but not in non-users. Currently used OC containing norethindrone or desogestrel progestins have little impact on blood pressure. Their correlated reduction in impact on insulin concentrations, though small, suggests common mechanisms through which OC affect blood pressure and insulin.