A multicentred Phase III clinical trial was conducted in 12 countries to compare HRP112 (depot-medroxyprogesterone acetate, 25 mg and estradiol cypionate, 5 mg) and HRP102 (norethisterone enantate, 50 mg and estradiol valerate, 5 mg) given every 28 days. Contraceptive efficacy and side-effects of both regimens were reported previously. Their effect on vaginal bleeding patterns is the object of this paper. A total of 2320 women were randomly assigned to each drug and 2000 of them provided a menstrual diary. The comparison of the bleeding patterns is made using a 90-day reference period approach and following the guidelines published by WHO. The analysis failed to identify any major difference in the vaginal bleeding patterns induced by both contraceptive preparations. For both drugs, the first bleeding episode following the first injection occurs early. For 70% of users, this is followed by a regular vaginal bleeding pattern similar to an untreated pattern. Others experience irregular bleeding and a few have either infrequent or frequent bleeding. The extremes of amenorrhea and prolonged bleeding are rare. There are no major trends in vaginal bleeding patterns with prolonged use of either preparation. Women with the worst vaginal bleeding patterns discontinue early in the clinical trial and the last three months of experience have the most influence in their decision to stop using the contraceptive method. The analysis suggests how the life-table analysis of discontinuation reasons underestimates the true incidence of vaginal bleeding irregularities in a clinical trial.