OBJECTIVE In patients having medical abortions with methotrexate and misoprostol: (1) to determine if giving ibuprofen or acetaminophen plus codeine with misoprostol (prior to onset of pain) prevented severe pain; and (2) to determine if there were predictors of medical abortion pain. METHODS A group of 281 women randomized to receive placebo, ibuprofen or acetaminophen with codeine. This was taken at home with the misoprostol 4-6 days after the methotrexate. RESULTS There were no significant differences between the three groups with respect to age, gestational age, parity, anxiety, depression, worst period pain score, and ethnicity. There was no significant difference with respect to rates of severe pain scores. The mean pain score for the entire group was 6.2 on a scale of 0 to 10. Severe pain (scores of 9 or 10) were reported by 23.4% of women and in this group, the mean maternal age was lower (P=0.05), parity was lower (P=0.01), worst period pain scores were higher (P=0.001), anxiety scores were higher (P=0.05) and satisfaction was lower (P=0.01). CONCLUSIONS The pain experienced in medical abortion causes significant distress and more research is needed to reduce it.