In a Swedish double-blind multicentre study, omeprazole (30 mg o.m.) was compared with the H2-receptor antagonist cimetidine (400 mg b.d.) in 152 patients. Clinical assessments and laboratory investigations were carried out at 2 and 4 weeks, and again at 6 weeks in unhealed patients. Endoscopy was performed at 2 weeks, and again at 4 and 6 weeks in unhealed patients. The patients in the two groups were well-matched prior to treatment. Omeprazole was superior to cimetidine in ulcer-healing rate after 2, 4 and 6 weeks. After 2 weeks of treatment, 66% of the omeprazole- and 45% of the cimetidine-treated patients were healed (P = 0.02), after 4 weeks 97 and 84% (P = 0.01), and after 6 weeks 100 and 92% (P = 0.02), respectively. There was a more pronounced improvement in the patients' symptoms in the omeprazole group after 2 weeks (P = 0.05). Both drugs were well-tolerated, but there was a high prevalence of patients with adverse events in the cimetidine group (51%, compared to 30% of the omeprazole group; P = 0.02). A total of 125 patients were followed for 6 months after healing. The patients were investigated by endoscopy after 6 months, or whenever symptoms occurred. There was no significant difference in the rate of relapse within 6 months between the two treatment groups: 54% relapsed in the omeprazole group and 52% in the cimetidine group. In conclusion, 30 mg of omeprazole, given once daily, is superior to 400 mg of cimetidine twice daily in duodenal ulcer healing; but ulcer relapse in the two groups appears to be equivalent.