Background: Epidemiological differences suggest that treatments for H. pylori eradication should be locally validated. Aim: To perform a cost benefit study of different treatment options for H. pylori infection. Patients and methods: One hundred and sixty-seven patients with active duodenal ulcer and H. pylori infection who completed a 2-week treatment with one of the following regimens were included: famotidine plus amoxycillin plus metronidazole (FAM), omeprazole plus amoxycillin plus tinidazole (OAT) or lansoprazole plus clarithromycin plus amoxycillin in 3 (LAC1) or 2 (LAC2) daily doses. We compared efficacy, adverse effects and cost. Results: Eradication rate was 74.6, 72.9, 96.4 y 91.7% for FAM, OAT, LAC1 and LAC2 respectively (p<0.05). Direct cost ranged from US$ 50 for FAM to US$ 220 for LAC1. A decision analysis was carried out in a model including direct and indirect costs and considering retreatment with antibiotics after the first treatment failure and one-year treatment with H2-blockers in case of a second failure. FAM was selected as the most cost-effective option, with an estimated cost of about US$ 300 ± 148 per patient. However, cost associated to LAC2 was very similar (US$ 320 ± 58) and the lower standard deviation suggests less variation. Sensitivity analyses, considering reasonable fluctuation in parameters such as eradication rate, cost and follow-up period suggest that a regimen containing a proton pump inhibitor, clarithromycin and amoxycillin may be the most cost-effective treatment. Conclusions: These results should be confirmed in other settings, specially in ordinary clinical practice, far from clinical research. (Rev Méd Chile 2000; 128: 367-77)