OBJECTIVE: Compare cost-effectiveness of rabeprazole (RAB) vs ranitidine (RAN) in acute and maintenance therapy of erosive esophagitis using symptom response, rather than endoscopic healing, as the outcome. METHODS: Cost-effectiveness was assessed in a decision-analysis model based on clinical-trial and published data. Costs reflected 1998 Medicare reimbursement, generic RAN, and estimated AWP for RAB. Cost per symptomatic recurrence prevented over one year of maintenance therapy is reported as average and incremental cost-effectiveness ratios (ACER, ICER). Patients were initially treated with RAB 20mg QD or RAN 150mg QID for eight weeks. Asymptomatic RAB patients received RAB 20mg QD maintenance, while asymptomatic RAN patients received RAN 150 BID. Symptomatic RAB patients received dose escalation, and symptomatic RAN patients were switched to RAB therapy. Non-responders were evaluated for surgery. Recurrences were treated with RAB 40mg or RAN 300mg BID. If RAN patients required RAB for acute therapy, they received RAN 300 BID maintenance. Symptomatic recurrences on RAN 300mg BID maintenance were treated with RAB, followed with RAB maintenance. Up to two symptomatic recurrences were allowed. RESULTS: Average per-patient costs were higher with RAB (RAB $2,020 vs RAN $1,917). RAB, however, prevented more symptomatic recurrences (RAB 74% vs RAN 41%; ACER RAB $2,748/recurrence prevented vs RAN $4,719/ recurrence prevented). Excess physician visits and procedures with RAN offset RAB’s higher cost. In sensitivity analysis, RAB remained more cost-effective than RAN despite varying key clinical and cost variables +/- 20%. The ICER of preventing one additional symptomatic recurrence with RAB rather than RAN is $313. CONCLUSION: RAB may be a cost-effective alternative to RAN when modeling symptom relief in the management of erosive esophagitis.