Upper gastrointestinal (UGI) symptoms are commonly-reported side effects in NSAID-users, often requiring physician intervention and utilization of healthcare resources (e.g. endoscopy, acid-suppression therapy). We hypothesize that a medication associated fewer UGI symptoms will result in significant clinical and healthcare utilization advantages over existing NSAIDs. OBJECTIVE: Quantify clinically-significant UGI distress among patients treated with celecoxib, a novel COX-2 inhibitor, compared to NSAIDs and placebo. METHODS: Combined data from 8-blinded celecoxib phase-III clinical trials of at-least 12-weeks duration were used (# subjects): celecoxib (3,216), placebo (1,136), and NSAIDs (2,427), divided among naproxen (1,366), ibuprofen (345) and diclofenac (716). A composite UGI-symptom measure, UGI distress, was developed using WHO adverse event codes (abdominal pain, nausea, and dyspepsia). To simulate clinically-significant adverse events requiring medical intervention, only moderate (some limitation of activities) and severe (inability to carry out activities) events were included. Survival analysis was used to calculate the cumulative percentage of patients with UGI distress over a 12-week period. RESULTS: Kaplan-Meier curves revealed a significant difference (p<0.01) between NSAIDs compared to placebo and/or celecoxib. At 12-weeks, the cumulative probability of patients to have UGI distress were: NSAIDs (12.9%), celecoxib (8.1%), and placebo (8.5%) (p<0.01 for NSAIDs vs. celecoxib or placebo; p=NS for celecoxib vs. placebo). CONCLUSION: NSAID-treated patients experience significant UGI distress beyond the background rate of UGI distress, as represented by placebo. These results demonstrate the clinical advantage of celecoxib over existing NSAIDs and suggest a potential healthcare utilization advantage due to decreased frequency of UGI distress and need for physician intervention.