The aim of this study was to evaluate the accuracy of symptomatology and esophagogastroduodenoscopy (EGD) in the diagnosis of proven gastroesophageal reflux disease (GERD). We evaluated the symptoms and EGD findings of 100 consecutive patients presenting with symptoms suggestive of GERD. Patients' symptoms were scored at their first visit with a standardized symptom scoring system (grades 0-3). Grade 3 symptoms were the most severe. EGD findings were classified according to the modified Savary-Miller scale. Esophageal acid exposure was quantified using 24-hour esophageal pH monitoring; a positive composite score was considered evidence of GERD. Fifty-seven patients had positive pH scores, and 43 were negative. The combination of grade 2 or 3 heartburn and/or regurgitation with erosive esophagitis or Barrett's esophagus on EGD had a 97 per cent specificity and 64 per cent sensitivity for accurately diagnosing GERD. It is concluded that, in the presence of moderate to severe symptoms and endoscopic injury, the diagnosis of GERD can be made without further studies. However, 24-hour esophageal pH monitoring is still indicated in patients with mild typical symptoms, atypical symptoms, or when the combination of heartburn and regurgitation, regardless of their severity, occurs in the absence of severe mucosal damage.