Prolonged intraesophageal pH monitoring is considered by some to be the most sensitive and specific test of gastroesophageal reflux. We prospectively examined the ability of the test to discriminate 64 hospitalized patients with typical reflux symptoms from 20 age-matched hospitalized control subjects. Patients were subdivided based on endoscopic findings into two groups: group 1, normal endoscopy (n = 30); group 2, erosive esophagitis (n = 34). Six different individual reflux variables and a scoring system were evaluated. Total esophageal acid exposure time and the number of reflux episodes requiring longer than 5 min to clear were each found to have greater discriminatory power than other variables and the scoring system. Although the 64 patients had significantly more acid reflux than controls, only 48% had abnormal results (defined as 2 SD from the control mean). Group 1 patients had significantly more reflux than controls, though only 21% had abnormal results. Group 2 patients were significantly different than both controls and group 1, but 29% had normal studies. Ninety-three percent of the group 1 patients with normal studies responded to antireflux therapy, and only 1 patient had another explanation for the symptoms. The finding that 24-h pH monitoring was normal in half of the individuals presenting with reflux symptoms and in 29% of the patients with erosive esophagitis indicates that negative test results must be interpreted with caution. The insensitivity of the test may relate to the manner in which the study has traditionally been performed in the hospital, and outpatient ambulatory monitoring may improve its reliability.