Objective: To assess three different methods of correlating symptoms with episodes of acid reflux in gastroesophageal reflux disease.
Methods: One hundred and one consecutive patients with heartburn were recruited. Patients were divided into those with and those without endoscopic esophagitis. Three different symptom indices were used, one of which was based on the binomial formula. It took into account not only the ratio of acid-related symptoms to all symptoms, but also the quantity of acid reflux.
Results: There was complete agreement between all three indices in 78% of the patients. All three indices were positive significantly more often (p < 0.001) in patients with esophagitis than in patients with no esophagitis.
Conclusions: All indices gave similar results. The binomial symptom index offers theoretical advantages, but a prospective study of response to acid suppression would be required to evaluate further its merits.