Dyspepsia is a common symptom and some selection process for endoscopy is required. This study seeks to determine if noninvasive tests for Helicobacter pylori could be useful as a screening test to help select patients for endoscopy. Consecutive patients attending for upper gastrointestinal endoscopy were interviewed prior to endoscopy and the endoscopic diagnoses was recorded. The presence of Helicobacter pylori infection was assessed by serology. Some patients also had a 13C urea breath test or rapid urease test (CLO test). 436 consecutive patients were evaluated. The endoscopy findings were normal in 44%, 29% had reflux oesophagitis, 18% had duodenal ulcer, duodenitis or gastric ulcer and 9% had other diagnoses. 54.8% of patients were positive for Helicobacter serology. Using either the CLO test or 13C urea breath test as the confirmatory test for the diagnosis of Helicobacter pylori, the sensitivity of the serology test was 96% and 91% respectively and the specificity was 66.6 and 82%. Patients with negative serology and no history of recent NSAID or aspirin use comprised 34% of the total with dyspepsia or reflux symptoms. There were no gastric or duodenal ulcers in this group. The serology test may have some potential or the initial evaluation of dyspepsia. These tests need to be prospectively evaluated in general practice.