It has been proposed that patients with dyspepsia can be classified into symptom groupings that may represent different pathophysiological entities; however, it remains to be shown that distinct symptom subgroups exist. To estimate the prevalence of dyspepsia (defined as upper abdominal pain) and dyspepsia subgroups, an age- and sex-stratified random sample of Olmsted County, Minnesota, residents, aged 30-64 years, were mailed a valid self-report questionnaire; 82% responded (n = 835). Subgroups were as follows: those with symptoms suggestive of peptic ulceration (ulcerlike dyspepsia), those with gastric stasis (dysmotilitylike dyspepsia), those with gastroesophageal reflux (refluxlike dyspepsia), and the remainder (unspecified dyspepsia). Ulcerlike dyspepsia was the commonest subgroup (prevalence, 16.0/100; 95% confidence interval, 13.4-18.5), but 43% of subjects with dyspepsia could be classified into more than one subgroup. Nearly one third of dyspeptics also had irritable bowel symptoms, but these were not confined to any particular dyspepsia subgroup. Although dyspepsia is very common in the community and the majority have ulcerlike symptoms, there is such overlap among the dyspepsia subgroups that a classification based on symptoms alone in uninvestigated patients may not be useful.