The aim of the study was to assess the association of abdominal symptoms in a random sample of a general population and to find whether the associations could be confirmed at follow-up 5 years later. The study population was a sex- and age-stratified random sample of people living in the western part of Copenhagen County, Denmark. Of 4807 eligible subjects 79% attended the study and filled in a questionnaire on abdominal symptoms. Five years later the study was repeated and 85% of the survivors participated. Data from both studies were analysed separately for sex, age group and the following pain variables: unspecified abdominal pain, pain located to the epigastrium, pain provoked by stress or hunger, pain relieved by eating and pain relieved by defecation. Three clusters of symptoms occurred in all the analyses: borborygmi/altering stool consistency/distension; acid regurgitation/heartburn and nausea/vomiting. Unspecified pain was associated with all three clusters, pain provoked by stress or hunger and pain relieved by defecation associated with the borborygmi/altering stool/distension cluster, whereas pain in the epigastrium and pain relieved by eating did not show consistent relationships to any of the clusters. Additionally, the clusters associated with each other more often than could be expected by chance. As a consequence of our findings we suggest that the three clusters of symptoms constitute three common abdominal syndromes