BACKGROUND: Irritable Bowel Syndrome (IBS) is a chronic/common condition that causes a significant effect on the individual (reduced quality of life), society (time lost off work) and health services. Comparison of studies evaluating the management of IBS has been hindered by the lack of a widely adopted validated symptom score. The aim of this study was to develop and validate a disease specific score to measure the symptoms of patients with IBS. METHODS: A self-administered 14-item symptom questionnaire (based on Rome II criteria) was mailed to 533 persons included in a prevalence study of IBS. The reliability of each underlying dimension identified was measured by Cronbach's alpha. Validity was assessed by comparing symptom scores with concurrent IBS specific quality of life (QoL) scores. Reproducibility was measured by the test-retest method and responsiveness measured by effect size. RESULTS: 379 (71%) questionnaires were returned. The underlying dimensions identified were pain, diarrhoea and constipation. Cronbach's alpha was 0.74 for pain, 0.90 for diarrhoea and 0.79 for constipation. Pain and diarrhoea dimensions had good external validity (r = -0.3 to -0.6), constipation dimension had moderate external validity (r = -0.2 to -0.3). All dimensions were reproducible (ICCs 0.75 to 0.81). Effect sizes of 0.27 to 0.53 were calculated for those with a reported improvement in symptoms. CONCLUSION: The Birmingham IBS Symptom Questionnaire has been developed and tested. It has been shown to be suitable for self-completion and acceptable to patients. The questionnaire has 3 internal dimensions which have good reliability, external validity and are responsive to a change in health status.