Background: Irritable bowel syndrome (IBS) prevalence estimates are essential for measuring the importance of the problem and the burden of the illness. However, IBS diagnosis is based on clinical criteria which have changed over time. The aim of this study was to assess how the use of different diagnostic criteria influences estimated IBS prevalence and, in particular, to compare how this prevalence varies using the new Rome II criteria and previous diagnostic criteria.
Methods: An epidemiological survey was conducted in the general population using personal interviews in a home setting. Two thousand individuals stratified by gender, age, social class and geographic location were randomly selected from the general population of Spain. Participants were classified as 'potential IBS subjects' (n = 281) or 'non-potential IBS subjects' (n = 1719) as a result of their response to a screening question regarding the presence of abdominal pain, constipation or diarrhoea. Thus, we estimated IBS prevalence according to the following diagnostic criteria: Manning, Rome I, Rome II, Drossman, Talley and Kay & Jorgensen.
Results: Prevalences of IBS according to the Manning and Rome I criteria were 10.3% and 12.1%, respectively. These were higher than the prevalences obtained with the other criteria used, which varied from 2.1% to 4.9%. IBS was more prevalent in females than males irrespective of the criteria used (the female:male ratio ranged from 2.33 to 4.33). Subjects who fulfilled the Rome II, Drossman, Talley or Kay & Jorgensen criteria also fulfilled the Rome I and Manning criteria. However, between 44% and 80% of subjects who fulfilled the Rome I or Manning criteria did not fulfil the more stringent diagnostic criteria.
Conclusions: The prevalence of IBS varies enormously depending on the diagnostic criteria employed. Criteria based on the frequency of symptoms, such as the Rome II criteria, produce much lower prevalence values compared to criteria based solely on the presence of symptoms. In fact, more than two-thirds of subjects who fulfilled the Rome I criteria would not have been diagnosed with IBS if Rome II criteria had been employed.