Irritable Bowel Syndrome in the United States: Prevalence, Symptom Patterns and Impact

A. P. S. Hungin; L. Chang; G. R. Locke; E. H. Dennis; V. Barghout


Aliment Pharmacol Ther. 2005;21(11):1365-1375. 

In This Article

Materials and Methods

A random-digit dialling technique as described previously[21] was used to identify study participants. This technique has proven feasibility and accuracy in health research[22,23,24] and enhances the representativeness of the study. Random-digit dialling allowed large numbers of people in the US to be easily identified while avoiding the possibility of cluster sampling often seen with multiple fixed-site surveys. Sampling limitations often noted with postal surveys were also avoided. Research supports this technique as a highly efficient way of yielding accurate data.[22,23,24]

This methodology assumed that more than 90% of the US households had a telephone and that approximately 25% of numbers were unlisted. Random-digit dialling sampling enabled numbers in the US to be assigned with a probability of selection. As previously described,[21] a random initial sample was drawn from all known listings using a 1 in n sampling interval and an algorithm was then utilized to randomize the last two digits, thus ensuring the inclusion of unlisted numbers. All telephone numbers were prescreened for a valid dialling tone to maximize data efficiency. All fieldwork was conducted on weekday evenings and weekends so that the working population was represented in the study. In order to enhance representativeness, quota sampling was applied in terms of age and gender. This survey was carried out by The Sample Surveys Research Group (now Synovate), West Malling, UK, and commissioned by Novartis Pharmaceuticals, New Jersey, US.

The survey included two questionnaires and was executed in two phases; the development of the original questionnaires has been previously described.[21] The first questionnaire was subjected to five stages of piloting and testing, including respondent validation. Further validation was conducted using clinicians and lay individuals in the US - terminology modification ensured respondent comprehension. The revised US version was piloted amongst 'live' respondents prior to commencement of fieldwork. The locally derived and adapted questionnaire was then tested in the US using the telephone interview technique.

In the first phase, a screening questionnaire was employed to identify subjects already diagnosed with IBS (medically diagnosed by their clinician) and to elicit information that identified non-diagnosed IBS sufferers according to the Manning, Rome I or Rome II diagnostic criteria. Subjects were classified as having undiagnosed IBS if they confirmed that they suffered from at least two of the Manning criteria and experienced abdominal pain/discomfort at least once in every 2 months in the previous 12 months. The screening questionnaire was presented to potential participants as being part of a general health project in order to avoid introducing bias and directing responses towards IBS or other GI conditions.

Patients were defined as having IBS with constipation (IBS-C) if they experienced one or more of the following: harder/more lumpy stools than usual, the need to pass stools less often than usual (<3 times/week), or periods of straining, which was not accompanied by the need to pass stools more often than usual (>3 times/day), looser/more watery stools than usual and periods of urgency. Patients were defined as having IBS with diarrhoea (IBS-D) if they experienced one or more of the following: looser/more watery stools than usual, the need to pass stools more often than usual (>3 times/day), or periods of urgency, which were not accompanied by the need to pass stools less often than usual (<3 times/week), harder/more lumpy stools than usual and periods of straining. Patients with other bowel habit patterns were defined as having alternating IBS (IBS-A). Differentiation into these specific IBS subtypes was based only on the symptoms suffered by each respondent, without reference to a specific time frame.

In the second phase of the survey, respondents who were identified as having IBS in phase I of the survey, either medically diagnosed or diagnosed using IBS criteria (Manning, Rome I or Rome II), were invited to participate in a more formal interview regarding their symptoms, general health factors, lifestyle and impact of IBS symptoms on their lives. Data on comorbid conditions, therapy/medication use and consultation patterns were also collected. Thus, three groups of respondents were identified: those with medically diagnosed IBS; those who had IBS that had not been medically diagnosed, but who fulfilled IBS diagnostic criteria; and a healthy comparator group identified from the screening questionnaire as not having IBS. Any individuals who had experienced IBS symptoms within the previous 12-month period were identified as current IBS sufferers.

For symptoms of constipation or diarrhoea, individuals were questioned (based onthe IBS diagnostic criteria) about: any periods needing to pass stools more often than usual (>3 times/day) or less often than usual (<3 times/week); any periods when stools were looser/more watery than usual or harder/more lumpy than usual; periods of straining when going to pass a stool; periods of urgency when going to pass a stool; feeling of incomplete evacuation after passing a stool; periods of bloating.

As in the European study,[21] the development of the questionnaire used in the US population incorporated the clinical experience of the clinicians involved with this study. Alternative diagnoses (not IBS) were excluded at an early stage of the questionnaire and pilot research was performed to ensure the Manning and Rome I/II criteria were effective in this setting. The questionnaire was designed and carefully validated to avoid order effects, such as suggestion based on earlier questions.

Based on the assumption of a conservative IBS prevalence rate of 5-10%, it was estimated that 5000 respondents would need to be contacted to reach 250-500 IBS sufferers in the US. Specific target age groups for the population were 18-34, 35-54 and over 55 years, with equal numbers contacted within these age groups.

The study was carried out in the US in a sample of subjects aged 18 years or over. Those previously diagnosed with Crohn's disease, ulcerative colitis, coeliac disease, diverticulitis, peptic ulcer or cancer (including colon and gynaecological cancers) were excluded from the results because of possible overlap with IBS symptoms.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.