Randomized patients in IBS research had different disease characteristics compared to eligible and recruited patients

Objective: The recruitment process may generate a selected patient sample, which may threaten the generalizability of trial results. This risk is particularly high in case disease and patient characteristics demonstrate a wide variation, such as in irritable bowel syndrome (IBS). We compared IBS patients who were selected, approached, and randomized to participate in a clinical trial assessing the efficacy of dietary fiber therapy in IBS. Study Design and Setting: Retrospective survey in primary care patients diagnosed with IBS by their general practitioner in the past 2 years selected and invited for participation in a trial. Characteristics were compared between randomized patients (n = 193) nonrandomized eligible patients (n = 371), and patients not eligible for participating in the trial (n = 724). Results: Of the 2, 100 IBS patients, 1,288 (61 %) returned the questionnaire. Randomized patients had a higher intensity of IBS abdominal pain as compared to the other groups, a higher consultation rate and a longer IBS disease history. Noneligible patients had less active IBS symptoms. Conclusions: Patients randomized do differ from those nonrandomized in IBS disease characteristics. These observations may have implications for the applicability of our research outcome. (C) 2008 Elsevier Inc. All rights reserved.