The Importance of Being Ernest (and Betty): First Name and Risk for Inflammatory Bowel Disease

Abstract

Listen

Translate article

Purpose: The unusual sequence of three unrelated patients with the same first name and a histopathologic diagnosis of inflammatory bowel disease (IBD) prompted us to design a study to determine whether patients with certain first names were more frequently affected by Crohn's disease, ulcerative colitis, or indeterminate forms of chronic inflammatory bowel disease. Methods: Using a nationwide database of colonic biopsy specimens from community-based endoscopy centers, we evaluated the frequency of a histopathologic diagnosis of Crohn's disease, ulcerative colitis, or indeterminate chronic inflammatory bowel disease in male and female subjects with the 20 most common names in the United States (www.ssa.gov/oact/babynames/background.html). For comparison, we also evaluated the frequency of microscopic colitis (collagenous or lymphocytic) and colonic adenomas. Results: Our database contained a total of 695,451 unique patients (median age 60 years; 50.4% female) from 43 states, DC and PR, who had colonoscopy with biopsies between 1-1-2008 and 12-31-2010. The lowest prevalence of IBD was found amongst males named Ernest (0.84%), while the name Jason was associated with the highest prevalence (12.71%; OR 17.2 95%CI 5.36 -55.06; p<.0001). Among females, the name Betty carried the lowest risk (1.15%) and Amanda the highest (9.43%; OR 8.92 95%CI 5.70 -13.95; p<.0001). The prevalence of microscopic colitis showed small non-significant variations amongst the name groups; in contrast, colonic adenomas were generally but not linearly, inversely associated to the prevalence of IBD. Age was a major determinant of the association between name and prevalence of IBD, as patients with low prevalence names were generally older than those with high prevalence. This association, however, was not universal; men named Benjamin and David had an identical median age (57 years) but the prevalence of IBD was 7.09% and 3.44%, respectively (p<.0001). All men of median age 57 and not named Benjamin had a prevalence of IBD of only 1.92% (OR 3.90 95%CI 2.74-553; p<.0001). Table 1 illustrates some representative prevalences.Table: No Caption available.Conclusion: These data show an apparently strong relationship between some common U.S. names and the prevalence of IBD. Secondary analyses provide some explanations for this odd phenomenon (mostly related to the different frequency of names in different decades and their respective reasons for colonoscopy) but there remain unexplained associations (such as the greater risk of IBD for Benjamin as compared to David), which subtler analyses (including socioeconomic status, racial or ethnic background, etc.) might help clarify. Our conclusions should serve as a warning that incomplete and sensationalistic analyses may yield misleading results.