Much of the inflammatory bowel disease (IBD) literature focuses on Caucasian populations even though several studies suggest that the incidence of IBD in African Americans is approaching the incidence in Caucasian populations. The current literature varies in its description of phenotype among non-Caucasians with IBD. Much of the data are derived from small case series' or retrospective data. Due to the variability between these studies, larger studies assessing a broader spectrum of variables for longer periods of time are necessary.Methods:
Adult patients with Crohn's disease were identified from Montefiore Medical Center (Bronx, NY) using the Clinical Looking Glass database from January 1, 2000 to December 31, 2013. Retrospective chart review identified data including demographics, laboratory values, medications, disease phenotype, surgery and hospitalizations. Bivariate associations were performed to determine the relationship between race and these variables. Logistic regression models were built to determine whether race was independently associated with surrogate markers of poor outcomes including perianal disease, IBD surgery and use of medications, after adjusting for covariates that were clinically relevant and associated with the outcome variable.Results:
A total of 1068 participants were included grouped by race: white (n = 391), black (n = 353), and other race (n = 324). Socioeconomic status, which was estimated from census tracting, was significantly lower in non-whites (P < 0.001). Median erythrocyte sedimentation rate (ESR) was significantly associated with race (P < 0.01); with higher values among blacks compared with whites and other races (25 versus 21 versus 20 mm/h, respectively). In the adjusted model, the odds of an ESR >20 mm/h was 1.59 times greater in blacks compared with whites (P = 0.01). Vitamin D levels were significantly associated with race (P = 0.01); the proportion of whites with levels <20 and <30 ng/mL was greater compared with non-whites. There was a statistically significant association between race and perianal disease. In the adjusted model, the odds of perianal disease among blacks and other races was 2.4 and 1.8, respectively, compared with whites (P < 0.01, P = 0.09). There was a statistically significant association between race and steroids (P < 0.001) and immunomodulators (P < 0.01), but not biologic therapy (P = 0.74). In the adjusted models, the odds of steroid use was 1.5 and 2.1 times greater in blacks and other races, respectively, compared with whites (P = 0.02, P < 0.001, respectively). In the adjusted model there was a non-significant trend towards blacks and other races having more IBD-related surgery compared with whites (OR 2.0, P = 0.11; OR 2.1, P = 0.08, respectively).Conclusions:
In this cohort of urban Crohn's disease patients, African American patients experienced more severe disease as evidenced by higher median ESR, higher prevalence of perianal disease, and a trend toward more IBD-related surgery. Despite these differences, there was no difference in biologic use. Non-Caucasians were more likely to be treated with corticosteroids which may contribute to the observed differences, including higher rates of surgery. Interestingly, non-Caucasian patients had higher median Vitamin D levels despite indicators of more severe disease. These data suggest that non-Caucasian patients may manifest a more severe disease phenotype and warrant prospective studies to further investigate these differences.