Abstract BACKGROUND Despite advances in the management of upper gastrointestinal bleeding (UGIB), hospital admission rates for patients with Inflammatory Bowel Disease (IBD) may not have declined. We analyzed the frequency and temporal trends in hospitalization rates for UGIB among those with IBD. METHODS Population-based, administrative healthcare databases were used to identify all individuals who visited the emergency department for a non-variceal UGIB in Alberta, Canada between fiscal year 2010/11 and 2019/20. Alberta Health Services is the only healthcare provider for an estimated 4.39 million Albertans as of 2020. International Classification of Diseases (ICD) identified patients admitted to hospital with UGIB and IBD ICD-10 codes at any diagnostic field. The frequency of hospitalizations in Alberta were stratified by sex, age group, and Charleston comorbidities (one, two, three or more). Hospitalization frequencies were also stratified by quintiles of the Pampalon index of social and material deprivation. The 1st quintile of the Pampalon index is the most privileged and the 5th quintile is the most socially or materially deprived. Differences in frequencies between covariates were measured using a goodness of fit Chi-square test and differences in proportions were measured using a z-test of proportions. Annual hospitalization rates were calculated using annual population estimates of Alberta. Average Annual Percentage Change (AAPC) in hospitalization rates and associated 95% confidence intervals (CIs) were calculated using Poisson regression, or negative binomial regression if data were overdispersed. RESULTS From 2010/11 to 2019/20, UGIB hospitalization rates among those with IBD decreased from 4.29 (95% CI: 2.63, 7.00) to 1.83 (95% CI: 0.92, 3.67) hospitalizations per 1,000,000 Albertans (AAPC = −7.52; 95% CI: −12.45, −2.32, p = 0.0049). Of the 159 individuals living with IBD, 120 had Crohn’s disease and 39 had ulcerative colitis. The proportion of UGIB hospitalizations was greater among males (62.26%), those 25-64 years of age (62.89%), and those with zero Charleston comorbidities (52.87%). Individuals who were more privileged materially or socially made up a greater proportion of UGIB hospitalizations especially among the 1st, 2nd, and 3rd quintiles in material deprivation and 1st, 3rd, and 4th quintiles in social deprivation. The mean length of stay for a UGIB among those with IBD was 6.31 days (95% CI: 5.23, 7.38). CONCLUSION Hospitalization rates for UGIB among those with IBD are decreasing in Alberta. However, there are differences in the frequency and proportion of hospitalization rates among those materially or socially deprived with more privileged quintiles making up a larger proportion of hospitalizations.
Buie et al. (Thu,) studied this question.