Association between country of birth and gastric intestinal metaplasia: a retrospective cohort study.

Abstract

As a precursor to gastric cancer, gastric intestinal metaplasia (GIM) represents a target for surveillance. US-based guidelines recommend surveillance of racial/ethnic minorities and immigrants from high incidence gastric cancer regions, yet there is marked variability in prevalence amongst these subgroups and within groups from high incidence regions. There is a paucity of information regarding country of birth as a risk factor for GIM and we sought to determine the association between country of birth and GIM in an ethnically and racially diverse US population. This was a retrospective cohort study of persons who underwent esophagogastroduodenoscopy (EGD) with gastric biopsy at University of Miami Hospital between 2011 and 2021. A natural language processing (NLP) algorithm was developed and implemented to extract diagnoses of GIM and <i>Helicobacter pylori</i> (HP) infection from endoscopic pathology reports. Multivariable logistic regression was performed to evaluate risk factors for GIM, accounting for important covariates, including country of birth. A total of 21,108 persons from 130 varying countries of birth were included in the study. A total of 1699 cases of GIM were identified yielding a prevalence of 8.0% (95% CI: 7.7-8.4%). Multivariable analysis was restricted to countries with at least 100 persons in the cohort, yielding 15 countries with 1208 cases of GIM. Country of birth (p < 0.0001), race/ethnicity (p = 0.026), active HP infection (p < 0.0001), and increasing age (p < 0.0001) were significantly associated with increased odds of GIM. Highest odds for GIM were among persons born in Ecuador (OR 2.34, 95% CI 1.56-3.50), Honduras (OR 2.34, 95% CI 1.65-3.34), and Peru (OR 2.17, 95% CI 1.58-2.99). We demonstrate that country of birth is a key risk factor for GIM. Not all countries that are thought to be in "high-risk" regions are associated with higher rates of GIM, underlining the importance of studying the under-investigated risk factor of country of birth. Guidelines should account for country of birth, in addition to other risk factors, to tailor screening/surveillance appropriately. Shida Haghighat, MD, MPH is supported by an NIH training grant T32 DK 11667805.