Risk factors for barrett’s esophagus identified in a large-scale Japanese community cohort study

Barrett’s esophagus (BE) is a well-established precursor to esophageal adenocarcinoma (EAC); however, population-based data from East Asia remain limited. Identifying region-specific risk factors for BE is essential for guiding early detection and surveillance strategies. We aimed to estimate the incidence of BE and identify the demographic, clinical, lifestyle, and pharmacologic factors independently associated with BE in a large Japanese cohort. We conducted a retrospective cohort study using a comprehensive administrative database integrating health insurance claims, demographic information, and annual health checkup records. Individuals enrolled in municipal health insurance programs between April 1, 2012, and September 30, 2021, with ≥ 12 months of continuous coverage and no prior BE diagnosis were included. Incident BE was defined as ICD-10 code K22.7 in claims data. Cox proportional hazards models were applied to estimate hazard ratios for incident BE, adjusting for age, sex, comorbidities, lifestyle factors, anatomical conditions (gastroesophageal reflux disease [GERD] and esophageal hiatal hernia [EHH]), Helicobacter pylori (H. pylori) infection, and proton pump inhibitor (PPI) or potassium-competitive acid blocker (P-CAB) and histamine-2 receptor antagonist (H2RA) use. Among 620,125 eligible individuals (median follow-up: 6.2 years), 1,577 incident cases of BE were identified, yielding an incidence rate of 46.4 per 100,000 person-years (95% confidence interval: 44.2–48.7). Independent risk factors included age 50–79 years, male sex, gastroesophageal reflux disease (GERD), esophageal hiatal hernia, peripheral vascular disease, liver disease, and acid-suppressant use. History of H. pylori infection was significant in the model including GERD. In contrast, lifestyle factors including body mass index, diabetes, current smoking, and heavy drinking were not significantly associated with BE in multivariable analysis. The associations with H. pylori infection and PPI/P-CAB. use may reflect reverse causality and should thus be interpreted with caution. These results emphasize the importance of region-specific risk assessment, and support targeted endoscopic surveillance for BE; however, given the inability to account for segment length and the predominance of short-segment BE in Japan, the implications for detecting long-segment BE or EAC remain to be established.