Background: Diabetes mellitus (DM) has been considered a potential risk factor for gastric cancer, but the evidence is conflicting. We evaluated the association of DM with incident gastric cancer in a large cohort of men and women with endoscopic assessment at baseline and during follow-up.

Methods: We performed a retrospective cohort study of 195,312 adult men and women who underwent upper endoscopy at baseline and during follow-up between 2003 and 2014. DM was defined as fasting serum glucose ≥ 126 mg/dL, self-reported history of DM or current use of antidiabetic medications. Gastric cancer was confirmed histologically.

Results: The prevalence of DM at baseline was 3.0% (n = 5774). Over 865,511 person-years of follow-up, 198 participants developed gastric cancer. The fully adjusted hazard ratio (HR) for incident gastric cancer comparing participants with and without DM at baseline was 1.76 [95% confidence interval (CI) 1.04-2.97; P = 0.033). When we evaluated DM as a time-varying covariate, the fully adjusted HR was 1.66 (95% CI 1.04-2.68; P = 0.036). The association between DM and incident gastric cancer did not differ by the presence of intestinal metaplasia (P for interaction = 0.61).

Conclusions: In this large cohort with endoscopic follow-up, DM was independently associated with increased gastric cancer incidence. The increased risk was independent of mucosal atrophy and intestinal metaplasia and was consistent in participants with newly developed DM during follow-up. Patients with DM may require more intensive endoscopic follow-up for gastric cancer screening.