An association between gastric atrophy and esophageal squamous cell carcinomas (ESCC) has been described. However, the mechanism of this association is unknown. In this study, we aimed to examine this relationship in a cohort of patients with varying grades of gastric atrophy to increase the understanding about the causality of the association. Patients diagnosed with gastric atrophy between 1991 and 2005 were identified in the Dutch nation wiiiie histopathology registry (PALGA). The incidence of ESCC and, presumably unrelated, small cell lung carcinomas (SCLC) observed in these patients was compared with that in the general Dutch population. Relative risks (RRs) and 95% confidence intervals were calculated by a Poisson model. At baseline histological examination, 97,728 patients were diagnosed with gastric atrophy, of whom 23,278 with atrophic gastritis, 65,934 with intestinal metaplasia and 8,516 with dysplasia. During follow-up, 126 patients were diagnosed with E, CC and 263 with SCLC (overall rates 0.19, respectively 0.39/1,000 person-years at risk). Compared with the general Dutch population, patients with gastric atrophy ran a RR of developing ESCC of 2.2 [95% Cl 1.8-2.6] and of SCLC of 1.8 [95% Cl 1.6-2.1]. The risk of ESCC did not increase with increasing severity of gastric atrophy (p = 0.90). In conclusion. this study found an association between gastric atrophy and both ESCC and SCLC, but the risk of ESCC did not increase with the severity of gastric atrophy. Therefore, a causal relationship seems unlikely. Confounding factors, such as smoking, may explain both associations. (C) 2008 Wiley-Liss, Inc.