Objective: To identify patient factors that are associated with emergency presentation of esophageal and gastric cancer, and further to evaluate long-term prognosis in this cohort.

Background: The incidence of emergency presentation is variable, with the prognosis of patients stabilized and discharged to return for elective surgery unknown.

Methods: The primary admission of patients with esophageal or gastric cancer within the Hospital Episode Statistics database (1997–2012) was used to classify as emergency or elective diagnosis. Multivariate regression analyses were used to identify patient factors associated with emergency diagnosis and prognosis.

Results: A total of 35,807 (29.4%) and 45,866 (39.6%) patients with esophageal and gastric cancer presented as an emergency over the study period. Age ≥70, female sex, non-white ethnicity, Charlson comorbidity index score ≥3 and more deprived Townsend index were independent predictors of emergency cancer diagnosis. Emergency diagnosis was an independent predictor of increased 5-year mortality for all patients with esophageal cancer [hazard ratio (HR) = 1.63, 95% confidence interval (CI) 1.61–1.65] and gastric cancer (HR = 1.20, 95% CI 1.16–1.23). Specifically patients receiving surgery on an elective follow-up admission with an initial emergency diagnosis had a poorer prognosis (esophageal cancer: HR = 1.35, 95% CI 1.27–1.44, gastric cancer: HR = 1.13. 95% CI 1.04–1.22), with a significant increase in liver recurrence (esophageal cancer: 7.1% vs 4.9%; P < 0.001, gastric cancer: 7.0% vs 4.8%; P < 0.001) compared to patients referred electively.

Conclusions: Emergency presentation of esophageal and gastric cancer is associated with a poor prognosis, due to the increased incidence of metastatic disease at diagnosis and a higher recurrence rate after surgery.