[Untitled]

Objectives: The aim of this study was to evaluate the reliability and adequacy of the existing system of cancer registration in the United Kingdom to monitor past and future trends in esophago-gastric cancer incidence. Methods: The Northern and Yorkshire UK Cancer Registry was interrogated for all cases of esophageal and gastric cancer occurring between 1984 and 1993. Data concerning year of registration, subsite, histology, sex, and ages were recorded and analyzed. Results: A total of 22,300 cases were identified from an estimated population of 6.7 million. The overall age- and sex-standardized incidence of gastric cancer fell over the 10-year period from 12.8 to 10.5 per 100,000 (p < 0.001) while esophageal cancer increased from 4.6 to 5.4 cases per 100,000 (p = 0.006). Adenocarcinoma of the gastric cardia increased in proportion from 29.1% to 52.2% (p < 0.0001), 70.4% of esophageal and 71% of gastric cancer registrations were recorded without details of subsite. For 25% of esophageal cancers and 36% of gastric cancers there was no histological information. Conclusions: While the trend toward an increasing incidence of adenocarcinoma at the esophago-gastric junction reported in earlier studies appears to be confirmed, the high incidence of imprecise subsite reporting of cancer registry data illustrated in this study should make us look critically at the findings of other cancer registry data. Recognition of cancer of the esophago-gastric junction as distinct from other gastric and esophageal subsites may improve accuracy of recording and allow cancer registry data to more accurately monitor the changes in esophago- gastric cancer incidence in subsequent analyses.