Counting cancers at the junction - a problem of routine statistics.

The increase over time in the incidence of cancer arising at the oesophagogastric junction has been the subject of many papers reviewing data obtained from cancer registries and other sources of routine statistics. The analysis of such data is beset with a number of problems, all of which compromise comparability over time and hence complicate interpretation. This makes it extremely difficult to assess with any degree of reliability the quantitative extent to which these cancers really are increasing. Some recent datasets, such as from the Eindhoven Cancer Registry, are now providing higher-quality information that can remedy this deficiency. In the absence of such routine information, useful insights can be obtained from analysis of appropriate clinical datasets that exist in Japan.