Abstract

Cancer registration data for a defined geographical area, covering a seven year period, were modified to include the variable "Asian ethnic origin." The data were then used to test the hypothesis that a difference would be found between Asians and non-Asians in the pattern of cancer by site. Whereas the incidence of cancer of all sites appeared to be significantly lower in Asians (p less than 0.05), after taking account of this, and adjusting for the very different age distributions of the two populations, a highly significant difference (p less than 0.0005) was found between the two groups in the distribution of cancer between sites. Although, given the size and young age structure of the Asian population, absolute numbers of cases were small, a significant excess of Asian cases (compared with the expected) occurred for cancer of the tongue, oral cavity, pharynx, and oesophagus. For most sites there were fewer Asian cases than would be expected, particularly so for the stomach, testis, and skin. The results indicate the need for formal epidemiological study to test specific aetiological hypotheses which may account for these apparent differences.