PubMed1986
Genealogies for the Mexican-American city of Laredo, Texas, have been assembled by computer from individual civil and church records of birth, marriage, and death. Documentation is available on vital events in the lives of over 300,000 individuals, about 80% of the city population from 1870-1981. These data were collected to determine the degree to which death from cancer is more clustered in families than would be expected by chance alone; methods specific to this data base have been developed to accomplish this task. A statistically significant excess of familial cancer was observed overall when all cancer sites were pooled, but no evidence was observed for excess familial risk at single sites except for breast cancer and perhaps for ovarian cancer. The excess of breast cancer risk is comparable to that observed in other populations. A few site-combinations manifest excess familial risk, most notably those involving and dominated by breast cancer and certain digestive system sites. We do not confirm the degree of familiarity observed elsewhere for cancers of the lung, colorectum, stomach, or other sites in this generally low-risk population. Even where we find evidence of excess risk, the degree of excess is small and the number of multiply affected families too small to test etiologic models by segregation analysis. The absence of excess familial risk does not appear to be due to inadequate numbers of cases, since breast cancer is familial with no more occurrences in Laredo than other sites. These results differ to some extent from those found in a similar study of Utah Mormons, but it is unclear whether this is because of differences in risk patterns or statistical properties of the analytic methods used in the two studies.