Family history of cancer in a female cohort population in Japan.

This study was aimed at examining the effect of familial predisposition on cancer incidence in a cohort population. Family history (FH) of cancer and reproductive history were obtained by a simple questionnaire from about 40,000 participants in a cervical cancer screening program visiting screening clinics in 1977. Number and age structure of family members were not known. About 40% of subjects reported positive FH which was not medically verified. Comparing the observed and expected numbers of pair-wise combination of various cancers observed among the family members, we found more intense cluster of cancers of the same than different sites. This population was followed until 1987 by using the population-based cancer registry and 638 cancer cases were detected. By matching residence and birth year, the same number of nonaffected controls was chosen from the original cohorts. Positive effect of FH was detected for large bowel and breast cancers, but not for stomach and cervical cancers. Older age at marriage and lower parity were noted for breast cancer cases.