Risk factors for breast cancer in women in northern Alberta, Canada, as related to age at diagnosis.

A population-based case-control study involving interviews with 577 female breast cancer patients and 826 controls in northern Alberta. Canada, revealed that some determinants of breast cancer varied according to age. Among women under age 45, risk factors included a younger age at menarche, late age at last birth, high parity, and recent use of oral contraceptives. At older ages risk was related to natural as opposed to surgical menopause, late age at first birth, low parity, late age at natural menopause, and tonsillectomy. At all ages there was an increased risk of breast cancer associated with difficulty in conceiving, benign breast disease, not having breast fed, and a history of breast cancer among mothers or sisters. For some variables the age differences were pronounced; the combination of low parity and late age at first birth was associated with a sevenfold increase in breast cancer risk at risk at ages 55-80 but a slight decrease at ages under 45. The effect of tonsillectomy steadily increased with age and represents a new lead, but certain features of the data suggest that the link to oral contraceptives among among younger women and the inverse relation to breast feeding at all ages may not be causal. Even though design limitations (cases interviewed in a different setting from controls) appeared not to influence conclusions, the results may have been subjected to interview bias and thus should be interpreted cautiously.