"Risk factors for benign breast disease".

Abstract

Brinton et al. report an inverse relationship between oral contraceptive (OC) use and benign breast disease. Their contention that the association is real and not due to selective bias, as has been suggested by Janerich et al., is difficult to accept from the data that are reported. Adequate control for both selection bias and confounding factors influenced by physicians' prescribing practices and the benign breast disease patients' choice of contraception has not been carried out in the majority of studies reported. In the study conducted by Brinton et al., careful control of influencing factors such as reason for discontinuing OC use, prior breast disease and social class was carried out, but there is a strong possibility of a selection bias affecting the results. Cases and controls were selected from participants in the Oxford-FPA study who were restricted to married women 25-39 years of age and who were currently using OCs, an IUD, or a diaphragm without previous OC use. Thus, women who, prior to 25 years of age, elected to discontinue OC use because of breast discomfort related to benign breast disease, or women whose physician discontinued prescribing OCs because of benign breast disease, would have been entirely selected out of the Oxford-FPA study. Such women would have no opportunity of being included in Brinton's study, reducing the overall likelihood that OC ever users and longterm users would be found among the women with benign breast disease. This could account for some or all of the deficit of OC ever users and longterm users among the benign breast disease cases found in Brinton's study. Brinton's thorough analysis has contributed to the understanding of several factors related to the risk of benign breast disease such as obesity and fertility factors. The association between parity and late age at 1st birth and benign breast disease must be sorted out before the interpretation of a "protective effect" of OCs on benign breast disease can be accepted.