Cigarette smoking and breast cancer: a case-control study of screening program participants

A case-control study of Canadian women aged 40-59 years (n = 123; 369) has been conducted on the relationship of cigarette-smoking and initial visit breast cancer detection within a multicenter screening program. Among premenopausal women, a relative risk of 2.1 (1.1, 4.0) was found for ever-smokers versus never-smokers. There was a dose-response gradient with relative risks of 1.0, 1.3, 2.5, and 3.5 among women with 0, 1-200, 201-500, and greater than 500 cigarette-years of exposure, respectively. No overall association was detected among postmenopausal women but a significant interaction with parity was present in this group. When menopausal status was ignored, there was a nonsignificantly elevated risk of 1.4 (0.9, 2.1) for ever- versus never-smokers but the dose-response was significant with relative risks of 1.0, 0.9, 1.7, and 2.0 in the above four exposure categories, respectively. These results persisted despite adjustment for several important variables. These data are consistent with an interaction of smoking and participation in the screening study or with a possible etiologic role for smoking. Even if smoking is not related to breast cancer etiologically, these data still suggest that smokers who visit screening centers have an elevated risk of breast cancer detection on initial visit, especially among premenopausal women. They do not support the hypothesis of a protective effect of smoking and underscore the need for further study into this important relationship.