Breast cancer and alcohol consumption. A study in weak associations

A total of 1467 cases of women with breast cancer and 10,178 sex- and age-matched hospital controls were examined for alcohol consumption and other potential risk factors. Effects of risk factors were assessed by odds ratios, and adjustment was made for confounding variables by using stratified analyses and logistic regression. Several factors were found to influence alcohol consumption, including age, religion, education, occupation, marital status, body mass, and cigarette smoking. Lean women (Quetelet index, less than 22) had elevated unadjusted odds ratios for breast cancer of 2.10, 1.71, and 1.41 associated with consuming less than 5, 5 to 15, and more than 15 g of alcohol per day, respectively. However, this pattern is not consistent with a dose response, and adjustment for a risk profile of confounding factors, including education and occupation (which are strong correlates of age at first pregnancy and parity), reduced these estimates to 1.40, 1.24, and 0.87, none of which differs significantly from 1.0. Among all subgroups, the odds ratios adjusted for pertinent confounders and interactions fluctuated randomly about 0.94 and showed no consistent trend with increasing alcohol consumption. While these results do not entirely rule out a weak association between breast cancer and alcohol in certain subgroups, neither do they provide compelling evidence that alcohol has a role in the genesis of this malignancy.