Alcohol
Although the association between alcohol consumption and breast cancer risk has been controversial, substantial evidencehas accumulated to support the existence of a positive association. In a meta-analysis of 38 case control and cohort studies, Longnecker estimated relative risks of 1.1 (95% confidence interval, 1.1 to 1.2) for 1 drink per day, 1.2 (1.1 to 1.3) for 2 drinks per day, and 1.4 (1.2 to 1.6) for 3 drinks per day. In the five largest prospective studies, all controlled for major breast cancer risk factors, relative risks for the highest category of alcohol consumption compared with zero consumption were 3.3 (95% confidence interval, 1.2 to 9.3) based on 303 cases, 1.6 (1.0 to 2.6) based on 2,933 cases, 1.5 (1.0 to 2.0) based on 493 cases, 1.2 (0.8 to 1.9) based on 519 cases, and 1.7 (p for trend = .05) based on 422 cases

In a pooled analysis of the six cohort studies examining data on alcohol and dietary factors that included 200 or more cases, the risk of breast cancer increased monotonically with increasing intake of alcohol, with no statistical evidence of heterogeneity among studies. The multivariate relative risk for a 10 g per day increase in alcohol was 1.09 (95% confidence interval, 1.04 to 1.13). In this and other analyses, adjustment for known breast cancer risk factors and dietary variables hypothesized to be related to breast cancer had little impact on the association with alcohol. In the collective literature, beer, wine, and liquor all contribute to the positive association, which strongly suggests that alcohol per se is responsible for the increased risk.

Whether reducing alcohol consumption in middle life decreases risk of breast cancer is an important practical issue. In a 1987 report, women who drank before age 30 years and later stopped experienced a similar elevation in risk to those who continued to drink. In a large study designed to address this issue, however, recent consumption of three or more drinks per day was associated with a relative risk of 2.2, whereas the relative risk was 0.9 for consumption of three or more drinks per day from ages 16 to 29 years. This suggests that recent adult drinking may be more important than drinking patterns earlier in life and that reductions in consumption in midlife should reduce risks of breast cancer.

In intervention studies, consumption of approximately two alcoholic drinks per day increased total and bioavailable estrogen levels in premenopausal women, and single doses of alcohol acutely increased plasma estradiol levels in postmenopausal women. These findings suggest a mechanism by which alcohol may increase breast cancer risk. In a cross-sectional study, alcohol intake was associated with elevated plasma levels of estrone sulfate, a long-term indication of estrogen status, which in turn was associated with future risk of breast cancer. In a large prospective analysis, high intake of folic acid appeared to mitigate completely the excess risk of breast cancer due to alcohol, although folic acid intake was not associated with breast cancer risk among nondrinkers. Because alcohol metabolites inactivate folic acid and low folate levels are associated with increased misincorporation of uracil into DNA, this finding suggests another mechanism for the adverse effects of alcohol.

Of all the associations between dietary factors and breast cancer risk, the relationship with alcohol is by far the most consistent. This association has been observed in many diverse populations and cultures, and rigorous attempts to account for this relationship by other variables have been unsuccessful. Moreover, the effect of alcohol on endogenous estrogen levels provides a plausible mechanism. Together, this body of data provides strong evidence for a causal relationship between alcohol consumption and breast cancer risk. The public health implications of this knowledge, however, are complicated by the fact that consumption of one to two alcoholic beverages per day is almost certainly protective against cardiovascular disease. Because cardiovascular disease is the leading cause of death among women, moderate drinking is associated overall with a modest reduction in total mortality. Although the situation is still complex, reduction of daily alcohol consumption appears to be one of relatively few methods of actively reducing breast cancer risk, whereas many methods exist for reducing the risk of cardiovascular disease.

Caffeine
Considerable speculation that caffeine may be a risk factor for breast cancer followed a report that women with benign breast disease experienced relief from symptoms after eliminating caffeine from their diets. Most case control studies, however, have not observed evidence of a positive association with breast cancer. In prospective studies, no increase in breast cancer risk has been seen, and in one a weak, but significant, inverse association between caffeine consumption and breast cancer risk was observed. Similarly, no evidence for an association between tea consumption and risk of breast cancer has been seen in epidemiologic studies. Thus, the epidemiologic evidence is not compatible with any substantial increase in breast cancer risk associated with drinking coffee or tea.