Statements about breast cancer risk from moderate alcohol consumed daily are based on relationship observed between self-reported daily alcohol consumption at the time of study entry (or 3 years after enrollment, in Allen et al) and the relative risk of breast cancer detected during the study. As noted above, temporal and mechanistic considerations demonstrate that drinking reported at the time of study enrollment could not have caused the cancers. Presumably, however, the women in these studies had been drinking alcohol for some period of time before entering the study. Previous studies (Longnecker et al., 1995b; Terry et al., 2006), support a relationship between lifetime alcohol consumption and breast cancer risk. Therefore, if alcohol acts as a cumulative carcinogen, and alcohol consumption by postmenopausal women is a surrogate measure of lifetime alcohol consumption, then assessing alcohol consumption in postmenopausal women monitors lifetime exposure to a presumably cumulative carcinogen.
