Calculation of population attributable risk for alcohol and breast cancer

Objectives: Because of increasing evidence that alcohol may be causally associated with breast cancer, we reconsider the population attributable risk (PAR) for alcohol and breast cancer for the US adult female population using an effect estimate from a meta-analysis and incorporating a revised perspective on measurement error correction. Methods: To estimate PAR, we employed a formula appropriate to use with an adjusted effect estimate. To estimate intermediate quantities needed to apply that formula, we used adjusted relative risk estimates from a previously published meta-analysis, as well as SEER cancer statistics and general population data from the third National Health and Nutrition Examination Survey. We used relative risk estimates uncorrected for measurement error. Results: The estimated age-adjusted PAR for alcohol and breast cancer was 2.1%. Conclusions: Because of the modest association between alcohol and breast cancer and the generally moderate level of alcohol intake among US women, the proportion of breast cancer attributable to alcohol intake is small. Widespread efforts to reduce alcohol consumption would not have a substantial impact on breast cancer rates in this population. While selected subgroups of women might benefit from decreasing alcohol consumption, specific profiles for such women have yet to be defined and defended.