Background: We describe an epidemiologic analytical study of the relationship between current diet and breast cancer risk.
Method: The study design is a case-control analysis. Cases were recruited from a mammography screening program used within the national health care system; the control subjects were selected from subjects free of breast cancer in the same population. A total of 380 cases and 525 control subjects, frequency-matched for age, month of mammography, and county of residence, were identified. Of these, 265 cases and 432 control subjects were included in this analysis. Odds ratios for breast cancer in relation to food and nutrient intake were the main outcome measures.
Results: Exposure in the highest quartile of beta-carotene intake gave an odds ratio of 0.6 (95% confidence interval, 0.4 to 1.0). No increased risk was noted with high fat intake. Breast cancer risk was associated with alcohol intake only when alcohol was analyzed in quartiles: odds ratio, 1.6 (95% confidence interval, 1.0 to 2.4) for the highest quartile of intake vs the lowest. Stratified analyses showed that a high fat intake might decrease the protective effect of beta-carotene intake. Risks did not change appreciably with adjustment for total energy intake or known breast cancer risk factors.
Conclusions: As in most other studies, no strong risk factors for breast cancer have been identified in the current diet. The negative association between breast cancer risk and beta-carotene intake may be supported by a plausible mechanism, but our finding concerning alcohol should be interpreted cautiously since there was no dose-response relationship and the biological mechanism for a threshold effect at very low levels of consumption is unclear.