Intake of fruits and vegetables and risk of breast cancer - A pooled analysis of cohort studies

Context Some epidemiologic studies suggest that elevated fruit and vegetabl e consumption is associated with a reduced risk of breast cancer. However, most have been case-control studies in which recall and selection bias may influence the results. Additionally, publication bias may have influenced t he literature on associations for specific fruit and vegetable subgroups. Objective To examine the association between breast cancer and total and sp ecific fruit and vegetable group intakes using standardized exposure defini tions. Data Sources/Study Selection Eight prospective studies that had at least 20 0 incident breast cancer cases, assessed usual dietary intake, and complete d a validation study of the diet assessment method or a closely related ins trument were included in these analyses. Data Extraction Using the primary data from each of the studies, we calcula ted study-specific relative risks (RRs) that were combined using a random-e ffects model. Data Synthesis The studies included 7377 incident invasive breast cancer ca ses occurring among 351825 women whose diet was analyzed at baseline. For c omparisons of the highest vs lowest quartiles of intake, weak, nonsignifica nt associations were observed for total fruits (pooled multivariate RR, 0.9 3; 95% confidence interval [CI], 0.86-1.00; P for trend =.08), total vegeta bles (RR, 0.96; 95% CI, 0.89-1.04; P for trend=.54), and total fruits and v egetables (RR, 0.93; 95% CI, 0.86-1.00; P for trend=.12). No additional ben efit was apparent in comparisons of the highest and lowest deciles of intak e. No associations were observed for green leafy vegetables, 8 botanical gr oups, and 17 specific fruits and vegetables. Conclusion These results suggest that fruit and vegetable consumption durin g adulthood is not significantly associated with reduced breast cancer risk .