LAUSR

OBJECTIVES Provegetarian diets, also known as predominantly plant-based (but not vegetarian or vegan) or plant-forward diets, have been associated with health benefits. However, a distinction is needed between high- and low-quality provegetarian dietary patterns (PVGs). We sought to examine potential associations between PVG indices and breast cancer (BC) incidence. METHODS We assessed 10 812 women in the Seguimiento Universidad de Navarra cohort. We calculated an overall PVG pattern from a validated semi-quantitative food-frequency questionnaire as proposed by Martínez-Gonzalez et al, assigning positive scores (based on quintiles) to plant foods and reversing the quintile scores for animal foods. Participants were categorized according to tertiles of the overall score. We also calculated a healthful PVG (hPVG) and unhealthful PVG (uPVG) as proposed by Satija et al. RESULTS: After a median of 11.5 years of follow-up, 101 incident BC cases, confirmed by medical records, were observed. A significant inverse association with BC (comparing tertile 2 vs. tertile 1, HR= 0.55; 95% confidence interval, 0.32-0.95) was identified for a modest overall PVG, but not for hPVG and uPVG separately. Nevertheless, the highest tertile was not associated with BC. CONCLUSIONS In this large prospective cohort study, a moderate adherence to a PVG might decrease the risk of BC. Further studies should replicate and expand these results to other racial, ethnic, and socioeconomic groups.