Abstract
BACKGROUND: The prevalence of obesity and associated diseases, including several cancers, continues to rise. OBJECTIVES: Given the growing evidence that the degree of food processing impacts health, we investigated the relationship between the degree of food processing and the risk of obesity-related cancers (ORCs). METHODS: We analyzed data from 17,756 participants in the "Seguimiento Universidad de Navarra" project, a prospective cohort of Spanish graduates. Dietary intake was assessed using a validated 136-item food frequency questionnaire at baseline and after 10 y, with foods classified according to the NOVA system. Tertiles of unprocessed or minimally processed food (UMPF) and ultraprocessed food (UPF) consumption were established based on the distribution of percentage energy intake from each group at baseline. ORC cases were self-reported and confirmed through blind review of medical charts and consultation of the National Death Index. Time-dependent Cox proportional hazard models adjusted for potential confounders were used to estimate hazard ratios (HRs). RESULTS: After a median follow-up of 14 y, 263 confirmed cases of ORC were identified. The HR for ORC in the highest baseline tertile of UMPF consumption was 0.70 [95% confidence interval (CI): 0.49, 1.00; P-trend = 0.042] as compared with the lowest tertile. These results remained unchanged using the cumulative mean of UMPF consumption after 10 y of follow-up: 0.69 (95% CI: 0.48, 0.99; P-trend = 0.036). We found no significant results for UPF consumption. CONCLUSIONS: In this cohort, a higher consumption of UMPF was associated with a 30% relatively lower risk of ORC in the long term. These results support the importance of food policies encouraging the consumption of UMPF for cancer prevention.