Association of the glycaemic index and glycaemic load with colorectal cancer in the population of Córdoba (Argentina): results of a case-control study using a multilevel modelling approach.

The glycaemic index (GI) and glycaemic load (GL) are involved in the aetiology of different diseases, and they could be related to the development of colorectal cancer (CRC). The aim of this study was to evaluate the association between the quality and quantity indicators of carbohydrates consumed by the population of Córdoba (Argentina) and the odds of developing CRC in 2008-2016 period. A case-control study was conducted with 492 participants (161/331 cases/controls), interviewed through a validated FFQ. Multilevel logistic regression models were used to assess the effect of GI, GL and the quantity or weekly intake of high-GI foods on CRC occurrence, following adjustment for individual/first-level covariates, and using level of urbanisation as the contextual variable. The models were stratified by sex. Participants in the highest v. lowest tertile of dietary GL and weekly intake of high-GI foods had increased odds of CRC presence in the entire sample (OR 1·64, 95 % CI 1·16, 2·34 and OR 1·11, 95 % CI 1·09, 1·14, respectively) and in women (OR 1·98, 95 % CI 1·24, 3·18 and OR 1·41, 95 % CI 1·09, 1·83, respectively). In men, the second tertile of GL and weekly intake of high-GI foods were associated with CRC (OR 1·44, 95 % CI 1·04, 1·99 and OR 1·48, 95 % CI 1·32, 1·65, respectively). Also, GI was associated with CRC in women (highest v. lowest tertile OR 2·12, 95 % CI 1·38, 3·27). In addition to the quantity and quality of carbohydrates intake, it is important to consider the frequency of consumption of high-GI foods in CRC prevention.