Abstract
AIMS/HYPOTHESIS: We investigated the association between gluten intake and long-term type 2 diabetes risk among Americans. METHODS: We followed women from the Nurses' Health Study (NHS, n = 71,602, 1984-2012) and NHS II (n = 88,604, 1991-2013) and men from the Health Professionals Follow-Up Study (HPFS, n = 41,908, 1986-2012). Gluten intake was estimated using a validated food frequency questionnaire every 2-4 years. Incident type 2 diabetes was defined as self-reported physician-diagnosed diabetes confirmed using a supplementary questionnaire. RESULT: Gluten intake was strongly correlated with intakes of carbohydrate components, especially refined grains, starch and cereal fibre (Spearman correlation coefficients >0.6). During 4.24 million years of follow-up, 15,947 people were confirmed to have type 2 diabetes. After multivariate adjustment, pooled HRs and 95% CIs for type 2 diabetes, from low to high gluten quintiles, were (p(trend) < 0.001): 1 (reference); 0.89 (0.85, 0.93); 0.84 (0.80, 0.88); 0.78 (0.74, 0.82) and 0.80 (0.76, 0.84). The association was slightly weakened after further adjusting for cereal fibre, with pooled HRs (95% CIs) of (p(trend) < 0.001): 1 (reference); 0.91 (0.87, 0.96); 0.88 (0.83, 0.93); 0.83 (0.78, 0.88) and 0.87 (0.81, 0.93). Dose-response analysis supported a largely linear inverse relationship between gluten intake up to 12 g/day and type 2 diabetes. The association between gluten intake and type 2 diabetes was stronger when intake of added bran was also higher (p(interaction) = 0.02). CONCLUSIONS/INTERPRETATION: Gluten intake is inversely associated with type 2 diabetes risk among largely healthy US men and women. Limiting gluten in the diet is associated with lower intake of cereal fibre and possibly other beneficial nutrients that contribute to good health.