Methyl Donor Nutrient Intake and Incidence of Type 2 Diabetes: Results From Three Large US Cohorts

OBJECTIVE We examined whether intake of methyl donor nutrients, including vitamins B-2, B-6, and B-12 and folate, from foods and/or supplements is associated with type 2 diabetes risk. RESEARCH DESIGN AND METHODS We included 203,644 women and men from the Nurses' Health Study (1984-2016), Nurses' Health Study 2 (1991-2017), and Health Professionals Follow-Up Study (1986-2016). Dietary data were collected every 2-4 years with use of semi-quantitative food-frequency questionnaires. Cox proportional hazards models with time-varying covariates were used to evaluate associations between each nutrient and type 2 diabetes risk. We combined cohort-specific hazard ratios (HRs) using inverse variance-weighted fixed-effects meta-analyses. RESULTS During 4,900,181 person-years of follow-up, we documented 19,475 incident type 2 diabetes cases. In multivariable-adjusted meta-analyses, participants in the highest quintiles of total vitamin B-2 and B-6 intakes had lower risk of diabetes compared with those in the lowest quintiles (HR 0.93 [95% CI 0.89, 0.98] for B-2 and 0.93 [0.89, 0.97] for B-6). With stratification by source, significant associations remained for B-2 from food but not from supplements. Neither association for B-6 from food nor association for B-6 from supplements attained significance. No association was observed between total B-12 intake and diabetes. However, B-12 from food was marginally associated with higher diabetes risk (1.05 [1.00-1.11]) but not after additional adjustment for red meat intake (1.04 [0.99-1.10]). No evidence of association was observed between intakes of folate and diabetes. CONCLUSIONS The results of our study suggest that higher intake of vitamin B-2 and B-6, especially B-2 from food sources, may be associated with a modestly lower type 2 diabetes risk.