Carbohydrate Quantity and Quality and Risk of Type 2 Diabetes: Results from Three Large Prospective US Cohorts
Abstract Objectives Carbohydrate intake has been reported to be associated with higher type 2 diabetes (T2D) risk, but high and low quality of carbohydrate may have different effects. Furthermore, these effects may differ depending on overall macronutrient composition. We aimed to examine associations of isocalorically substituting high quality carbohydrates (HQC) and low quality carbohydrates (LQC) with other macronutrients on T2D risk. Methods We included 69,949 women from the Nurses’ Health Study, 90,239 women from the Nurses’ Health Study 2, and 40,539 men from the Health Professionals Follow-up Study. Dietary data were collected every 2–4 years using a semi-quantitative food frequency questionnaire, from which dietary intake of LQC and HQC were calculated. LQC was defined as the percentage of energy (E%) from carbohydrates from refined grains, sugary foods, and potatoes, and HQC as E% from carbohydrate from whole grains. Hazard ratios were estimated using Cox proportional hazard regression analyses with time-varying covariates. Results During 4389,996 years of follow-up, we documented 11,872 T2D cases. In pooled multivariable-adjusted analyses, substituting 5E% from saturated fat with 5E% from LQC was associated with a higher T2D (HR[95% CI] = 1.05[1.00–1.09]) while substitution with HQC was associated with a lower T2D risk (HR[95% CI] = 0.93[0.87–0.99]). Isocaloric substitution of other macronutrients with LQC was not associated with T2D risk. On the other hand, isocaloric substitution of HQC was associated with lower T2D risk when the replacement nutrient was monounsaturated fat (HR[95% CI] = 0.88[0.83–0.94]), polyunsaturated fat (HR[95% CI] = 0.92[0.86–0.98]), trans fat (HR[95% CI] = 0.90[0.85–0.97]), animal protein (HR[95% CI] = 0.88[0.83–0.93]), and vegetable protein (HR[95% CI] = 0.90[0.84–0.96]). Conclusions Our study suggests that higher intake of HQC, especially from whole grains, is associated with a lower T2D risk, irrespective of the macronutrient it replaces. In contrast, a higher intake of LQC is associated with a higher risk of type 2 diabetes, but only when it replaces saturated fat. Our findings highlight the importance of making a distinction between carbohydrate from high and low quality sources and taking into account different substitutions when examining macronutrients. Funding Sources National Institutes of Health.