Intakes of Vegetables and Potatoes and Associations with Incident Diabetes

OBJECTIVES: Higher vegetable intakes are hypothesized to reduce the risk of developing diabetes but the findings are weak. Using Diet, Cancer and Health cohort, we aimed to 1) analyze the relationship between intakes of total vegetable, vegetable subgroups and incident diabetes, 2) analyze separately the association between intakes of total potatoes (excluding fries/chips) and incident diabetes, and 3) understand whether the relationship between vegetable intake and incident diabetes is mediated by baseline BMI. METHODS: Cross-sectional associations between exposures (baseline intakes of total vegetables, vegetable subgroups, and potatoes, estimated from a food frequency questionnaire), and baseline BMI were assessed by multivariable-adjusted linear regression models. Non-linear associations between exposures and incident diabetes were examined by multivariable-adjusted Cox proportional hazards models with restricted cubic splines. Mediation by BMI was quantified through natural direct and indirect effects. RESULTS: Among 54,984 participants (median age: 56 years), 7836 diabetes cases were recorded during median follow-up of 16.3 years. Participants in the highest total vegetable intake quintile (median: 319 g/d) had a 0.29 kg/m(2) (95% CI: −0.39, −0.18) lower BMI and a 20% (95% CI: 14, 25%) lower risk of incident diabetes after multivariable adjustments, compared to those in the lowest quintile (median: 67 g/d). Baseline BMI mediated 20% (95% CI: 13, 31%) of the association between vegetable intakes and diabetes. Conversely, those in the highest potato intake quintile (median: 256 g/d) had a 9% (95% CI: 2, 16%) higher risk of diabetes after multivariable adjustments and compared to those in the lowest quintile (median: 52 g/d). Of the vegetable subclasses, higher intakes of green leafy, cruciferous, yellow/orange/red, other vegetables and legumes were significantly associated with a lower risk of diabetes. Stratified analyses on the absolute scale suggest that those at a higher risk of diabetes (BMI ≥30) may benefit the most from higher vegetable intakes. CONCLUSIONS: The findings provide evidence that a higher vegetable, but not potato, intake might help to mitigate diabetes risk, partly by reducing BMI. FUNDING SOURCES: Pratik Pokharel is supported by an Edith Cowan University Higher Degree by Research Scholarship Australia.