Carbohydrate Intake Variability, Weight Change and Type 2 Diabetes Risk: A Prospective Cohort Study

ABSTRACT Aim To investigate whether variability in carbohydrate intake is associated with body weight change and incident type 2 diabetes (T2D). Materials and Methods This prospective study used 24‐h dietary data from the UK Biobank. Variability in total carbohydrate, dietary fibre and sugar intake was assessed by the coefficient of variation (CV) from at least three assessments. Multivariable linear regression models were used to assess the association between the CVs and body weight change. Cox proportional hazards models were performed to estimate the association between the CVs and incident T2D. Mediation analysis was conducted to assess whether weight change mediated the associations between the CVs and incident T2D. Results Among 53 691 participants (mean age, 56 years), each 1‐standard deviation (SD) increment in the CV of total carbohydrate, dietary fibre and sugar intake was associated with 0.20–0.24 kg greater weight gain over a median 6.6‐year follow‐up (all p < 0.001). During a median follow‐up of 10.3 years, each 1‐SD increase in fibre and sugar intake variability was associated with a higher risk of incident T2D (hazard ratio [HR], 1.05; 95% confidence interval [CI], 1.00–1.10 and HR, 1.09; 95% CI, 1.04–1.14, respectively). Weight gain mediated 7% and 5% of the associations between fibre and sugar intake variability and incident T2D, respectively. Conclusion Greater variability in carbohydrate‐related intake was associated with modestly greater long‐term weight gain and a modestly higher risk of incident T2D. These findings suggest that carbohydrate intake regularity may represent an additional dimension of diet relevant to long‐term cardiometabolic health, particularly from a population‐health perspective.