Background: The carbohydrate-insulin model of obesity postulates that hyperinsulinemia induces weight gain through increased hunger and lower energy expenditure. We investigated the extent to which genetic risk for hyperinsulinemia associates with energy intake and dietary composition.

Methods: We conducted a population-based study of 148,494 unrelated individuals from the UK Biobank with available genetic and dietary data. Insulin resistance (IR) and secretion (IS) were captured by a 10 and 19-variant genetic risk score (GRS) respectively. Dietary intake was evaluated using a web-based 24-hour diet recall. Linear regression models were used to examine the associations of IR-GRS and IS-GRS with total energy intake and dietary composition after adjusting for confounders. We tested for effect modification by age, sex, and body mass index. In sensitivity analyses we excluded individuals with diabetes.

Findings: The IR-GRS was associated with higher energy (β=131 kcal/day [95% CI 73 to 189] P=0.02) and polyunsaturated fat intake (β=0.52% of energy [95% CI 0.27 to 0.77] P=0.03). We noted interactions between the IR-GRS and age on carbohydrate (Pint=0.03) and sugar intake (Pint=0.02), showing that age magnified the effect of the IR-GRS on carbohydrate (top decile of age compared to the bottom decile: β=2.62% of energy[95% CI -0.02 to 5.26] vsβ=-4.24% of energy [95% CI -7.24 to -1.24]) and sugar intake (β=2.45% of energy[95% CI 0.18 to 5.12] vs. β=-3.12% of energy[95% CI -5.60 to -0.64]). These findings were consistent in individuals free of diabetes (n=140,969). There were no significant findings for the IS-GRS.

Interpretation: These data suggest that genetically driven hyperinsulinemia is associated with dietary intake, and that age could modify the association between genetic risk and carbohydrate intake. Our findings underline the importance of hyperinsulinemia on eating behavior and may serve as preliminary evidence to inform novel strategies for the prevention of common complex metabolic diseases.


J. Merino: None. J.B. Cole: None. E. Lim: None. K. Yates: None. C. Liu: None. J.C. Florez: None.


European Commission

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at