Potato consumption, polygenic scores, and incident type 2 diabetes: An observational study

Abstract

Whether the consumption of different processed potatoes is detrimental to type 2 diabetes (T2D) is highly debated. This study aimed to assess the relations between potato consumption and the risk of T2D and whether the relationship was modified by the genetic predisposition to T2D. We included 174,665 participants from the UK Biobank at baseline. Potato consumption was evaluated using the 24-hour dietary questionnaire. The genetic risk score (GRS) was calculated based on 424 variants associated with T2D. After adjustment for demographic, lifestyle, and dietary factors, the consumption of total potatoes was significantly and positively associated with T2D risk [hazard ratio (HR) comparing two or more servings/day with non-consumers was 1.28 (95% CI: 1.13–1.45)]. HRs (95% CIs) of T2D for each 1-SD increment in boiled/baked potatoes, mashed potatoes, and fried potatoes were 1.02 (0.99–1.05), 1.05 (1.02–1.08), and 1.05 (1.02–1.09), respectively. There were no significant interactions between the consumption of total or different processed potatoes and overall GRS on T2D risk. Theoretically, replacing one serving/day of total potatoes with the same amount of non-starchy vegetables was related to a 12% (95% CI: 0.84–0.91) lower T2D risk. These results showed the positive associations of the consumption of total potatoes, mashed potatoes or fried potatoes and genetic risk with higher incident T2D. An unhealthy potato-based diet is associated with higher diabetes risk regardless of genetic risk.