Potato consumption and risk of type 2 diabetes: A dose–response meta-analysis of cohort studies

Summary Background & aims High potato intake has been suggested as a risk factor for the development of type 2 diabetes. We aimed to investigate the association between potato consumption and risk of type 2 diabetes. Methods A systematic review was conducted on PubMed and Embase from the database commencement until September 2017 (updated by June 2018) following the MOOSE guidelines. The random effect model dose–response meta-analysis method of Greenland and Longneck was used to estimate the maximally adjusted log hazard ratio (HR) for a unit (serving per day) increment of potato consumption. A restricted cubic spline model with three knots was used to evaluate the potential non-linear relationship. Results A total of 3544 citations were retrieved from the databases, of which six prospective cohort studies including 4545230 person-year of follow-up and 17,758 diabetes cases met the inclusion criteria. The pooled dose–response HR per an increment of 1 serving/day of total potato consumption was 1.20 (95% CI 1.13 to 1.127, P 2  = 27.1%, P for heterogeneity = 0.23) both in men and women. The larger risk were observed for 2 serving/day (HR 1.44, 95% CI 1.28 to 1.63) and 3 serving/day (HR 1.74, 95% CI 1.45 to 2.09). We found significant evidence of a non-linear association between total potato consumption and risk of type 2 diabetes (X 2  = 17.5, P for linearity  Conclusion Long-term high consumption of potato (each serving a day increase) may be strongly associated with increased risk of diabetes. These findings suggest that diet–health policy may be of importance in the prevention of diabetes.