A Diabetes-Related Dietary Pattern Is Associated with Incident Diabetes in Obese Men in the Korean Genome Epidemiology Study

Abstract

Diet plays an important role in both the development and management of diabetes. Using data from the Korean Genome Epidemiology Study, we assessed dietary patterns associated with the clinical indicators of diabetes. This study included 7255 subjects aged 40-69 y. Individuals with chronic diseases were excluded. The daily intakes of specific food items were assessed using a dish-based semiquantitative food-frequency questionnaire comprising 103 items; the food items were then grouped into 26 food groups. Dietary patterns were analyzed by the reduced rank regression method using glycated hemoglobin, the homeostasis model of insulin resistance, and fasting glucose concentrations as dependent variables. We investigated the associations between dietary patterns and incident diabetes using the Cox proportional hazards model. During an 11.5-y follow-up, the incidence of diabetes was 11.8/1000 person-years. The dietary pattern related to selected biomarkers of diabetes was characterized by a relatively high intake of kimchi, beef, other meat, fish, and coffee in men and a high intake of rice, kimchi, and fruit in women. In men, the association of dietary patterns with incident diabetes was significant only in the obese group, and those in the top quartile of the dietary pattern score had a 1.72 times (95% CI: 1.15, 2.56 times) greater risk of incident diabetes than those in the bottom quartile. Conversely, dietary patterns in women were not associated with incident diabetes. Using reduced rank regression, we identified dietary patterns related to selected biomarkers of diabetes in a long-term study with follow-up data in Korea.