Prospective Study of Optimal Obesity Index Cutoffs for Predicting Development of Multiple Metabolic Risk Factors: The Korean Genome and Epidemiology Study

Abstract

Background: In this prospective cohort study, we estimated the risk of developing more than I metabolic risk factor, using different obesity indices. In addition, we investigated the relative usefulness of the obesity indices for predicting development of such risk factors and calculated optimal cutoffs for the obesity indices. Methods: The cohort comprised 10 038 representative residents of a small city and a rural county who were recruited in 2001-2002. Follow-up examinations were conducted every 2 years. Among the 3857 participants without metabolic syndrome at baseline, 1102 new cases occurred during the 6-year follow-up. Receiver operating characteristic (ROC) curves for the obesity indices were plotted to compare the usefulness of the obesity indices. Results: The numbers of new cases of multiple metabolic risk factors among people in the highest quintiles of body mass index (BMI), waist circumference (WC), waist-hip ratio (WEIR), and waist-height ratio at the baseline examination were 2 to 3 times those in the lowest quintiles. The area under the ROC curve for WHR was significantly higher than that for BMI. The optimal BMI cutoff was 24 kg/m(2) in men and women, and the optimal WC cutoffs were 80 cm and 78cm in men and women, respectively. Conclusions: Both overall obesity and central obesity predicted risk of developing multiple metabolic risk factors, and WHR appeared to be a better discriminator than BMI. To prevent development of metabolic diseases among Koreans, it might be useful to lower the cutoff for abdominal obesity, as defined by WC.