Obesity has known as the main cause of type 2 diabetes (T2DM). Many simple surrogate indices of obesity, such as body mass index (BMI), waist circumference (WC), waist-to-height ratio (WtHR) has been used, however, there is controversy about which indicator is most useful. Recently a new index (a body shape index, ABSI) was developed as a more reliable surrogate index than BMI in determination of the association between cardiovascular events and body composition. We aim to assess whether or not ABIS is better than BMI and WC to predicting T2DM incidence in Korea elderly population. The CMC study is a prospective cohort study of metabolic disease. A total of 5,390 subjects (2,034 men, 3,356 women) age 40 and older, free of T2DM at baseline (2003-2006) were included in the present study, and we estimated the cumulative incidence of T2DM after 4 years follow-up. ABSI was calculated using waist circumference, weight, height (WC/(BMI2/3height1/2)). Adjusted relative risk(RR) for diabetes was estimated by Poisson regression model where index values were computed separately for each control, and receiver operating characteristic (ROC) curve analysis was applied. There were no associations between ABIS and T2DM incidence in both sexes. In men, BMI and WC did not show significant associations either, and in women, the RR of highest quartile relative to the lowest quintile was significant (1.95 (1.27-3.01)) only in WC. AUCs (area under the ROC curve) and 95% CIs of WC were 0.613(0.591-0.634) in men and 0.642 (0.626-0.659) in women, and showed significant differences compared with the AUCs of ABSI (0.567 (0.545-0.588) for men, 0.580(0.563-0.597) for women) and BMI (0.578 (0.556-0.599) for men, 0.601 (0.584-0.618) for women) in both sexes. In this study, ABSI was not better than a WC in predicting the ability for T2DM, and WC was a good surrogate index of obesity to T2DM incidence especially in Korea elderly women.


S. Lim: None. J. Lee: None. K. Yoon: None.

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 http://www.diabetesjournals.org/content/license.