Comparison of different anthropometric indices for identifying dyslipidemia in school children

Abstract Background Anthropometric indices have been associated with dyslipidemia. Objective To compare the abilities of BMI, waist circumference (WC) and WC/height to identify children's dyslipidemia. Methods Students 1261 (639 male) age 9.5 ± 2.1 years. were examined for anthropometry and lipid levels. Triglycerides ≥ 1.69 mmol/L and high-density lipoprotein cholesterol (HDL-C) Results The prevalence of abnormal triglycerides was 4.1% and HDL-C 8.1%. The areas under the receiver operator curves (ROC) were: BMI = 0.87, WC = 0.83, and WC/height = 0.84 in predicting both low HDL-C and high triglycerides. Multiple regression analyses showed that the odds ratios (OR) were highest for WC [6.5], followed by WC/height [5.4], and BMI [4.9], for dyslipidemia. Conclusions The results suggest that WC, WC/height, and BMI similarly predicted dyslipidemia, using ROC analyses. However, regression analyses showed that WC followed by WC/height was most predictive of dyslipidemia.