Purpose: The purpose of this study was to identify how accurately body mass index (BMI) and waist circumference (WC) detect obesity in young adult women. Method: Measurements of height, weight, WC, and percent body fat (% BF) were obtained and bioelectrical impedance analysis was used to estimate body fat in 329 female college students. The sensitivity and specificity to screen obesity by BMI and WC were determined using SPSS. Received operating characteristic (ROC) curve analysis was used to assess the appropriate BMI and WC predicting % BF. Results: % BF-defined obesity $(geq30%)$ had higher prevalence than BMI-defined obesity $(geq25kg/m^2)$ and WC-defined obesity $(geq80cm)$. BMI $geq25kg/m^2$ and WC $geq80cm$ had high specificity (both, 100%), but low sensitivity (respectively, 13.5% and 22.9%) in detecting % BF-defined obesity. The BMI and WC cutoff values corresponding to % BF-defined obesity were 21.2kg/$m^2$ and 73cm, which were lower than recommended reference values for Korean women. These values decreased specificity but increased sensitivity to detect obesity. The areas under the ROC curve were good (0.84, 0.86) for BMI and WC. Conclusion: BMI and WC have good specificity but miss more than $77sim86%$ of people with excess fat. Therefore, BMI and WC cutoff values need to be revised and body fat should be considered when screening for obesity in young adult women.
PDF Links
Full text via DOI 
Download Citation 
