Research Article

Sensitivity and Specificity Improvement in Abdominal Obesity Diagnosis Using Cluster Analysis during Waist Circumference Cut-Off Point Selection

Table 7

Waist circumference cut-offs based on ROC Curves. Sensitivity, Specificity, Youden’s Index, Positive Likelihood, and Distance to the ROC Curve. Maracaibo City Metabolic Syndrome Prevalence Syndrome, 2013.

WC (cm)Sensitivity (%)Specificity (%)Youden’s IndexDistance to ROCLR+

6 phenotypes
Women90.7575.265.30.410.422.16
73.368.50.420.412.32
91.7573.368.40.420.412.32
Men97.7576.567.80.440.40 2.37
98.1574.869.70.450.392.46
98.40 74.569.70.440.40 2.45

4 phenotypes
Women90.7582.275.90.580.293.41
91.580.179.30.600.283.86
92.2578.483.00.610.284.61
Men97.50 85.780.20.660.24 4.32
98.1583.882.30.660.234.73
98.40 83.582.30.660.24 4.71

2 phenotypes
Women90.7594.392.60.870.0912.74
91.593.493.70.880.0914.82
92.592.994.60.880.0917.20
Men97.00 96.098.50.950.0464.00
98.1596.099.50.960.04192.00
98.6595.599.50.950.05 191.00

6 phenotypes: HNW, MONW, MHO, MDO, and Overweight healthy and sick groups
4 phenotypes: HNW, MONW, and Overweight healthy and sick groups.
2 phenotypes: HNW and MDO.
() Selected cut-off (in cm) based on Sensitivity, Specificity, Youden Index, and Positive Likelihood Ratios (LR+), giving emphasis to highest sensitivity values.
() Cutpoint 1, asserted using the maximum Youden Index.
() Cutpoint 2, obtained from the point closet to the ROC (0.1).