Skip to main content
. 2020 Nov 24;29(4):281–291. doi: 10.7570/jomes20071

Table 4.

AUCs, ideal cut-offs, sensitivity, and specificity of surrogate markers for IR and insulin secretion for identifying MetS in the ROC analysis

Variable AUC (95% CI) P Cut-off value Sensitivity Specificity Youden index
HOMA-IR 0.851 (0.784–0.904) 0.001 ≥ 2.86 86.67 70.67 0.5733
HOMA2-IR 0.787 (0.712–0.849) 0.001 ≥ 1.69 80.01 65.33 0.4533
HOMA-AD 0.846 (0.778–0.899) 0.001 ≥ 6.26 68.14 85.33 0.5333
HOMA-TG 0.755 (0.678–0.822) 0.001 ≥ 0.72 53.33 81.35 0.3467
TyG index 0.836 (0.767–0.891) 0.001 ≥ 9.88 76.00 88.00 0.6400
FIGR 0.548 (0.465–0.630) 0.335 ≥ 2.41 85.33 46.67 0.3200
FIRI 0.768 (0.669–0.850) 0.001 ≥ 2.96 83.67 70.47 0.4414
HOMA-1%S 0.850 (0.783–0.903) 0.001 ≤ 28.2 82.74 77.33 0.5731
HOMA-2%S 0.779 (0.704–0.843) 0.001 ≤ 53.9 69.33 74.67 0.4400
QUICKI 0.844 (0.776–0.898) 0.001 ≤ 0.32 93.33 61.33 0.5467
Bennet index 0.795 (0.721–0.856) 0.001 ≤ 1.2 89.30 74.62 0.4534
McAuley index 0.815 (0.743–0.874) 0.001 ≤ 6.05 78.76 92.12 0.4937
Raynaud index 0.742 (0.664–0.810) 0.001 ≤ 3.10 76.01 62.67 0.3867
1/Insulin 0.735 (0.657–0.804) 0.001 ≤ 0.08 86.67 50.15 0.3733
GI ratio 0.547 (0.464–0.629) 0.343 ≤ 7.45 85.33 55.81 0.2928
FISI 0.797 (0.724–0.858) 0.001 ≤ 6.96 83.67 70.37 0.4414
SPISE 0.822 (0.751–0.880) 0.001 ≤ 3.79 73.47 90.70 0.5067
HOMA-1%B 0.784 (0.680–0.875) 0.001 ≤ 94.74 71.43 74.42 0.4585
HOMA-2%B 0.842 (0.773–0.896) 0.001 ≤ 87.12 93.62 76.74 0.6800

Cut-off values correspond to the highest Youden index.

AUC, area under the curve; IR, insulin resistance; MetS, metabolic syndrome; ROC, receiver operating characteristic; HOMA, homeostatic model assessment; AD, adiponectin; TG, triglycerides; TyG, triglyceride-glucose; FIGR, fasting insulin to glucose ratio; FIRI, fasting insulin resistance; %S, insulin sensitivity; QUICKI, quantitative insulin sensitivity check index; 1/Insulin, reciprocal insulin; GI, glucose to insulin; FISI, fasting insulin sensitivity index; SPISE, single-point insulin sensitivity estimator; %B, beta-cell function.