Table 4. ROC curve analysis results.
Sensitivity | Specificity | PPV | NPV | Area under the ROC curve | P | Cutoff | ||||
---|---|---|---|---|---|---|---|---|---|---|
Area | Standard error | Confidence interval | ||||||||
Total score | 90.00 | 95.83 | 84.91 | 97.35 | 0.967 | 0.011 | 0.946 | 0.988 | < 0.001 | 4.5 |
Question 1 | 90.00 | 72.40 | 45.92 | 96.53 | 0.812 | 0.023 | 0.768 | 0.856 | < 0.001 | 1.0 |
Question 2 | 71.00 | 52.86 | 28.17 | 87.50 | 0.619 | 0.031 | 0.559 | 0.679 | < 0.001 | 0.5 |
Question 3 | 91.00 | 60.16 | 37.30 | 96.25 | 0.825 | 0.023 | 0.779 | 0.870 | < 0.001 | 0.5 |
Question 4 | 27.00 | 83.85 | 30.34 | 81.52 | 0.554 | 0.033 | 0.489 | 0.620 | 0.094 | 0.5 |
Question 5 | 70.00 | 93.75 | 74.47 | 92.31 | 0.819 | 0.028 | 0.763 | 0.875 | < 0.001 | 1.0 |
Total score (without Q4) | 97.00 | 87.50 | 66.90 | 99.12 | 0.964 | 0.012 | 0.941 | 0.987 | < 0.001 | 3.5 |
ROC = receiver operating characteristic; PPV = positive predictive value; NPV = negative predictive value.