Skip to main content

Table 4 Sensitivity, specificity, and odds ratios of the three clinical risk factor algorithms for CIN3+ or CIS/SCC/AC detection

From: Combined clinical and genetic testing algorithm for cervical cancer diagnosis

Target genes Cutoff Sensitivity (%) (95 % CI) Specificity (%) (95 % CI) AUC (%) (95 % CI) Odds ratio (95 % CI) P value
CIN3+ lesion
 Algorithm 1 220 89.87 (84.18–93.67) 75.95 (70.72–80.50) 82.9 (78.9–86.9) 28.02 (15.65–50.17) <0.001
 Algorithm 2 220 86.08 (79.82–90.62) 80.76 (75.84–84.87) 83.4 (79.3–87.5) 25.94 (15.17–44.36) <0.001
 Algorithm 3 220 92.41 (87.19–95.60) 72.51 (67.11–77.32) 82.5 (78.5–86.4) 32.09 (16.88–61.00) <0.001
CIS/SCC/AC
 Algorithm 1 220 96.51 (90.24–98.81) 64.46 (59.41–69.21) 78.8 (74.2–83.5) 50.19 (15.55–161.98) <0.001*
 Algorithm 2 220 100.0 (94.87–100.0) 70.80 (65.92–75.24) 84.0 (80.4–87.6)
 Algorithm 3 220 100.0 (94.87–100.0) 61.43 (56.33–66.30) 79.5 (75.3–83.7)
  1. P value determined by chi-squared test and *Fisher’s exact test; odds ratio for CIN3+ or CIS/SCC/AC
  2. CI confidence interval, Gene m methylated gene