TABLE 2.
Sample | Test | No. testing positive (sensitivity [95% CI] [%])c | No. testing negative (specificity [95% CI] [%])d | PPV (%) | NPV (%) |
---|---|---|---|---|---|
CS | AC2 | 27 (93.1 [81.1-97.7]) | 257 (98.5 [9.6-99.3]) | 87.1 (74.1-94.1) | 99.2 (97.7-99.7) |
L-Pap | AC2 | 29 (100 [91.4-100]) | 258 (98.9 [97.2-99.5]) | 90.6 (78.7-96.2) | 100 (99.0-100) |
ACT | 29 (100 [91.4-100]) | 256 (98.1 [96.1-99.1]) | 85.3 (72.7-92.7) | 100 (98.9-100) | |
AMP | 25 (86.2 [72.5-93.7]) | 238 (91.2 [87.9-93.7]) | 52.1 (40.4-63.5) | 98.3 (96.4-99.3) | |
PT | 21 (72.4 [57.3-83.7]) | 242 (92.7 [89.6-95.0]) | 52.5 (39.7-64.4) | 96.8 (94.4-98.2) |
PPV, positive predictive value; NPV, negative predictive value.
Including AC2 and ACT.
Comparison of the sensitivities of the five tests found highly significant differences (P < 0.001 by the Cochrane Q test). In pairwise comparisons, PT was inferior to AC2 or ACT (P = 0.01 by the McNemar test).
Comparison of the specificities of the five tests found highly significant differences (P < 0.001 by the Cochrane Q test). In pairwise comparisons, AMP was inferior to AC2 or ACT (P < 0.001 and P = 0.001, respectively, by the McNemar test) and PT was inferior to AC2 or ACT (P = 0.001 and P = 0.007, respectively).