Table 1. Test characteristics of the assays.
|
Unsatisfactory samples
|
Accuracy for CIN2+
|
|||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Test method | Positive criterion | Test positivity rate N (%) | D+ | D− | True positive | False negative | False positive | True negative | Sensitivity | Specificity | Positive predictive value | Negative predictive value |
| PAP | ⩾ASCUS | 21/305 (6.9) | 1 | 7 | 10 | 5 | 11 | 279 | 0.667 | 0.962 | 0.476 | 0.982 |
| ⩾LSIL | 16/305 (5.2) | 1 | 7 | 9 | 6 | 7 | 283 | 0.600 | 0.976 | 0.563 | 0.979 | |
| ⩾HSIL | 6/305 (2.0) | 1 | 7 | 5 | 10 | 1 | 289 | 0.333 | 0.997 | 0.833 | 0.967 | |
| LBC | ⩾ASCUS | 20/301 (6.6) | 1 | 11 | 11 | 4 | 9 | 277 | 0.733 | 0.969 | 0.550 | 0.986 |
| ⩾LSIL | 16/301 (5.3) | 1 | 11 | 9 | 6 | 7 | 279 | 0.600 | 0.976 | 0.563 | 0.979 | |
| ⩾HSIL | 7/301 (2.3) | 1 | 11 | 5 | 10 | 2 | 284 | 0.333 | 0.993 | 0.714 | 0.966 | |
| Colposcopy | ⩾Abnormal | 103/252 (40.9) | 2 | 59 | 11 | 3 | 92 | 146 | 0.786 | 0.613 | 0.107 | 0.980 |
| ⩾High-grade | 16/252 (6.3) | 2 | 59 | 4 | 10 | 12 | 226 | 0.286 | 0.950 | 0.250 | 0.958 | |
| HR-HPV DNA (14 types) | 1 of 14 types + | 58/313 (18.5) | 0 | 0 | 16 | 0 | 42 | 255 | 1.000 | 0.859 | 0.276 | 1.000 |
| NASBA mRNA (9 types) | 1 of 9 types + | 33/313 (10.5) | 0 | 0 | 15 | 1 | 18 | 279 | 0.938 | 0.939 | 0.455 | 0.996 |
| PreTect HPV-Proofer | 1 of 5 types + | 23/313 (7.3) | 0 | 0 | 13 | 3 | 10 | 287 | 0.813 | 0.966 | 0.565 | 0.990 |
The column ‘Positive criterion’ specifies the different criteria used within each method; ‘Test positivity rate’ gives the number of positive of total number of samples given each specified criterion for each method; The column ‘Unsatisfactory samples’ specifies number of unsatisfactory samples for each method with a CIN2+ (D+) and without a CIN2+ (D−) diagnosis. Unsatisfactory samples have been excluded from the accuracy calculations for detection of CIN2+.