TABLE 4.
Estimated sensitivity of endocervical diagnostic tests for C. trachomatis stratified by subjects' characteristics
| Characteristics of subjects tested | % Sensitivity (95% CI)a for cervical test type:
|
|||
|---|---|---|---|---|
| PACE2
|
LCR
|
PCR
|
||
| Culture-PCR | Culture-LCR | Culture-PCR | Culture-LCR | |
| All subjects | 60.5 (53.9, 67.0) | 60.2 (50.5, 69.8) | 83.5 (80.2, 86.8) | 74.7 (70.8, 78.5) |
| Age (yr) | ||||
| <20 | 66.3 (59.6, 73.1) | 65.8 (57.7, 73.9) | 88.9 (84.5, 93.4) | 78.9 (70.5, 87.3) |
| ≥20 | 56.3 (46.2, 66.3) | 55.0 (43.5, 66.6) | 79.8 (71.1, 88.4) | 71.6 (66.5, 76.7) |
| Mucopurulent endocervical discharge | ||||
| Present | 72.6 (55.3, 89.9) | 70.2 (56.3, 84.0) | 87.5 (75.7, 99.4) | 83.9 (77.2, 90.6)c |
| Absent | 56.0 (48.4, 63.6) | 56.0 (44.9, 67.2) | 80.9 (73.6, 88.2) | 71.1 (65.4, 76.7) |
| Current menses | ||||
| Yes | 60.7 (34.0, 87.5) | 60.4 (33.6, 87.2) | 87.1 (73.0, 100) | 74.3 (59.4, 89.1) |
| No | 60.3 (55.2, 65.4) | 60.0 (50.7, 69.3) | 83.1 (78.9, 87.3) | 74.7 (70.5, 78.8) |
| N. gonorrhoeae detected at cervix | ||||
| Yes | 66.4 (47.9, 84.9) | 62.2 (41.4, 83.0) | 91.2 (83.4, 98.9)b | 76.8 (66.7, 86.9) |
| No | 59.2 (51.5, 66.9) | 58.3 (46.2, 70.4) | 81.0 (74.5, 87.4) | 71.3 (66.1, 76.6) |
| Hematuria | ||||
| Yes | 67.0 (56.6, 77.4) | 66.0 (53.3, 78.8) | 88.8 (81.8, 95.7) | 81.5 (73.4, 89.7)b |
| No | 55.7 (46.7, 64.7) | 54.4 (44.8, 63.9) | 81.6 (76.1, 87.2) | 70.9 (65.5, 76.2) |
| Urine leukocyte esterase | ||||
| Positive | 63.2 (57.2, 69.3) | 61.7 (55.5, 67.9) | 85.5 (81.0, 89.9) | 79.6 (74.4, 84.8)b |
| Negative | 57.0 (48.9, 65.2) | 55.3 (42.6, 67.9) | 81.7 (75.3, 88.1) | 66.2 (54.1, 78.2) |
a
Adjusted for variation in test performance across centers. Performance was defined as described in the text using two reference standards (culture-PCR and culture-LCR); estimates are reported separately for each. CI, confidence interval.
b
P = 0.05 for comparison of test sensitivities in the presence versus absence of subject characteristic noted.
c
P = 0.004 for comparison of test sensitivities in the presence versus absence of subject characteristic noted.