Table 3.
Associations of demographic, behavioral and clinical characteristics with persistent T. vaginalis*
| Persistent TV | Parasitologic cure | Unadjusted | Adjusted** | |
|---|---|---|---|---|
| N=77 | N=493 | OR (95% CI) | OR (95% CI) | |
| HIV-seropositive | 42 (55) | 240 (49) | 1.27 (0.75, 2.12) | 1.14 (0.70, 1.87) |
| Age (years) | 35 ±6 | 35 ± 7 | 1.00 (0.96, 1.03) | |
| Educational level (years) | 7 ±3 | 7 ± 3 | 0.97 (0.90, 1.04) | |
| Duration of sex work (years) | 9 ±6 | 9 ± 9 | 1.00 (0.97, 1.02) | |
| Sexual activity in the past week | 39 (51) | 270 (55) | 0.85 (0.53, 1.36) | |
| Number of sexual episodes in the past week | ||||
| 1 sex act | 19 (25) | 111 (23) | 1.00 (0.54, 1.86) | |
| 2 or more sex acts | 20 (26) | 159 (32) | 0.74 (0.41, 1.34) | |
| Number of partners in the past week | ||||
| 1 partner | 29 (38) | 207 (42) | 0.82 (0.49, 1.38) | |
| 2 or more sex partners | 10 (13) | 63 (13) | 0.93 (0.44, 1.96) | |
| Unprotected sex in the last week1 | 11 (14) | 96 (19) | 0.69 (0.34, 1.41) | 0.64 (0.30, 1.36) |
| Hormonal contraceptive use | ||||
| Oral contraceptive pills | 4 (5) | 38 (8) | 0.64 (0.24, 1.73) | |
| DMPA | 12 (16) | 78 (16) | 0.94 (0.43, 2.05) | |
| Hormonal contraception2 | 16 (21) | 121 (25) | 0.81 (0.42, 1.55) | |
| Vaginal washing in the last week | 66 (86) | 442 (90) | 0.70 (0.29, 1.69) | |
| Bacterial vaginosis at diagnosis | 44 (57) | 209 (43) | 1.81 (1.12, 2.92) | 1.90 (1.16, 3.09) |
| Bacterial vaginosis at assessment of cure | 36 (47) | 201 (41) | 1.28 (0.81, 2.02) | |
| Yeast present on wet prep at diagnosis | 8 (10) | 44 (9) | 1.18 (0.52, 2.67) | |
| Yeast present on wet prep at assessment of cure | 4 (5) | 58 (12) | 0.41 (0.14, 1.17) | |
| Gonorrhea at diagnosis | 1 (1) | 15 (3) | 0.42 (0.05, 3.29) | |
| Gonorrhea at assessment of cure | 1 (1) | 19 (4) | 0.33 (0.04, 2.51) | |
| Cervicitis at diagnosis | 3 (4) | 48 (10) | 0.38 (0.12, 1.19) | |
| Cervicitis at assessment of cure | 6 (7) | 30 (6) | 1.07 (0.41, 2.82) | |
N (%) or mean ± SD presented.
N represents the number TV infections contributed by 360 unique participants. Eleven participants contributed infections while HIV-seronegative and HIV-seropositive. Demographic characteristics and hormonal contraceptive use reported at diagnosis. All other characteristics reported at the assessment of cure unless specified otherwise.
All variables were assessed for inclusion in the model. Variables that were associated with HIV status (p<0.10) and either produced a 10% in coefficient or had p-values <0.10 were included in the final model. Unprotected sex and calendar year were included in the model as a priori potential confounders.
Unprotected sex vs. 100% condom use or no sex.
Use of oral contraceptive pills, injectables or norplant compared to use of non-hormonal methods or no contraception (i.e. used nothing, condoms, diaphragm, tubal ligation, spermicide or had a hysterectomy)