Table 3.
Correlates of over-reporting recent product use per FTFI, and by administration route
| Oral Group N=314 | Vaginal Group N=158 | |||||||
|---|---|---|---|---|---|---|---|---|
| Baseline Variables | Univariable Adjusted* | Multivariable Adjusted* | Univariable Adjusted* | |||||
| Country (adjusted for time) | RR | 95% CI | p | RR (95% CI) | p | RR | 95% CI | p |
| Uganda vs. South Africa | 0.74 | 0.55 to 0.99 | 0.04 | 1.02 (0.72 to 1.44) | 0.91 | 0.78 | 0.38 to 1.62 | 0.51 |
| Zimbabwe vs. South Africa | 0.80 | 0.64 to 0.99 | 0.04 | 1.16 (0.81 to 1.64) | 0.42 | 0.69 | 0.41 to 1.18 | 0.18 |
| Time (months) | 1.01 | 1.00 to 1.03 | 0.04 | 1.01 (1.00 to 1.03) | 0.07 | 1.03 | 0.96 to 1.1 | 0.42 |
| Age over 25 vs. 25 or younger | 0.88 | 0.83 to 1.63 | 0.12 | 1.20 | 0.81 to 1.78 | 0.36 | ||
| Primary vs. some secondary school or more | 1.16 | 0.75 to 1.79 | 0.39 | 1.26 | 0.63 to 2.66 | 0.51 | ||
| Earns own income; Yes vs. no | 1.15 | 0.98 to 1.35 | 0.08 | 1.08 (0.92 to 1.26) | 0.36 | 1.06 | 0.73 to 1.52 | 0.77 |
| Married vs. not married | 0.61 | 0.45 to 0.85 | <0.02 | 0.68 (0.49 to 0.94) | 0.02 | 1.08 | 0.59 to 1.98 | 0.79 |
| Has a primary partner vs. no primary partner | 0.86 | 0.49 to 1.48 | 0.58 | 1.03 | 0.32 to 3.30 | 0.96 | ||
| Living with partner vs. not living with partner | 0.85 | 0.70 to 1.04 | 0.12 | 1.24 | 0.78 to 1.97 | 0.36 | ||
| Other male sex partners, past 3 months | 0.12 | 0.64 | ||||||
| One vs. zero | 0.94 | 0.77 to 1.15 | 0.78 | 0.45 to 1.33 | ||||
| 2+ vs. zero | 1.45 | 0.98 to 2.17 | 1.06 | 0.34 to 3.33 | ||||
| Condom use, last vaginal sex4 | ||||||||
| Yes vs. no | 1.11 | 0.92 to 1.34 | 0.28 | 0.96 | 0.63 to 1.45 | 0.84 | ||
| Anal sex, past 3 months | ||||||||
| Yes vs. no | 1.01 | 0.72 to 1.26 | 0.73 | 1.19 | 0.77 to 1.85 | 0.43 | ||
| Sex work in past year4 | ||||||||
| Yes vs. no | 1.13 | 0.83 to 1.52 | 0.44 | 1.1 | 0.59 to 2.04 | 0.76 | ||
| Contraception method3 | ||||||||
| Oral contraceptive vs. not | 0.89 | 0.72 to 1.08 | 0.24 | 1.15 | 0.74 to 1.79 | 0.52 | ||
| Baseline diagnosis of a curable STI1 | ||||||||
| Yes vs. No | 0.98 | 0.81 to 1.19 | 0.82 | 1.15 | 0.74 to 1.81 | 0.53 | ||
| HSV-2 seropositivity2 Yes vs. no | 0.96 | 0.82 to 1.12 | 0.63 | 1.18 | 0.82 to 1.72 | 0.37 | ||
| Worried about getting HIV in next year | <0.01 | 0.04 | 0.02 | |||||
| Very worried vs. not-at-all worried | 1.38 | 1.13 to 1.70 | 1.30 (1.06 to 1.6) | 0.73 | 0.48 to 1.10 | |||
| Somewhat worried vs. not-at-all worried | 1.18 | 0.91 to 1.53 | 1.14 (0.88 to 1.48) | 0.42 | 0.23 to 0.77 | |||
| HIV risk score >6 vs. score of 6 or less | 0.96 | 0.82 to 1.14 | 0.66 | 0.94 | 0.62 to 1.42 | 0.77 | ||
We used a Generalized Linear Mixed Model (GLMM) with random intercept and included all available follow-up quarterly visits with PK measures, adjusting for time (in months) between baseline and PK visit and country. Baseline correlates examined include demographic, sexual behaviour and HIV risk variables, and preselected prior to analysis, as per Table 1.
All univariable and multivariable analyses are presented, controlling for country and time. Marginal Wald Chi-Square Test p-value is provided. If more than one risk factor had a p<0.1, they were entered into a multivariable model.
Collected by face-to-face Interviews (FTFI) on case report forms.
Collected by audio computer-assisted self-interviewing (ACASI).