Table 5.
Univariable and multivariable negative binomial models of the number of sex acts in the last month among partnerships that were ongoing in the month before interview (N = 640 partnerships, 515 participants, 2370 observations)a.
| Variable | Unadjusted IRRb | (95% CI) | Adjusted IRRc | 95% CI | P |
| ART group | |||||
| Pre-ART | 1.00 | 1.00 | |||
| ART-eligible | 0.93 | (0.83, 1.03) | 0.77 | (0.65, 0.91) | 0.002 |
| Time (months) | |||||
| First report of partnership | 1.00 | 1.00 | |||
| ≥6 months after first report | 1.54 | (1.41, 1.68) | 1.22 | (1.07, 1.38) | 0.002 |
| Interaction: ART group × timed | 1.26 | (1.07, 1.50) | 0.007 | ||
| Sex | |||||
| Female | 1.00 | 1.00 | |||
| Male | 1.14 | (1.02, 1.27) | 1.02 | (0.92, 1.14) | 0.71 |
| Age (years) | |||||
| 18–21 | 0.85 | (0.66, 1.11) | 0.89 | (0.78, 1.13) | 0.220 |
| 22–29 | 1.00 | 1.00 | |||
| 30–39 | 1.09 | (0.96, 1.23) | 1.00 | (0.89, 1.12) | |
| 40+ | 1.02 | (0.89, 1.17) | 0.89 | (0.78, 1.01) | |
| Gender normse | |||||
| Lowest quartile | 1.00 | 1.00 | <0.001 | ||
| Second quartile | 0.78 | (0.71, 0.85) | 0.85 | (0.78, 0.93) | |
| Third quartile | 0.72 | (0.65, 0.80) | 0.76 | (0.69, 0.84) | |
| Fourth quartile | 0.74 | (0.67, 0.82) | 0.80 | (0.72, 0.90) | |
| Frequency of condom use with this partner in past 6 months | |||||
| Always | 1.00 | 1.00 | <0.001 | ||
| Never/sometimes | 0.74 | (0.69, 0.80) | 0.86 | (0.79, 0.93) | |
| Partner insists to have sex when participant does not want to | |||||
| No | 1.00 | 1.00 | 0.044 | ||
| Yes | 0.86 | (0.78, 0.94) | 0.91 | (0.83, 1.00) | |
| Partner had sex with others in past 6 months | |||||
| No | 1.00 | 1.00 | <0.001 | ||
| Yes/I think so | 0.75 | (0.70, 0.81) | 0.86 | (0.80, 0.93) | |
| Partner lives | |||||
| With participant | 1.00 | 1.00 | <0.001 | ||
| Not with participant | 0.72 | (0.66,0.78) | 0.75 | (0.68, 0.82) | |
| Want more children | |||||
| No | 1.00 | 1.00 | 0.003 | ||
| Yes | 1.11 | (1.02, 1.20) | 1.13 | (1.04, 1.23) | |
| Partnership type | |||||
| Ongoing at enrolment | 1.00 | 1.00 | <0.001 | ||
| New | 1.18 | (1.05, 1.32) | 1.24 | (1.11, 1.39) |
ART, antiretroviral therapy; CI, confidence interval; IRR, incidence rate ratio.
aThe number of observations for each variable is reported in Table 3 as the same partnership observations contribute to both analyses.
bVariables found significant in univariable analysis only: higher incidence rate ratio (IRR) associated with being male, knowing anyone on antiretroviral drugs before enrolment, higher levels of perceived stigma, complete antiretroviral drug knowledge, spending little or no time with family, knowledge of partner's HIV status, having disclosed HIV status to partner, more than three lifetime partners and higher relationship quality scores. Lower IRR was associated with greater reliance on family/friends when having a serious problem and the participant had ever performed a physical act of violence to their partner.
cAlso adjusted for clinic where recruitment for the study occurred. Estimated random-effects variance = 0.098.
dThe estimated incidence rate ratio of the number of sex acts in the last month for the ART-eligible group after first report compared to the pre-ART group after first report is aIRR 0.97, 95% CI (0.88, 1.08); calculated by exponentiating the sum of the ART group main effect and interaction effect coefficients.
eThe highest quartile represents the 25% of the cohort with the most equitable gender norms.