Table 3. Regression analyses of infant length-for-age Z-scores by duration of in utero exposure to tenofovir.
| Univariable mixed-effects regression (N=461) | Multivariable mixed-effects regression1 (N=323) | |||
|---|---|---|---|---|
| Duration of in utero exposure to tenofovir | Difference in LAZ (95% CI) | p-value | Difference in LAZ (95% CI) | p-value |
| > 22 weeks | 0.05 (-0.22 to 0.31) | 0.72 | -0.12 (-0.47 to 0.23) | 0.49 |
| 12- 22 weeks | -0.02 (-0.25 to 0.21) | 0.87 | -0.06 (-0.35 to 0.24) | 0.71 |
| < 12 weeks (reference) | 0 | - | 0 | - |
95% CI – 95% confidence interval; LAZ – length-for-age Z-scores, calculated using Fenton (birth) and World Health Organization (postnatal) growth reference standards (adjusted for sex and gestational age);
1
Adjusted for maternal characteristics at initiation of antiretroviral treatment (height, log10 HIV viral load, hemoglobin, cd4 cell count, socio-economic situation), preterm delivery, hazardous drinking in the early postpartum period, and time-varying, current breastfeeding