Table 3.
Kaplan-Meier (unadjusted) and Cox proportional regression (adjusted) analyses for age at attainment of milestones for HIV-infected infants stratified by ART responses and ART regimen and compared with HIV-unexposed uninfected infants
| Unadjusted b | Adjusted c | ||||
|---|---|---|---|---|---|
| Na | Median (IQR) Months | P | HR (95% CI) | P | |
| Walking unsupported | |||||
| HUU-reference | 69 | 13 (11, 14) | |||
| HIV+, 6-mo VL <1000 c/mL | 28 | 15 (13, 16) | 0.0009 | 0.58 (0.36, 0.92) | 0.02* |
| HIV+, 6-mo VL ≥1000 c/mL | 25 | 16 (13, 20) | 0.0001 | 0.43 (0.24, 0.76) | 0.004 |
| HIV+, 6-mo CD4% ≥25% | 27 | 14 (13, 15) | 0.02 | 0.73 (0.46, 1.16) | 0.2* |
| HIV+, 6-mo CD4% <25% | 27 | 17 (13, 20) | <0.0001 | 0.35 (0.19, 0.61) | <0.001 |
| HIV+, 6-mo WAZ ≥ −2 | 38 | 15 (13, 16) | 0.001 | 0.59 (0.39, 0.90) | 0.014* |
| HIV+, 6-mo WAZ < −2 | 15 | 17 (13, 22) | <0.0001 | 0.26 (0.13, 0.53) | <0.001 |
| Monosyllabic speech | |||||
| HUU-reference | 72 | 12 (10, 13) | - | - | |
| HIV+, 6-mo VL <1000 c/mL | 28 | 15 (12, 16) | <0.0001 | 0.28 (0.16, 0.46) | <0.001* |
| HIV+, 6-mo VL ≥1000 c/mL | 24 | 15 (14, 19) | <0.0001 | 0.20 (0.11, 0.36) | <0.001 |
| HIV+, 6-mo CD4% ≥25% | 27 | 14 (12, 16) | <0.0001 | 0.33 (0.21, 0.55) | <0.001* |
| HIV+, 6-mo CD4% <25% | 26 | 16 (14, 19) | <0.0001 | 0.14 (0.08, 0.27) | <0.001 |
| HIV+, 6-mo WAZ ≥ −2 | 37 | 14 (13, 16) | <0.0001 | 0.26 (0.16, 0.42) | <0.001* |
| HIV+, 6-mo WAZ < −2 | 15 | 16 (15, 19) | <0.0001 | 0.15 (0.07, 0.29) | <0.001 |
HUU HIV-unexposed uninfected, HIV+ HIV-infected, IQR interquartile range, CI confidence interval, VL plasma viral load (HIV RNA), WAZ weight - for - age Z-score
aN, number of events
b Comparison of each group of HIV-infected infants vs HIV-unexposed uninfected infants (HUU) using log - rank tests
c Cox regression analyses adjusted for monthly household rent. For comparisons of effective viral load and CD4%, adjustment was also made for WAZ < −2 at 9 months of age (approximately 6 months after initiating ART)
*Using Bonferroni to account for 6 multiple comparisons, P-values <0.008 were considered statistically significant