Table 2. Maternal and infant health and well-being outcomes from birth to 12 months post-birth (N = 913) grouped by intervention condition: Enhanced Intervention (EI, N = 405) vs. Standard Care (SC, N = 508).1 .
| EI (N = 405) n (%) | SC (N = 508) n (%) | Estimated odds ratio, EI vs. SC, (95% CI)2 | 2-sided p-value2 | |||||
| HIV transmission-related behaviors | ||||||||
| Told sexual partner about HIV status3 | 1.00 | (0.96, 1.03) | 0.862 | |||||
| Asked sexual partner to test for HIV3 | 1.05 | (0.99, 1.10) | 0.107 | |||||
| Always used a condom3 | 1.03 | (0.91, 1.16) | 0.645 | |||||
| Gave infant cotrimoxazole (6 months) | 222 | (99.6) | 261 | (98.5) | 4.39 | (0.54, 35.82) | 0.167 | |
| Took infant to 6-week or 6-month HIV PCR test and fetched results | 206 | (72.5) | 247 | (71.8) | 1.07 | (0.58, 1.97) | 0.836 | |
| One feeding method first 6 months: formula or breastfeeding | 279 | (91.8) | 297 | (78.6) | 3.02 | (1.20, 7.60) | 0.019 | * |
| Infant health status and bonding | ||||||||
| Weight-for-age z-score≥−23 | 1.08 | (1.01, 1.16) | 0.035 | * | ||||
| Height-for-age z-score≥−23 | 0.99 | (0.90, 1.08) | 0.759 | |||||
| Weight-for-height z-score≥−23 | 0.84 | (0.76, 0.94) | 0.002 | x | ||||
| Normal development according to WHO motor milestones (using 50th percentile ages)3 | 1.08 | (0.95, 1.24) | 0.238 | |||||
| Breastfed exclusively for at least 6 months | 57 | (71.3) | 50 | (52.1) | 2.38 | (1.04, 5.44) | 0.040 | * |
| Brockington Postpartum Bonding Total Score ≤39 (lower score = better bonding, 12 months) | 100 | (98.0) | 175 | (98.9) | 0.57 | (0.06, 5.65) | 0.631 | |
| Parental Stress Index Score ≤ median of 85 (12 months) | 50 | (48.5) | 95 | (53.7) | 0.84 | (0.54, 1.30) | 0.426 | |
| Healthcare and health monitoring | ||||||||
| At least one postnatal clinic visit (6 months) | 156 | (63.2) | 119 | (44.9) | 1.88 | (0.84, 4.23) | 0.127 | |
| Mother knows CD4 cell count at 12 months | 59 | (59.6) | 89 | (53.9) | 1.26 | (0.75, 2.12) | 0.373 | |
| On average, missed no HAART medication doses, past week (12 months) | 18 | (94.7) | 25 | (78.1) | 5.57 | (0.48, 64.92) | 0.170 | |
| Depression | ||||||||
| Not depressed (GHQ<7)3 | 1.08 | (1.03, 1.13) | 0.002 | * | ||||
| Social support | ||||||||
| Receiving child support grant (1.5, 6, or 12 months) | 251 | (63.4) | 345 | (69.0) | 0.78 | (0.46, 1.31) | 0.343 | |
| Larger than the median baseline social network (>25)3 | 0.98 | (0.96, 1.01) | 0.250 | |||||
. Sample size reflects participants with at least one follow-up assessment. Sample sizes for each assessment (note that a variable may be among a subset of the full sample, have missing values, or both, leading to a smaller effective sample size): 1.5 months post-birth: EI (N = 377), SC (N = 466), total (N = 843); 6 months: EI (N = 304), SC (N = 378), total (N = 682); 12 months: EI (N = 106), SC (N = 181), total (N = 287).
. Random effects logistic regression, controlling for clinic clustering. Models include an indicator for intervention status.
. Longitudinal outcome. In addition to the indicator for intervention status, models include time (months since the baseline assessment) and an intervention-time interaction. The interaction term was the effect of interest. See Table 3 for observed values over time.
EI significantly better than SC. 2-sided p-value<0.05 (binomial test: 1-sided, upper-tail p-value<0.025).
SC significantly better than EI. 2-sided p-value<0.05.