Table 1.
Baseline Characteristics of Children Infected and Uninfected with HIV (n = 1,343)
| All Children (n = 1,343) [Median (IQR) or n/N (%)] | Children Infected with HIV (n = 299) [Median (IQR) or n/N (%)] | Children Uninfected with HIV (n = 1,044) [Median (IQR) or n/N (%)] | P Value* | |
|---|---|---|---|---|
| Age, mo | 58.6 (27.3 to 102.7) | 46.4 (25.0 to 72.8) | 61.8 (29.0 to 108.5) | <0.0001 |
| Male | 645/1,343 (48) | 152/299 (51) | 493/1,044 (47) | 0.2701 |
| Race | ||||
| African/black | 451/1,343 (34) | 220/299 (74) | 231/1,044 (22) | |
| Mixed-race | 887/1,343 (66) | 78/299 (26) | 809/1,044 (78) | <0.0001 |
| Indian/Asian/white | 5/1,343 (0.4) | 1/299 (0.3) | 4/1,044 (0.4) | |
| Height–age z score | −1.57 (−2.31 to −0.81) | −1.94 (−2.77 to −1.18) | −1.46 (−2.19 to −0.70) | <0.0001 |
| Weight–height z score | 0.51 (−0.22 to 1.27) | 0.68 (−0.02 to 1.47) | 0.45 (−0.31 to 1.14) | 0.0002 |
| Reported recent worm infection | 84/1,340 (6) | 23/298 (8) | 61/1,042 (6) | 0.2775 |
| Prior TB treatment | 162/1,340 (12) | 113/297 (38) | 49/1,043 (5) | <0.0001 |
| BCG vaccination† | 1,165/1,343 (87) | 261/299 (87) | 904/1,044 (87) | 0.7526 |
| cART at enrollment‡ | 251/289 (88) | |||
| CD4 count | 1,317 (867 to 1,804) | |||
| CD4 percentage | 30 (23 to 36) | |||
| Viral load§ | 1,882 (140 to 20,176) | |||
| Current TB contact | 836/1,340 (62) | 54/298 (18) | 782/1,042 (75) | <0.0001 |
| TB contact scoreǁ | 3 (0 to 5) | 0 (0 to 0) | 4 (0 to 6) | <0.0001 |
| TB contact score¶ | 5 (4 to 7) | 4 (2 to 6) | 5 (4 to 7) | 0.0051 |
| Recent household death | 118/1,338 (9) | 26/298 (9) | 92/1,040 (9) | 1.0000 |
Definition of abbreviations: BCG = bacillus Calmette-Guérin; cART = combination antiretroviral treatment; IQR = interquartile range; TB = tuberculosis.
Represents comparison of proportions (chi-square test) or medians (Wilcoxon rank-sum test) between children infected and not infected with HIV.
Children were considered BCG vaccinated if a BCG scar was noted or a record of vaccination was available.
Among the 38 children infected with HIV not receiving cART at baseline, five had started cART by the 3-month follow-up.
Viral load was lower than detectable limits in 126 children and missing in 31 children. Hence, the median and IQR were calculated in 142 children with detectable viral load values.
Calculated among all children.
Calculated only in children with a current TB contact.