Table 2. Associations with GBS carriage among 340 pregnant women with and without HIV infection, Malawi, 2008–2010*.
Variable | GBS carriage, no. positive for variable/no. total (%) | OR (95% CI) | Adjusted OR (95% CI) |
---|---|---|---|
All study participants | 390/1,857 (21.0) | – | – |
HIV-negative participants | 313/1,441 (21.7) | Reference | Reference |
HIV-positive participants |
77/397 (19.4) |
0.88 (0.66–1.17) |
0.84 (0.63–1.12) |
Age group, y† | |||
16–19 | 68/345 (19.7) | 2.29 (0.68–7.76) | 1.41 (0.40–4.92) |
20–24 | 122/582 (21.0) | 2.48 (0.74–8.28) | 1.96 (0.58–6.63) |
25–29 | 113/498 (22.7) | 2.74 (0.82–9.18) | 2.43 (0.72–8.18) |
30–34 | 60/261 (23.0) | 2.79 (0.82–9.48) | 2.59 (0.76–8.88) |
35–39 | 24/122 (19.7) | 2.29 (0.64–8.15) | 2.03 (0.56–7.33) |
>40 |
3/31 (9.7) |
Reference |
Reference |
First pregnancy | 137 (24.0) | 1.30 (1.01–1.65) | 1.62 (1.19–2.19) |
Second or subsequent pregnancy |
244 (19.6) |
Reference |
Reference |
Recruited during rainy season | 235 (22.4) | 1.18 (0.94–1.50) | – |
Recruited during nonrainy season |
155 (19.6) |
Reference |
– |
Birth weight <2.5 kg | |||
No | 307 (21.5) | Reference | – |
Yes |
71 (19.4) |
0.88 (0.65–1.17) |
– |
Multiple birth | |||
No | 381 (21.3) | Reference | – |
Yes |
9 (18.0) |
0.81 (0.34–1.71) |
– |
HIV-positive subgroup‡ | |||
Antiretroviral treatment | |||
No | 67 (20.4) | Reference | Reference |
Yes | 15 (18.1) | 0.91 (0.45–1.75) | 1.14 (0.58–2.26)§ |
CD4 cell count/mm3¶ | |||
<200 | 9 (14.1) | Reference | Reference |
200–499 | 34 (17.2) | 1.27 (0.57–2.81) | 1.30 (0.58–2.92)§ |
>500 | 33 (28.2) | 2.40 (1.07–5.41) | 2.55 (1.10–5.90) |
*GBS, group B streptococcus; OR, odds ratio; CI, confidence interval; –, not applicable. Adjusted OR derived from logistic regression by use of backward stepwise approach. Final model retained variables with p value <0.1 along with age group and parity because of their known association with GBS carriage. †χ2 test for linear trend, p = 0.93. ‡χ2 test for linear trend, p = 0.01. §Adjusted model includes age group, first pregnancy vs. second or subsequent pregnancies; antiretroviral treatment, and CD4 count for the HIV-positive subgroup. ¶19 participants had no CD4 count available.