Skip to main content
. Author manuscript; available in PMC: 2018 Mar 1.
Published in final edited form as: Int J Tuberc Lung Dis. 2017 Mar 1;21(3):256–262. doi: 10.5588/ijtld.16.0478

Table 2.

Correlates of maternal WHO TB screen positivity among HIV+ women

Maternal WHO TB Screen
Negative
N=333§
Positive
N=165§
Univariate Analysis

Correlate n or Median (% or IQR) n or Median (% or IQR) Coefficient or ORι (95% CI) p^
Mothers
 Age (years) 28 (24–32) 28 (24–33) 1.01 (0.97–1.04) 0.74
 Years of education 8 (7–12) 8 (7–10) 0.97 (0.91–1.03) 0.28
 Number of people in home 4 (4–6) 5 (4–6) 1.14 (1.00–1.29) 0.05
 Partner HIV positive 154/215 (71.6) 74/107 (69.2) 0.88 (0.49–1.60) 0.68
 CD4 count (cells/mm3) 498 α (348–668) 454ε (339–603) 1.00 (1.00–1.00) 0.06
 Combination ART 214 (64.3) 99 (60.0) 0.83 (0.56–1.24) 0.37
 History of TB 31 (9.3) 24 (14.5) 1.66 (1.03–2.68) 0.04
  Years since TB 3β (2–5) 4ζ (3–6) 1.18 (0.97–1.43) 0.09
 TB exposure 32/328 (9.8) 20/156 (12.8) 1.36 (0.72–2.56) 0.34
  Isoniazid preventive therapy 6/32 (18.8) 2/20 (10.0) 0.48 (0.04–5.49) 0.55
Infants
 Sex (male) 160 (48.0) 84 (50.9) 0.89 (0.64–1.25) 0.50
 Birth weight (kg) 3.2γ (2.8–3.5) 3.25η (2.8–3.5) +0.04 (−0.08–0.16) 0.53
 BCG vaccination 329/332 (99.1) 164 (99.4) 1.50 (0.15–14.74) 0.73
 BCG scar 305/332 (91.9) 144 (87.3) 0.61 (0.31–1.19) 0.15
 HIV infected 9/319 (2.8) 12/158 (7.6) 2.83 (1.16–6.90) 0.02
 Breastfed (primarily) 189/327 (57.8) 94/162 (58.0) 1.00 (0.70–1.47) 0.96
 Weight-for-age Z score −0.27δ (−1.12–0.51) −0.53θ (−1.22–0.46) −0.09 (−0.35–0.17) 0.50
 Hospitalization 18 (5.4) 17 (10.3) 2.01 (0.84–4.80) 0.12
  Tuberculosis 0/18 (0.0) 0/17 (0.0) Not run
 Cough 86/332 (25.9) 73 (44.2) 2.27 (1.34–3.84) 0.003
 Fever 62/332 (18.7) 49 (29.7) 1.84 (0.99–3.40) 0.05
 Difficulty breathing 23/332 (6.9) 26 (15.8) 2.51 (1.23–5.11) 0.01
 Anorexia 17/332 (5.1) 19 (11.5) 2.41 (1.30–4.46) 0.005
§

Total N, except where otherwise noted;

α

N=216;

β

N=31;

γ

N=279;

δ

N=332;

ε

N=104;

ζ

N=24;

η

N=141;

θ

N=165;

ι

OR, Odds ratio;

^

p value by logistic or linear regression as appropriate.