Skip to main content
. 2016 Oct 25;19(1):20926. doi: 10.7448/IAS.19.1.20926

Table 3.

Bivariate and multivariate analysis of loss to follow-up

Bivariate analysis Multivariate analysis


Variable Category OR Confidence interval p aOR Confidence interval p
Gender Male 1.15 0.71–1.88 0.558
Female 1
Age 0.050 0.011
≤24 5.61 0.68–45.41 0.110 13.58 1.48–62.68 0.021
25–39 2.45 0.33–18.04 0.379 4.59 0.58–36.71 0.150
40–54 1.68 0.22–12.92 0.618 3.00 0.36–25.05 0.310
≥55 1 1
Provider group 0.002 0.002
<50 patients 13.35 3.00–59.30 0.001 15.83 3.44–72.82 <0.001
50–250 patients 1 1
>250 patients 7.93 1.92–32.69 0.004 8.70 2.70–36.57 0.003
WHO stage 0.549
I 1.81 0.68–4.85 0.233
II 1.02 0.32–3.22 0.966
III 1.44 0.60–3.42 0.404
IV 1
ART status <0.001 <0.001
Pre-ART ineligible 30.93 13.77–69.46 <0.001 33.80 14.20–80.46 <0.001
Pre-ART eligible 11.55 6.73–19.84 <0.001 13.51 7.71–23.68 <0.001
On ART 1 1
TB diagnosis No 1.05 0.63–1.76 0.830
Yes 1
CD4 at baseline 0.200
≤50 1
51–200 1.51 0.68–3.34 0.301
201–350 2.37 1.11–5.04 0.025
351–500 1.51 0.57–3.98 0.396
>500 1.59 0.33–7.57 0.554

aOR, adjusted odds ratio; ART, antiretroviral therapy; CD4, cluster of differentiation 4; OR, odds ratio; TB, tuberculosis.