Skip to main content

Table 4 Logistic regression model for assessing the effect of TB co-infection on clinical failure, December 2011–August 2012, JUSH, Ethiopia

From: Outcomes of highly active antiretroviral therapy and its predictors: a cohort study focusing on tuberculosis co-infection in South West Ethiopia

Logistic regression

Follow up time in months

At 6 months of follow up visits

At 12 months of follow up visits

Univariate analysis

Multivariate analysis

Univariate analysis

Multivariate analysis

OR (95 % CI)

P

AOR (95 % CI)

P

AOR (95 % CI)

P

AOR (95 % CI)

P

Variables

 TB status at base line

  Absent

Reference

Reference

Reference

 

Reference

 

  Present

3.0 (1.90–4.64)

0.001

2.90 (1.41–6.09)

0.028

4.62 (2.79–7.64)

<0.001

2.93 (1.41–6.09)

0.004

 Employment status at base line

  Work fulltime

Reference

Reference

Reference

Reference

  Work parttime

4.62 (1.89–11.74)

0.005

5.48 (1.96–15.39)

0.042

5.68 (2.01–16.13)

0.001

6.07 (1.63–22.65)

0.007

 Drug addiction behavior at base line

  Absent

Reference

Reference

 

Reference

Reference

  Present

1.24 (0.98–4.55)

0.076

1.20 (0.78–3.65)

0.086

1.71 (1.09–2.67)

0.020

1.76 (1.02–3.03)

0.041

 Base line WHO stage

  I/II

Reference

Reference

Reference

  

  III

16.30 (7.87–33.70)

0.001

17.83 (7.44–42.7)

<0.001

8.72 (4.01–19.03)

<0.001

6.58 (3.03–16.08)

<0.001

  IV

16.44 (7.05–38.37)

0.001

17.06 (5.12–56.80)

<0.001

19.17 (7.82–46.97)

<0.001

6.98 (2.09–22.64)

0.002

 Baseline functional status

  Working

Reference

Reference

Reference

  

  Ambulatory

3.74 (2.49–5.62)

0.001

1.86 (1.05–3.28)

0.034

3.58 (2.26–5.67)

0.001

2.56 (3.47–8.98)

0.075