Skip to main content

Table 4 Bivariate and multivariate cox proportional hazards model of incident TB among HIV infected individuals on pre-ART follow-up in, JUSH, 2013, N = 584

From: The effect of isoniazid preventive therapy on incidence of tuberculosis among HIV-infected clients under pre-ART care, Jimma, Ethiopia: a retrospective cohort study

Variables

Category

Crude hazard ratio (95% CI)

Â¥Adjusted hazard ratio (95% CI)

Age group (years)

15-29

1.00

 

30-39

0.88 (0.48 1.62)

 

39+

0.38 (0.15 0.98)*

 

Sex (N = 587)

Female

1.00

 

Male

0.61 (0.35 1.06)

 

Marital status

Never married

1.00

 

Married

0.56 (0.28 1.12)

 

Separated

0.83 (0.29 2.35)

 

Divorced

0.72 (0.27 1.91)

 

Widowed

0.23 (0.05 1.02)

 

Level of education (N = 585)

Primary

1.00

 

No education

1.85 (0.95 3.63)

 

Secondary

1.38 (0.67 2.85)

 

Tertiary

0.63 (0.15 2.74)

 

People per household (N = 583)

<=2

1.00

 

>2

1.54 (0.76 3.09)

 

Tobacco use

Yes

1.00

 

No

0.66 (0.35 1.25)

 

Alcohol use

Yes

1.00

 

No

0.68 (0.37 1.23)

 

Past opportunistic infection (N = 587)

No

1.00

1.00

Yes

2.57 (1.43 4.61)**

2.41 (1.33 4.34)

WHO clinical stage

Stage I/ II

1.00

1.00

Stage III/ IV

3.39 (1.90 6.02)**

3.05 (1.61 5.81)**

Functional status

Working

1.00

 

Ambulatory

1.29 (0.58 2.88)

 

Bedridden

5.65 (0.77 41.32)

 

IPT prophylaxis®

Yes

1.00

1.00

No

1.95 (1.02 3.74)

2.02 (1.04 3.92)*

CD 4 + cells count (cells/μl) (N = 585)

> = 500

1.00

1.00

350 – 499

3.06 (1.47 6.37)**

2.87 (1.37 6.03)**

<350

2.34 (1.02 5.39)*

3.16 (1.04 3.92)**

  1. *Statistically significant at p ≤ 0.05 **statistically significant at p ≤ 0.01.
  2. ®Exchanging the reference at adjusted level of analysis results in (aHR = 0.5).
  3. ¥Global test of proportional-hazards assumption for predictors fitted to Cox proportional hazard model was not significant (df = 5, ch2 = 7.04, p = 0.218).