Skip to main content

Table 4 Logistic regression predicting likelihood of not having DOT at home

From: Community-based directly observed treatment for TB patients to improve HIV services: a cross-sectional study in a South African province

Control variable

Odds ratio

Confidence interval

Significance

Sex

 Female

1.00

–

NS

 Male

1.28

(0.89,1.83)

 

Age (continuous)

0.98

(0.96, 0.99)

**

Employment

 Employed

1.00

–

NS

 Unemployed

0.65

(0.42, 1.01)

 

Marital status

 Married or cohabiting

1.00

–

*

 Unmarried and not cohabiting

1.47

(1.02, 2.11)

 

Housing quality

 Not in very poor housing

1.00

–

**

 In very poor housing

1.84

(1.13, 2.99)

 

Education

 Secondary school or higher

1.00

–

NS

 No or only primary school

1.15

(0.74, 1.79)

 

Undergoing retreatment for TB

 No

1.00

–

NS

 Yes

1.42

(0.90, 2.27)

 
  1. Note 1: *P < 0.05; *P < 0.01; NS = Not significant
  2. Note 2: Includes all patients in the sample, N = 1101
  3. Note 3: Computed using SPSS procedure Binary Logistic