Table 4 Factors associated with TB disease development in bivariable and multivariable cox regression models.

From: A longitudinal study on latent TB infection screening and its association with TB incidence in HIV patients

Ā 

Bivariable model

adjusted by HAART status (time dependent)

adjusted by LTBI treatment

HR

95% C.I.

aHR

95% C.I.

aHR

95% C.I.

Socio-demographics

Male gender

0.88

0.51 to 1.5

0.88

0.51 to 1.5

0.90

0.53 to 1.55

Local residence

0.20

0.13 to 0.31

0.23

0.15 to 0.37

0.21

0.14 to 0.33

Chinese

0.34

0.23 to 0.51

0.38

0.25 to 0.57

0.37

0.24 to 0.55

Baseline conditions

HIV diagnosis year

0.90

0.84 to 0.96

0.92

0.86 to 0.98

0.89

0.83 to 0.95

Age at HIV diagnosis (years old)

0.998

0.98 to 1.02

1.01

0.99 to 1.03

1.001

0.98 to 1.02

First CD4 (/Ī¼L)

0.998

0.997 to 0.999

0.996

0.995 to 0.998

0.998

0.997 to 0.999

First CD4/CD8 ratio

0.09

0.03 to 0.27

0.03

0.01 to 0.11

0.10

0.03 to 0.29

First log10 viral load (/mL)

1.58

1.24 to 2.02

1.93

1.48 to 2.51

1.58

1.25 to 2.01

Diabetes mellitus diagnosisĀ§

1.05

0.33 to 3.32

1.25

0.39 to 3.97

1.54

0.37 to 6.43

Body mass index^

Ā Ā underweight

ref

Ā 

ref

Ā 

ref

Ā 

Ā Ā normal

0.56

0.3 to 1.02

0.53

0.29 to 0.97

0.56

0.3 to 1.02

Ā Ā marginal

0.21

0.08 to 0.58

0.20

0.07 to 0.55

0.22

0.08 to 0.6

Ā Ā overweight

0.35

0.14 to 0.87

0.33

0.13 to 0.82

0.37

0.15 to 0.91

Ā Ā obese

0.53

0.12 to 2.34

0.48

0.11 to 2.13

0.59

0.13 to 2.62

First ADI diagnosisĀ§

5.59

3.71 to 8.41

12.46

7.66 to 20.24

11.11

5.91 to 20.88

LTBI testing

Ever tested LTBI

0.14

0.09 to 0.21

0.15

0.1 to 0.23

/

Ā 

Ever tested LTBI +ve

2.60

1.51 to 4.48

2.54

1.47 to 4.37

/

Ā 

First LTBI test +ve

6.47

3.77 to 11.10

6.18

3.57 to 10.73

/

Ā 

Subsequent LTBI test +ve

0.68

0.26 to 1.79

0.68

0.26 to 1.78

/

Ā 

Received LTBI treatment&

0.33

0.17 to 0.64

0.34

0.18 to 0.65

/

Ā 

LTBI test result and LTBI treatment

Ā Ā LTBI test -ve and without LTBI treatment

ref

Ā 

ref

Ā Ā Ā 

Ā Ā LTBI test +ve but without LTBI treatment

10.71

5.88 to 19.5

10.50

5.75 to 19.17

/

Ā 

Ā Ā LTBI test +ve and with LTBI treatment

1.01

0.48 to 2.12

1.01

0.48 to 2.11

/

Ā 

never tested and without LTBI treatment

11.47

6.85 to 19.19

10.87

6.34 to 18.61

/

Ā 

Conditions at HAART initiation

HAARTĀ§

0.45

0.29 to 0.71

/

Ā 

0.46

0.3 to 0.73

Age (years old)

1.02

0.99 to 1.05

/

Ā 

1.02

0.99 to 1.05

CD4 (/Ī¼L)

0.995

0.99 to 0.998

/

Ā 

0.99

0.99 to 0.997

CD4/CD8 ratio

0.01

0.001 to 0.13

/

Ā 

0.007

0.001 to 0.08

log10 viral load (/mL)

1.10

0.84 to 1.43

/

Ā 

1.09

0.83 to 1.43

  1. Bold ā€“ variables have a significant association (pā€‰<ā€‰0.05) with the outcome.
  2. ADI ā€“ AIDS defining illness; HAART ā€“ highly active antiretroviral therapy; LTBI ā€“ latent TB infection; IQR ā€“ interquartile range.
  3. HR ā€“ hazard ratio estimated in bivariable cox regression model: model eventā€‰=ā€‰TB disease, time variableā€‰=ā€‰from HIV diagnosis to the end point (either TB date or latest in care date (largest date of CD4 collection, PPD collection, death, LTBI date, TB date, HAART initiation date), whichever earlier).
  4. aHR ā€“ adjusted hazard ratio estimated in multivariable cox regression model with the confounder.
  5. ^BMI categories ā€“ underweight (<18.5); normal (18.5ā€“22.9); marginal (23ā€“24.9); overweight (25ā€“29.9); obese (ā‰„30). Reference: http://www21.ha.org.hk/smartpatient/MiniSites/en-US/bmi/BMI-Normal/.
  6. &LTBI treatment refers to the initiation of isoniazid regimen (300ā€‰mg daily) only, regardless of the exact duration of treatment as a proportion of patients had received LTBI treatment in a separate clinical service (e.g. TB and Chest Clinic) the data of which were not merged with the research dataset.
  7. Ā§Occurrence of the respective factor after TB disease development excluded.
Back to article page