| |
Univariate model
|
Multivariable model
|
|---|
|
Predictor
|
OR
|
95% CI
|
P
|
OR
|
95% CI
|
P
|
|---|
|
Intercept 1
|
—
|
—
| |
0.146
|
(0.090, 0.24)
| |
|
Intercept 2
|
—
|
—
| |
0.400
|
(0.25, 0.65)
| |
|
Age 2–5 vs 0-2
|
1.43
|
(0.88, 2.32)
|
0.15
|
1.34
|
(0.82, 2.20)
|
0.24
|
|
Age 5–15 vs. 0-2
|
1.24
|
(0.82, 1.89)
|
0.31
|
1.04
|
(0.68, 1.60)
|
0.85
|
|
Age 15+ vs 0-2
|
0.553
|
(0.37, 0.84)
|
< 0.01
|
0.431
|
(0.28, 0.66)
|
< 0.01
|
|
No. of Windows
|
1.145
|
(1.09, 1.21)
|
< 0.01
|
1.097
|
(1.02, 1.18)
|
< 0.01
|
|
Sleep in Same Room
|
0.559
|
(0.43, 0.72)
|
< 0.01
|
0.66
|
(0.48, 0.89)
|
< 0.01
|
|
People in Home
|
1.070
|
(1.04, 1.10)
|
< 0.01
|
1.033
|
(1.00, 1.07)
|
< 0.01
|
- PTST- = Persistently negative tuberculin skin test; TSTC = Tuberculin skin test converter; TST + = tuberculin skin test positive at baseline; CI = confidence interval. Univariate model results show results for each variable one at a time, while the multivariable model includes all variables. Model parameterized as logit(Pr(y ≤ j)) = α + β
x for j = 1 (PTST-), 2 (TSTC). Each OR shows the odds or PTST vs TSTC and also TSTC vs TST + .
