Table 2. Multivariable model for the effect of cART and baseline covariates on study outcomes.
Variable | Death (165,164 Person-Days, 49 Events) | Death, Default, Lost to Follow-Up (165,164 Person-Days, 81 Events) | Death, Hospitalization, Serious Opportunistic Infection (140,687 Person-Days, 102 Events) | |||
Hazard Ratio [95% CI] | p-Value | Hazard Ratio [95% CI] | p-Value | Hazard Ratio [95% CI] | p-Value | |
“On cART” | 0.3 [0.1, 1.1] | 0.06 | 0.3 [0.1, 0.6] | 0.0005 | 0.4 [0.2, 0.8] | 0.02 |
Weeks “on cART”a | 1.0 [0.9, 1.0] | 0.005 | 1.0 [0.9, 1.0] | 0.009 | 1.0 [0.9, 1.0] | 0.10 |
First CD4 cell count (per 20-cell increase, linear) | 0.8 [0.7, 0.9] | 0.002 | 0.9 [0.9, 1.0] | 0.003 | 0.9 [0.8, 1.0] | <0.001 |
Interaction between “on cART” and value of first CD4 cell count≤350 cells/µl | 1.2 [1.0, 1.4] | 0.03 | —b | 1.1 [1.0, 1.2] | 0.03 | |
Missing a CD4 cell count at TB treatment start | 0.6 [0.3, 1.4] | 0.25 | 1.2 [0.6, 2.2] | 0.62 | 0.9 [0.5, 1.6] | 0.65 |
Weeks between TB treatment start and first CD4 cell count≤350 cells/µl, if positivea | 1.0 [1.0, 1.1] | 0.55 | 1.0 [1.0, 1.1] | 0.57 | 1.0 [0.9, 1.0] | 0.48 |
Age≥43 y (75th percentile) | 1.7 [0.9, 3.4] | 0.12 | 1.1 [0.6, 2.0] | 0.69 | 1.5 [0.9, 2.4] | 0.10 |
Female gender | 1.6 [0.8, 3.2] | 0.20 | 1.1 [0.6, 1.8] | 0.84 | 1.3 [0.8, 2.0] | 0.32 |
Rural treatment site | 1.4 [0. 7, 3.0] | 0.36 | 0.9 [0.5, 1.6] | 0.71 | 0.8 [0.4, 1.3] | 0.32 |
In-patient at health facility at TB start | 2.1 [1.1, 4.3] | 0.04 | 1.4 [0.9, 2.5] | 0.17 | 1.8 [1.1, 2.8] | 0.02 |
Estimates are adjusted for all other variables in model and follow-up day, most recent CD4 cell count, and current hospitalization.
Days “on cART” and days to first CD4 cell count≤350 cells/µl transformed to the week scale.
The interaction between “on cART” and value of first CD4 cell count≤350 cells/µl was not statistically significant for the combined endpoint of death, default, and lost to follow-up and was therefore not included in this multivariable model.