Table 3.
Unadjusted and adjusted hazard ratios of virologic failure (n= 6264) and treatment switching (n= 6269)
| Univariate analysis | Multivariate analysis | ||||||
|---|---|---|---|---|---|---|---|
|
|
|
||||||
| cHR | 95% CI | P | aHR | 95% CI | p | PHA, VIF | |
| Virologic failure | 0.531, 1.01–1.20 | ||||||
| Timing of VLa | 0.990 | 0.986–0.995 | <0.001 | 1.09 | 1.02–1.15 | 0.008 | |
| Treatment switching | 0.197, 1.01–1.23 | ||||||
| Timing of VLab | 0.992 | 0.987–0.997 | 0.003 | 1.13 | 1.05–1.21 | <0.001 | |
The timing of first VL done after ART initiation was included as a continuous time variable, indicating the number of months since ART initiation.
cHR, crude hazard ratio; aHR, adjusted hazard ratio; PHA, p value for the global proportional-hazards assumption; VIF, range of variance inflation factors.
a
HR of virologic failure or treatment switch for per month of delay of the first VL done since ART initiation (between 2.5 and 9 months);
b
the variable age was categorized because the proportional-hazards assumption was violated with age as a continuous variable (p=0.045).