Table 3. Stratified results for multivariable accelerated failure time model of association between Xpert MTB/RIF use and time to MDR TB treatment initiation for patients in MDR TB treatment cohorts, Latvia, 2009–2012*.
| Exposure | Stratifier | Multivariable regression, time ratio (95% CI) | p value |
|---|---|---|---|
| Xpert MTB/RIF | Previous TB | ||
| Not conducted | No | Baseline | |
| Conducted, rifampin-resistant TB | No | 0.16 (0.12−0.21) | <0.001 |
| Not conducted | Yes | Baseline | |
| Conducted, rifampin-resistant TB |
Yes |
0.34 (0.21–0.55) |
<0.001 |
| Previous TB | Xpert MTB/RIF | ||
| No | Not conducted | Baseline | |
| Yes | Not conducted | 0.80 (0.62–1.04) | 0.09 |
| No | Conducted, rifampin-resistant TB | Baseline | |
| Yes | Conducted, rifampin-resistant TB | 1.69 (1.03–2.76) | 0.04 |
*Association between use of Xpert MTB/RIF and time to MDR TB treatment initiation was stratified by the effect modifier previous tuberculosis. Association between previous tuberculosis and time to MDR TB treatment initiation was stratified by Xpert use. There were 372 patients in the model. All results derived from the final model shown in Table 2. Median in baseline strata of Xpert MTB/RIF and previous tuberculosis was 41 days. MTB, Mycobacterium tuberculosis; RIF, rifampin. MDR TB, multidrug-resistant tuberculosis; NA, not applicable.