Table 2. Predictors of favorable interim treatment outcome using exact logistic regression.
| Characteristic | Favorable N = 16 | PoorN = 7 | Unadjusted odds ratio [95% CI] | Exact p-value |
|---|---|---|---|---|
| Age, years | ||||
| <30 | 2 (12) | 2 (29) | referent | |
| 30–49 | 11 (69) | 4 (57) | 2.75 [0.28–26.6] | p = 0.56 |
| ≥50 | 3 (19) | 1 (14) | 3.0 [0.15–59.9] | p = 0.99 |
| Gender, male | 10 (63) | 6 (86) | 0.28 [0.03–2.9] | p = 0.37 |
| Baseline Body Mass Index, mean % ±SD | 19.4 ±2.0 | 19.6 ±4.6 | p = 0.68 | |
| HIV infected | 6 (38) | 0 | n/c | p = 0.12 |
| Smoking | 3 (19) | 4 (57) | 0.17 [0.03–1.22] | p = 0.14 |
| Alcohol | 6 (38) | 4 (57) | 0.45 [0.45–2.74] | p = 0.65 |
| Prior TB treatment episodes | ||||
| None | 3 (19) | 0 | referent | |
| One | 3 (19) | 2 (29) | n/c | P = 0.46 |
| Two or more | 10 (62) | 5 (71) | n/c | P = 0.52 |
| Pretreatment MGIT TTP a , mean hours ±SD | 261 ±155 | 244 ±108 | p = 0.93 | |
| Proportion with pretreatment MGIT TTP <216 hours | 9 (56) | 3 (43) | 1.71 [0.29–10.3] | p = 0.67 |
| Pretreatment to week 4 change in MGIT TTP a , mean hours ±SD | 562 ±360 | 433 ±324 | p = 0.46 | |
| Week 2 TDA, mean ±SD | 2.2 ±0.67 | 1.9 ±0.54 | p = 0.38 | |
| Proportion with week 2 TDA > 2log killing, (%N) | 9 (56) | 3 (43) | 1.71 [0.29–10.3] | p = 0.67 |
| Week 4 TDA mean | 2.6 ±0.75 | 1.9 ±0.53 | p = 0.033 | |
| Proportion with week 4 TDA > 2log killing, (%N) | 13 (81) | 2 (28) | 1.71 [0.29–10.3] | p = 0.026 |
| Proportion with increase in TDA from week 2 to week 4 | 15 (94) | 3 (43) | 20.0 [1.61–247.98] | p = 0.017 |
aGreater change in time-to-positivity (TTP) reflects greater decrement in sputum bacterial burden.