| Cases | Adjusted odds of treatment success versus treatment failure/relapse/death# | |
| Male (versus female)¶ | 4653 | 1.0 (0.9–1.1) |
| Older age (per 10-year increment) ¶ | 6724 | 0.8 (0.8–0.9) |
| HIV infected (versus not HIV infected) ¶ | 615 | 0.3 (0.2–0.4) |
| Extensive disease (versus not extensive) ¶ | 4792 | 0.5 (0.4–0.6) |
| Previous TB treatment¶ | ||
| None | 1275 | 1.0 (Reference) |
| First-line drugs only | 4410 | 0.6 (0.5–0.8) |
| First-line and second-line drugs | 618 | 0.2 (0.15–0.3) |
| MDR-TB patient group+ | ||
| MDR-TB only | 4763 | 1.0 (Reference) |
| MDR+INJr | 1130 | 0.6 (0.5–0.7) |
| MDR+FQr | 426 | 0.3 (0.2–0.4) |
| XDR-TB | 405 | 0.2 (0.2–0.3) |
| Pulmonary resection surgery performed (versus no pulmonary resection surgery)+ | 373 | 1.5 (0.9–2.6) |
| Experienced a serious adverse event (versus no serious adverse event)+ | 1511 | 1.0 (0.8–1.2) |
Data are presented as n or adjusted OR (95% CI). TB: tuberculosis; MDR-TB: multidrug-resistant TB (resistance to at least isoniazid and rifampicin); MDR-TB only: MDR-TB, but susceptible to fluoroquinolones, amikacin/kanamycin and capreomycin (at least one second-line injectable drug tested); MDR-TB+INJr: MDR-TB plus resistance to amikacin/kanamycin and/or capreomycin, but susceptible to fluoroquinolones; MDR-TB+FQr: MDR-TB plus resistance to any fluoroquinolone, but susceptible to amikacin/kanamycin and/or capreomycin (at least one second-line injectable drug tested); XDR-TB: extensively drug-resistant tuberculosis (MDR-TB plus resistance to any fluoroquinolone and any second-line injectable drug (amikacin/kanamycin and/or capreomycin)). #: odds ratios of treatment success (cure and completion) versus treatment failure/relapse/death adjusted for age, sex, HIV infection, previous TB treatment, previous MDR treatment (treatment for >1 month with two or more second-line drugs) and extent of disease. See Methods and [17] for treatment outcome definitions. ¶: estimate adjusted for all other covariates (characteristics) shown. +: each of these parameters estimated separately, and adjusted for age, sex, HIV, extent of disease and previous treatment with first- or second-line TB drugs. Statistical significance is represented by bold type.