Table 5. Multivariate analyses of association between active disease caused by a Beijing strain of Mycobacterium tuberculosis and relapse among race/ethnicity groups while controlling for other risk factors for relapse, Tuberculosis Trials Consortium Study 22*.
| Characteristic | Asian–Pacific Islander
(n = 50) |
Non-Hispanic black
(n = 148) |
Non-Hispanic white
(n = 63) |
Hispanic
(n = 79) |
|||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| OR (95% CI) | p value | OR (95% CI) | p value | OR (95% CI) | p value | OR (95% CI) | p value | ||||
| Infected with Beijing strain | 15.8 (1.3–192) | 0.03 | 1.8 (0.5–6.5) | 0.35 | 1.0 (0.1–7.7) | 0.98 | 1.0 (0.1–13) | 0.97 | |||
| Underweight at tuberculosis diagnosis | 3.1 (0.3–34) | 0.35 | 2.9 (0.8–6.3) | 0.15 | 11 (2.4–48) | <0.01 | 4.6 (0.9–24) | 0.07 | |||
| Pulmonary cavitation | 2.1 (0.1–33) | 0.60 | 4.0 (0.8–19) | 0.09 | 2.7 (0.5–15) | 0.25 | 6.6 (0.7–61) | 0.09 | |||
| Bilateral pulmonary disease | 5.2 (0.4–69) | 0.21 | 1.6 (0.5–4.8) | 0.44 | 1.2 (0.2–9.9) | 0.84 | 2.1 (0.3–15) | 0.46 | |||
| Two-month sputum culture positivity† | – | – | 3.3 (1.1–9.7) | 0.03 | 3.5 (0.6–20) | 0.16 | 4.6 (0.5–40) | 0.17 | |||
*The Tuberculosis Trials Consortium Study enrolled patients during 1995–1998. Participants in the case–control study were selected from among 1,004 HIV-infected participants. OR, odds ratio; CI, confidence interval. †Not included for Asian–Pacific Islander patients because none were culture positive at 2 months.