Table 2. Risk of tuberculosis attributable to diabetes or to HIV infection in study sites in South Texas and north-eastern (NE) Mexico, 2006–2008.
| Age (years) | Diabetes |
HIV infection |
|||||
|---|---|---|---|---|---|---|---|
| RR (95% CI) | ARexposed (%)a | ARpopulation (%)b | RR (95% CI) | ARexposed (%)a | ARpopulation (%)b | ||
| South Texas | |||||||
| 20+ (n = 61) | 2.7 (1.6–4.4) | 63 | 26 | 17.8 (6.5–9.0) | 94 | 5 | |
| 20–34 (n = 20) | 0.9 (0.1–6.8) | −9 | 1 | 34.4 (8.0–147.7) | 97 | 6 | |
| 35–64 (n = 32) | 5.1 (2.6–10.2) | 80 | 48 | 12.2 (2.9–50.9) | 92 | 5 | |
| 65+ (n = 9) | 1.7 (0.5–5.8) | 41 | 22 | 0c | NA | NA | |
| NE Mexico | |||||||
| 20+ (n = 172) | 3.1 (2.3–4.2) | 68 | 24 | 16.0 (7.5–34.0) | 94 | 3 | |
AR, attributable risk; CI, confidence interval; HIV, human immunodeficiency virus; NA, not applicable; RR, relative risk.
a Risk of tuberculosis attributable to exposure among exposed patients, ARexposed = 1 – 1/RR.
b Risk of tuberculosis attributable to exposure among the corresponding general population, ARpopulation = p(exposed)(RR − 1)/[1 + p(exposed)(RR − 1)].
c There were no HIV-positive patients in this age group.