Table 1.
Baseline characteristics
| Entire cohort (n=15 501) | São Paulo state population 2010* | |
|---|---|---|
| Age, years† | ||
| 15–25 | 2930 (19%) | 21% |
| >25–35 | 4055 (26%) | 23% |
| >35–45 | 3247 (21%) | 19% |
| >45–55 | 2605 (17%) | 16% |
| >55–65 | 1527 (10%) | 11% |
| >65 | 1118 (7%) | 10% |
| Sex | ||
| Female | 4683 (30%) | 52% |
| Male | 10 818 (70%) | 48% |
| Self-reported skin colour‡ | ||
| White | 7129 (55%) | 64% |
| Brown or mixed | 3989 (31%) | 29% |
| Black | 1488 (12%) | 6% |
| Asian or Indigenous | 237 (2%) | 1% |
| Years of schooling§ | ||
| Illiterate | 521 (4%) | 4% |
| 1–3 | 1465 (12%) | .. |
| 4–7 | 4620 (38%) | .. |
| 8–11 | 4518 (37%) | .. |
| 12–14 | 798 (6%) | 23% |
| ≥15 | 379 (3%) | 41% |
| Incarcerated | ||
| Yes | 1609 (10%) | <1% |
| No | 13 892 (90%) | >99% |
| Homeless | ||
| Yes | 391 (3%) | .. |
| No | 15 110 (97%) | .. |
| Alcohol use | ||
| Yes | 2053 (13%) | .. |
| No | 13 448 (87%) | .. |
| Drug use | ||
| Yes | 1019 (7%) | .. |
| No | 14 482 (93%) | .. |
| Diabetes | ||
| Yes | 880 (6%) | 13% |
| No | 14 621 (94%) | .. |
| Mental disorder | ||
| Yes | 336 (2%) | .. |
| No | 15 165 (98%) | .. |
| HIV status | ||
| Negative | 11 155 (72%) | .. |
| Positive | 1874 (12%) | <1% |
| Unknown | 2472 (16%) | .. |
| Immunosuppression from causes other than HIV infection | ||
| Yes | 113 (1%) | .. |
| No | 15 388 (99%) | .. |
| Anatomical classification | ||
| Pulmonary tuberculosis | 12 458 (80%) | .. |
| Pulmonary tuberculosis and extrapulmonary tuberculosis | 409 (3%) | .. |
| Extrapulmonary tuberculosis | 2280 (15%) | .. |
| Miliary or disseminated | 354 (2%) | .. |
| Microbiological status¶ | ||
| Positive | 10 956 (78%) | .. |
| Negative | 3010 (22%) | .. |
| Place of diagnosis‖ | ||
| Primary or outpatient care | 9349 (61%) | .. |
| Emergency or urgent care facility | 3300 (22%) | .. |
| Hospitalised | 2465 (16%) | .. |
| Upon autopsy | 159 (1%) | .. |
| Treatment outcome | ||
| Treatment success | 12 227 (79%) | .. |
| Treatment failure | 254 (2%) | .. |
| Death | 1247 (8%) | .. |
| Loss to follow-up | 1537 (10%) | .. |
| Not evaluated | 236 (2%) | .. |
*
Data sources (appendix p 4) were the 2010 census (age, sex, education, and self-reported skin colour), the primary care database (diabetes), InfoPen (incarceration), and Notifiable Diseases Information System (HIV status). We did not find a population-based estimate for some variables and we could not redistribute the data for education from the 2010 census on all category levels present in the tuberculosis data.
†
Missing data n=19 (<1%).
‡
Missing data n=2658 (17%).
§
Missing data n=3200 (21%).
¶
Missing data n=1535 (10%).
‖
Missing data n=228 (2%).