Table 1.
Risk factors for dying because of SARS in mainland China derived from univariate and multivariate logistic regression
| Risk factor | Probable cases | Deaths (CFR%) | Univariate | Multivariate | ||
|---|---|---|---|---|---|---|
| Crude odds ratio (95% CI) | P‐value | Odds ratio (95% CI) | P‐value | |||
| All | 5327 | 343 (6.4) | ||||
| Sex | ||||||
| Female | 2607 | 147 (5.6) | 1.00 | – | 1.00 | – |
| Male | 2720 | 196 (7.2) | 1.30 (1.04, 1.62) | 0.020 | 0.81 (0.64, 1.02) | 0.072 |
| Age (categories) | ||||||
| 0–19 years | 476 | 8 (1.7) | ||||
| 20–39 years | 2844 | 70 (2.4) | ||||
| 40–59 years | 1457 | 130 (8.9) | ||||
| 60–79 years | 486 | 124 (25.5) | ||||
| 80–93 years | 62 | 11 (17.7) | ||||
| Age (continuous, 10 years) | 1.80 (1.69, 1.92) | <0.001 | 1.86 (1.74, 2.00) | <0.001 | ||
| Occupation* | 0.006 | |||||
| HCW | 1021 | 34 (3.3) | 0.55 (0.37, 0.82) | 0.003 | 0.76 (0.52, 1.15) | 0.18 |
| Non‐HCW | (4306) | 309 (7.2) | – | – | 1.00 | – |
| HES | 644 | 43 (6.7) | 1.14 (0.79, 1.65) | 0.47 | ||
| LES | 1834 | 108 (5.9) | 1.00 | – | ||
| Unknown | 793 | 51 (6.4) | 1.10 (0.78, 1.55) | 0.59 | ||
| Others | 1035 | 107 (10.3) | – | – | ||
| Location | <0.001 | <0.001 | ||||
| Beijing | 2522 | 184 (7.3) | 1.00 | – | 1.00 | – |
| Tianjin | 175 | 21 (12.0) | 1.73 (1.07, 2.80) | 0.025 | 1.79 (1.07, 2.99) | 0.027 |
| Guangdong | 1504 | 54 (3.6) | 0.47 (0.35, 0.65) | <0.001 | 0.13 (0.08, 0.23) | <0.001 |
| Shanxi | 449 | 24 (5.4) | 0.72 (0.46, 1.11) | 0.14 | 0.71 (0.45, 1.13) | 0.15 |
| Inner Mongolia | 282 | 28 (9.9) | 1.40 (0.92, 2.13) | 0.11 | 1.34 (0.86, 2.08) | 0.19 |
| Other provinces | 395 | 32 (8.1) | 1.12 (0.76, 1.66) | 0.57 | 1.30 (0.86, 1.98) | 0.21 |
| Period of the SARS epidemic (categories) | ||||||
| November–December | 19 | 1 (5.3) | ||||
| January | 164 | 14 (8.5) | ||||
| February | 745 | 23 (3.1) | ||||
| March | 546 | 44 (8.1) | ||||
| April | 3198 | 229 (7.2) | ||||
| May | 655 | 32 (4.9) | ||||
| Time since the first SARS case (continuous, 30 days) | 1.08 (0.97, 1.21) | 0.16 | 0.59 (0.48, 0.71) | <0.001 | ||
| Duration from onset to admission | ||||||
| 0–1 days | 1758 | 115 (6.5) | ||||
| 2–3 days | 1401 | 81 (5.8) | ||||
| 4–5 days | 835 | 56 (6.7) | ||||
| 6–7 days | 617 | 46 (7.5) | ||||
| 8–9 days | 263 | 19 (7.2) | ||||
| 10–11 days | 160 | 16 (10.0) | ||||
| ≥12 days | 267 | 10 (3.7) | ||||
| Missing | 26 | 0 (0.0) | ||||
| Duration from onset to admission (continuous, days) | 1.00 (0.97, 1.02) | 0.80 | ||||
*In univariate analysis, occupation has been divided into four groups: HCW is health care worker; HES is high economic status such as businessman, civil servant and teacher; LES is low economic status such as farmer, day‐care worker/nanny, waiter/waitress and worker. ‘Others’ includes three groups with a strong correlation with age, i.e. retired people, students and children, and these were not considered in univariate analysis. In the multivariate logistic regression model, only two groups have been introduced: HCWs and all non‐HCWs combined.