Table 1. Distribution of selected variables between severe H7N9 and H1N1pdm patients.
| Selected variables | H7N9 a | H1N1pdm a | P b | Adjusted P c |
|---|---|---|---|---|
| Median age by years (IQR d ) | 59.0 (46.5–72.5) | 27.0 (7.00–48.8) | <0.0001 | NA |
| Male, FREQ d (%) | 27 (73.0) | 89 (48.4) | 0.006 | NA |
| Median BMI d (IQR) | 23.9 (22.7–25.3) | 22.9 (19.4–25.5) | 0.084 | 0.499 |
| Chronic medical condition, FREQ (%) | ||||
| Pulmonary disorders (including asthma) | 6 (16.7) | 17 (9.60) | 0.336 | 0.127 |
| Cardiovascular disorders (excluding hypertension) | 9 (25.0) | 13 (7.30) | 0.004 | 0.801 |
| Metabolic disorders (including diabetes mellitus) | 5 (13.9) | 11 (6.20) | 0.213 | 0.838 |
| Other disorders | 7 (19.4) | 31 (17.9) | 0.829 | 0.678 |
| Any disorder | 20 (55.6) | 59 (34.5) | 0.018 | 0.466 |
| Treatment, FREQ (%) | ||||
| Neuraminidase inhibitor | 27 (84.4) | 153 (83.6) | 0.913 | 0.571 |
| Antibiotics | 31 (96.9) | 179 (97.8) | 0.557 | 0.673 |
| Glucocorticoid | 23 (71.9) | 132 (72.5) | 0.939 | 0.592 |
| Clinical outcome, FREQ (%) | ||||
| ICU admission | 28 (84.8) | 120 (65.2) | 0.026 | 0.245 |
| ARDS | 23 (71.9) | 80 (44.4) | 0.004 | 0.004 |
| Respiratory failure | 28 (87.5) | 110 (60.1) | 0.003 | 0.064 |
| Heart failure | 9 (28.1) | 31 (17.0) | 0.138 | 0.132 |
| Liver dysfunction | 14 (43.8) | 33 (18.5) | 0.002 | 0.009 |
| Renal dysfunction | 14 (43.8) | 18 (10.0) | <0.0001 | 0.004 |
| Death | 16 (45.7) | 27 (15.3) | <0.0001 | 0.011 |
| Time duration | ||||
| Interval from onset of illness to neuraminidase inhibitor treatment, Median days (IQR) | 9.00 (5.00–13.0) | 5.00 (3.00–8.00) | <0.0001 | 0.014 |
| Interval from onset of illness to first medical consultation, Median days (IQR) | 3.00 (1–6.00) | 1.00 (0–4.00) | 0.002 | 0.108 |
| Interval from onset of illness to hospitalization, Median days (IQR) | 6.00 (4.00–8.50) | 4.00 (1.00–6.00) | <0.0001 | 0.082 |
| Interval from onset of illness to death, Median days (IQR) | 24.5 (20.0–37.3) | 13.0 (7.00–22.0) | 0.001 | 0.019 |
a Patients with missing data were excluded for each variable.
b Pearson chi-square test was used for comparing proportions and continuity correction or Fisher’s Exact Test was used if appropriate. Mann-Whitney U test was used for comparing medians.
c Age and gender were adjusted using General Linear Model (for BMI) or Cox proportional hazards model (for selected time durations) for continuous variables and Logistic regression model for categorical variables.
d Interquartile range (IQR), Frequency (FREQ), Body mass index (BMI)