Skip to main content

Table 7 Logistic regression model for hospital mortality in critically ill H1N1 patients

From: Acute kidney injury among critically ill patients with pandemic H1N1 influenza A in Canada: cohort study

Variable Deaths Survivors Crude OR P value Adjusted OR P value
(n = 113) (n = 432) (95% CI) (95% CI)
H1N1 details, demographics, and comorbidities
Confirmed H1N1 104 (92.0) 364 (84.3) 2.16 (1.04-4.47) 0.04 1.59 (0.66-3.80) 0.30
Days of symptoms before hospitalization 5.5 (5.5) 5.2 (4.6) 1.01 (0.97-1.06) 0.55   
Days of symptoms before ICU admission (n = 534, 110, 424) 7.2 (6.5) 6.2 (5.1) 1.03 (1.00-1.07) 0.08 1.04 (1.00-1.09) 0.06
Age (yr) 51.8 (15.6) 47.0 (14.6) 1.02 (1.01-1.04) 0.003 1.02 (1.00-1.04) 0.03
Female sex 55 (48.7) 234 (54.2) 0.80 (0.53-1.22) 0.30 1.00 (0.58-1.71) 0.99
Obesity 42 (37.2) 106 (24.5) 1.82 (1.17-2.82) 0.008 1.49 (0.85-2.62) 0.16
Diabetes mellitus 24 (21.2) 116 (26.9) 0.74 (0.45-1.21) 0.23   
Chronic obstructive pulmonary disease 20 (17.7) 75 (17.4) 1.02 (0.60-1.76) 0.93   
Alcohol abuse 9 (8.0) 64 (14.8) 0.50 (0.24-1.03) 0.06 0.77 (0.34-1.75) 0.54
Chronic kidney disease 15 (13.3) 33 (7.6) 1.85 (0.97-3.54) 0.06 1.35 (0.50-3.61) 0.55
Day 1 support, physiology, and laboratory parameters
APACHE II (n = 497, 101, 396) 25.0 (10.2) 20.0 (8.6) 1.06 (1.04-1.09) <0.0001 1.05 (1.01-1.08) 0.006
Invasive mechanical ventilation (n = 537, 111, 426) 82 (73.9) 289 (67.8) 1.34 (0.84-2.14) 0.22   
Inotrope or vasopressor 52 (46.0) 186 (43.1) 1.13 (0.74-1.71) 0.57   
Mean arterial pressure (mmHg) (n = 506, 103, 403) 75.1 (20.9) 74.2 (18.6) 1.02 (0.91-1.14) 0.70   
PaO2/FiO2 (n = 480, 101, 379) 126 (84) 161 (99) 0.95 (0.93-0.98) 0.002 0.95 (0.92-0.99) 0.007
Creatinine (μmol/L) (n = 521, 109, 412) 159 (127) 115 (118) 1.03 (1.01-1.04) 0.002 1.001 (0.98-1.03) 0.95
Platelets (109/L) (n = 504, 105, 399) 151 (93) 196 (91) 0.94 (0.91-0.97) <0.0001 0.92 (0.89-0.96) <0.0001
Other
Acute kidney injury 85 (75.2) 244 (56.5) 2.34 (1.47-3.73) 0.0004 1.35 (0.74-2.48) 0.33
  1. Categorical data are presented as n (%) and continuous data as mean (SD). Mortality status was missing for 17 patients. Heart rate, temperature, creatine kinase, WBC, albumin, urea, and urine output were excluded due to a high proportion of missing data. SOFA score was excluded because it was highly collinear with APACHE II (Pearson r = 0.6). For each variable, we specify n when it differs from the first row total. The final model included 414 patients and did not have evidence of lack of goodness of fit (p = 0.49). ORs for continuous variables refer to an increment of 1 unit (age, APACHE II) or 10 units (mean arterial pressure, PaO2/FiO2, platelets, creatinine).